From Roger.Kynaston at lbhf.gov.uk Tue Apr 1 06:26:04 2003 From: Roger.Kynaston at lbhf.gov.uk (Kynaston Roger) Date: Tue, 1 Apr 2003 12:26:04 +0100 Subject: SUMMARY Silly date question Message-ID: <03Apr1.122910bst.119043@lbhfinet.lbhf.gov.uk> As ever the sun gurus come through in record time. Thanks to all who responded. date -a does advance/retard the clock very slowly. Consensus is that it works at around 1/15 second. I have therefore set it in progress and should be finished sometime tomorrow. Regards Roger Roger Kynaston UNIX Systems Analyst IT Services Division Policy & Administration London Borough Hammersmith & Fulham Tel. 0208 753 2919 The information contained in this e-mail is intended for the recipient or entity to whom it is addressed. It may contain confidential information that is exempt from disclosure by law and if you are not the intended recipient, you must not copy, distribute or take any act in reliance on it. If you have received this e-mail in error, please notify the sender immediately and delete from your system. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From edsbicapps58 at dlo.gsi.gov.uk Tue Apr 1 09:09:02 2003 From: edsbicapps58 at dlo.gsi.gov.uk (Procter, Paul Mr (EDS)) Date: Tue, 1 Apr 2003 15:09:02 +0100 Subject: SUMMARY: Sparc 20 issues Message-ID: Thanks to Richard Skelton for his gentle 'nudge' to remind me that the symptoms are quite useful with the solution. Anyway, 2nd attempt....... The problem was: Hopefully someone will have come across these issues before and can help. We have a Sparc 20 that has died and we have a Sparc 5 that we could use to get us up and running, however on inserting the boot disk and attempting a boot it is unable to find le0 and/or motherboard SCSI controllers. Is there anyway that /dev and /devices an be rebuilt so that we can use the Sparc 5 disks to boot from. The solution: Many thanks to the following for their contributions. Casper Dik, Michael Auria, Steve Maher, Charles Homan, Marc Sheldon, Steven Hogarth, Pablo Jejcic, Matthew Darcy, Pascal Grostabussiat, Peter Stokes, Roman Pestka, Adam Kirby, Ian Pease. It would appear that the Sun website has a slightly different solution to the one that Casper supplied - but, Casper, that's the one that worked - thanks. The solution appeared to be: =========================================================== echo '#path_to_inst_bootstrap1' > /etc/path_to_inst reboot -- -r =========================================================== For your infomation: Sun list it as "echo '#path_to_inst_bootstrap_1' > /etc/path_to_inst" (Note the extra underscore between bootstrap and 1) Again many thanks as you all were telling me roughly the same things. Regards, Paul Procter _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From blanham at cleartrack.com Tue Apr 1 11:07:48 2003 From: blanham at cleartrack.com (Brett Lanham) Date: Tue, 1 Apr 2003 10:07:48 -0600 Subject: SUMMARY: FC HBA model number discovery Message-ID: <102BA6EB88A7D448B144013E7A6EDBE106BD0A@email.cleartrack.com> Thanks to Joe Fletcher, Cyril Plisko, Bertrand Hutin, Pascal Brostabussiat, Bryan Moore, Roman Pestka, Bob Green, and Juergen Waiblinger. Several people mentioned using lputil ( usually in /usr/sbin/lpfc/ ). I tried that but it says no adaptors found. Not sure why because I'm 99% sure there are emulex HBAs in the box. We actually have 2 of the using ATF (application transparent failover) which maybe has something to do with why lputil doesn't show them? If anyone has more info on that please feel free to share. Another suggestion someone mentioned was prtdiag (/usr/platform/`uname -i`/sbin/prtdiag). I had tried this before but in my haste I looked right past the fact that the model was listed as LP8000. Excuse me while I hide behind my embarassment. Anyway if it weren't for the list I might not have gone back so thanks to everyone who replied and anyone who has ever helped anyone else on this list. Other suggestions where: prtconf modinfo sysdef and a utility someone wrote in perl for system discovery: http://www.azoria.com/whatabout Brett > -----Original Message----- > From: Brett Lanham [mailto:blanham at cleartrack.com] > Sent: Tuesday, April 01, 2003 9:09 AM > To: Sunmanagers (E-mail) > Subject: FC HBA model number discovery > > > I need some help figuring out what hardware is on a remote 220R. > Specifically I need to determine what the model of a FC HBA > is. I know the > package for the adapter is lpfc and pkginfo tells me it is an emulex > lightpulse but it does not give the model number. I'm alos > looked through > prtconf but that doesn't give that detail of info. Is there > any way for me > to get more detailed info on the card remotely (via ssh)? > Thanks. I'm sure > this question has been asked before and I apologize for the > redundancy I'm > just in a pich right now to supply this info. Thanks for > your patience. > > Brett > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From george at yorku.ca Tue Apr 1 13:02:00 2003 From: george at yorku.ca (George Kahler) Date: Tue, 01 Apr 2003 13:02:00 -0500 (EST) Subject: SUMMARY: NIS+ domains In-Reply-To: <20030326.113444.56463549.george@yorku.ca> References: <20030326.113444.56463549.george@yorku.ca> Message-ID: <20030401.130200.127157173.george@yorku.ca> As Sal Serafino & Tim Villa pointed out to me, it is not possible for a single client to belong to two different domains at the same time. Thanks George On Wed, 26 Mar 2003 11:34:44 -0500 (EST), George Kahler wrote: > Hi, could someone please tell me if it is possible to have a client > that belongs to two different NIS+ domains ? > > If so, how do you tell it what the search order should be ? > > Thanks, > George > > > ------------------------------------------------------------------------------- > George Kahler e-mail: george at yorku.ca > Sr. Systems Administrator humans: (416) 736-2100 x.22699 > Computing and Network Services machines: (416) 736-5830 > Ontario, Canada, M3J-1P3 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From aronthal at cirba.com Tue Apr 1 13:31:48 2003 From: aronthal at cirba.com (Adam Ronthal) Date: Tue, 01 Apr 2003 13:31:48 -0500 Subject: SUMMARY: SSH RSA Key Authentication on Solaris Message-ID: <5.2.0.9.2.20030401132511.028fad10@mail.netsurf.net> Many people responded, and as always, I am impressed with the quality and volume of responses from the list! Kudos go to Crist Clark for pointing out a typo in my configuration.... I had added the public file to $HOME/.ssh/authorized_hosts instead of $HOME/.ssh/authorized_keys Shame on me for not double checking *everything*.... even the simple stuff you know you can't screw up! Other suggestions included checking the permissions on the .ssh directory and files therein (they were ok), running the sshd daemon in debug mode (gives more useful info than the client) and an interesting (though in this case not relevant) reference to key interoperability between Sun/OpenSSH and SSH.COM implementations (http://www.netsys.com/cgi-bin/display_article.cgi?1254). Many thanks to the following for their responses: John Julian Len Rose Ian Wallace Ryan Bishop Dan Lowe Jesse Trucks Ben Green Darren Dunham Crist Clark -Adam ------------ Original Question ------------- >Hi All... > >Been banging my head on this one for a while.... perhaps someone out there >can offer some advice. > >In the past, I've set up ssh key-based authentication using RSA keys with >empty passwords for restricted accounts and scriptable scp file >copies. Last time I had to set this up, it was between a Solaris 8 box >with OpenSSH and an OpenBSD box. The whole process took about 10 minutes >to get working. > >I have a situation here with Solaris 8 and Solaris 9 servers where I'm >trying to set up the same sort of thing. I know that the stock ssh on Sol >9 doesn't support key-based auth, so I replaced it with OpenSSH 3.5p1 - >same version that is installed on my Solaris 8 box. > >I generate rsa keys: > >$ ssh-keygen -t rsa > >Copy the id_rsa.pub file to the target box and add the contents of the >file to the .ssh/authorized_hosts file. > >That *should* be sufficient, and yet, it doesn't work: > >debug1: authentications that can continue: >publickey,password,keyboard-interactive >debug1: next auth method to try is publickey >debug1: userauth_pubkey_agent: testing agent key >/export/home/catrep/.ssh/id_rsa >debug1: authentications that can continue: >publickey,password,keyboard-interactive >debug1: try privkey: /export/home/catrep/.ssh/identity >debug1: try privkey: /export/home/catrep/.ssh/id_rsa >debug1: read PEM private key done: type RSA >debug1: authentications that can continue: >publickey,password,keyboard-interactive >debug1: try privkey: /export/home/catrep/.ssh/id_dsa >debug1: next auth method to try is keyboard-interactive >debug1: authentications that can continue: >publickey,password,keyboard-interactive >debug1: next auth method to try is password > > >The same behavior is evident when using ssh-agent first: > >$ eval `/usr/local/bin/ssh-agent` >$ /usr/local/bin/ssh-add > > >My sshd_config file contains: > >RSAAuthentication yes >PubkeyAuthentication yes >AuthorizedKeysFile .ssh/authorized_keys > > > >Any ideas? > >Thanks in advance, will summarize to the list. > >-Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mikelist at sky.net Tue Apr 1 15:00:54 2003 From: mikelist at sky.net (Mike's List) Date: Tue, 1 Apr 2003 14:00:54 -0600 (CST) Subject: SUMMARY: root's /bin/false Message-ID: Thanks to: Brett Monroe, Casper Dik, topher, John Timon, Rich Teer, Crist Clark, Michael Peppard, Debbie Tropiano... All agreed that root's shell should not be change, ie. if the system reboot and needs fsck or single user mode, you'll have to boot into cdrom --too much of a hassle. Casper and Crist both mentioned that and changing root's shell will not prevent the exploit because it's executable w/in sendmail, dns, etc. and not user's shell. One mentioned that passwd -l might be safer than changing root's shell. One mentioned using qmail and dnscache, of course there's also postfix, but it's a major overhaul moving to a different mail software... Leaving root's shell alone for now and just keeps patching away. - Mike ---------- original message ---------- Maybe I'm not thinking right today or overlook something simple, but... Anyone implementing root's shell to be /bin/false? Since most of us here at work do sudo, and with so many bugs/patches required for sendmail, DNS, and the likes, I can't keep up in a timely manner. I was just wondering if anyone decided to have root shell as /bin/false --ie. easy enough to boot the system via cdrom and recover, edit root's entry should it be necessary. ...or maybe another shell not sure if anything will be broken/requires for root's shell to be /sbin/sh or not. Note, shell changes only, I'm not suggesting passwd -l the account; but then no one needs to login as root in the first place. Thanks. - Mike [ In addition to www.sunfreeware.com more packages at ftp.patriots.net ] ----- "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- Benjamin Franklin, Historical Review of Pennsylvania. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From b.j.smith at ieee.org Tue Apr 1 16:55:31 2003 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue, 01 Apr 2003 16:55:31 -0500 (EST) Subject: [SUMMARY/NEW PROBLEM] Accomodating VLSM subnets in Solaris 8 ... In-Reply-To: <1049231469.3e8a006d7e6cf@webmail.smithconcepts.com> References: <1049231469.3e8a006d7e6cf@webmail.smithconcepts.com> Message-ID: <1049234131.3e8a0ad371a3c@webmail.smithconcepts.com> Quoting "Bryan J. Smith" : > [ My apologizes in advance if this is answered in a FAQ somewhere. ] > Limitation (Assumed for Solaris 8): > - /etc/inet/netmasks only supports non-CIDR netmasks (/8, /16, /24) > Problem: > - My subnet is a CIDR (/26). First off, I used CIDR when I meant as shorthand for VLSM (I use the two terms far too interchangably). So when I wrote "/26" I meant it as shorthand for "255.255.255.192" (which is in my /etc/inet/netmask). Secondly, an endless number of people (wow!) have E-mailed off-list and stated that Solaris 8 does, in fact, support VLSM netmasks in /etc/inet/netmask. Okay, that's good to now confirm. Which brings me to why I made that incorrect assumption. Because 1) my /etc/inet/netmasks already has: XX.YYY.ZZ.64 255.255.255.192 [ Where IP = XX.YYY.ZZ.71 and ZZ = class A ] And 2) after a Google search that came back with several pages that said SunOS 5 doesn't support CIDR/VLSM in /etc/inet/netmasks (I'll now assume those are dated / for older SunOS 5.x versions). So I incorrectly assumed the previous administrator** overlooked this and that was the root cause. I'm going to have to check to see where and why the above /etc/inet/netmask is not working on my system. I'll further summarize why it didn't work. If anyone has any suggestions, please let me know. Thanx. -- Bryan **NOTE: I just changed employers and caught this nagging issue off-the-bat where select clients could not connect to this one Solaris system's web services (because they were all using the same major carrier as I quickly discovered ;-). netstat -r was reporting XX.0.0.0 as the subnet and ifconfig was reporting ff000000 as the mask with XX.255.255.255 as the broadcast address. Sure enough, a quick "ifconfig set XX.YYY.ZZ.71 netmask 255.255.255.192 broadcast XX.YY.ZZ.127" took care of the problem completely (can't belive my new employer had been putting up with this little issue for that long -- the previous admin blamed it on Solaris!). -- Bryan J. Smith (suffix-free title for your protection) mailto:b.j.smith at ieee.org http://www.thebs.org ---------------------------------------------------------- community : free will communism : will free _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From s.vaka at cox.net Tue Apr 1 17:32:27 2003 From: s.vaka at cox.net (s.vaka at cox.net) Date: Tue, 1 Apr 2003 17:32:27 -0500 Subject: Summary: Swap Filling up: /tmp full on a E4500 Message-ID: <20030401223227.DGVC1060.fed1mtao08.cox.net@smtp.west.cox.net> Thanks to Amaresh, Ben Green, Vincent for quick help. It turned out be Oracle Intelli agent process 'dbsnmp' hogging several GB of memory. Special thanks to Ben Green for pointing out this as a known bug and availability of oracle 9.2 patches from Metalink. Other useful ways to figure out to use 'top', 'lsof', /usr/bin/ps -el (and other options ps -e -o vsz,osz,user,comm) -----Original Message----- Subject: Swap Filling up: /tmp full on a E4500 the problem has been caused by swap being filled up resulting in /tmp full mesasge. (WARNING: /tmp: File system full, swap space limit exceeded) System has huge amount of physical memory and swap available aroung 8 Gb. My question is how would i identify what process is causing this and clean up without having to reboot. Suspecting oracle shared memory ? Any other ideas? Thanks S Vaka _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From georgep at nmr.mgh.harvard.edu Tue Apr 1 21:03:33 2003 From: georgep at nmr.mgh.harvard.edu (George P.) Date: Tue, 1 Apr 2003 21:03:33 -0500 (EST) Subject: SUMMARY: Attach PC monitor to an Ultra-10 (Solaris 2.7) In-Reply-To: Message-ID: The resolution settings on my framebuffer were not supported by my PC monitor. I ended up using m64config to change them. On a remote terminal (as root), I did the following: m64config -res 1280x1024x75 try I would like to thank the following people for their replies: Ric Anderson Hichael Morton On Tue, 1 Apr 2003, George P. wrote: > Hi all, > > How do I attach a PC monitor to an Ultra-10 running Solaris 2.7? (monitor > model is 18" NEC/Mitsubishi LCD1850E LCD Flat-Panel Display 1280x1024) > > I connected the PC monitor to the SUN PCI card, but the computer doesn't > "see" it until I physically remove the SUN Creator 3D graphics card. Is > there a way to be avoid the removal of the SUN graphics card? After > removing the SUN card, the SUN booting messages appear properly on the PC > monitor, but right before the system is up, I get the following warning: > > Warning: There are no 'nf' devices plumbed and up > Unable to start NF FDDI SMT/SNM daemon. > Warning: There are no 'pf' devices plumbed and up > Unable to start PF FDDI SMT/SNM daemon. > > Also, when I try to load openwindows, the screen flickers and at that > point I just have to kill the process remotely. > > Also, how do I attach a PC monitor to an Ultra-5 running Solaris 2.7? > > Thank you very much for your input. Please, let me know if you need me to > provide additional info. > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Ors.Tiszay at eth.ericsson.se Wed Apr 2 02:15:55 2003 From: Ors.Tiszay at eth.ericsson.se (Ors Tiszay (ETH)) Date: Wed, 2 Apr 2003 09:15:55 +0200 Subject: SUMMARY: Taking part of a disk out of Veritas Message-ID: Hi, Thanks for all who replied. There seems to be no consensus on this issue, I got an equal amount of replies saying "it's not possible" and "yes, this is the way to do it". Since as of writing this I have not had the time to actually try and follow the suggested procedures and see the result, I quote here both types of replies. Here follows the original posting and the replies that I got (in order of appearance :-)). Ors ---------------------------------------------------------------------------- Original posting: > I have a Netra T1405 with 1G of RAM running Solaris 8 and > VxVM. The root disk is encapsulated, the rest of the disks > are also under Veritas control, in two disk groups. Since > much of the available literature advises against having the > dump device on a disk controlled by Veritas, I thought of > setting it up on a portion of disk which is not under Veritas > control (unfortunatelly installing one more disk is not an option). > The question is: is it possible to take a part of one of the > disks out of Veritas, so that it becomes accessible by its > "conventional" Unix device? In all the setups that I have > seen so far Veritas controlled the entire disk, eg. c0t0d0s2. > Is this doable, and if so, how? ---------------------------------------------------------------------------- Alan Pae: Vertitas is all or nothing, either it controls the disk or it does not ---------------------------------------------------------------------------- Michael Maciolek: This page gives instructions on how to make part of a disk available to Veritas. Although the instructions are intended for a different final goal - creating a small rootdg - they should allow you do do what you need as well. http://www.carumba.com/talk/veritas/singleslicerootdg.shtml It looks as though you will need to take one disk completely out of Veritas, partition it, and then return just a piece of it to Veritas. ---------------------------------------------------------------------------- Paulo Fessel: It's possible to do this - here we set up rootdg in one slice of each disk we have under Solstice DiskSuite control. However, if I understood you allright, you'll have to vxevac one of the disks to take it off the DG and make it available again to Solaris. After this has been done, use one of the slices to do your memory dump (1 GB partition as you're using 1 GB RAM, right?). Then, use the remaining space of the disk as a single slice and make it available again to Veritas. In short: 1. /etc/vx/bin/vxevac -g dgname diskname 2. format 3. select 4. create slices - 1 for dump, 1 for Veritas 5. add the Veritas slice back to the DG: # vxdctl add disk type=simple (ignore WARNING - the disk is added to /etc/vx/volboot) # vxdisk -f init type=simple (initializes disk) # vxdg -g dgname adddisk (adds to dgname) # vxdctl enable (activates all modifications) ---------------------------------------------------------------------------- Gary Chambers: VxVM requires exclusive use of any devices under its control. Have you considered using Solstice DiskSuite for mirroring the root filesystem? ---------------------------------------------------------------------------- Pascal Grostabussiat: As far as I know this is not possible. When Veritas take over a disk it overwrites the label of the disk (eventually moves the boot-block if necessary to not loose it in case of encapsulation). There is no better granularity than disk-wise. Another point, the only reason for encapsulating the boot-disks is to mirror them. But encapsulating the boot-disks increases complexity. From my own experience, encapsulating the boot-disks with Veritas is some kind of "religion", you feel you have to do it or not. Here in Sweden the religion seems to be "encapsulate" (maybe because Veritas tells people to do it). In other european contries the religion is "no encapsulation" because not necessary and provides higher complexity. If you want to mirror the boot-disks, you could still use Solstice DiskSuite and mirror partitions/slices but not whole disks so that you could still have one or a few partitions on their own. But whatever you use the partitions/slices for, think about possible performance issues. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From flegal at seo-toulouse.com Wed Apr 2 04:03:19 2003 From: flegal at seo-toulouse.com (=?iso-8859-1?Q?Fran=E7ois?= Legal) Date: Wed, 02 Apr 2003 11:03:19 +0200 Subject: Summary : NFS tar and chown Message-ID: <3E8AA757.67903939@seo-toulouse.com> Hi all. Thanks for all who replied. The problem was that users were unable to "chown" and "tar -xo" on Sunos 4.1.3 nfs clients. For chown, it seems that the client machine needs root access on an nfs share, for users to use chown. But then I've got a new problem. Is it possible to use root= option of share in SEC_SYS security mode. Then for tar -xo I got no solution.' Any suggestions ? --Frangois _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From RWilliams at B-Line.ca Wed Apr 2 08:57:03 2003 From: RWilliams at B-Line.ca (Richard Williams) Date: Wed, 2 Apr 2003 08:57:03 -0500 Subject: Summary root alternative Message-ID: <014E75437724D511A60F0050BAAB3BFB3C7DA6@MAIL> Thanks to all that replied. I will go with the example below. Thanks Allan. -----Original Message----- From: Alan Bradley - CPX WC [mailto:AlanB at ComparexAfrica.co.za] Sent: April 2, 2003 8:46 AM To: 'Richard Williams' Subject: RE: root alternative Hi Richard, Make a user id and then give it the same uid and gid as root by manually editing the /etc/passwd file. For example if you edit /etc/passwd and the root entry is as follows: root:x:0:1:The Super User:/home/root:/usr/bin/ksh and your user account line is as follows: rrabbit:x:1001:10:Roger Rabbit:/home/rrabbit:/bin/ksh Then, you would edit your user account line to be: rrabbit:x:0:1:Roger Rabbit:/home/rrabbit:/bin/ksh Hope that helps. BTW I wouldn't disable the root user personally. Regards, Alan. NOTICE: The contents of this message and any attachments are personal to the sender and, whilst having been sent using the Comparex Africa (Pty) Ltd e-mail system, are not, and should not be construed as having been sent on behalf of Comparex Africa (Pty) Ltd. -----Original Message----- From: Richard Williams [mailto:RWilliams at B-Line.ca] Sent: Wednesday, April 02, 2003 3:36 PM To: Sunmanagers (E-mail) Subject: root alternative Hi Managers, I am using a Solaris 8 environment and wish to use an alternate account with all root privileges. How do I disable root and create a back up account to work with. Richard _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From PHackett at talisman-energy.com Wed Apr 2 13:19:21 2003 From: PHackett at talisman-energy.com (Hackett, Peter) Date: Wed, 2 Apr 2003 11:19:21 -0700 Subject: SUMMARY: swap size option in vfstab Message-ID: <8DE364DF318FD411956900508B5AA44506CBC0AB@appsrv3> Thanks to respondants. Yes it is valid. This option limits the size of the /tmp FS. Further info @ man mount_tmpfs Regards, > -----Original Message----- > What is the purpose of option in this vfstab entry from a Sol 7 server? > > swap - /tmp tmpfs - yes size=512m > > Is the "size=512m" a valid options with sol 7 (and above)?? > > If this is a valid option, is it appropriate for a server with 10+ gigs of > system memory? How can one test and measure to find the appropriate value? > > Also, if someone could link me to some docs with info on this option, that > would be appreciated as well. > > Thanks. > > Peter _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From s.evans at TriCorInd.com Wed Apr 2 14:15:29 2003 From: s.evans at TriCorInd.com (Evans, Shawn) Date: Wed, 2 Apr 2003 13:15:29 -0600 Subject: Summary: T3 Lun and Veritas-vxvm Message-ID: Thank's to everyone who answered! The short, sweet, and simple answer to my question is: You can't! A solution provided by Sun: Switch to SDS... --- - Shawn L. Evans - - mailto:s.evans at tricorind.com - - Phone: 1.618.632.9252 x247 - -----Original Message----- From: Evans, Shawn [mailto:s.evans at TriCorInd.com] Sent: Tuesday, April 01, 2003 2:43 PM To: sunmanagers at sunmanagers.org Subject: T3 Lun and Veritas-vxvm I am attempting to take a lun from a T3 show below and configure it within format to have the slices shown below. Then I want to take this same disk, and import it to Veritas, but, I only want to import slice 3 leaving 0 and 1 untouched. Is this possable, and if so, how? 4. c4t1d0 /ssm at 0,0/pci at 1d,600000/pci at 1/SUNW,qlc at 4/fp at 0,0/ssd at w50020f2300007f61,0 Part Tag Flag Cylinders Size Blocks 0 usr wm 1 - 512 1024.00MB (512/0/0) 2097152 1 usr wm 513 - 1024 1024.00MB (512/0/0) 2097152 2 unassigned wu 0 0 (0/0/0) 0 3 usr wu 1025 - 1057 66.00MB (33/0/0) 135168 4 unassigned wu 0 0 (0/0/0) 0 5 unassigned wm 0 0 (0/0/0) 0 6 unassigned wm 0 0 (0/0/0) 0 7 unassigned wm 0 0 (0/0/0) 0 The reason is I am attempting to 'simulate' a T3+ and slice 0 and 1 are for SNDR. --- - Shawn L. Evans - - mailto:s.evans at tricorind.com - - Phone: 1.618.632.9252 x247 - _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From georgep at nmr.mgh.harvard.edu Wed Apr 2 15:33:40 2003 From: georgep at nmr.mgh.harvard.edu (George P.) Date: Wed, 2 Apr 2003 15:33:40 -0500 (EST) Subject: SUMMARY: Disk replacement on Ultra-10 (Solaris 2.7) Message-ID: Hi all, As soon as the network issues were resolved, the installation continued normally. Thanks to all who responded. Justin Stringfellow Russ Poffenberger Hichael Morton Richard Hampshire Joon Martin Hansen Justin Stringfellow Mark S Cohen ---------- Forwarded message ---------- Date: Wed, 2 Apr 2003 12:09:07 -0500 (EST) From: George P. To: sunmanagers at sunmanagers.org Subject: UPDATE: Disk replacement on Ultra-10 (Solaris 2.7) I would like to add a couple of things. - I waited for about 12 hours (I left it overnight)! - We are using NIS, and there was a network issue, but I know this can be fixed by running ypinit -c, after installation is complete. Should I just disable NIS, and see if it proceeds? ---------- Forwarded message ---------- Date: Wed, 2 Apr 2003 11:42:09 -0500 (EST) From: George P. To: sunmanagers at sunmanagers.org Subject: Disk replacement on Ultra-10 (Solaris 2.7) Hi all, I replaced a defective hard drive on an Ultra-10, and booted using a Solaris 2.7 CD-ROM. I can see the disk ok, and then proceeded with partitioning (format), and filesystem creation (newfs). I did not run into any problems. The issue is that I am unable to complete the Solaris installation on the new disk. After filling out the network and timezone info, the console reports that the system identification is complete, but it does not proceed with installation. If I try to restart the installation, it reports that the instalation program is already running. I have done this several times before, and I don't recall running into this. Is there a way that I can manually force it to proceed after the system identification is done? Thank you in advance for your input. Please, let me know if you need me to provide additional info. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zaigui at yahoo.com Wed Apr 2 17:22:13 2003 From: zaigui at yahoo.com (Zaigui Wang) Date: Wed, 2 Apr 2003 14:22:13 -0800 (PST) Subject: Summary: Sun hardware for printer server -- Recommendation? In-Reply-To: <20030326065820.56414.qmail@web12302.mail.yahoo.com> Message-ID: <20030402222213.24997.qmail@web12305.mail.yahoo.com> I got a few responses, including one response regarding software suggestion for printing queue management (CUPS). Most pointed out that print server resource requirements are pretty minimal. It is not CPU-intensive, but one needs to be rather generous on disk space and RAM. Network capacity/bandwidth may also play an important role. In some cases, people uses Ultra 1 as a print server to serve 30+ printer spools. In another case, a sparc 5 was used as print server and no major issues. As far as redundancy goes, getting two system so that the primary one has a secondary to fail over to. But if you are stuck with one system, disk mirroring should definitely be used to prevent against disk failure. Thanks all for helping, especially, Kevin Metzger Lars Hecking Vincent Ribeaud Tim Chipman arjan.van.herwijnen at lgphilips-displays.com Zaigui --- Zaigui Wang wrote: > Managers, > > In an effort to consolidte and centralize our > printing > spool management on Sun servers, I am planning to > set > up a printer spool server (with HP jetdirect spool > installer program) and need some suggestion as to > what > kind of sun hardware is suitable for this. > > The print server will support around 160 printing > spools, which include printers from a variety of > locations (U.S., China, Korea, U.K, etc.) We have > about 40 UNIX servers, most of them have separate > printing spool setup. > > I am looking at the SunFire 280R with the minimum > configuration (1GB mem, 36GB F-CAL drive, 1x900MHz). > I > don't think a print server needs a whole lot of CPU > power. What do you guys suggest? If you have other > ideas regarding what is the best way of doing this > (redundancy, fault tolerance setup), please also let > me know. > > Thanks. > > Zaigui > Yahoo! Platinum - Watch CBS' NCAA March Madness, > live on your desktop! > http://platinum.yahoo.com > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers Yahoo! Tax Center - File online, calculators, forms, and more http://tax.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From netcomrade at bookexchange.net Wed Apr 2 18:50:56 2003 From: netcomrade at bookexchange.net (NetComrade) Date: Wed, 2 Apr 2003 18:50:56 -0500 (EST) Subject: SUMMARY: e450 disks hot swappable/pluggable? In-Reply-To: Message-ID: Thanks to all of you who responded. The answer is yes disks are hot swappable/pluggable, just run disks;vxdctl enable once you're done. And indeed it worked on a local e450. On Mon, 31 Mar 2003, NetComrade wrote: > Can i plug/swap new drives w/o rebooting? > > System is running VM 3.1 > > Thanks _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Wed Apr 2 19:09:03 2003 From: martini at mrpeabody.llnl.gov (Dave Martini) Date: Wed, 2 Apr 2003 16:09:03 -0800 (PST) Subject: SUMMARY:Comparison of Sparc to Intel Message-ID: <200304030009.h33093S03686@raider.llnl.gov> Seems that price and performance goes to the Intel box. I did forget to mention this will be an Oracle server. My original question: What would be an equivalent SUN system that would match the > processing power of this: > > 2 xeon processor, 2 gig ram, linux and 1 Terrabyte of disk storage. > > Are there benchmarks or sites that compare Intel to Sparc? > Many suggest this spec site for benchmarks: http://www.specbench.org/cpu2000/results/cpu2000.html Most said it was like comparing apples to oranges and also depended on the application. I received a few detailed responses one from Tim Chipman AFAIK, there are relatively few, informal resources that compare sparc to intel as you are seeking. typically, the "water is muddy" because it depends heavily on what the intended use of the systems are. For example, -> Database server: Lots of I/O and memory access will render CPU speed less critical ... -> Computation server (bioinformatics, etc) - less IO, tons of CPU requirements render memory/bandwidth less critical. We've got a few dual-athlon systems here "in production", one is the "head node" of a "beowulf cluster" that we use for computationally intensive (not! IO/intensive) work. The dual athlon by itself (2x1600mhz) is ~2x faster in raw CPU power than our 4x400mhz e3500, suggesting that for *this type of CPU intensive work*, 1600mhz of athlon power is approximately equal to 1600mhz of sunUltraSparcII power. The beowulf cluster operating on appropriate CPU intensive problems that are split across the cluster has ~20 ghz of CPU power and clearly outguns the e3500 by a large margin. Additionally, IMHO, if you have nice SCSI (FCAL?) subsystems for disk on your X86 hardware, you can get very reasonable disk IO performance and overall end up with very reasonable performing systems. Ultimately, I think few people would argue that you can get "more bang for your buck" with X86 hardware (as compared to Sun, HP, IBM) "big iron / single dedicated vendor" Unix implementations. The difference tends to be, - concerns over support, integration of tools onto platform? - concerns over maintenance, troubleshooting, hardware failure ? - concerns over budget (or lack thereof ? :-) So. It will all depend on what kind of use you have in mind for the gear, and how $$ influences the decision, since there are many factors in the equation. Hope this helps a bit, --Tim And another on from Jay Lessert Well, the storage is irrelevant, right? If you've got 1TB, it's SCSI or FC, probably FC, and a SPARC/Solaris box will talk to it the same as an X86/Linux box. The 2GB RAM is irrelevant, too, right? Not very much RAM, hard to find a SPARC box with a max capacity *less* than that. If you're running a single CPU-bound application with graceful memory access patterns (good locality), you should lose the second processor, and a 2.8GHz Xeon is somewhere between 1.5X and 2.0X faster than a 1.05GHz US-III. But don't waste money on the Xeon, then. Get a 3GHz P4. Cheaper and faster. And another from Ben Green It greatly depends on what application(s) you will be running. Vendors, such as Oracle, can shed some light with benchmarks. In my experience with Oracle and well laid out Hitachi/EMC SANs, a Compaq or Dell with two P4 Xeons running at 2.8 GHz will crucify Sun servers up to a V880, with Oracle tuned well in both cases. Hyperthreading is a huge gain with those processors, too. It really depends what you are going to do, but as a general rule of thumb -- the Linux box will win and will cost one half or one third of the SUN's price. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Ko at yahoo.com Wed Apr 2 21:36:06 2003 From: Ko at yahoo.com (Lee-Whei McGhee) Date: Thu, 3 Apr 2003 02:36:06 GMT Subject: =?ISO-8859-3?B?c3VtbWFyaWVzLFNsZWVwIGJldHRlciwgaW1wcm92ZSB2aXNpb24gYW5kIG1lbW9yeSA0NjkzZHp1SDEtOTczTVRLQS0xNiA=?= Message-ID: <20030403023740.3475B2084A@sunmgrs.cs.toronto.edu> An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030403/8f0589bb/attachment.html From 6xym79zng at aol.com Wed Apr 2 20:40:59 2003 From: 6xym79zng at aol.com (Selma Thurman) Date: Thu, 03 Apr 03 01:40:59 GMT Subject: wex dwmonstration Message-ID: An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030403/a5bed880/attachment.html From edsbicapps58 at dlo.gsi.gov.uk Thu Apr 3 05:26:39 2003 From: edsbicapps58 at dlo.gsi.gov.uk (Procter, Paul Mr (EDS)) Date: Thu, 3 Apr 2003 11:26:39 +0100 Subject: SUMMARY:Sendmail problems Message-ID: Thanks to all who replied to this one but the solution was not as difficult as I had thought It was to do with the host files and NIS. The Hosts file was rebuilt removing the 'duff' entries and then distributed accordingly. The next step was to clear out the old emails as they could be hold out-of-date information and then stop and start the sendmail daemon to refresh it with the new data. This appears to have solved the problems. Thanks Paul Procter ============================================================================ ============= The Question was: This is probably something simple and I am probably overlooking something fundamental but here goes: We are running sendmail on Solaris 8. NIS is configured. When a message is sent the following errors are seen: SYSERR:(root):mail loops back to me (MX Problem?) sendmail [4038] hdkmsmskjakk losing ./ asdfakdjlkjhqlwe sendmail panic SYSERR:(root):sendmail:cannot save rejected mail anywhere. Has anyone seen these types of errors and can possibly give me a clus on how to fix them? Many thanks. Paul Procter _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers The original of this email has been scanned for viruses by the Government Secure Intranet (GSI) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs. This E-mail was sent from an E-mail address under the control of the MOD. Privileged / In Confidence Information may be contained in this message. If you have received this E-mail in Error, please destroy all copies in your possession or control and kindly notify the sender by reply e-mail. The addressee (or individual responsible for delivery of the message to such person) may not forward, discolse or copy this E-mail to any third party, without the prior consent of the MOD. Furthermore, MOD does not accept liability for any changes made to this E-mail after it was sent. Please advise immediately if you or your employer do not consent to Internet E-mail for messages of this kind. If you take offence to the contents of this E-mail, you can contact the System Management team at qcissmt at dlo.gsi.gov.uk The original of this email has been scanned for viruses by the Government Secure Intranet (GSI) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs. ........................................................ This E-Mail has been originated from the Internet. If you do not wish to receive E-mail from this address, please reply to this E-mail, and request that the sender remove your E-mail address from their system. QCIS SMT .............................................................. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From yakov.lerner at kavado.com Thu Apr 3 05:57:43 2003 From: yakov.lerner at kavado.com (Yakov Lerner) Date: Thu, 03 Apr 2003 12:57:43 +0200 Subject: SUMMARY: mouse (and color) support for vim in xterm References: <20021203201918.21873.64302.Mailman@sunmgrs.cs> Message-ID: <3E8C13A7.7000600@kavado.com> Thanks to all those who replied, and especially to Walter Heukles: Justin Stringfellow, Joe Fletcher, Michael Auria , I figured it out with your help. The editor in question is vim in xterm box. I wanted color xterm + mouse support. /bin/vi is not capable of mouse-in-xterm support. dtterm does not seem to support mouse-in-vim. But xterm does have both color and 'mouse reporting' with proper TERM setting. The solution was to insert 'set mouse=vn' in .vimrc To get color+mouse, the solution was to copy the 'xterm-xfree86' file (or 'xterm-vt220') from /usr/local/share/terminfo/x/. to /usr/share/lib/terminfo/x/. and to set TERM to 'xterm-xfree86' (or 'xterm-vt220') (Terminfo as shipped with solaris8 does not seem to have entry with both color and mouse support). Jacob L. -----.............. % echo 'scale=80; a(1)*4' | bc -l .....--------------- sunmanagers-request at sunmanagers.org wrote: > Send sunmanagers mailing list submissions to > sunmanagers at sunmanagers.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > or, via email, send a message with subject or body 'help' to > sunmanagers-request at sunmanagers.org > > You can reach the person managing the list at > sunmanagers-admin at sunmanagers.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of sunmanagers digest..." > > > Today's Topics: > > 1. V480 serial connection (Hubert Chu) > 2. Netscape Messaging 3.6 - Production Issue (Markham, Richard) > 3. sun blade 150 (=?iso-8859-1?q?ur=20friend?=) > 4. automatic starting of tomcat (Ramaswamy, Satish) > 5. automatic starting of tomcat (Ramaswamy, Satish) > 6. A5200 performance with a lot of small (2-8k) files (NetComrade) > 7. summary of automatic starting of tomcat (Ramaswamy, Satish) > 8. E6800 domain reboot problem (send_mondo_set) (Kerekes, Ed) > 9. Smarter install script for patch clusters? (Lars Hecking) > 10. summary of automatic starting of tomcat (Ramaswamy, Satish) > 11. summary of automatic starting of tomcat (Ramaswamy, Satish) > > --__--__-- > > Message: 1 > Date: Tue, 3 Dec 2002 12:01:38 -0500 (EST) > From: Hubert Chu > To: > Subject: V480 serial connection > > We're experiencing problems connecting to a new V480 through a serial > connection. I'm connecting directly to the 480's serial port with a null > modem RJ-45 cable. I can see the POST, boot sequence, and login prompt > just fine, but I'm unable to transmit any commands. I'm also unable to > send a break signal. > > I'm almost certain the cabling is correct, as we set up our other 480s in > the same manner. Has anyone ever experienced this problem? Could it be a > hardware problem with the serial port? > > TIA > > --__--__-- > > Message: 2 > From: "Markham, Richard" > To: "Sunman (sunmanagers at sunmanagers.org)" > Subject: Netscape Messaging 3.6 - Production Issue > Date: Tue, 3 Dec 2002 12:26:09 -0500 > > I am getting thousands of similar occurences in my log file. > Same error value 105. > > --log message-- > The mail system on xxx.xxx.com encountered the following error: > > The following module was unable to complete its task due to some > problem. Please contact server at netscape.com with the following > information: > > > Module Name: SMTP-Deliver > Message Name: 20021201071826.AAB23769 at xxx.xxx.com-1-SMTP > Process ID: 7972 > Exit Value: 105 > --- > Can anyone please describe exit value 105 or point to me where I can > find documentation for this particular release. It does not seem > to be readily available. Thanks much in advance. > > --__--__-- > > Message: 3 > Date: Tue, 3 Dec 2002 17:37:08 +0000 (GMT) > From: =?iso-8859-1?q?ur=20friend?= > Subject: sun blade 150 > To: sunmanagers at sunmanagers.org > > Hi sun managers, > > I have a sun blade 150 machines running with solaris 8 > 0202 with OBP 4.6.9 and with all latest patches. The > CDE hangs just few seconds after booting up > > The CDE display gets released when the machine is > disconnected from the network. M64 graphics card > patch is also applied > > Can anybody provide solution to this problem > > In few of the other sun blade 100 machine cpu usage is > very high even though there is no application running. > These systems are installed with solaris 8 0401 with > obp 4.6.5 with latest patches > > Any solution > > thanks > > UR Izer > > > ________________________________________________________________________ > Missed your favourite TV serial last night? Try the new, Yahoo! TV. > visit http://in.tv.yahoo.com > > --__--__-- > > Message: 4 > Subject: automatic starting of tomcat > Date: Tue, 3 Dec 2002 11:51:45 -0600 > From: "Ramaswamy, Satish" > To: > > hi, i have a solaris e-450 server and each time i'd like to find out how to > start tomcat automatically when the server reboots. > > i've written a wrapper script called autotomcat as follows > > #!/bin/sh > #written by satish ramaswamy > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > PATH=/usr/bin:/usr/sbin:/usr/dt/bin:/usr/openwin/bin:/usr/ucb:$TOMCAT_HOME/bi > n:. > export PATH #chang made 12/03/02 > cd $TOMCAT_HOME/bin > pscount=`ps -ef |grep "[j]ava"| wc -l` > #echo $pscount > if [ $pscount -eq 0 ]; then > cd $TOMCAT_HOME/bin > #echo "inside if" > tomcat.sh start > fi > > i've tried couple of things. > 1) renamed this script S98autotomcat and placed it in /etc/rc3.d/ directory > but this doesn't seem to work when the server reboots. > 2) in cron 10,20,30,40,50 * * * * /scripts/autotomcat > /dev/null have the > script called every 10 minutes. the reason i do this is because i have a > backup script and don't know how long the backup takes and at the end i > reboot the server. > > but this didn't work either. however, i changed my autotomcat script and am > now exporting the PATH but i don't see why i should since i'm not using > the variables in the script outside the script. > > thanks in advance. > > > > satish > ph:630-875-8263 > > --__--__-- > > Message: 5 > Subject: automatic starting of tomcat > Date: Tue, 3 Dec 2002 12:43:03 -0600 > From: "Ramaswamy, Satish" > To: > > i'm sending this again as i got a weird response that the header was > suspicious. > > > -----Original Message----- > From: Ramaswamy, Satish > Sent: Tuesday, December 03, 2002 12:33 PM > To: 'sunmanagers at sunmanagers.org' > Subject: RE: automatic starting of tomcat > Importance: High > > > yes. the script is executable and there are no errors in the cron log file > /var/cron/olog file. > > randy, i know grep something always returns something. > but if you use > grep "[1st letter in the grep expression] rest of pattern", you don't have to > use grep -v grep. you can test it out. i use it all the time. > > darren, my other scripts in the crontab work. i understand *not working* is > a generic term but there are no errors in the cron log file. that's what is > bothering me. and i feel surely someone out there is running tomcat in a sun > server and i've tried echo "started" /tmp/autotomcat and the file does get > created. > > also i tried, JAVA_HOME=/usr/bin/java (i did type java to find out where > java is located) > export JAVA_HOME > > but i get java exceptions. and i my tomcat startup script is named > startup.sh. > > > -----Original Message----- > From: Bablak, Steve [mailto:Steve.Bablak at Marconi.com] > Sent: Tuesday, December 03, 2002 12:15 PM > To: Ramaswamy, Satish > Subject: RE: automatic starting of tomcat > > > is the script executable? i.e.) permissions 7xx > > -----Original Message----- > From: Ramaswamy, Satish [mailto:Satish.Ramaswamy at uscellular.com] > Sent: Tuesday, December 03, 2002 12:52 PM > To: sunmanagers at sunmanagers.org > Subject: automatic starting of tomcat > > > hi, i have a solaris e-450 server and each time i'd like to find out how to > start tomcat automatically when the server reboots. > > i've written a wrapper script called autotomcat as follows > > #!/bin/sh > #written by satish ramaswamy > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > PATH=/usr/bin:/usr/sbin:/usr/dt/bin:/usr/openwin/bin:/usr/ucb:$TOMCAT_HOME/b > i > n:. > export PATH #chang made 12/03/02 > cd $TOMCAT_HOME/bin > pscount=`ps -ef |grep "[j]ava"| wc -l` > #echo $pscount > if [ $pscount -eq 0 ]; then > cd $TOMCAT_HOME/bin > #echo "inside if" > tomcat.sh start > fi > > i've tried couple of things. > 1) renamed this script S98autotomcat and placed it in /etc/rc3.d/ > directory > but this doesn't seem to work when the server reboots. > 2) in cron 10,20,30,40,50 * * * * /scripts/autotomcat > /dev/null have the > script called every 10 minutes. the reason i do this is because i have a > backup script and don't know how long the backup takes and at the end i > reboot the server. > > but this didn't work either. however, i changed my autotomcat script and > am > now exporting the PATH but i don't see why i should since i'm not using > the variables in the script outside the script. > > thanks in advance. > > > > satish > ph:630-875-8263 > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > --__--__-- > > Message: 6 > Date: Tue, 3 Dec 2002 14:07:24 -0500 (EST) > From: NetComrade > To: sunmanagers at sunmanagers.org > Subject: A5200 performance with a lot of small (2-8k) files > > Dear Sun Gurus, > > We have an A5200 with VM 3.5 on top of it configured with a couple of > RAID5 volumes (one is 500G), which house a lot of very small > files. Customer has been complaining about how slow the setup is (file > system is vxfs, block size is 4k, stripe size is 128k). Customer is happy > about performance with large files. A5200 is connected to an E420r via > fibre, arbitrated loop. > > I don't know where to start to try to optimize IO. Anything we can do at > all? > > Any advice or pointers appreciated. > > Thanx. > > (will summarize) > > --__--__-- > > Message: 7 > Subject: summary of automatic starting of tomcat > Date: Tue, 3 Dec 2002 13:03:33 -0600 > From: "Ramaswamy, Satish" > To: > > i just exported the PATH variable today and i can't test it out until the eve/ > late at night today when i can reboot the server. > > also i've set and exported the PATH variable in /etc/profile which is read > first for the korn/bourne shells hence didn't feel the need of exporting the > PATH again in my script but have done so now. > > > > -----Original Message----- > From: Ryan.Bishop at exim.gov [mailto:Ryan.Bishop at exim.gov] > Sent: Tuesday, December 03, 2002 12:49 PM > To: Ramaswamy, Satish > Subject: Re: automatic starting of tomcat > > > > You have to export the path since this is being run as a shell script, > hence it is running within a shell. Anytime that you are writing a "shell > script" rather you are using sh, bash, tcsh or anything else, it is always > good practice to export your paths so that the shell will have the correct > envioronment variables that you expect it to have. > > When it runs in the shell, the user of the shell, in this case root (?) > will use the default environment variables within /etc/skel/profile for the > path (which by the way is only /usr/bin and /usr/sbin i believe). > > Question: Did the script work correctly once you started exporting the > PATH variable to the shell process? If so, then the proof is in the > pudding, as they say. > > I hope that this helps. > > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > =-=-=-= > > Ryan Bishop > Sr. Unix Administrator/Security Administrator > > "In three words I can sum up everything I've learned about life: It goes > on." -- Robert Frost > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > =-=-=-= > > > > "Ramaswamy, Satish" > > lular.com> cc: (bcc: Ryan > Bishop/IMT/EXIMBANK) > Sent by: Subject: automatic > starting of tomcat > sunmanagers-admin at sunma > nagers.org > > > 12/03/2002 12:51 PM > > > > > > > hi, i have a solaris e-450 server and each time i'd like to find out how > to > start tomcat automatically when the server reboots. > > i've written a wrapper script called autotomcat as follows > > #!/bin/sh > #written by satish ramaswamy > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > PATH=/usr/bin:/usr/sbin:/usr/dt/bin:/usr/openwin/bin:/usr/ucb:$TOMCAT_HOME/bi > > n:. > export PATH #chang made 12/03/02 > cd $TOMCAT_HOME/bin > pscount=`ps -ef |grep "[j]ava"| wc -l` > #echo $pscount > if [ $pscount -eq 0 ]; then > cd $TOMCAT_HOME/bin > #echo "inside if" > tomcat.sh start > fi > > i've tried couple of things. > 1) renamed this script S98autotomcat and placed it in /etc/rc3.d/ > directory > but this doesn't seem to work when the server reboots. > 2) in cron 10,20,30,40,50 * * * * /scripts/autotomcat > /dev/null have the > script called every 10 minutes. the reason i do this is because i have a > backup script and don't know how long the backup takes and at the end i > reboot the server. > > but this didn't work either. however, i changed my autotomcat script and > am > now exporting the PATH but i don't see why i should since i'm not using > the variables in the script outside the script. > > thanks in advance. > > > > satish > ph:630-875-8263 > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > --__--__-- > > Message: 8 > Subject: E6800 domain reboot problem (send_mondo_set) > Date: Tue, 3 Dec 2002 14:37:21 -0500 > From: "Kerekes, Ed" > To: > > We are having an issue with some of our domains rebooting with messages > similar to: > > Dec 3 13:17:34 SUNW,UltraSPARC-III+: [ID 563681 kern.notice] send > mondo > timeout [735294 NACK 0 BUSY] > Dec 3 13:17:34 IDSR 0x1 aids: > Dec 3 13:17:34 SUNW,UltraSPARC-III+: [ID 823475 kern.notice] c > Dec 3 13:17:34 SUNW,UltraSPARC-III+: [ID 100000 kern.notice] > Dec 3 13:17:34 unix: [ID 836849 kern.notice] > Dec 3 13:17:34 ^Mpanic[cpu13]/thread=3000f676940: > Dec 3 13:17:34 unix: [ID 152620 kern.notice] send_mondo_set: timeout > Dec 3 13:17:34 unix: [ID 100000 kern.notice] > Dec 3 13:17:34 genunix: [ID 723222 kern.notice] 000002a101a44600 > SUNW,U > ltraSPARC-III+:send_mondo_set+238 (1, 1, 2a, 3, 5b35173bfd4, > 5b35173bfe1) > Many lines follow > > This has happened on 4 different domains on 2 different 6800s several > times and we have replaced CPUs and per Sun's recommendation we have > also turned off predictive branching on the cpus. This has us stumped. > We are running oracle 9i on all domains. If anyone has any ideas or has > seen this before it would be greatly appreciated. > > Ed Kerekes > STERIS Corporation > > --__--__-- > > Message: 9 > Date: Tue, 3 Dec 2002 20:07:20 +0000 > From: Lars Hecking > To: sunmanagers at sunmanagers.org > Subject: Smarter install script for patch clusters? > > Hopefully someone from Sun is reading this list :-) > > I was going to upgrade an old SS10 with a SUN1.05 (1GB) disk to > Solaris 8 02/02. After End User plus OpenGL, I end up with > > Filesystem kbytes used avail capacity Mounted on > /dev/dsk/c0t3d0s0 123231 28668 82240 63% / > /dev/dsk/c0t3d0s6 593446 431580 102522 81% /usr > (swap 256MB, no /export/home) > > Trying to install the current Recommended patch cluster, the installer > told me I don't have enough disk space to save old files: > > Insufficient space in /var/sadm/patch to save old files. > Space required in bytes: 125292 > Space available in bytes: 94562 > > So I removed all patches from patch_order that are already installed, > and also removed the corresponding patch directories. Still no joy: > > Insufficient space in /var/sadm/patch to save old files. > Space required in bytes: 118153 > Space available in bytes: 94560 > > However, after moving /var/sadm/patch (which was empty save .mu_applied) > aside and linking it to a fs with enough free space, the patch cluster > applied fine, and it turns out that /var/sadm/patch itself occupies less > than 1MB of disk space, so I removed the link and copied the patch directory > back to /var/sadm. > > Filesystem kbytes used avail capacity Mounted on > /dev/dsk/c0t3d0s0 123231 62543 48365 57% / > /dev/dsk/c0t3d0s6 593446 452058 82044 85% /usr > > Doing the math, the patch install has taken 57353kB, a lot less than > 118153kB as the install script suggested. > > Now, I was wondering where the cluster install script can be made more > intelligent, as it is obviously possible to install the cluster within > these disk space constraints. Currently, it simply uses the size of the > patch cluster itself as an estimate, and compares it to the free space > on the filesystem with holds /var/sadm/patch: > > BASESIZE=`du -sk $PATCHDIR | cut -f1` > SUPPLEMENT_SIZE=`expr $BASESIZE / 2` > bytes_avail=`df -b /var/sadm/patch | tail -1` > bytes_avail=`echo $bytes_avail | sed 's/.*\ //'` > if [ ${SUPPLEMENT_SIZE} -gt ${bytes_avail} ]; then > > Yes, I realise that this is a sufficiently obscure problem :-) > > --__--__-- > > Message: 10 > Subject: summary of automatic starting of tomcat > Date: Tue, 3 Dec 2002 14:10:22 -0600 > From: "Ramaswamy, Satish" > To: > > all: thanks for your input. i was given a script which i'm enclosing after > i've made changes to it according to my environment. this was provided by > david foster. i'll implement and test it later on tonight. > > so hopefully, the scottish chap who also has tomcat installed can use this > instead of manually starting it up. > > i've placed it in /etc/init.d/tomcat and made it executable. i'll create a > symbolic link in the /etc/rc3.d/ dir > > cd /etc/rc3.d > ln -s S98autotomcat /etc/init.d/tomcat though david instructed me to do this > i the /etc/rc2.d dir but i think it shouldn't make a difference. > > i'm also sending it as a text attachement david's orginal script. > > satish > > #!/bin/sh > # Make sure that /export/home/webmstr/ is mounted > [ ! -d /export/home/webmstr/jakarta-tomcat-3.3 ] && exit 1 > > case "$1" in > 'start') > if [ -x /export/home/webmstr/jakarta-tomcat-3.3/bin/startup.sh ]; > then > echo "Starting tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > /export/home/webmstr/jakarta-tomcat-3.3/bin/startup.sh > fi > ;; > > 'stop') > if [ -x /export/home/webmstr/jakarta-tomcat-3.3/bin/shutdown.sh ]; > then > echo "Stopping tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > /export/home/webmstr/jakarta-tomcat-3.3/bin/shutdown.sh > fi > > ;; > > *) > echo "Usage: $0 { start | stop }" > exit 1 > ;; > esac > exit 0 > #!/sbin/sh > # Make sure that /usr/local/ is mounted > [ ! -d /usr/local/jakarta-tomcat-4.0.5 ] && exit 1 > > case "$1" in > 'start') > if [ -x /usr/local/jakarta-tomcat-4.0.5/bin/startup.sh ]; then > echo "Starting tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > CATALINA_HOME=/usr/local/jakarta-tomcat-4.0.5 > export CATALINA_HOME > /usr/local/jakarta-tomcat-4.0.5/bin/startup.sh > fi > ;; > > 'stop') > if [ -x /usr/local/jakarta-tomcat-4.0.5/bin/shutdown.sh ]; then > echo "Stopping tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > CATALINA_HOME=/usr/local/jakarta-tomcat-4.0.5 > export CATALINA_HOME > /usr/local/jakarta-tomcat-4.0.5/bin/shutdown.sh > fi > ;; > > *) > echo "Usage: $0 { start | stop }" > exit 1 > ;; > esac > exit 0 > > --__--__-- > > Message: 11 > Subject: summary of automatic starting of tomcat > Date: Tue, 3 Dec 2002 14:18:24 -0600 > From: "Ramaswamy, Satish" > To: > > a correction the syntax of the link command should be ln -s > /etc/init.d/tomcat Stomcat ... > > sorry. i always get confused and hence check my unix in a nutshell. > > s > > > -----Original Message----- > From: Ramaswamy, Satish > Sent: Tuesday, December 03, 2002 2:10 PM > To: 'sunmanagers at sunmanagers.org' > Subject: summary of automatic starting of tomcat > > > all: thanks for your input. i was given a script which i'm enclosing after > i've made changes to it according to my environment. this was provided by > david foster. i'll implement and test it later on tonight. > > so hopefully, the scottish chap who also has tomcat installed can use this > instead of manually starting it up. > > i've placed it in /etc/init.d/tomcat and made it executable. i'll create a > symbolic link in the /etc/rc3.d/ dir > > cd /etc/rc3.d > ln -s S98autotomcat /etc/init.d/tomcat though david instructed me to do this > i the /etc/rc2.d dir but i think it shouldn't make a difference. > > i'm also sending it as a text attachement david's orginal script. > > satish > > #!/bin/sh > # Make sure that /export/home/webmstr/ is mounted > [ ! -d /export/home/webmstr/jakarta-tomcat-3.3 ] && exit 1 > > case "$1" in > 'start') > if [ -x /export/home/webmstr/jakarta-tomcat-3.3/bin/startup.sh ]; > then > echo "Starting tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > /export/home/webmstr/jakarta-tomcat-3.3/bin/startup.sh > fi > ;; > > 'stop') > if [ -x /export/home/webmstr/jakarta-tomcat-3.3/bin/shutdown.sh ]; > then > echo "Stopping tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > TOMCAT_HOME=/export/home/webmstr/jakarta-tomcat-3.3 > export TOMCAT_HOME > /export/home/webmstr/jakarta-tomcat-3.3/bin/shutdown.sh > fi > > ;; > > *) > echo "Usage: $0 { start | stop }" > exit 1 > ;; > esac > exit 0 > #!/sbin/sh > # Make sure that /usr/local/ is mounted > [ ! -d /usr/local/jakarta-tomcat-4.0.5 ] && exit 1 > > case "$1" in > 'start') > if [ -x /usr/local/jakarta-tomcat-4.0.5/bin/startup.sh ]; then > echo "Starting tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > CATALINA_HOME=/usr/local/jakarta-tomcat-4.0.5 > export CATALINA_HOME > /usr/local/jakarta-tomcat-4.0.5/bin/startup.sh > fi > ;; > > 'stop') > if [ -x /usr/local/jakarta-tomcat-4.0.5/bin/shutdown.sh ]; then > echo "Stopping tomcat server ..." > JAVA_HOME=/usr/j2se; export JAVA_HOME > CATALINA_HOME=/usr/local/jakarta-tomcat-4.0.5 > export CATALINA_HOME > /usr/local/jakarta-tomcat-4.0.5/bin/shutdown.sh > fi > ;; > > *) > echo "Usage: $0 { start | stop }" > exit 1 > ;; > esac > exit 0 > > > --__--__-- > > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > End of sunmanagers Digest _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From joe.fletcher at btconnect.com Thu Apr 3 06:18:00 2003 From: joe.fletcher at btconnect.com (joe.fletcher at btconnect.com) Date: Thu, 3 Apr 2003 12:18:00 +0100 Subject: SUMMARY: V880 "memory" error Message-ID: <3e8c1868.5351.0@btconnect.com> Hi, Tentative summary this. It appears it was actually a bad DIMM. However there's a bug in the firmware such that it reports the wrong module as being faulty. Patches will follow naturally. I'm still curious as to why it was the disk traffic that provoked the error but presumably the copy operations exercised the specific location which had the fault. ------------ originally ------------------- Anyone got any ideas on this? System is a V880 fully packed patched upto 108528-18, OBP 4.7.0. It's almost certainly not an error in physical memory and we've had the DIMMs swapped out. Errors get clocked as soon as we start loading the disks. Looks to me like it might be a PCI issue but I'm open to suggestions. Cheers Joe _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rbaldus at e-one.com Thu Apr 3 07:45:28 2003 From: rbaldus at e-one.com (Baldus, Marc) Date: Thu, 3 Apr 2003 07:45:28 -0500 Subject: Summary: OT: Free RAM to Good Home Message-ID: Thanks for all of the responses to this e-mail. I have given all of the RAM away. Marc B. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rcruz at cesa.co.cr Thu Apr 3 09:46:52 2003 From: rcruz at cesa.co.cr (Randall Cruz) Date: Thu, 03 Apr 2003 08:46:52 -0600 Subject: SUMMARY: Tunning of TCP on Solaris 9 Message-ID: <3E8C495C.6050305@cesa.co.cr> Thanks a lot to all the Gurus who replied, so the summary about this topic is the following. Here are some of the links that help me, for tunning the TCP Stack. http://docs.sun.com/db/doc/806-7009 (for solaris 9) http://docs.sun.com/db/doc/816-0607 (for solaris 8) http://www.sean.de/Solaris/soltune.html http://www.sun.com/blueprints Thanks to all and especially to Nicolas Dorfsman, Steve Chapman, Karyn Williams, SANTHAKUMAR Siva, Kevin Buterbaugh and Moamer O. Mohsin. Have a nice day! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From R.K.Hillman at bton.ac.uk Thu Apr 3 09:43:42 2003 From: R.K.Hillman at bton.ac.uk (Ray Hillman) Date: Thu, 03 Apr 2003 15:43:42 +0100 Subject: SUMMARY: Gnome messed up my keyboard References: <3E8C3525.1020003@bton.ac.uk> Message-ID: <3E8C489E.2000609@bton.ac.uk> I haven't had any replies to this but after reporting the problem to Sun UK some days ago, asking the questions below on Sunmanagers only recently and tearing my hair out, I have managed to fix the problem. Below, is an expurgated version of my email to Sun: You will probably be pleased to know that I've fixed this problem. In /etc/dt/config I found I file dated Mar 25 16:19 named Xservers. In that file there is the line :0 Local local_uid at console root /usr/openwin/bin/Xsun :0 -nobanner +kb I checked on other Solaris 9 systems and found that I didn't have this file and infact the /usr/dt/config/Xservers file has the line: :0 Local local_uid at console root /usr/openwin/bin/Xsun :0 -nobanner which I assume is the default? I don't really know too much about X. However, I decided to rename /etc/dt/config/Xservers to /etc/dt/config/Xservers.old and reboot. Hey presto, keyboard is now o.k. I cannot be sure whether the Gnome installation created this or whether I did by clicking a button, but it was around the time I installed Gnome 2. Ray Ray Hillman wrote: Hi, Perhaps somewhat foolishly, I installed Sun's version of Gnome 2 on my Blade 100 workstation last week. I decided that I would take a look at this later and switched back to CDE, but now find that I have keyboard problems - notably I cannot type the tilde or hash character and that keys which repeated no longer do so. I'm using a type 6 USB keyboard (UK). Can anyone advise me how to get the default behaviour back? I don't feel that I should start fiddling around modifying key mappings. I've just go this idea that the Gnome installation has tweaked one or more files at the system level, which are common to other desktops. I can still type the characters at the command line prompt, but not in any desktop application. Thanks Ray -- RFC-822 : R.K.Hillman at bton.ac.uk Tel : +44 1273 600900 Ext. 2644 or +44 1273 642644 (direct) Fax : +44 1273 642666 Http : Systems Manager, University of Brighton Information Services, Watts Building, Moulsecoomb, Brighton, East Sussex, UK, BN2 4GJ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From RDinwid at templeinland.com Thu Apr 3 10:20:22 2003 From: RDinwid at templeinland.com (Dinwiddie, Ron (TIFPC)) Date: Thu, 3 Apr 2003 09:20:22 -0600 Subject: Summary: NIS, ypbind & maybe Oracle Message-ID: <78448300A1B3D21182100008C75DC421136CFBC3@mail.temple.com> All, Original problem stated below. I had a couple of responses and suggestions from the following: - Casper Dik who asked if I had run the ypinit -c as the error message stated (I had and the problem persisted) - Cathy Bodenstedt who suggested taking a look at the Oracle kernel parameters Here's the action I finally took with a final suggestion from Sun Tech Support (Jose Martinez) that has apparently resolved this issue: 1) asked one of our Oracle DBA's to check the file identified in the Sunsolve article below - which he did and the entry is correct 2) stopped NIS/yp on both the master and slave 3) "destroyed" NIS on the master (after backing up - I didn't just fall of the turnip truck *S* - I deleted the /var/yp directory) 4) recreated the NIS master "ypinit -m" 5) took the same action on the slave, except used "ypinit -s " 6) the problem came back after about 8 minutes - which is when I contacted Sun 7) Jose had me redo the steps above with the same result 8) the one thing that he had me do that had not previously been done, and apparently resolved the issue was to rename the following file: /var/yp/binding/`domainname`/ypservers 9) stopped and restarted NIS (/var/lib/netsvc/yp/ypstop & /var/lib/netsvc/yp/ypstart) Once I renamed the file in step 8 above and restarted NIS, the error messages stopped. A new issue was introduced that may be related to the renaming of the file when a late-night reboot of the Master showed that NIS successfully failed over to the slave, but the master also bound itself to the slave during the reboot phase. This was easily resolved by stopping the yp processes on both the master and the slave, starting them on the master - which bound itself back to itself as it should - and the restarted them on the slave - which also bound itself back to the master. A 2 minute "fix". Ron D -----Original Message----- From: Dinwiddie, Ron (TIFPC) [mailto:RDinwid at templeinland.com] Sent: Tuesday, April 01, 2003 3:13 PM To: sunmanagers at sunmanagers.org Subject: NIS, ypbind & maybe Oracle All, Please bear with me as this may be fairly long as I try to describe our issue. Setup: 2 E420R's - 1 set up as NIS Master and 1 as NIS slave/client OS - both running Solaris8 Oracle 8i - installed on NIS Master Both systems are on the same subnet Problem: Although NIS appears to be running properly and successfully, /var/adm/messages is growing extremely large with the following error message that repeats every 5 seconds: Apr 1 14:03:42 NISmaster ypbind[28563]: [ID 379826 daemon.error] service not installed, use /usr/sbin/ypinit -c The code inside the [] directly behind ypbind changes, but the ID number remains the same, as does the rest of the message. As I mentioned before, NIS appears to be running properly and all yp processes are running. I can execute a "ypcat hosts" on either system and get the map. I can run "/usr/ccs/bin/make" on the master and it pushes the maps out successfully. Things I have modified/checked include: - stopping & restarting yp (/usr/lib/netsvcyp/ypstop & ypstart) When I do this, the error message stops for approximately 10 minutes, then starts up again - editting /var/yp/Makefile and removing all references to maps that we aren't using in the NIS environment (we are only pushing out passwd, hosts, & group in this instance) - checked /etc/nsswitch.conf to ensure the proper entries were "files nis" - they are - recreating both the master & slave - cd /var/yp - ypinit -m (for master) - ypinit -s NISmaster (for the slave) - made both systems client using ypinit -c - found the following article concerning Oracle and asked one of our DBA's to check it out (he said that the recommended solution is already in place on the master): - http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=finfodoc%2F18622&zone_32=ypb ind - this article addresses the following: If something (client or server side) formulates a query which that leads to it attempting to contact a non existent database, the setup is likely to send a query off to NIS/YP. It is at this stage, that the NIS/YP domainname in ypbind gets corrupt. We have another sub-net running NIS successfully and the only difference I can see between the two is that the "problem" NIS master has Oracle installed on it while the "good" NIS master doesn't. Does anyone have any ideas that I haven't mentioned above? Thanks, Ron D _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bounce at trafficmagnet.com Thu Apr 3 11:40:34 2003 From: bounce at trafficmagnet.com (Sarah Williams) Date: Fri, 4 Apr 2003 00:40:34 +0800 (CST) Subject: www.sunmanagers.org Message-ID: <1954CR1000038421@p1j2m3a4.pdhost.com> Hi I visited www.sunmanagers.org, and noticed that you're not listed on some search engines! I think we can offer you a service which can help you increase traffic and the number of visitors to your website. I would like to introduce you to TrafficMagnet.com. We offer a unique technology that will submit your website to over 300,000 search engines and directories every month. You'll be surprised by the low cost, and by how effective this website promotion method can be. To find out more about TrafficMagnet and the cost for submitting your website to over 300,000 search engines and directories, visit us at: http://p1j2m3a4.pdhost.com/pdsvr/www/r?1000038421.1954.15.xRAV0Eta2FAxXN I would love to hear from you. Best Regards, Sarah Williams Sales and Marketing E-mail: Sarah_Williams at trafficmagnet.com http://www.TrafficMagnet.com This email was sent to summaries at sunmanagers.org. We apologize if this email has reached you in error. We honor all removal requests. Please go to the link below to be removed from our mailing list. http://p1j2m3a4.pdhost.com/pdsvr/www/optoutredirect?UC=Lead&UI=7612483 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030404/eb45c3b0/attachment.html From cbar44 at tsg.cbot.com Thu Apr 3 14:43:20 2003 From: cbar44 at tsg.cbot.com (Christopher L. Barnard) Date: Thu, 03 Apr 2003 13:43:20 -0600 Subject: SUMMARY: stopping broadcast boot on jumpstart? In-Reply-To: Message from "Christopher L. Barnard" of "Tue, 01 Apr 2003 14:27:44 CST." <200304012027.h31KRiI20040@shalmaneser.tsg.cbot.com> Message-ID: <200304031943.h33JhKI04693@shalmaneser.tsg.cbot.com> I asked: > We have a jumpstart server available to do flash-updates on other servers. > When a client does a "boot net install" everything works fine and we have > a built and configured server. The problem is that when this is done every > other system on that network that filters rpcbind reports a failed rpc > connection. This sends our monitoring applications and the people watching > them into a tizzy. When I asked Sun about this problem, they said I just > have to live with it ... the net part of the boot is a broadcast and > there is nothing I can do about it. I do not like the phrase "there is > nothing you can do about it". So, has anyone figured out a way to do a > jumpstart without doing an initial broadcast boot from the client > machine? If I could somehow tell the client to boot from a specific > IP, all would be good with the world. The answer: There is nothing I can do about it. "boot net" does a broadcast, and that is built into the OBP. I have to live with the rpcbind failure messages. Several people mentioned doing this on an isolated network, which is what we do when it is feasible. But when doing a live upgrade to an existing server, we can't move the server to a lab or string another network cable to the box temporarily. This is done over the network dedicated for backups, so it doesn't interfere with anything on the production networks. But since every server is on the backup network they all see and report the violation. Thanks to the many folks who took the time to respond. +-----------------------------------------------------------------------+ | Christopher L. Barnard O When I was a boy I was told that | | cbarnard at tsg.cbot.com / \ anybody could become president. | | (312) 347-4901 O---O Now I'm beginning to believe it. | | http://www.cs.uchicago.edu/~cbarnard --Clarence Darrow | +----------PGP public key available via finger or PGP keyserver---------+ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From plamadeleine at lightbridge.com Thu Apr 3 16:26:15 2003 From: plamadeleine at lightbridge.com (Paul LaMadeleine) Date: Thu, 03 Apr 2003 16:26:15 -0500 Subject: Final Summary - possible 2gig limit on usfdump on individual files? Message-ID: <5.2.0.9.2.20030403162343.05110fe8@jupiter.credtech.com> Greetings again, Well, I looked on sunsolve to see if the patch was there yet and it was. Here is a link to it. http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=fpatches%2F110387&zone_32=110387 I'll be downloading and installing it sometime this month, but I've been running on the beta ufsdump executable for a while now and have not had an issue. If you have any questions, feel free to email me. thanks, Paul >OK, > > I received a test binary from Sun and it fixed my problems. They > are planning on releasing a patch in 4-6 weeks, probably with the next > patch cluster release. > > I'll be applying the patch when it comes out and if any out there > want's to know how it went please let me know. > > thanks, > > Paul > >>I've opened a software support call on this and Sun is working on fixing >>it. There is a bug track that someone else opened on sunsolve.sun.com. >> >>When I get a patch, I'll re-summarize. >> >> thanks, >> >> Paul >> >>At 10:26 AM 1/13/2003 -0500, Paul LaMadeleine wrote: >>>Greetings, >>> >>> I'm trying to backup database files using ufsdump (we don't >>> want the full file system for some reason) and any file over 2 gig >>> seems to get this type of error: >>> >>>>04:39:26 01/12/2003 Backing up /db1/alaa/ala_temp01.dbf >>>> DUMP: Writing 32 Kilobyte records >>>> DUMP: Date of this level 0 dump: Sun Jan 12 04:39:27 2003 >>>> DUMP: Date of last level 0 dump: the epoch >>>> DUMP: Dumping /dev/md/rdsk/d80 (alawal:/db1) to /dev/rmt/0cn. >>>> DUMP: Mapping (Pass I) [regular files] >>>> DUMP: Cannot find filesystem mount point for `/db1/alaa/ala_temp01.dbf' >>>> DUMP: The ENTIRE dump is aborted. >>> >>> I've done some looking around and from what I've read is that >>> this is do to a large file issue with ufsdump on individual files as >>> opposed to the full file system. >>> >>> Can anyone confirm this or know of a patch to fix it? >>> >>> thanks, >>> >>> Paul LaMadeleine _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sunmngrs at mzserver.com Thu Apr 3 17:31:19 2003 From: sunmngrs at mzserver.com (Sun Manager) Date: Thu, 3 Apr 2003 17:31:19 -0500 Subject: SUMMARY: Veritas and Root Disk Setup Message-ID: Hello Sun Managers, I'd like to thank everyone for their replies: Kevin Buterbaugh, Hichael Morton, Bertrand Hutin, Aaron M. Hirsch, Jay Lessert, Ken Zachmann, Darren Dunham, Bryan Moore, Ben Rockwood, and Preiszl Matthias. There were several different answers I received (see my question at the bottom of this email): 1. About half of the answers I got were to just use DiskSuite and use Veritas for the data only. 2. Use a manual cloning/mirroring script. 3. Make a copy of the original disk (to be paranoid about a problem), encapsulate, and mirror, as mentioned in the Sun Blueprints. Most people said I should not worry about encapsulation of the root disk failing since it is very reliable and it updates all the vfstab and boot stuff automatically (and its faster than my idea). This page's suggestion also seemed interesting: http://unixway.com/vm/veritasvm/rootdg.html Also, if it's set up correctly, you should be able to see your disks without the Veritas drivers. See this page: http://www.sunhelpdesk.com/users/john/Mirror-boot-disk-with-VxVM.htm What a pain in the #@! all this stuff can be...but that's why we're all Sun Managers, right? I don't know about you, but I would just like technology to work as easy and painlessly as advertised...OK OK, that's enough of my philosophical input. ORIGINAL QUESTION: -------------------------------------------------- I posted to the list a while back and I am now at the point where I have the hardware to mirror my root disk; I was just wondering what everyone's feedback is on this particular "case study" and whether my "plan" will work: I have a SunFire 280R (one primary disk NOT under Veritas control with /, /usr, /usr/local, and /var slices) and one new internal, untouched 36GB disk. I also have a D1000 with 5-36GB drives and 7-9GB drives. The last 9GB drive is the only one in 'rootdg' (the rest are in '/export/home/xyz' user data disk groups). Is the following feasable (see my reasons why I want to do it this way at the end): 1. Initialize the new disk into Veritas control under rootdg with 1 (root) partition 2. Boot from CD-ROM and ufsdump each of the original disk's slices. 3. Boot into single user mode (so I have the Veritas drivers loaded) and ufsrestore each of the partitions into the newly initialized internal disk which is now under Veritas control. 4. I assume I have to do something else to make this new disk bootable? 5. Reboot into OpenBoot and boot off new disk. 6. If everything looks good, continue with the mirroring setup (initialize old disk, etc.) The reason I would like to do it this way is so that I don't have to mess around with encapsulating my root disk (so if something goes wrong, it's no big deal). Only once everything is working, will I go ahead and initialize my original disk and make it the mirror. Perhaps someone can tell me if this is possible to do? Also, on a quick sidenote: If your disk and your data are all under Veritas control, and some disaster happenes and you want to boot from a CD-ROM -- you wouldn't be able to access any of your data because there are no Veritas drivers on the CD-ROM, correct? Is there any way to get around this? (I've heard/read about custom Jumpstart -- but how about from a CD-ROM?) -------------------------------------------------- --END TRANSMISSION-- ;-) _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sun_consultant at hotmail.com Thu Apr 3 17:57:22 2003 From: sun_consultant at hotmail.com (Sun Consultant) Date: Thu, 3 Apr 2003 14:57:22 -0800 Subject: SUMMARY: Scanning Software for Solaris References: Message-ID: SANE seems to be the scanning software of choice: http://www.mostang.com/sane/ ----- Original Message ----- From: "Sun Consultant" To: "Sun Managers" Sent: Thursday, April 03, 2003 10:52 AM Subject: Scanning Software for Solaris > Hi-- > > Are there any good scanning software packages for Solaris? > (For use with Epson SCSI scanner) > > Thanks > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From UAdmnSUN at netscape.net Thu Apr 3 18:29:01 2003 From: UAdmnSUN at netscape.net (UAdmnSUN at netscape.net) Date: Thu, 03 Apr 2003 18:29:01 -0500 Subject: Summary:Spaces in SSH traffic Message-ID: <4EE1442A.191D9F03.007AE1D3@netscape.net> Hi, First thanks to the following: Crist Clark" Tim Chipman William Yodlowsky "Walt Sullivan" Haywood, Steven" Martynas Buozis and special thanks to "Crispin Harris" for the detailed description ( attched below). I have ended up rounding the double quotes with single quotes ('), escapeing the double quotes with \ did not help as my calling script didn't like it. Thanks UnixAdmin Crispin Harris reply: The quotes are being stripped in the multiple shells. /usr/local/bin/ssh /usr/local/bin/snpp -m '"Call Me"' or /usr/local/bin/ssh /usr/local/bin/snpp -m \"Call Me\" This is because the local shell (in which this command is executed) sees the command line and does this: (single quotes (') are used to denote the start and end of the string contained by the variables named) ARGV[0] = '/usr/local/bin/ssh' ARGV[1] = '' ARGV[2] = '' ARGV[3] = '/usr/local/bin/snpp' ARGV[4] = 'Call Me' ARGV[5] = '' The way that ssh works is that it looks for all arguments _UP_TO_ and _INCLUDING_ a hostname to define it's behaviour. It then passes all the remaining arguments to the shell on the remote end for execution. This means that on '' the a shell is executed with '' privilege with the command: /usr/local/bin/snpp Call Me The command probably needs to be: /usr/local/bin/snpp "Call Me" Somehow you need to the the double quotes through to the far end. There are two ways to do this: 1: enclose the double quotes in single quotes: '"Call Me"' this makes ARGV[4] = '"Call Me"' which thus makes the remote command: /usr/local/bin/snpp "Call Me" 2: escape the double quotes with a backslash (\). This prevents the first shell from interpreting the quotes. The arguments thus become: ARGV[0] = '/usr/local/bin/ssh' ARGV[1] = '' ARGV[2] = '' ARGV[3] = '/usr/local/bin/snpp' ARGV[4] = '"Call' ARGV[5] = 'Me"' ARGV[6] = '' The remote command is thus assembled like this: /usr/local/bin/snpp "Call Me" and the remote shell (and thus the perl script) do what you expect. You want ARGV[4] to equal '"Call Me" "Call Me" and passes it as the single variable Crispin Harris Systems Manager Thin Technologies Limited Vox: +(61) 8 9488 3000 Fax: +(61) 8 9488 3030 Mob: +(61) 413 629 298 E-mail: charris at thintechnologies.com Web: www.thintechnologies.com > -----Original Message----- > From: sunmanagers-admin at sunmanagers.org > [mailto:sunmanagers-admin at sunmanagers.org]On Behalf Of > UAdmnSUN at netscape.net > Sent: Thursday, April 03, 2003 2:46 PM > To: sunmanagers at sunmanagers.org > Subject: OT: Spaces in SSH traffic > > > Hi All, > > My appologies for asking this question here. > We have a host which can send pager messages. Other hosts > connect to this host via SSH to send any pager messages. We > use perl-snpp to page. When we send page from the pager host > no messages are sent ok (spaces between the words do not > disappear). But when sent from the remote servers messages > are truncated. Ex: > /usr/local/bin/ssh /usr/local/bin/snpp -m > "Call Me" > ( user is set to go with no password) > > Message received is: > Call (only) > > I believe this is happening in SSH tunnel. Any SSH gurus out > there who can give me a clue. All the hosts are Solaris and > OpenSSH (latest) version 3.51 > > Thanks > UnixAdmin > > __________________________________________________________________ > Try AOL and get 1045 hours FREE for 45 days! > http://free.aol.com/tryaolfree/index.adp?375380 > > Get AOL Instant Messenger 5.1 for FREE! Download Now! > http://aim.aol.com/aimnew/Aim/register.adp?promo=380455 > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > __________________________________________________________________ Try AOL and get 1045 hours FREE for 45 days! http://free.aol.com/tryaolfree/index.adp?375380 Get AOL Instant Messenger 5.1 for FREE! Download Now! http://aim.aol.com/aimnew/Aim/register.adp?promo=380455 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From himanshukhona at indiatimes.com Thu Apr 3 23:50:30 2003 From: himanshukhona at indiatimes.com (himanshukhona) Date: Fri, 04 Apr 2003 10:20:30 +0530 Subject: Summary: tape to tape dd on different workstations Message-ID: <200304040418.JAA20886@WS0005.indiatimes.com> Hi, Below is the mail attached which has similar answers to many others. Thanks to all of them. AS the subject suggests i needded to duplicate tapes which would be in tape drives connected to different hosts. The answer was pretty much simple: From osamaomar at link.net Fri Apr 4 02:32:57 2003 From: osamaomar at link.net (Osama Omar) Date: Fri, 4 Apr 2003 09:32:57 +0200 Subject: Summary Mail on Solaris 8 In-Reply-To: <7F09903EF5CBD211B5A20008C739C02C66F8AB@GVMAIL01> Message-ID: <001201c2fa7c$6a2f5210$c79e939d@onshore.pgs.com> RE: Mail on Solaris 8Hi Many thanks to all who replied I have tried this way and gave me what I need In my /etc/hosts file I have the following entry: host ip address mailhost.mydomain.com mailhost Osama Omar Ali PGS Data Processing Middle East IT Dept. (002)(02) 2761353 (002)(02) 2761353 (002) (010) 6008263 -----Original Message----- From: Vincent Ribeaud [mailto:Vincent_Ribeaud at adp.com] Sent: Wednesday, April 02, 2003 11:18 AM To: 'osamaomar at link.net' Subject: RE: Mail on Solaris 8 Hi, In my /etc/hosts file I have the following entry: 172.xx.xx.xx mailhost.mydomain.com mailhost Vincent -----Original Message----- From: Osama Omar [mailto:osamaomar at link.net] Sent: Wednesday, April 02, 2003 11:17 AM To: 'Sunmanagers (E-mail)' Subject: Mail on Solaris 8 Hi How can I configure my Solaris 8 dns server to send mails to internet mail accounts using the "mail" command from its clients ? eg: # mail osamaomar at link.net test ^D how can this reach my mail account while I dont have a unix mail server but want the Unix DNS to point to an exchange server? Regards Osama Omar Ali PGS Data Processing Middle East IT Dept. (002)(02) 2761353 (002)(02) 2761353 (002) (010) 6008263 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers This message is confidential and may also be legally privileged. If you are not the intended recipient, please notify postmaster at adpwilco.com immediately. You should not copy it or use it for any purpose, nor disclose its contents to any other person. The views and opinions expressed in this e-mail message are the author's own and may not reflect the views and opinions of ADP Wilco. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Schaper at media-saturn.com Fri Apr 4 05:11:23 2003 From: Schaper at media-saturn.com (Schaper, Soeren) Date: Fri, 4 Apr 2003 12:11:23 +0200 Subject: SUMMARY Remote System control problems Message-ID: <9E86F3E9F5C5D6119A5D00508BE1286B9FA94F@exchange.media-saturn.com> Hi all, thanks to Kelly G. McDonald, Mark Preston, Charlott Ratliff, Andrew Maddox, Joe Fletcher, Jim Musso, Thomas Autry and Dirk Boenning who all pointed out to me that there is an error with the sun documentation and to redirect the console you have to set input-device and output device to rsc-console and not as described in the dokumentation just to rsc. again, thanks to all who replied Soeren -----Urspr|ngliche Nachricht----- Von: Schaper, Soeren Gesendet: Donnerstag, 3. April 2003 16:50 An: 'sunmanagers' Betreff: Remote System control problems Hi, I'am in the process of configuring the RSC on our newest V880. So far all went well, I can connect and issue all rsc commands. What I don't get running is having the console redirected to the rsc to see the boot messages and if I send a break, nothing happens. The OBP command diag-output-to rsc does not seem to work. Does anyone knows whats wrong here ? Soeren _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From flegal at seo-toulouse.com Fri Apr 4 05:54:40 2003 From: flegal at seo-toulouse.com (=?iso-8859-1?Q?Fran=E7ois?= Legal) Date: Fri, 04 Apr 2003 12:54:40 +0200 Subject: Summary : sendmail smarthost Message-ID: <3E8D6470.FC45FD51@seo-toulouse.com> Hi managers. Thanks for all your answers. My problem is still not fixed, and I will give you more details : Then question was how to configure sendmail for relaying all emails to a smarthost. I'am in the domain mysub.bogus.com. There is one mail server for domain bogus.com ( with a cname mailhost.bogus.com in the dns system) All mails for the outside arrive by this host. In my subdomain mysub.bogus.com, I've got another mail server ( called mail-server ), and I want to use this one as my smarthost. When I specify as DSmail-server or DSmail-server.mysub.bogus.com, the sendmail log show that it tries to send outgoing mail to bogus.com directly. If I leave DS empty, the outgoing mails are sent to mailhost.bogus.com The only way I've found is to indicate DS[] -- Cordialement Frangois Legal S.E.O. - Test automatique Tel : 05 34 55 40 86 Fax : 05 34 55 40 95 E-mail : flegal at seo-toulouse.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From kevin at pmimail.com Fri Apr 4 08:50:09 2003 From: kevin at pmimail.com (Kevin Metzger) Date: Fri, 4 Apr 2003 08:50:09 -0500 (EST) Subject: SUMMARY: verify commands: /usr/bin/passwd, etc In-Reply-To: Message-ID: Before I begin, this was a duplicate question to "Hacked Help" that got held for a questinable subject line. The summary is also a copy of the same, but I will copy it here for completeness. First, thanks to the 56 people who responded. By far the most frequent response was to check md56 fingerprints at http://sunsolve.Sun.COM/pub-cgi/fileFingerprints.pl . I downloaded the md5 from http://sunsolve.sun.com/md5/md5.tar.Z . One suggestion for http://www.sun.com/blueprints/0501/Fingerprint.pdf . Another suggestion was to get and run chkrootkit from http://www.chkrootkit.org which I did first. That suggested the t0rn trojan, but that is know to atack DNS servers that this server is not running. The other suggestion that was often repeated is to reload or some conbination of remove the server from the network _now_ and boot from CDROM or remove the drive to another Solarib server and run diagnostics while the compromised drive is mounted in /a . My solution thus far is to run the chkrootkit first then check the md5 fingerprints of everything in /usr/bin and /usr/sbin. If it didn't check and I didn't know what it was it got mv-ed and if it were a solaris binary, copied it from the CD. I also wrote all the md5's to a log file and cron a job that creates a file nightly and diff the two for all files in /usr/bin and /usr/sbin . And just for fun a added three lines to my .profile that do a last for root, bin and adm for my inspectin each and every time I login. I have disabled root logins except for the console adn verified that users bin and adm have the NP no password set. I have not seen any further suspicious logins. I plan to impliment the noshell script below. For anyone interested the Solaris binaries compromised were /usr/bin/du /usr/bin/ls /usr/bin/passwd /usr/bin/find /usr/bin/netstat /usr/bin/su and the replacements were exactly the same size and date as the origional. Tim Wort went on to say: > >As for the accounts: adm and all system accounts should not have passwords >or shells configured, they should be locked with the shell replaced, I >would replace the shell with a script called noshell (from Titan.). > >noshell: > >trap "" 1 2 3 4 5 6 7 8 9 10 12 15 19 > >HOSTNAME=`uname -n` >USER=\`id | awk '{print $1}' | awk -F= '{print $1}'\` > >/bin/cat /dev/null |mailx -s "Attempted access by ${USER} on host >${HOSTNAME}" root@${HOSTNAME} & > >echo "Sorry, you are not allowed to logon." > >exit > -- Kevin Metzger Systems Administrator Progressive Medical, Inc. 800 777-3574 x2686 desk 614 378-6396 mobile 614 389-0740 fax Recieved Fri, 21 Mar 2003, from Kevin Metzger: -Date: Fri, 21 Mar 2003 15:43:21 -0500 (EST) -From: Kevin Metzger -To: sunmanagers mailing list -Subject: verify commands: /usr/bin/passwd, etc - -I think I've been hit by invaders and want to verify commonly molested -commands before running them. systems are Solaris 7 and 8 on sparc and x86. - -The evidence is the output from last: -dwayne console Fri Mar 21 14:15 - 14:19 (00:04) -root pts/17 host130-220.pool Fri Mar 21 13:56 - 14:01 (00:04) -adm pts/16 host130-220.pool Fri Mar 21 13:56 - 13:57 (00:00) - -Dwayne is my tech support, but how is someone getting in as adm directly? One -other machine here shows root, adm and bin logging in. I'm seeing this on -three seperate machines. - -thanks in advance and I summarize. - --- -Kevin Metzger -Systems Administrator -Progressive Medical, Inc. - -800 777-3574 x2686 desk -614 378-6396 mobile -614 389-0740 fax -_______________________________________________ -sunmanagers mailing list -sunmanagers at sunmanagers.org -http://www.sunmanagers.org/mailman/listinfo/sunmanagers - _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rob.delanghe at telindus.be Fri Apr 4 08:23:20 2003 From: rob.delanghe at telindus.be (Rob De Langhe) Date: Fri, 4 Apr 2003 15:23:20 +0200 Subject: SUMMARY: Netscape 7 on Solaris9 helper apps do not start Message-ID: Hi, I found the problem myself, it was from the RTFM category : the help pages from Netscape Communicator 7 explain how to add a helper application: you need to specify the path to the application that it must launch for a certain MIME type / extension. It nowhere tells you to add the argument '%s' behind that file path, as it uses to be in all other Netscape versions so far (see .mailcap, e.g.) Well done of myself, and this on a Friday afternoon. Imagine... Rob -----Original Message----- From: Rob De Langhe To: 'sunmanagers at sunmanagers.org' Sent: 04-04-2003 14:05 Subject: Netscape 7 on Solaris9 helper apps do not start Hi, I have Netscape Communicator 7.0 on Solaris-9, and want to configure a helper app to view PostScript files, but they fail with a popup (from NS) "/tmp/SL210_211_OH.ps could not be opened, because an unknown error occured. Sorry about that. Try saving to disk first and then opening the file." The file "/tmp/SL210_211_OH.ps" exists on disk, and in my preferences of NS I have put Extension: ps File-type: application/ps Handled by: /usr/local/bin/gv %s When I try to view this PS file from my browser, the browser shows a popup telling me that it's downling the PS file and will open it with the help application '/usr/local/bin/gv %s', but then I get the popup error msg, and if I run "ls -lu /usr/local/bin/gv" I see that this program is not even accessed! I even did launch my browser from the command line with the "-debug" option, but that doesn't print any messages after the browser started up. How to debug this any further? Rob _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From kevin at pmimail.com Fri Apr 4 08:46:45 2003 From: kevin at pmimail.com (Kevin Metzger) Date: Fri, 4 Apr 2003 08:46:45 -0500 (EST) Subject: SUMMARY: hacked help In-Reply-To: Message-ID: First, thanks to the 42 people who responded. By far the most frequent response was to check md56 fingerprints at http://sunsolve.Sun.COM/pub-cgi/fileFingerprints.pl . I downloaded the md5 from http://sunsolve.sun.com/md5/md5.tar.Z . One suggestion for http://www.sun.com/blueprints/0501/Fingerprint.pdf . Another suggestion was to get and run chkrootkit from http://www.chkrootkit.org which I did first. That suggested the t0rn trojan, but that is know to atack DNS servers that this server is not running. The other suggestion that was often repeated is to reload or some conbination of remove the server from the network _now_ and boot from CDROM or remove the drive to another Solarib server and run diagnostics while the compromised drive is mounted in /a . My solution thus far is to run the chkrootkit first then check the md5 fingerprints of everything in /usr/bin and /usr/sbin. If it didn't check and I didn't know what it was it got mv-ed and if it were a solaris binary, copied it from the CD. I also wrote all the md5's to a log file and cron a job that creates a file nightly and diff the two for all files in /usr/bin and /usr/sbin . And just for fun a added three lines to my .profile that do a last for root, bin and adm for my inspectin each and every time I login. I have disabled root logins except for the console adn verified that users bin and adm have the NP no password set. I have not seen any further suspicious logins. I plan to impliment the noshell script below. For anyone interested the Solaris binaries compromised were /usr/bin/du /usr/bin/ls /usr/bin/passwd /usr/bin/find /usr/bin/netstat /usr/bin/su and the replacements were exactly the same size and date as the origional. Tim Wort went on to say: > >As for the accounts: adm and all system accounts should not have passwords >or shells configured, they should be locked with the shell replaced, I >would replace the shell with a script called noshell (from Titan.). > >noshell: > >trap "" 1 2 3 4 5 6 7 8 9 10 12 15 19 > >HOSTNAME=`uname -n` >USER=\`id | awk '{print $1}' | awk -F= '{print $1}'\` > >/bin/cat /dev/null |mailx -s "Attempted access by ${USER} on host >${HOSTNAME}" root@${HOSTNAME} & > >echo "Sorry, you are not allowed to logon." > >exit > -- Kevin Metzger Systems Administrator Progressive Medical, Inc. 800 777-3574 x2686 desk 614 378-6396 mobile 614 389-0740 fax Recieved Fri, 21 Mar 2003, from Kevin Metzger: -Date: Fri, 21 Mar 2003 14:52:26 -0500 (EST) -From: Kevin Metzger -To: sunmanagers mailing list -Subject: hacked help - -I think I've been hacked and am goign to change the root and other passwords. -How can I verify that my passwd command has not been compromised? - -Thanks and I will summarize. - --- -Kevin Metzger -_______________________________________________ -sunmanagers mailing list -sunmanagers at sunmanagers.org -http://www.sunmanagers.org/mailman/listinfo/sunmanagers - _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From debbie at icus.com Fri Apr 4 12:01:34 2003 From: debbie at icus.com (Debbie Tropiano) Date: Fri, 4 Apr 2003 11:01:34 -0600 (CST) Subject: SUMMARY: Is it safe to remove SEAM? In-Reply-To: from "debbie" at Mar 20, 2003 12:20:57 PM Message-ID: <200304041701.h34H1Ygl079257@ds9.icus.com> Folks - Well, I removed the SEAM packages and rebooted the system and things seemed to be OK, so I would surmise that it is "safe". However, the MIT Kerberos still didn't inter-operate with the Cisco networking devices as well as we needed, so we're going to use TACACS instead and have reverted to SEAM on the Solaris side of things. Debbie My original message: > Folks - > > I'm working on installing/configuring MIT Kerberos on my Solaris 9 > system and keep running into the Solaris SEAM stuff, thus I would > like to just remove all of the SEAM packages. Before I mess up > my system, I'd like opinions on whether or not that would be "safe". > > What do y'all think? > > Thanks in advance for any help, > Debbie -- + Debbie Tropiano -- debbie at icus.com -- http://www.icus.com/personal.html + | Mommy to Nathan b: 8/17/1995, ^Sara^ b: 10/25/2000 d: 11/7/2000 & | | Leah b: 10/17/2001 a: 9/26/2002 "God shows His opposition to cancer and | | birth defects, not by eliminating them or making them happen only to bad | | people (He can't do that), but by summoning forth friends and neighbors | + to ease the burden and to fill the emptiness." -- Harold S. Kushner + _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ypismerov at tucows.com Fri Apr 4 13:19:32 2003 From: ypismerov at tucows.com (Yura Pismerov) Date: Fri, 04 Apr 2003 13:19:32 -0500 Subject: SUMMARY: Spectralogic library and Legato References: <3E8DBBD0.DE8803C4@tucows.com> Message-ID: <3E8DCCB4.70A084F1@tucows.com> Thanks to Luc I. Suryo for reply. But it turned out that sgen driver was interfering with Legato lus. After I removed sgen (rem_drv sgen), everything went smoothly. Yura Pismerov wrote: > > Hi dear sunmanagers, > > I'm hitting a wall trying to get to work Spectralogic TreeFrog (Spectra > 2k) AIT-2 library to work with Legato on Solaris 9. The library is > currently working with Amanda backup with no problem - I use mtx based > changer script that utilizes sgen driver. But when I try to configure it > with Legato jbconfig it fails to attach the device, though it detects > the tapes and the changer (see below). I searched Google with no luck. > Does anybody know what to do ? > > TIA. > > Here is some relevant info. > > bash-2.05# uname -a > SunOS xxxx.com 5.9 Generic_112233-04 sun4u sparc SUNW,Sun-Fire-280R > > ------------------------------------8<---------------------------------- > > bash-2.05# pkginfo -l LGTOserv > PKGINST: LGTOserv > NAME: NetWorker for Solaris (Backup/Recover) Server > CATEGORY: application > ARCH: sparc > VERSION: 6.1.3.Build.428 > BASEDIR: /usr > VENDOR: Legato Systems, Inc. > DESC: NetWorker for Solaris (Backup/Recover) Server > PSTAMP: gelato0211212144 > INSTDATE: Apr 03 2003 09:31 > HOTLINE: Please contact your local service provider > STATUS: completely installed > FILES: 31 installed pathnames > 1 shared pathnames > 29 executables > 43890 blocks used (approx) > > -------------------------8<------------------------------------ > > bash-2.05# /usr/sbin/jbconfig > 1) Install a SmartMedia Jukebox. > 2) Install an Autodetected SCSI Jukebox. > 3) Install an Autodetected NDMP SCSI Jukebox. > 4) Install an SJI Jukebox. > 5) Install an STL Silo. > > What kind of Jukebox are you installing? [1] 2 > These are the SCSI Jukeboxes currently attached to your system: > 1) scsidev at 1.6.0: Spectralogic > Which one do you want to install? 1 > Installing an 'Spectralogic' jukebox. > Name you would like to assign to the jukebox device? jb-1 > A brief description of the jukebox device? [scsidev at 1.6.0: Spectralogic] > Pathname of the control port for the jukebox device? [scsidev at 1.6.0] > Unable to access device "Spectralogic" on control port "scsidev at 1.6.0". > > Have you already installed the "drivers" package on this machine? > (yes/no) yes > There appears to be something incorrect about your driver > installation. Please consult your driver documentation > to help you troubleshoot your installation. > > -----------------------------------8<------------------------------------------- > > bash-2.05# /etc/LGTOuscsi/inquire > scsidev at 0.6.0:TOSHIBA DVD-ROM SD-M14011009|CD-ROM, /dev/rdsk/c0t6d0s2 > scsidev at 1.4.0:SONY SDX-500C 01 |Tape > scsidev at 1.5.0:SONY SDX-500C 01 |Tape > scsidev at 1.6.0:SPECTRA 215 1008|Autochanger (Jukebox) > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From dakotak1d at netscape.net Fri Apr 4 18:14:51 2003 From: dakotak1d at netscape.net (dakotak1d at netscape.net) Date: Fri, 04 Apr 2003 18:14:51 -0500 Subject: SUMMARY: xhost/X Display question Message-ID: <607B12F5.74890D72.02E7D20C@netscape.net> Thanks to all who responded, In the end I choosed Xvfb - X Virtual Frame Buffer. So far it's working perfectly! Xvfb uses shared memory to create a virtual frame buffer to (for example) DISPLAY=:1.0 Because of this I wonder if at some point I'll have to increase shared memory segments, I guess we'll see. Sincere thanks to the following who suggested Xvfb: David Foster Jonathan Burelbach Robert N Wood Matthew Stier Other suggestions: VNC Server www.realvnc.com: "Vandevegt, James Matthew (Jim)" "Roetman, Paul" dtautologin: "Heilke, Rainer" ORIGINAL QUESTION: SunGods, I have a requirement to automate a process upon startup (probably a script in rc3.d) that will do: xhost + su - appluser -c "export DISPLAY=satori:0.0;xhost +;/usr/dt/dtterm" (I know, I know...X security...) Or perhaps I need to do something else....the goal is to allow an application that runs via cron can open a dtterm session which it needs & uses without manual intervention. If this workstation gets rebooted, etc. the job always fails unless someone manually logs in as the application user and opens a dtterm & does the export DISPLAY, xhost, etc. I can't seem to get this set via startup scripts...it complains with: Xlib: connection to "satori:0.0" refused by server Xlib: Invalid MIT-MAGIC-COOKIE-1 key xhost: unable to open display "satori:0.0" Is the problem because the workstation goes to CDE login screen & cannot be 'sent' a dtterm until it's actually logged in? TIA, Kevin __________________________________________________________________ Try AOL and get 1045 hours FREE for 45 days! http://free.aol.com/tryaolfree/index.adp?375380 Get AOL Instant Messenger 5.1 for FREE! Download Now! http://aim.aol.com/aimnew/Aim/register.adp?promo=380455 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rob.delanghe at telindus.be Mon Apr 7 04:28:57 2003 From: rob.delanghe at telindus.be (Rob De Langhe) Date: Mon, 7 Apr 2003 10:28:57 +0200 Subject: SUMMARY: problems with StarOffice 6.0 Message-ID: Hi, here we are again with the solution to our problem : Thanks to Brian Riley, Gerard Henry, Frank Smith, John D Groenveld, Mario Williams, Tim Chipman & Tony van Lingen for replying, but a very special thanks to Ed Murphy who pinpointed the exact cause of the problem: if Staroffice finds the library "libXrender" in the LD_LIBRARY_PATH variable, it will use it in order to speed up the X-display. But a minor (...) side-effect is that the menu labels do not display anymore. So I simply edited the startup script "/opt/staroffice6.0/program/soffice" and modified the setting of the LD_LIBRARY_PATH in the "case" statement (around line 144). Just before the line(s) export LD_LIBRARY_PATH I included the following line to remove the directory /usr/sfw/lib (containing the nasty libXrender library) from the variable contents: LD_LIBRARY_PATH=`echo $LD_LIBRARY_PATH|sed -e 's/\/usr\/sfw\/lib//g'` export LD_LIBRARY_PATH That's all. Everything is working fine from then on. We have NOT applied the patch from SunSolve yet, waiting to see if I get other problems with the native version. By the way, this patch 112885-03 is listed to apply on Sol-7 and Sol-8, nothing said about Sol-9. So cross our fingers and wait&see. Regards to this spectacular group of gurus! Rob A very special thanks to -----Original Message----- From: Rob De Langhe To: 'sunmanagers at sunmanagers.org' Sent: 04-04-2003 12:00 Subject: UPDATE: problems with StarOffice 6.0 Hi, I got a few remarks that I needed to invoke the "setup" script from the CD with the option "-net". I followed the procedure at docs.sun.com (http://docs-pdf.sun.com/816-7522/816-7522.pdf) to render SO6 available to end-users. After this, as a user, I launch the command $HOME/staroffice6/soffice This displays a main SO6 window, but with a popup window that's titled "Registration", but any text in this window is blanked out (fonts? colors?) Also all the menu options in the main SO6 window do not carry labels, in these pull-down menu options again all text labels are blank. What did I do wrong ? What's missing ? Rob -----Original Message----- From: Rob De Langhe To: 'sunmanagers at sunmanagers.org' Sent: 04-04-2003 08:31 Subject: problems with StarOffice 6.0 Hi, we are running Solaris-9, and I just wanted to give StarOffice a try for our users. I installed the bundled CD "StarOffice 6.0", it goes into /opt/staroffice6.0 When an ordinary user launches "/opt/staroffice6.0/soffice some.doc" (MS-Word document) it shows an initial window-border (altough empty) and in there a small popup-window (again empty) with just a yellow triangle with exclamation mark in it and a non-labeled button that I can press. When this button is pressed, the program crashes and leave a core dump, not a single message. How about that ? Where do I start to troubleshoot ? All reactions are very welcome ! Rob _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Martin.Gabel at ffm3.siemens.de Mon Apr 7 09:16:41 2003 From: Martin.Gabel at ffm3.siemens.de (Gabel Martin) Date: Mon, 7 Apr 2003 15:16:41 +0200 Subject: SUMMARY: 13W3M pin allocation Message-ID: <30888BF0456F6C41AD65E5415EC1A8130DC450@ffmr222a.ffm3.siemens.de> Thanks to all who have replied. I got several valuable links and pdf files: http://sunsolve.sun.com/handbook_pub/Devices/Monitor/MONITOR_Connector_Pinou ts.html http://csgrad.cs.vt.edu/~tjohnson/pinouts/#video I'll write a more detailed essay when I've successfully tested my cable cu may _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From hbp4c at virginia.edu Mon Apr 7 11:02:10 2003 From: hbp4c at virginia.edu (Howard Powell) Date: Mon, 7 Apr 2003 11:02:10 -0400 (EDT) Subject: SUMMARY: $PATH in sun Gnome2 Message-ID: A couple good answers to my problem came up. The best answer I found was to modify the ~/.gnomerc which was not present in the account to add the $LD_LIBRARY_PATH and $PATH to it. That should convince Gnome to see what I want it to see. The other good ideas were to use my $LDFLAGS variable to set the -L and -R directorys for linking and runtime libraries, and then recompile the software I want to add to the menu. Another was to copy or link the .bash_profile I was using to ~/.bashrc to be sure that bash was correctly reading my environment. Thanks Mario Williams, Dave Uhring, and the sysadmin at Stockholm Observatory for your kind help. Howard --- "This document was supplied free of defects in spelling, punctuation or grammar. Check with your provider for possible errors in reception." Howard Powell Undergraduate Astronomy UVA Assistant LSP - Astronomy _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From flegal at seo-toulouse.com Mon Apr 7 11:42:37 2003 From: flegal at seo-toulouse.com (=?iso-8859-1?Q?Fran=E7ois?= Legal) Date: Mon, 07 Apr 2003 17:42:37 +0200 Subject: Summary : automount on sunos 4.1.3 Message-ID: <3E919C6D.D1625186@seo-toulouse.com> Thanks for all the answers. None of them really fixed my problem, so here is the solution I used I changed my maps by username1 file-server:/export/home/username1 This solution will do a mount request for /export/home each time a new user will log in. Frangois Legal _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From reinaldo.ono at copel.com Mon Apr 7 15:47:38 2003 From: reinaldo.ono at copel.com (reinaldo.ono at copel.com) Date: Mon, 7 Apr 2003 16:47:38 -0300 Subject: Summary : Diferent Colors in Motif Message-ID: Sun Managers, nobody give me the right answer to my question : Original Question : > Hi Sunmanagers gurus, > I need some help to solve a problem. > I have two servers. One have Solaris 2.6 and have a Motif aplication that shows me some colors. The second . > have Solaris 8 and the same Motif aplication shows different colors. Both servers are accessed by Hummingbird Exceed using 256 colors X11 emulator for Windows. > I don't know what can I do to the second server aplication get the same colors from the first server. > Could you help me with this question ? The problem was the same for dtterm and xterm. When I compare dtterm or Xterm opens from Solaris 8 with Solaris 2.6 or 9, the colors were different. But Sun's Support answers me : > This problem is documented in bugId 4489804: xterm -bg blue produces > different shades of blue on 2.6 and 8 > > Please check the workarounds available in 4489804: > > setenv XCMSDB /dev/null (for one process) > > -or- > > rm /usr/openwin/lib/X11/Xcms.txt (for all processes) > > -or- > > replace /usr/openwin/lib/X11/Xcms.txt with the xc/lib/X11/Xcms.txt file from > > the XFree86 distribution from www.xfree86.org or the Xcms.txt file from patch > > 105618-01 (for all processes) > > The fix for this bug is integrated in Solaris 9. > There is no patch for this fix on Solaris 7 and 8. If customer needs an > official patch, you will have to open an escalation against this bug. > Best regards, > - Eric _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bodenste at mailcenter.csap.af.mil Mon Apr 7 17:04:01 2003 From: bodenste at mailcenter.csap.af.mil (Cathy M. Bodenstedt) Date: Mon, 07 Apr 2003 16:04:01 -0500 Subject: SUMMARY:Finding out Location of drives in an A1000 Message-ID: <3E91E7C0.5B09EE91@mailcenter.csap.af.mil> Found it myself. Thanks for Shawn Evans and Matthew Stiers' lreplies. /usr/lib/osa/bin/drivutil -i c1t5d0 |awk '{print $1}' -- Phone: 210-925-3244 DSN: 945-3244 [demime 0.99c.7 removed an attachment of type text/x-vcard which had a name of bodenste.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From tpetrasek at internap.com Mon Apr 7 17:04:19 2003 From: tpetrasek at internap.com (Trent Petrasek) Date: Mon, 7 Apr 2003 17:04:19 -0400 Subject: SUMMARY: u60 Red State Exception In-Reply-To: <20030407203427.GB73094@tpetrasek.acs.internap.com> References: <20030407203427.GB73094@tpetrasek.acs.internap.com> Message-ID: <20030407210419.GE73094@tpetrasek.acs.internap.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Greetings All!! Success! I have fixed the problem. I realize I said nothing of memory in my original post. I did, however, swap out memory several times, but it was always "used" memory. I installed four clean modules right out of the static bags and this solved the problem! Much thanks to Roger Leonard, Gary Trail and Steven Faulconer for the memory suggestions! Thanks all, - ---------------------------------------------------------- Trent Petrasek Internap Network Services Systems Engineer IT Services tpetrasek at internap.com 404.302.9742 On Mon, Apr 07, 2003 at 04:34:27PM -0400, Trent Petrasek wrote: > > Greetings Sun Managers, > > This morning I started work on an older model U-60 desktop system. The system has two 450 MHz cpu modules, two 256M memory modules, and two 9G drives. > > When I first powered up the system, it offered many "Red State Exceptions" to me. No problem, I've seen this before on a Netra 1125 which had a CPU problem. So I yank out one of the two CPU modules and power her back up. I get the same error. Ok, I think, it must be the other CPU. So I yank out CPU 2, replacing CPU 1. No go, I get the same error. At this point I've come to the end of my experience with "red state exceptions." > > I google the problem, searching for help. There is not much out there, and the majority of it is related to CPU. Not wanting to believe I have two bad CPU modules, I hunt down and locate a third 450MHz module and drop it in the box. Unfortunately, as you may have guessed, this did not work either. > > Now I'm frustrated because the problem is not readily apparent. Luckily, I have another u60 box that is down due to power problems. I borrow the motherboard from the machine, installing it into the broken u60. I populate it with working CPUs from the powerless u60 and power it up and I get the same error. > > Ok, I think to myself, it is either the power unit or the SCSI backplane. I unplug the SCSI backplane and all of the miscellaneous front end wiring from the motherboard except power switch and related functions. Upon powering it up, I get the same exact error. > > Here is the error itself: > > RED State Exception > > TL=0000.0000.0000.0005 TT=0000.0000.0000.0064 > TPC=0000.0000.efff.f9e4 TnPC=0000.0000.efff.f9e8 TSTATE=0000.0044.5604.1400 > TL=0000.0000.0000.0004 TT=0000.0000.0000.0064 > TPC=0000.0000.efff.f9e4 TnPC=0000.0000.efff.f9e8 TSTATE=0000.0044.5604.1400 > TL=0000.0000.0000.0003 TT=0000.0000.0000.0064 > TPC=0000.0000.efff.f9e4 TnPC=0000.0000.efff.f9e8 TSTATE=0000.0044.5604.1400 > TL=0000.0000.0000.0002 TT=0000.0000.0000.0064 > TPC=0000.0000.efff.e334 TnPC=0000.0000.efff.e338 TSTATE=0000.0044.5604.1400 > TL=0000.0000.0000.0001 TT=0000.0000.0000.0064 > TPC=0000.0000.efff.e764 TnPC=0000.0000.efff.e768 TSTATE=0000.0044.5600.0400 > > [rinse .. repeat] > > By this point, I am convinced the problem is the power unit, but I an interested in hearing your opinions about this as well. This hardware is out of warranty and there is no hope of securing a new contract on it. Your guidance is much appreciated. > > Thanks!!! > > > ---------------------------------------------------------- > Trent Petrasek Internap Network Services > Systems Engineer IT Services > tpetrasek at internap.com 404.302.9742 > > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQE+kefT1CLn4SP2qlMRAvb1AJ9wPCCk8YkkCD4eA09MJBS5C0AaNACg1drj fGs8i19DbBB3Skoa6An6qL0= =Ct4f -----END PGP SIGNATURE----- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From debbie at icus.com Mon Apr 7 17:52:56 2003 From: debbie at icus.com (Debbie Tropiano) Date: Mon, 7 Apr 2003 16:52:56 -0500 (CDT) Subject: SUMMARY: SEAM: Error while propagating the database Message-ID: <200304072152.h37Lqu5S029119@enterprise.icus.com> Hello - As it turns out the proper entries were in the /etc/inet/inetd.conf file, but were commented out. Changing the file and SIGHUPing inetd got me past it and on to just boring old permission problems which appear to be more straightforward and easier to fix (now just to figure out who at Sun to complain to about their documentation -- I've found several errors/omissions in the two Solaris 9 guides that I've been using). Debbie PS I did get one response but ultimately found someone else in the google archives of comp.protocols.kerberos that had had the same problem with the MIT version on RH. He gave me a few hints and I was able to figure it out. Original message: > Hello - > > I'm configuring SEAM (Kerberos 5) on two Solaris 9 hosts and > while trying to propagate the database from the master to the > new slave system using the command: > > /usr/lib/krb5/kprop_script slavekdc.example.com > > I keep getting the error: > > /usr/lib/krb5/kprop: Unknown code 146 in call to connect > while opening connection to slavekdc.example.com > > The slave system is able to change the password (using kadmin) for the > root principal on the master system, so it does appear that the setup > is correct otherwise. > > Both systems have the Feb/11/03 Solaris 9 Recommended Patch Cluster > installed and I am using the instructions from the Sun documents > (System Administration Guide: Security Services -- the problem is > at step 9 on page 99 which is the same as step 9 on the webpage > http://docs.sun.com/db/doc/806-4078/6jd6cjrul?a=view#setup-74). > > Any help would be appreciated, > Debbie -- + Debbie Tropiano -- debbie at icus.com -- http://www.icus.com/personal.html + | Mommy to Nathan b: 8/17/1995, ^Sara^ b: 10/25/2000 d: 11/7/2000 & | | Leah b: 10/17/2001 a: 9/26/2002 "God shows His opposition to cancer and | | birth defects, not by eliminating them or making them happen only to bad | | people (He can't do that), but by summoning forth friends and neighbors | + to ease the burden and to fill the emptiness." -- Harold S. Kushner + _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From meg991 at yahoo.com Mon Apr 7 23:12:29 2003 From: meg991 at yahoo.com (Meg Wall) Date: Mon, 7 Apr 2003 20:12:29 -0700 (PDT) Subject: Summary: Hardware failure? In-Reply-To: <10eef1644e.1644e10eef@west.gecems.com> Message-ID: <20030408031229.78478.qmail@web13501.mail.yahoo.com> Hi managers, Thanks for all of you replied, too many to list here! I've got all kind of suggestions, all of them are very informative, so I decide to call Sun. Sun's explaination is very similar to Colin Bigam's email, he's the winner! they did mention "replace the DIMM after 3 persistent memory error in 24 hours" rule. Thanks!! Meg --- Colin Bigam wrote: > Hi Meg; > > We see errors like this quite often on various systems, especially > on the newer sunfires. > > Memory errors come in three types: Intermittent, persistent, and > sticky. When the computer detects a single-bit memory error, it > will go and refetch the data from memory. If it's correct the second > time, then the error was transient, occurring randomly in the path > from memory to CPU. If it's still in error, then the error is > persistent--the system recalculates and rewrites the data back to > memory. Then it reads it again--if the read-after-rewrite is still > in error, then the error is labelled sticky. > > Sticky errors indicate bad memory, and should be replaced. Sun's > recommendation for persistent errors is to replace the DIMM if you > get more than three persistent errors in 24 hours, or if there's > a steady trend of increasing persistent errors. Transient errors > are almost completely random (usually caused by cosmic rays!), and > are not serious unless you start to get them steadily, in which case > you probably have a bad system board. > > There are also numerous patches which apply to memory errors. Make > sure you have a fairly recent patch cluster for your OS version on > the box, especially making sure that you have the 'memory scrubber' > patch. (you can search for this on sunsolve.sun.com) > > Hope this helps, > Colin > -- > | Colin Bigam, Senior Unix analyst > > ----- Original Message ----- > From: Meg Wall > Date: Monday, April 7, 2003 11:32 am > Subject: Hardware failure? > > > Hi managers, > > > > I just got the following messages, do I have a > > hardware failure here? What these mean? Thanks!! > > I will work on my summaries today. > > > > Meg > > > > Apr 7 12:12:02 server32 pcipsy: [ID 854591 kern.info] > > NOTICE: correctable error detected by pci0 (upa mid > > 1f) during > > Apr 7 12:12:02 server32 DVMA read transaction > > Apr 7 12:12:02 server32 pcipsy: [ID 750218 kern.info] > > AFSR=40230000.7f800000 AFAR=00000000.1c9d0a58, > > Apr 7 12:12:02 server32 double word offset=3, > > Memory Module U0701 id 31. > > Apr 7 12:12:02 server32 pcipsy: [ID 916270 kern.info] > > syndrome bits 23 > > Apr 7 12:12:02 server32 SUNW,UltraSPARC-II: [ID > > 354824 kern.info] [AFT0] errID 0x0003cc92.01cb251e > > Corrected Memory Error on U0701 is Intermittent > > Apr 7 12:12:02 server32 SUNW,UltraSPARC-II: [ID > > 376402 kern.info] [AFT0] errID 0x0003cc92.01cb251e ECC > > Data Bit 33 was in error and corrected > > Yahoo! Tax Center - File online, calculators, forms, and more > > http://tax.yahoo.com > > _______________________________________________ > > sunmanagers mailing list > > sunmanagers at sunmanagers.org > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers Yahoo! Tax Center - File online, calculators, forms, and more http://tax.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From meg991 at yahoo.com Mon Apr 7 23:29:13 2003 From: meg991 at yahoo.com (Meg Wall) Date: Mon, 7 Apr 2003 20:29:13 -0700 (PDT) Subject: Summary: Root cause of swap space full In-Reply-To: <20030404153929.12011.qmail@web13505.mail.yahoo.com> Message-ID: <20030408032913.45252.qmail@web13506.mail.yahoo.com> Hi Managers, I've got only 4 replies, I'm still not very sure what the root cause was. It's hard to do forensic analysis when the problem can't be repeated. (well, thank goodness). Thanks for all of you! Here's what I've got: From: Andrew_Cordova The only way I could see that this particular file system causing the crash would be if it had not been mounted. If you had say 4 disks, two active w/2 mirrors. /weblogs is mounted from active disk 2 and for some reason it is offline. what you have now is just the /weblogs mount point, w/ no disk mounted under it. The app is stupid, it just writes to where it is told, which is /weblogs, which in this scenario is just the mount point w/ no underlying disk. if /weblogs was full, then effectively / is full. just as if /home was not a separate file system and a user was to fill it up, then again, w/o quotas enabled, they have filled /. i have not seen this scenario w/ a fs created on rootdg, but have seen it when a datadg mount that we had lost the disk, the app just kept writting to /mountpoint till it ran us out of space. /tmp eventually runs out of space because it is effectively part of / and you have a down system, reboot clears /tmp - you login and clean space. an unusual scenario for sure, but as you know in this business the unusual is the norm. one other item to chk is that if / is still unusually full, then you may have /weblogs mounted over the data still written under /weblogs the mount point. From flegal at seo-toulouse.com Tue Apr 8 04:39:23 2003 From: flegal at seo-toulouse.com (=?iso-8859-1?Q?Fran=E7ois?= Legal) Date: Tue, 08 Apr 2003 10:39:23 +0200 Subject: Summary2 : Sendmail smarthost Message-ID: <3E928ABB.DE0EFF84@seo-toulouse.com> Thanks for all the answers, especially to Crist Clark who gave me many informations. The problem was that I set up my DNS MX record incorrectly. I just corrected the MX records in my DNS files and now ( almost ) everything is perfect. Thanks Frangois Legal _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bob at dtcc.edu Tue Apr 8 14:45:48 2003 From: bob at dtcc.edu (Bob Rahe) Date: Tue, 8 Apr 2003 14:45:48 EDT Subject: Summary: Sun Pci II and mouse (final) Message-ID: <200304081845.h38Ijm911361@hobbes.dtcc.edu> Well, thanks to Bryan Hodgson for some ideas which unfortunately didn't help the problem but he did warn me away from putting in any of the SiS driver updates into the XP machine(!). He was the ONLY person to comment. I opened a case with Sun since what was actually happening was that the mouse problem would only occur after openwindows had been running for some amount of time. So things might LOOK fixed but... Sun tried a couple of things including a patch (113616-01) which changed the problem. Now, it acts like it is missing mouse-ups. I.e. click on a close (x) box and until you move the mouse, even a tiny bit, it just stays 'pushed' but nothing happens. Sun tried some other stuff on that including changing the SunPCI card. Nothing seems to help. Interestingly, it DOESN'T happen when running under CDE. So we're going to leave it at that - the problem is pretty much innocuous and I can live with it until I convert to CDE eventually. Thanks. Bob Original message was: ----------------------------- I've installed a SUNpci II pro card in my Sunblade 1000 running Solaris 8, installed Windows XP and SP1 on it, upgraded the Sunpci software to 2.3.2 and run the update drivers script. I'm running it to a window on the SunBlade. The problem is the mouse. Clicks seem to act in ways that are non- obvious. For example, a single click (not click and hold, just a simple click) on the background acts like I'm clicking and holding. I.e. if I click once and then move the mouse it starts making rectangles. Similarly if I click on something (single) and move the mouse it starts to move the object around. Various other bizarre behavior in trying to get single clicks on some buttons to do anything and double clicks seem to require furious clicking to get to work. I've checked various places in the Windows control panels that might actually CAUSE this - there's an option that actually does that - sticky clicks or something - but it's turned off. I even tried turning it on and it oddly made NO difference.... Also, once the pc is booted up to windows the 'attach mouse' key/menu option seems to have no effect. As in nothing happens or changes when you select it or hit the meta-M. Ideas anyone? Thanks Bob -------------------------------------- -- ---------------------------------------------------------------------------- |Bob Rahe, Delaware Tech&Comm Coll. / | |Computer Center, Dover, Delaware / | |Internet: bob at dtcc.edu (RWR50) / | ---------------------------------------------------------------------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Warchol at harthosp.org Tue Apr 8 15:34:16 2003 From: Warchol at harthosp.org (Dave Warchol) Date: Tue, 08 Apr 2003 15:34:16 -0400 Subject: SUMMARY: Snap Clone HowTo Message-ID: Thanks to Adam Zimmerman and Bertrand Hutin for responding: From gtobares at redmegatone.com Tue Apr 8 15:45:07 2003 From: gtobares at redmegatone.com (Gustavo Tobares) Date: Tue, 8 Apr 2003 16:45:07 -0300 Subject: SUMMARY: tracking mail sending Message-ID: Thanks to Len Smith, They suggest to add the next line (or similar) to the syslog.conf mail.debug /var/log/mail (Make sure to use tabs between the two columns) Thank's a lot. Gustavo Tobares Administrador de Sistemas y DBA Centro de Computos - Red Megatone TE: 0342-4500972 - Fax: 0342-4500940 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From vaxzilla at jarai.org Tue Apr 8 16:21:20 2003 From: vaxzilla at jarai.org (Brian Chase) Date: Tue, 8 Apr 2003 13:21:20 -0700 (PST) Subject: SUMMARY: Problems with NFS automount of Linux filesystem under Solaris 8. In-Reply-To: Message-ID: On Tue, 8 Apr 2003, Brian Chase wrote: > I've an odd problem on a Solaris 8 client of mine. Its unable to > automount an NFS filesystem served off of a Linux system. The same > system doesn't have problems automounting any of the other filesystems > being served off of our SGI and NetApp file servers. More puzzling > still, I'm able to manually NFS mount the filesystem with the `mount' > command. > > The automount master file entry looks like this: > > /net /etc/auto.map -intr > > The relevant /etc/auto.map file entry looks like this: > > vol606 -retry=5000 linuxnfs:/mnt/vol606 > > This should automount the Linux server's filesystem under /net/vol606 on > the Solaris client system. When I try to `cd /net/vol606', the command > hangs for a very long time. The one error message I get is the > following: > > Apr 8 12:00:58 client1 nfs: [ID 333984 kern.notice] NFS server linuxnfs > not responding still trying > > However, if I run a `mount linuxfs:/mnt/vol606 /mnt2' on the same > Solaris client, that command succeeds and I can access the contents of > the filesystem just fine. > > Any ideas? Thanks go to Hendrik Visage, Wolf Schaefer and Roger Leonard for pointing me in the correct direction on this one. The Linux NFS server seems to be falsely advertising the NFS protocols it supports: # rpcinfo -p madmax | grep nfs 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs I think the automounter tries using NFSv3 tcp and I guess the manual mount command is smart enough to fall back on (or it defaults to) one of the udp versions. I found that if I tried to manually mount the filesystem with either version of NFS with `-o proto=tcp', the Linux server was unable to handle the request. I added a `proto=udp' option to the relevant entry in my auto.map file, and now it works. In hindsight, I should've realized it was something like this. I used to have a very similar problem with using Linux as an NFS server for diskless NetBSD/vax clients a number of years ago. It hadn't occurred to me that Linux would /still/ be broken in this way. Thanks for all the quick replies. -brian. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From alejandro.vega at tecinfocon.com.mx Tue Apr 8 17:55:56 2003 From: alejandro.vega at tecinfocon.com.mx (Angel Alejandro Vega Soto) Date: 08 Apr 2003 15:55:56 -0600 Subject: SUMMARY: [Fwd: E450 URGENT Assistance! not powering up] Message-ID: <1049838956.5415.18.camel@apolo.tic.com.mx> Geeez the people of this list are awesome, thanks to: Matthew Stier Sean Berry Jayashankar Nelamane Srinivasarao Colin Bigam 4jet at overnite.com Specially to: Pascal Grostabussiat who was kind enough for a more detail schematic help The Answer in brief by Greg Hanks: The cover in the rear engages two different micro switches. These switches are very delicate, but disable the power. Look carefully at them and ensure that the little plastic points on the cover engage the switches. If in question, unscrew them through the access holes on the side rails, remove the micro switches and ensure that the actuateing levers are in the proper location. These bend easily so be careful. I Think is a tipicall murphys law case: A switch of 50 cents. make a 15k dollar machine not to work. Thanks again to all of you ! The original post is below ! Return-Path: Delivered-To: tecinf01 at localhost Received: (qmail 13487 invoked by alias); 8 Apr 2003 14:19:39 -0000 Delivered-To: alias-tecinf-alejandro.vega at tecinfocon.com.mx Received: (qmail 13484 invoked from network); 8 Apr 2003 14:19:39 -0000 Received: from sunmgrs.cs.toronto.edu (128.100.27.219) by ipfarm03.uservers.net with SMTP; 8 Apr 2003 14:19:39 -0000 Received: from sunmgrs.cs (localhost [127.0.0.1]) by sunmgrs.cs.toronto.edu (Postfix) with ESMTP id 716492086D; Tue, 8 Apr 2003 10:17:55 -0400 (EDT) Delivered-To: sunmanagers at sunmanagers.org Received: from SMTP-01.X0L.uSERVERS.NET (unknown [148.245.15.3]) by sunmgrs.cs.toronto.edu (Postfix) with SMTP id 265C920862 for ; Tue, 8 Apr 2003 10:16:52 -0400 (EDT) Received: (qmail 17116 invoked from network); 8 Apr 2003 14:09:49 -0000 Received: from unknown (HELO apolo.tic.com.mx) (200.67.175.161) by smtp.x0l.uservers.net with SMTP; 8 Apr 2003 14:09:49 -0000 Subject: E450 URGENT Assistance! not powering up From: Angel Alejandro Vega Soto To: Sunmanagers List Content-Type: text/plain Organization: Tecnologia Informatica y Consultoria Message-Id: <1049815016.2392.14.camel at apolo.tic.com.mx> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.4 Sender: sunmanagers-admin at sunmanagers.org Errors-To: sunmanagers-admin at sunmanagers.org X-BeenThere: sunmanagers at sunmanagers.org X-Mailman-Version: 2.0.10 Precedence: bulk List-Unsubscribe: , List-Id: The Sun Managers Mailing List List-Post: List-Help: List-Subscribe: , List-Archive: X-Original-Date: 08 Apr 2003 09:16:56 -0600 Date: 08 Apr 2003 09:16:56 -0600 Content-Transfer-Encoding: 7bit HEllo managers, I have a very strange case. I recently installed two more processors with their correspondig DCDC on a E450. The power on the machine with no problem , same day 1 hour later I was asked to replace the existing DDS3 tape reader which was causing trouble with a new DDS3. I replace the DDS3 but now the machine doesnt power up, or power on. I reassembly everithing in place, all covers, every screw but the machine doesnt power up. The power plug is connected, the key switch is to off the to on and nothing, also connected the keyboard and repeat the steps and press the power button, and nothing not even the fans start. ??? WHAT IS IT ?? i remove the new DCDC and the CPUS, nothing. I reinstall them, the same, remove cards, disks, but nothing. I check docs.sun.com at http://docs.sun.com/db/doc/805-0429-10/6j0fk0g44?q=power+supply&a=view The only thing that Ive notice is that the 3er led (from left to right) on the power supply, the DC status is always off, anyone has ever seen anithing like this. The machine warranty time is over I ask sun for a time and material report but are very slow, any ideas ? how can I identify the failure component ? I WILL SUMMARIZE ! Thanks ! -- Angel Alejandro Vega Soto Tecnologia Informatica y Consultoria _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From osamaomar at link.net Thu Apr 3 06:40:40 2003 From: osamaomar at link.net (Osama Omar) Date: Thu, 3 Apr 2003 13:40:40 +0200 Subject: Summary RE: Mail on Solaris 8 In-Reply-To: <7F09903EF5CBD211B5A20008C739C02C66F8AB@GVMAIL01> Message-ID: <000c01c2f9d5$dc4784a0$c79e939d@onshore.pgs.com> RE: Mail on Solaris 8Hi Many thanks to all who replied I have tried this way and gave me what I need Osama Omar Ali PGS Data Processing Middle East IT Dept. (002)(02) 2761353 (002)(02) 2761353 (002) (010) 6008263 -----Original Message----- From: Vincent Ribeaud [mailto:Vincent_Ribeaud at adp.com] Sent: Wednesday, April 02, 2003 11:18 AM To: 'osamaomar at link.net' Subject: RE: Mail on Solaris 8 Hi, In my /etc/hosts file I have the following entry: 172.xx.xx.xx mailhost.mydomain.com mailhost Vincent -----Original Message----- From: Osama Omar [mailto:osamaomar at link.net] Sent: Wednesday, April 02, 2003 11:17 AM To: 'Sunmanagers (E-mail)' Subject: Mail on Solaris 8 Hi How can I configure my Solaris 8 dns server to send mails to internet mail accounts using the "mail" command from its clients ? eg: # mail osamaomar at link.net test ^D how can this reach my mail account while I dont have a unix mail server but want the Unix DNS to point to an exchange server? Regards Osama Omar Ali PGS Data Processing Middle East IT Dept. (002)(02) 2761353 (002)(02) 2761353 (002) (010) 6008263 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers This message is confidential and may also be legally privileged. If you are not the intended recipient, please notify postmaster at adpwilco.com immediately. You should not copy it or use it for any purpose, nor disclose its contents to any other person. The views and opinions expressed in this e-mail message are the author's own and may not reflect the views and opinions of ADP Wilco. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Karl.Rossing at Federated.CA Fri Apr 4 14:08:37 2003 From: Karl.Rossing at Federated.CA (Karl.Rossing at Federated.CA) Date: Fri, 4 Apr 2003 13:08:37 -0600 Subject: Summary Re: LDAP password aging In-Reply-To: Message-ID: Thanks to Zaigui Wang for pointing me in the right direction. I ended up placing a service call with SUN. There will be two options in the future that fix this problem. Both options require SUN ONE Directory server 5.1 1) Sun is testing patch. The patch number is T108993-16 and it's a backport of the Solaris 9 LDAP client. Which will support passwd aging. 2) Sun is testing and might be releasing a windows password sync program for DS 5.1. This might work for me. Hope this helps some of you out. sunmanagers-admin at sunmanagers.org wrote on 04/03/2003 01:55:54 PM: > As i understand it, there is no password aging functionality in LDAP. > > All our users usually make shell connections to our Solaris box. I'm > wondering if there are any perl/shell scripts to: > 1) check when a users passwd was last changed. I'm not sure how to query > this in ldap or if it's possible. > 2) force the user to change their passwd is older the x days. Something > simple would be via /etc/profile. If /export/home/username/.passwd-changed > is older than x days, force the user the passwd prompt or else continue > with the login script. > > Any thoughts? > > Thanks > Karl > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From anshuman at expertcity.com Fri Apr 4 16:06:03 2003 From: anshuman at expertcity.com (Anshuman Kanwar) Date: Fri, 4 Apr 2003 13:06:03 -0800 Subject: Resolved: RE: 555 on /home : summary Message-ID: Super ! Leonard, Roger [Roger.Leonard at marconi.com] && Andres Martinez [andres.martinez at tecno.com.mx] && Gurugunti, Mahesh [MGurugunti at us.nomura.com] && Crist Clark [crist.clark at globalstar.com] && Casper Dik [Casper.Dik at Sun.COM] && Michael H. Auria [michael.auria at philips.com]&& unico [unico at benedetto.homelinux.org] Many thanks for the quick solution, that was the problem. #------summary-- Edit /etc/auto_master and comment out : # /home auto_home -nobrowse /etc/init.d/autofs stop /etc/init.d/autofs start #--------------- Thomas Autry [tautry at jcpenney.com] && Michael Peppard [mpeppard at impole.com] Thank you too. :) /home is just a directory under / not a fs. NFS not running. Regards, Anshuman Kanwar Unix SysAdmin Expertcity Inc. -- (805) 690-5714 [off] ansh at expertcity.com (805) 895-4231 [cel] 5385 Hollister Ave (805) 690-6471 [fax] Goleta, CA. 93111 > -----Original Message----- > From: Crist Clark [mailto:crist.clark at globalstar.com] > Sent: Friday, April 04, 2003 12:39 PM > To: Anshuman Kanwar > Subject: Re: 555 on /home > > > Anshuman Kanwar wrote: > > > > Hi everyone, > > > > One dumb question: > > > > bash-2.05# uname -a > > SunOS v880-1 5.9 Generic_112233-04 sun4u sparc SUNW,Sun-Fire-880 > > bash-2.05# ls -ld /home > > dr-xr-xr-x 1 root root 1 Apr 1 15:49 /home > > bash-2.05# id > > uid=0(root) gid=1(other) > > bash-2.05# mkdir /home/log > > mkdir: Failed to make directory "/home/log"; Operation not > applicable > > bash-2.05# chmod 755 /home > > chmod: WARNING: can't change /home > > > > How do I undo the 555 permissions on /home ? > > You're running the automounter, aren't you? You can't modify > /home while > it's running. > -- > Crist J. Clark > crist.clark at globalstar.com > Globalstar Communications > (408) 933-4387 > > The information contained in this e-mail message is confidential, > intended only for the use of the individual or entity named above. > If the reader of this e-mail is not the intended recipient, or the > employee or agent responsible to deliver it to the intended recipient, > you are hereby notified that any review, dissemination, > distribution or > copying of this communication is strictly prohibited. If you have > received this e-mail in error, please contact > postmaster at globalstar.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Anand.Varada at consulting.fujitsu.com Tue Apr 8 17:17:51 2003 From: Anand.Varada at consulting.fujitsu.com (Anand.Varada at consulting.fujitsu.com) Date: Tue, 8 Apr 2003 14:17:51 -0700 Subject: SUMMARY- Problems Installing PatchPro on Solaris 8 - SUMMARY Message-ID: Thanks to all replied.. fastest responses.. amazing : Peter Ondruska ( Could able to complete my todays task.. successfully configured..Thanks ) You need to add SUNW certificates before starting pprosetup or smpatch. Details are at http://sunsolve.sun.com/patches/spag.html How to Import Sun Certificates into the Keystore :GertJan ( Could able to do immediate dianostics on failures... thanks) try to truss the command to see what it's doing (it may try to write to an error log elsewhere). ----- Original Message ----- From: To: Sent: Tuesday, April 08, 2003 10:16 PM Subject: Problems Installing PatchPro on Solaris 8 > Hello gurus, > > Trying to install patchpro on solaris 8. Installation was succeeded without > any errors. When trying to run the command "pprosetup" ..it fails with > below error. > > # /opt/SUNWppro/bin/pprosetup -H > > Problem detected during PatchPro initialization process. Please check the > log file. Exiting. > > NOTE: No errors found from log file. > > Somebody... please throw some light on this.. will summarize. > Best Regards, > Anand Varada _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From fwilde at chethams.org.uk Wed Apr 9 05:50:18 2003 From: fwilde at chethams.org.uk (Fergus Wilde) Date: Wed, 9 Apr 2003 10:50:18 +0100 Subject: SUMMARY: Ultra5 Sunvts gives disturbing error without apparent problems In-Reply-To: <3e8d5185.773.0@btconnect.com> References: <3e8d5185.773.0@btconnect.com> Message-ID: <200304091050.18084.fwilde@chethams.org.uk> Thanks to all those who wrote with helpful suggestions, Justin Stringfellow at Sun cracked it by notifying me that there was a known issue with VTS generating memory failures on Solaris 8 with certain patchlevels: "When running SunVTS 4.1, 5.1 with Solaris 8 and kernel update patch 108528-16, you will see error messages of: FATAL mem: "Unable to read pointer to memlist structure." Problem Solution Problem is due to the fact that the symbol "max_ce_err" was deleted from the kernel with the 108528-16 patch. Workaround: Install patch 113689 from http://sunsolve.sun.com Run with kernel patch 108528-15 Do not run pmemtest with 108528-16 or later." I solved the problem by 1) updating to SUNWvts 5.1 and its associated packages 2) Installing patch 1113689-10 This summary was a little delayed while I registered with sunsolve to get this non-public patch. I suppose if you don't have a support contract you would be obliged either not to run pmemtest (still leaves another memory test available) or backout to a a kernel patchlevel < 108528-16 Thanks also to Joe Fletcher for reminding me that you need to do cat /etc/release in order to get the version number, e.g: $ cat /etc/release Solaris 8 10/01 s28s_u6wos_08a SPARC Copyright 2001 Sun Microsystems, Inc. All Rights Reserved. Assembled 12 September 2001 and to David M. Dennis for helpful input about RAM. Please forgive me if I've missed anyone out, I've been collecting emails on a succession of different machines. My original query: In the spirit of hope triumphing over experience, I just wanted to ask if anyone can see a reason not to call the Sun Doctor (the machine is on Silver Service) after a stop a few seconds into a run of SunVtS with this error: 04/03/03 15:36:05 server SunVTS4.5: VTSID 8004 pmemtest. FATAL mem: "Unable to read pointer to memlist structure." uname -a# SunOS server 5.8 Generic_108528-19 sun4u sparc SUNW,Ultra-5_10 SunVTS version 4.5. Sun's own docs do seem to tell you to call them out, but I was just hoping someone knows a reason why the error might be in ... er ... error. Machine has up-to-date OBP, and is giving no signs of trouble, logs are void of doom-laden messages. However, since it's done very little work recently except fit its own recommended patch set, am I inviting trouble by doing nothing if I should need it soon? Fergus -- Fergus Wilde Chetham's Library Long Millgate Manchester M3 1SB Tel: +44 161 834 7961 Fax: +44 161 839 5797 http://www.chethams.org.uk _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From dario.principe at iol.it Wed Apr 9 07:30:13 2003 From: dario.principe at iol.it (=?iso-8859-1?Q?dario.principe@iol.it?=) Date: Wed, 9 Apr 2003 13:30:13 +0200 Subject: =?iso-8859-1?Q?SUMMARY:_Restricted_Shell_Question?= Message-ID: First of all, thanks to each of you who gave me suggestions: JULIAN, JHON C Andrew J Caines Alan Pae Wolf Schaefer Suresh Vuthamaraju Tom Husemann, Harald And thanks to the other who will respond in the next few hours. The things that will help me is using PATH variable inside .profile, and a good idea is using a script to select which log to view. Most of the suggestions was to use sudo.... but I cannot do that (policy questions....) A great idea we intended to apply was to use a web server, but again I cannot to that.... Thank you very much, Dario _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bodenste at mailcenter.csap.af.mil Wed Apr 9 10:01:11 2003 From: bodenste at mailcenter.csap.af.mil (Cathy M. Bodenstedt) Date: Wed, 09 Apr 2003 09:01:11 -0500 Subject: SUMMARY:script Message-ID: <3E9427A7.268E8DD2@mailcenter.csap.af.mil> Thanks to all that responded. I apologize if I missed a few names. Mike D. Kail Steve Chapman Rob McCauley Angel Alejandro Vega Soto Stephen Barnett Soeren Schaper Alan Bradley Shawn Evans Joel Fisher Bryan Moore cat install_log |grep "clone at 0" |grep link | awk -F: '{ print $2 }' grep "^add link .*/clone at 0" install_log | awk -F: '{print $2}' more install_log |grep "clone at 0" |grep link | cut -d: -f2 But this what I was really looking for, Thanks to Joel. Joel Fisher wrote: > Hey Cathy, > > I would think the better bet would be to do the below. > > $ grep network /etc/path_to_inst > "/pci at 1f,4000/network at 1,1" 0 "hme" > "/pci at 1f,2000/network at 1" 0 "ge" > > That will also give you in instance number. > -- Phone: 210-925-3244 DSN: 945-3244 [demime 0.99c.7 removed an attachment of type text/x-vcard which had a name of bodenste.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From gtobares at redmegatone.com Wed Apr 9 11:38:16 2003 From: gtobares at redmegatone.com (Gustavo Tobares) Date: Wed, 9 Apr 2003 12:38:16 -0300 Subject: SUMMARY: tracking mail sending Message-ID: Thank's to Crist Clark and Jay Lesser They give me two way to solve the problem a. Activate the mail.debug in syslog.conf b. See the properties of the e-mail received (outlook) Both options solve my problem. Thank's again. Gustavo Tobares Administrador de Sistemas y DBA Centro de Computos - Red Megatone TE: 0342-4500972 - Fax: 0342-4500940 >Gustavo Tobares wrote: > > Hi all, > > Im using SOLARIS 8 with sendmail, relaying outgoing mails to > exchange over NT 4. The mails sended from the solaris are in the form > > username at santafe.megatone > > One user named "hello" (hello at redmegatone.com) tell me with has > received one e-mail from pruebas at santafe.megatone in you (exchange) outlook > inbox. > > The user pruebas (pruebas at redmegatone.com) on the SOLARIS say with > her not send e-mails to any user. > > They are a way to know who send the e-mail from the SOLARIS using > pruebas at santafe.megatone directed to hello? > > They are logs in the Solaris sendmail system? how I can active it? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Rick.Brashear at ercgroup.com Wed Apr 9 11:48:42 2003 From: Rick.Brashear at ercgroup.com (Rick.Brashear at ercgroup.com) Date: Wed, 9 Apr 2003 10:48:42 -0500 Subject: SUMMARY: /dev/random for Solaris 8 Message-ID: <2857747EB2E7D311AC1A00508B6F9C7C0B1FC40D@kans020cercge.erc.gecapital.com> Thanks for the overwhelming response! Great list! The consensus was install patch 112438-01 and reboot. Other suggestions: * SUNWski on the solaris easy access cds * andirand (use google) * prngd (use google) Download it from Andreas Maier site: http://www.cosy.sbg.ac.at/~andi/SUNrand/ Thanks (in order received) to: Steven Haywood Gary Chambers Eric Trager Randy Romero Gavin Harris Chistopher L. Barnard Cathy M. Bodenstedt Mario Williams Rob De Langhe Casper Dik Lonnie Ratliff Mike Salehi Yakov Lerner Hichael Morton Sean Timmins William Yodlowsky Eric Forgette Tim Thomas Brian Pardy Celso G. Lima Kendall Lloyd Luc I. Suryo Preiszl Matthias Suresh Vuthamaraju Lars Hecking and also to those that are still reponding.... g ERC _______________________________ Rick Brashear Server Systems Information Technology Department Employers Reinsurance Corporation 5200 Metcalf Overland Park, Kansas 66201 * 913 676-6418 * rick.brashear at ercgroup.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnrams at cox.net Wed Apr 9 12:09:33 2003 From: johnrams at cox.net (johnrams at cox.net) Date: Wed, 9 Apr 2003 12:09:33 -0400 Subject: Summary: Sendmail Queue mode only Message-ID: <20030409160933.MKON22214.fed1mtao02.cox.net@smtp.west.cox.net> Thanks to Robert Tseng, David Meissner, Randy Chatfield, Ric Anderson, Crist Clark, Mark, Jeff Horwitz, Mario Henley, Israel Novelo Zel, William Enestvedt, Angel Rivera, Christopher Barnard. Unanimous conclusion is to use following approach. http://www.sendmail.org/~ca/email/doc/op-sh-4.html link is good source. In Solaris 8, add MODE="" in /etc/default/sendmail (create this file if it does not exist already) In Solaris 2.6, your have three choices: a. borrow /etc/init.d/sendmail from Solaris 8 b. directly edit /etc/init.d/sendmail to remvoe -bd c. remove/rename /etc/rc2.d/S88sendmail, then add an entry in root's crontab: 0,15,30,45 * * * * /usr/lib/sendmail -q (or any interval of your choice) After you're done above modification, bounce sendmail /etc/init.d/sendmail stop /etc/init.d/sendmail start Hi: If incoming mail is not to be allowed, how would i force sendmail to run in queue mode only. > > What would i do to the startup script? Turning off '-bd' flag is sufficient and restart. > > Thanks > John _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From pete_tukfxd at aol.com Wed Apr 9 06:09:24 2003 From: pete_tukfxd at aol.com (Peter Turner) Date: Wed, 09 Apr 2003 22:09:24 +1200 Subject: Want to meet up soon?.... 3285AKAG7-290CgBz835-19 Message-ID: <001101d6cd67$dee67565$20758104@bmmgfcr.ims> An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030409/eb5cc3f3/attachment.html From ellipses at society86.com Wed Apr 9 23:33:18 2003 From: ellipses at society86.com (ellipses) Date: Wed, 9 Apr 2003 22:33:18 -0500 Subject: SUMMARY:DNS Wierdness Message-ID: <004601c2ff12$079a6b90$01a8a8c0@rancid3> Thanks for all of the replies, I got a million it seems, I learned some new things I wasn't aware of before. I was misinformed. Here's my scenario to shed some more light on the subject. The Solaris machine is performing monitoring services for the internal and external networks. It has an internal address that has been given NAT capabilities through the firewall. The internal DNS boxen are Win2k Active Directory machines. The internal network is dominated by Win2k, just one Solaris box, and that's me. The external servers are Sprints public DNS machines. I plan to install two Solaris Boxes with Bind 9 when I get the appropriate hardware to place on our external network. Some issues I have. A lot of the replies said to have the internal forward or cache, I want them to but the policy is of right now that they will not. The current management wants them to only do internal DNS, the only external addresses are fed to clients through the proxy. So until that is settled the stuff wont forward or anything. The external DNS that we use know nothing of the internal domain. So I have to make use of bother sets of servers to resolve on both sides. With the restraints I have been given anyways. Some more weirdness... I went today and implemented the resolv.conf as I had sent it in the original post (I inly had internal servers in it), the weird part is that it works! I'm not aware of anything changing ,so... it just got weirder... FYI, my nsswitch.conf is set to "files dns" so that wasn't the problem. I am in agreement that my problem lies with the fact that so many pointed out the the DNS servers are giving me a "host not found" and not failing over. That sounds logical. Also I probably should fix my broke resolv.conf since it is not syntactically correct. (DOH!) But, I'm gonna leave it for a while and see what happens... weird... Thanks again for the many replies, and for fun I keep everyone up to date on this twisted tale... Thanks!! PS: Sorry about the book I just wrote... :/ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zoo11 at hst.co.kr Thu Apr 10 01:03:45 2003 From: zoo11 at hst.co.kr (Joohyun Cha) Date: Thu, 10 Apr 2003 14:03:45 +0900 Subject: SUMMARY: Blade 100 with firewire HDD Message-ID: <20030410140345.A11129@mail.hst.co.kr> Hi. managers. Thanks for your replies. Original question is below. It seems that Blade 100 can't be configured with other firewire devices except Sun's cam. Hichael Morton said that "if there is no mention in the user guide or on sun website, it isn't implemented." Thank you again. Original Posting: Hi. I'm trying to use fireware port (IEEE 1394) HDD. How can I do this? I couldn't find any reference in my blade 100 user's guide. Thanks in advance. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Nils.Wuerfel at TU-Ilmenau.de Thu Apr 10 04:00:37 2003 From: Nils.Wuerfel at TU-Ilmenau.de (Nils Wuerfel) Date: Thu, 10 Apr 2003 10:00:37 +0200 Subject: SUMMARY(2):FIRE280+A1000 needs too much boot time Message-ID: <3E9524A5.5090409@TU-Ilmenau.de> Problem solved. Many thaks for your suggestions. Please excuse my late Summary, I'm often out of office this time... I commented out some lines in /kernel/drv/rdriver.conf, which were automatically created by RaidManager 6.22 installation. See example below. My A1000 only uses target=0 / LUN 0. It seems that the box waits for timeouts on all devices(?) which are not really exist. regards NILS EXAMPLE: {root}$ cat rdriver.conf name="rdriver" target=1 parent="/pseudo/rdnexus at 0" probe=1 qtime=5; name="rdriver" module=1 lun=0 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=1 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=2 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=3 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=4 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=5 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=6 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" module=1 lun=7 target=0 parent="/pseudo/rdnexus at 1" mname="cl101_003" nexusbus=1 path="A"; name="rdriver" parent="rdnexus" target=0 lun=0; #name="rdriver" parent="rdnexus" target=0 lun=1; #name="rdriver" parent="rdnexus" target=0 lun=2; #name="rdriver" parent="rdnexus" target=0 lun=3; #name="rdriver" parent="rdnexus" target=0 lun=4; #name="rdriver" parent="rdnexus" target=0 lun=5; #name="rdriver" parent="rdnexus" target=0 lun=6; #name="rdriver" parent="rdnexus" target=0 lun=7; #name="rdriver" parent="rdnexus" target=4 lun=0; #name="rdriver" parent="rdnexus" target=4 lun=1; #name="rdriver" parent="rdnexus" target=4 lun=2; #name="rdriver" parent="rdnexus" target=4 lun=3; #name="rdriver" parent="rdnexus" target=4 lun=4; #name="rdriver" parent="rdnexus" target=4 lun=5; #name="rdriver" parent="rdnexus" target=4 lun=6; #name="rdriver" parent="rdnexus" target=4 lun=7; #name="rdriver" parent="rdnexus" target=5 lun=0; #name="rdriver" parent="rdnexus" target=5 lun=1; #name="rdriver" parent="rdnexus" target=5 lun=2; #name="rdriver" parent="rdnexus" target=5 lun=3; #name="rdriver" parent="rdnexus" target=5 lun=4; #name="rdriver" parent="rdnexus" target=5 lun=5; #name="rdriver" parent="rdnexus" target=5 lun=6; #name="rdriver" parent="rdnexus" target=5 lun=7; _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From harald.husemann at Materna.DE Thu Apr 10 07:49:01 2003 From: harald.husemann at Materna.DE (Husemann, Harald) Date: Thu, 10 Apr 2003 13:49:01 +0200 Subject: Summary: Using OnStream ADR in a Blade100 Message-ID: <73B855B4F72FD411A1CF009027B113C404C867AA@taxis.materna.de> Hi agn, got only one response so far, from Hichael Morton, who pointed me to sun's hcl. Thanks, I'll check it out! Hichael also gave me the hint to put the ADR on a seperate IDE-chain, and (of course) do a reconfig boot after adding the drive. Hm, seems no-one did this before, so I'm again a pioneer, .-) No problem, I'll try it - and inform the list abt. success or failure. Thanks to Hichael, have a nice hackin', Harald ============================================ Harald Husemann Systems Engineer Teammanager Unix administration and Configuration Management Materna Gmbh - Vo_kuhle 37 - D-44141 Dortmund, Germany Phone: +49-231-5599-8684 Mobile: +49-179-2300651 -----Urspr|ngliche Nachricht----- Von: Husemann, Harald [mailto:harald.husemann at Materna.DE] Gesendet: Mittwoch, 9. April 2003 15:55 An: sunmanagers at sunmanagers.org Betreff: Using OnStream ADR in a Blade100 Hi managers, I have to backup some data, and have an ONSTREAM ADR 30 (30 GB, IDE-Interface) laying around... I'd like to use it in a SUN Blade 100 (running Solaris 8), since the SUN has an IDE-Interface, it should work. But, before I disassemble my Blade: Anyone made experiences with this? The ADR runs perfectly well under LINUX , but I didn't find anything abt. using the drive with Solaris... Thanks in advance, will summarize, Harald ============================================ Harald Husemann Systems Engineer Teammanager Unix administration and Configuration Management Materna Gmbh - Vo_kuhle 37 - D-44141 Dortmund, Germany Phone: +49-231-5599-8684 Mobile: +49-179-2300651 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From George.Monappallil at dialog.com Thu Apr 10 11:55:01 2003 From: George.Monappallil at dialog.com (Monappallil, George) Date: Thu, 10 Apr 2003 11:55:01 -0400 Subject: Veritas VM docs: Summary Message-ID: <75014BFE3D0905438C4B6EF4C49E5FC6CA372B@usbumail2> Thanks for all of those who replied. Most pointed me to support.veritas.com. The following sites were also suggested http://www.filibeto.org/sun/lib http://www.cuddletech.com/veritas/ However, these sites dwell mostly on how to navigate VVM through the vmsa gui. Is there a site that has a command line doc? My machine is not allowed to have a gui, besides I prefer typing the commands in. Thanks in advance. -GM _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From dario.principe at iol.it Thu Apr 10 14:58:22 2003 From: dario.principe at iol.it (=?iso-8859-1?Q?dario.principe@iol.it?=) Date: Thu, 10 Apr 2003 20:58:22 +0200 Subject: =?iso-8859-1?Q?SUMMARY:_LTO_Ultrium_Question?= Message-ID: Too many people to thanks for replying to my post. Thanks to all of you. It seems that most of you will use Xcbn (I need b for Bsd compatibility as suggested by Legato) and this is what I did up till now and what I will do with Ultrium. Regards, Dario _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From SunAdmi at netscape.net Thu Apr 10 19:08:17 2003 From: SunAdmi at netscape.net (SunAdmi at netscape.net) Date: Thu, 10 Apr 2003 19:08:17 -0400 Subject: SUMMARY: Outgoing packet errors Message-ID: <4987DB94.7FCD5325.001A96BD@netscape.net> Got an answer that makes sense. Thanks to Jay Lessert and Frank Smith. The problem was with my own understanding of netstat -i. I now understand that it shows data since the last boot. (yes, I should have known this. I also should have noticed the errors and collisions were not growing) The number of errors and collisions has NOT increased since the system was set on the fly to 100fdx and has not been rebooted since. etc/system has the correct entries to retain after a reboot. These errors are left over from when the system was still at 10hdx. Thank you. Original message... Hello managers I am having a problem on one of my servers. It is a SUN Ultra-60, running Solaris 8 2/02 revision -17. The problem is this... The output from netstat is showing a high number of Oerrs. and collisions. As I am rather new to system administration I am not sure where to begin looking for a problem. Any suggestions would be appreciated, especially helpful would be a reference or link to a troubleshooting guide of some kind. Thank you in advance. Output from netstat... Name Mtu Net/Dest Address Ipkts Ierrs Opkts Oerrs Collis Queue lo0 8232 loopback localhost 748 0 748 0 0 0 hme0 1500 randomhost1 randomhost1 413758315 0 227619382 329543 609137 0 Forgot to add that both the nic and the switch are set at 100fdx. __________________________________________________________________ Try AOL and get 1045 hours FREE for 45 days! http://free.aol.com/tryaolfree/index.adp?375380 Get AOL Instant Messenger 5.1 for FREE! Download Now! http://aim.aol.com/aimnew/Aim/register.adp?promo=380455 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zoo11 at hst.co.kr Thu Apr 10 20:56:07 2003 From: zoo11 at hst.co.kr (Joohyun Cha) Date: Fri, 11 Apr 2003 09:56:07 +0900 Subject: SUMMARY: root disk clonning with dd Message-ID: <20030411095607.A18074@mail.hst.co.kr> Hi. Managers. Thanks to all your responses from Steve Maher Frangois Legal Yakov Lerner Espen Martinsen Hargrave, Mark Charles Gagnon Homan, Charles McCaffity, Ray Wingfield, John Hichael Morton Almost every managers said that it's normal if I dd with unmounted slice. I should first boot from cdrom and use dd command then there are no errors. but in my case (OS disk clonning with live system) I just can't shut it down and boot from cdrom so when I need to use clone disk by dd or ufsdump command I should do sync several times before excuting command and make a system quiet as possible as I can. And also again I should do fsck. Espen Martinsen and Homan, Charles pointed me a good resource for this purpose. http://pogostick.net/~esm/ftp/ESMclon.bin http://www.samag.com/documents/s=1148/sam0107j/0107j.htm. Thanks againg!! even if my english is so poor. Alwaye be well. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From flegal at seo-toulouse.com Fri Apr 11 07:52:28 2003 From: flegal at seo-toulouse.com (=?iso-8859-1?Q?Fran=E7ois?= Legal) Date: Fri, 11 Apr 2003 13:52:28 +0200 Subject: Summary : Unable to compile after patch cluster Message-ID: <3E96AC7C.BEF219AA@seo-toulouse.com> Hi all, and thanks for the answers. I fixed the problem by myself, using some informations from SOLDC forums. I removed the file /usr/ccs/bin/ad, and gcc is working again. -- Frangois Legal Hi Francois I had problems on 2 servers after applying the March rec patch cluster on a 2.6 platform. The problem for me was that patch 105693-13 overwrote /etc/inet/inetd.conf which made some of our site-specific services. It did not save an old copy, and there were no warnings or special installation instructions. Sun have now logged this issue as BugID # 4843569. If you had made changes to inetd.conf, then I suggest you restore the file from a backup taken before you applied the patches. If this fixes your problem, please post my reply in your summary, so others in the same position may benefit. Good luck Dennis Martens Unix Administrator, BCISU Queensland Health , Brisbane Australia 617 32340729 Fax 32341289 >>> Frangois Legal 8/04/03 19:28:31 >>> Hi all. I'm running Solaris 2.6 with GCC 3.1 Since I installed the last patch cluster from sun, I am unable to compile. I get the output : collect2: ld terminated with signal 9 [Killed], core dumped ld.so.1: /usr/ccs/bin/ad: fatal: libld.so.2: version `SUNWprivate_2.8' not found (required by file /usr/ccs/bin/ad) Anybody knows about that ? -- Frangois Legal E-mail : flegal at seo-toulouse.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zshouben at nianet.org Fri Apr 11 09:33:39 2003 From: zshouben at nianet.org (Shouben Zhou) Date: Fri, 11 Apr 2003 09:33:39 -0400 (EDT) Subject: summary: ypbind takes so long to up Message-ID: <200304111333.h3BDXdH02793@goby.nianet.org> Very thanks to: "Mike Demarco" He gives me the right answer! I have to change /etc/net/ti*/hosts files as well. -- Shouben Zhou National Institute of Aerospace (NIA) 144 Research Dr., Hampton, VA 23666 Tel: (757) 766-1809, Fax: (757) 766-1401 Cell: (757) 636-6303, Email: zshouben at nianet.org > > I use clone method to install Solaris system on all desktops, since they > are all identical. After cloning, chnage the IP, nodename and other few > files and bring the system with -r option. Everything works very well. > > Recently I have noticed when system reboot, it takes from 15 to 40 seconds > to bring the ypbind up. I have tried to remove the NIS client and redo it > from the single user mode ( domainname, ypinit -c ). It doesn't help much. > I don't know whatelse I need to check. Usually it takes no time to bring > the ypbind up. > > > -- > Shouben Zhou > National Institute of Aerospace (NIA) > 144 Research Dr., Hampton, VA 23666 > Tel: (757) 766-1809, Fax: (757) 766-1401 > Cell: (757) 636-6303, Email: zshouben at nianet.org > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From LWood at ALLEGROMICRO.com Fri Apr 11 10:05:38 2003 From: LWood at ALLEGROMICRO.com (Wood, Lawrence N) Date: Fri, 11 Apr 2003 10:05:38 -0400 Subject: SUMMARY - Arbitrary(?) start of OpenWindows Message-ID: <253CE41437DBD31180BF00D0B74452D204682F67@exchange_ma.allegromicro.com> Users can select their desired environment using the "Options -> Sessions" pull down menu on the login screen. Thanks to: Carl Schelin Charlotte Ratliff Brian Catlin Cathy M. Bodenstedt Ron Didwiddie Mike Demarco Original Post: One of my users, who was normally using CDE, logged in this morning and the system started OpenWindows instead of CDE. I've looked around his home directory but haven't been able to determine how this happened or how to get CDE back for him. Can anyone help? Larry Wood _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From e.bendriss at menara.ma Fri Apr 11 12:52:51 2003 From: e.bendriss at menara.ma (Bendriss Elmehdi) Date: Fri, 11 Apr 2003 16:52:51 -0000 Subject: how to logout someone after a period of inactivity 2 : SUMMARY Message-ID: <012101c3004a$cab5d540$cb0fd9d4@iamdg.net.ma> hi some one asked me to summarize : many told me to put an : export TMOUT=xx (xx in seconds) in /etc/profile : usefull above all when users use xterm many others recommend the "autologout" environment variable (csh) that do the same thing, but users may avoid it by putting something like : unset autologout in their own profile file some one sent me a link : http://access1.sun.com/codesamples/kstat.html there's a source code to get mouse and keyboard idle time : doesn't suite me because my users connect using telnet/ssh Beavers, Reginald sent me this shell code that i haven't try yet : as he said : logicaly it should do the job: ################################################### #!/bin/sh # egrep those who are immune who -u | egrep -v 'root|me' |while read line; do pid=`echo $line | awk '{print $7}'` idle=`echo $line | awk '{print $6}' | awk -F: '{print $2}'` if [ $idle >= 10 ]; kill -9 $pid fi done #####################################################" and finaly, the C source code that John Adams sent me; i attached it to to me email. that's it. thank you all for your help Cordialement/ Best Regards BENDRISS Elmehdi Administrateur Systhme Tel : +212 37 71 88 73 [demime 0.99c.7 removed an attachment of type application/octet-stream which had a name of autologout.c] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From e.bendriss at menara.ma Fri Apr 11 13:11:42 2003 From: e.bendriss at menara.ma (Bendriss Elmehdi) Date: Fri, 11 Apr 2003 17:11:42 -0000 Subject: how to logout someone after a period of inactivity 2 : SUMMARY 2 Message-ID: <013801c3004d$6cf5b670$cb0fd9d4@iamdg.net.ma> hi some one asked me to summarize : many told me to put an : export TMOUT=xx (xx in seconds) in /etc/profile : usefull above all when users use xterm many others recommend the "autologout" environment variable (csh) that do the same thing, but users may avoid it by putting something like : unset autologout in their own profile file some one sent me a link : http://access1.sun.com/codesamples/kstat.html there's a source code to get mouse and keyboard idle time : doesn't suite me because my users connect using telnet/ssh Beavers, Reginald sent me this shell code that i haven't try yet : as he said : logicaly it should do the job: ################################################### #!/bin/sh # egrep those who are immune who -u | egrep -v 'root|me' |while read line; do pid=`echo $line | awk '{print $7}'` idle=`echo $line | awk '{print $6}' | awk -F: '{print $2}'` if [ $idle >= 10 ]; kill -9 $pid fi done #####################################################" and finaly, the C source code that John Adams sent me; /* "@(#) autologout.c by David Dickson" */ /* modifications for exempt list processing by John Adams */ #include #include #include #include #include #include #define GRACE 60 /* grace time (sec) for user reply */ #define KWAIT 20 /* time to wait after kill (in sec.) */ #define WARNING 1 /* a warning message */ #define LOGOFF 2 /* a log-off message */ #define NOLOGOFF 3 /* a log-off failure message */ #define SEC_M 60 /* seconds per minute */ #define TIMEMAX 20 /* maximum idle time (minutes) */ /* uncomment this to not fork during processing */ /* #define NOFORK */ /* uncomment this to send mail to users when they go idle */ /* #define MAIL_USER */ struct utmp *utmpp; /* pointer to utmp file entry */ char *ctime(); /* returns pointer to time string */ struct utmp *getutent(); /* returns next utmp file entry */ typedef struct listent { char user[8]; struct listent *next; } LIST; /* exempt list */ LIST *head; LIST *tail; LIST *list_append(LIST *tail,char *item) { LIST *lp = (LIST *)malloc(sizeof(struct listent)); if (tail == NULL) { tail = lp; strcpy(tail->user,item); tail->next = NULL; } else { /* append to end */ tail->next = lp; strcpy(lp->user,item); lp->next = NULL; } /* return a pointer to the end */ return lp; } list_free(LIST *h) { /* walk a list, disposing of elements */ LIST *p = h; LIST *next; while (p != NULL) { next = p->next; free(p); p = next; } } int list_search(LIST *h,char *uname) { LIST *p; p = h; /* start at beginning */ while (p != NULL) { if (strcmp(p->user,uname) == 0) return 1; p = p->next; } /* not found */ return 0; } void init_exempt() { FILE *f; char readbuf[255]; head=NULL; tail=NULL; if ((f = fopen("/etc/autologout.exempt","r")) != NULL) { while (!feof(f)) { fgets(readbuf,255,f); /* hack off newline */ if (readbuf[strlen(readbuf)-1] == '\n') readbuf[strlen(readbuf)-1] = '\0'; /* strip comments and blanks */ if (readbuf[0] != ' ' && readbuf[0] != '#' && ( *readbuf != '\0' )) { tail = list_append(tail,readbuf); if ( head == NULL ) head = tail; } } } fclose(f); } main() { #ifndef NOFORK if (fork()) /* the parent process */ exit(0); /* exits */ #endif /* load up the exempt list */ init_exempt(); /* the child processes all utmp file entries: */ while ((utmpp = getutent()) != (struct utmp *) NULL) check_idle(); /* free the list */ list_free(head); exit(0); /* done, so bye */ } check_idle() /* select utmp entries older than TIMEMAX */ { char dev[24], name[12]; struct stat status, *pstat; time_t idle, pres_time, start, time(); pstat = &status; /* point to status structure */ if (utmpp->ut_type != USER_PROCESS) /* if not user process */ return(0); /* skip the utmp entry */ strcpy(dev, "/dev/"); /* add "/dev/" directory prefix */ strcat(dev, utmpp->ut_line); /* append basename of port */ if (stat(dev, pstat)) /* if can't get status for port */ bailout("can't get status of user's terminal", 1); pres_time = time((time_t *)0); /* get current time */ strncpy(name, utmpp->ut_user, 8); /* else get user name */ name[8] = '\0'; /* null terminate user name string */ /* should we even bother? */ if (list_search(head,name)) return; /* exempt. forget it. */ /* idle time is current time less last access time: */ idle = (pres_time - pstat->st_atime) / SEC_M; if (idle < TIMEMAX) /* if user was recently active */ return(0); /* ignore this utmp entry */ mesg(WARNING, name, dev, (int)idle); /* send warning to user */ if (stat(dev, pstat)) bailout("can't get status of user's terminal", 2); start = pstat->st_atime; /* start time for countdown */ sleep(GRACE); if (stat(dev, pstat)) bailout("can't get status of user's terminal", 3); if (start < pstat->st_atime) /* user did something */ return(0); else { /* user abandoned terminal */ if (!killit(utmpp->ut_pid)) mesg(NOLOGOFF, name, dev, (int)idle); /* couldn't kill */ else mesg(LOGOFF, name, dev, (int)idle); /* okay */ return(0); } } mesg(flag, name, dev, idle) /* mail to user and log message */ int flag, idle; /* flag indicates message type */ char *name, *dev; { char mbuf[256]; /* message buffer */ time_t tvec; FILE *fopen(), *fp, *log, *mprog; time(&tvec); /* store time in tvec */ if ((log = fopen("/usr/adm/logoutlog", "a")) == (FILE *) NULL) bailout("can't open log file", 4); if (flag == WARNING) { /* process warning message */ if ((fp = fopen(dev, "w")) == (FILE *) NULL) bailout("can't open user's terminal", 5); fprintf(fp, "%s: You've been idle for %d min.\07\n", name, idle); fprintf(fp, "you'll be logged off in 60 sec. unless you hit return:"); fclose(fp); fprintf(log, "** WARNING ** %s %s (%d min idle time) %s", name, dev+5, idle, ctime(&tvec)+3); } if (flag == LOGOFF) { /* process log-off message */ fprintf(log, "** LOGOFF ** %s %s (%d min idle time) %s", name, dev + 5, idle, ctime(&tvec) + 3); sprintf(mbuf, "/bin/mail %s", name); #ifdef MAIL_USER /* open pipe to mail program for writing */ if ((mprog = popen(mbuf, "w")) == (FILE *) NULL) bailout("can't use /bin/mail program", 6); fprintf(mprog, "Subject: Excess Idle Time\nLogged \ off - excess idle time - %s %s\ntty = %s\n", name, ctime(&tvec), dev + 5); fclose(mprog); #endif /* MAIL_USER */ } if (flag == NOLOGOFF) { /* send mail to root if can't kill */ fprintf(log, "** LOGOFF-FAIL ** %s (pid = %d) \ %s (%d min idle time) %s", name, utmpp->ut_pid, dev+5, idle, ctime(&tvec)+3); sprintf(mbuf, "/bin/mail root"); if ((mprog = popen(mbuf, "w")) == (FILE *) NULL) bailout("can't use /bin/mail program", 7); fprintf(mprog, "Subject: Can't logoff %s\nCan't Log off \ - %s %s\ntty = %s\n", name, name, ctime(&tvec), dev+5); fclose(mprog); } fclose(log); return(0); } killit(pid) /* terminate process using SIGHUP, then SIGKILL */ int pid; { kill(pid, SIGHUP); /* first send "hangup" signal */ sleep(KWAIT); if (!kill(pid, 0)) { /* SIGHUP might be ignored */ kill(pid, SIGKILL); /* then send sure "kill" signal */ sleep(KWAIT); if (!kill(pid, 0)) return(0); /* failure--refuses to die! */ else return(1); /* successful kill with SIGKILL */ } else return(1); /* successful kill with SIGHUP */ } bailout(message, status) /* display error message and exit */ int status; /* exit status */ char *message; /* pointer to the error message */ { fprintf(stderr, "autologout: %s\n", message); exit(status); } that's it. Cordialement/ Best Regards BENDRISS Elmehdi Administrateur Systhme Tel : +212 37 71 88 73 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From alek at ast.lmco.com Fri Apr 11 13:29:02 2003 From: alek at ast.lmco.com (Alek O. Komarnitsky (N-CSC)) Date: Fri, 11 Apr 2003 11:29:02 -0600 (MDT) Subject: SUMMARY: Why doesn't prtdiag return temperature information available via lom? Message-ID: <200304111729.LAA02196@goose.ast.lmco.com> I asked earlier this week: I'm using the "gettemp" program to monitor/notify - this is available at: http://www.komar.org/ -> Misc. Tech Stuff -> gettemp (disclaimer: I wrote it! ;-) which basically uses the "prtdiag" command to get temperature info. This is available on a number of machines, and is VERY handy to keep an eye not only on the ambient (or close to ambient) temperatures, but also CPU/etc. temperatures and auto-notify on overheats. We've used this for years and it works pretty darn well; even though the code is kinda ugly. I have some spanking new V120's (nice boxes) that I have loaded Solaris2.9 on and also added the LOM package.. The "prtdiag" command does NOT appear to return any temperature info (even with the -v option), but if one does a "lom -t", you get some data. So obviousely it *IS* available from a hardware point of view; but it would sure be nice if the prtdiag command returned it. I have done some misc. searching, but have not come across a patch or option for "prtdiag" ... has anyone else and/or if someone from Sun is lurking, this would be a REALLY NICE PATCH! ;-) Thanx, alek Unfortunately, I did not get a "good" answer - as mentioned above, there *IS* a hardware temperature sensor available on the V120 and it *IS* available when running Solaris (using the "lom" command), so it is surprising that the "prtdiag" command doesn't simply use the same code/interface to obtain this and then display it. One person said you must have picld running to help get environmental data, but that is running here - it's a fairly "stock" Solaris2.9 load. Another person mentioned that the SunFire hardware requires you get it from the RSC card (which appears to be the case on V280's for the ambient temperature), but as mentioned above, on this V120; it is clearly obtainable "directly" from Solaris. Hopefully a future patch/release of prtdiag will include this functionality which would be helpful/complete and probably not too difficult to add. In the meantime, I think I'll modify gettemp to call the lom command instead of prtdiag for those machines that require this. Thanx all that responded (see list below), alek David Foster Hichael Morton William Yodlowsky Ryan A. Krenzischek _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jeremyr at society86.com Wed Apr 9 22:26:10 2003 From: jeremyr at society86.com (Jeremy Russell) Date: Wed, 9 Apr 2003 21:26:10 -0500 Subject: SUMMARY:DNS Wierdness Message-ID: <003e01c2ff08$c24f3fb0$01a8a8c0@rancid3> Thanks for all of the replies, I got a million it seems, I learned some new things I wasn't aware of before. I was misinformed. Here's my scenario to shed some more light on the subject. The Solaris machine is performing monitoring services for the internal and external networks. It has an internal address that has been given NAT capabilities through the firewall. The internal DNS boxen are Win2k Active Directory machines. The internal network is dominated by Win2k, just one Solaris box, and that's me. The external servers are Sprints public DNS machines. I plan to install two Solaris Boxes with Bind 9 when I get the appropriate hardware to place on our external network. Some issues I have. A lot of the replies said to have the internal forward or cache, I want them to but the policy is of right now that they will not. The current management wants them to only do internal DNS, the only external addresses are fed to clients through the proxy. So until that is settled the stuff wont forward or anything. The external DNS that we use know nothing of the internal domain. So I have to make use of bother sets of servers to resolve on both sides. With the restraints I have been given anyways. Some more weirdness... I went today and implemented the resolv.conf as I had sent it in the original post (I inly had internal servers in it), the weird part is that it works! I'm not aware of anything changing ,so... it just got weirder... FYI, my nsswitch.conf is set to "files dns" so that wasn't the problem. I am in agreement that my problem lies with the fact that so many pointed out the the DNS servers are giving me a "host not found" and not failing over. That sounds logical. Also I probably should fix my broke resolv.conf since it is not syntactically correct. (DOH!) But, I'm gonna leave it for a while and see what happens... weird... Thanks again for the many replies, and for fun I keep everyone up to date on this twisted tale... Thanks!! PS: Sorry about the book I just wrote... :/ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mdb at cwiz.com Sat Apr 12 13:50:21 2003 From: mdb at cwiz.com (Martin D. Baldenegro (RRNC)) Date: Sat, 12 Apr 2003 11:50:21 -0600 (MDT) Subject: SUMMARY: Disable Echo Reply Message-ID: <200304121750.h3CHoJq7003202@dosmanos.cwiz.com> Sun Managers, Thanks to all the folks that replied, especially Sean Quaint who's reply headed me in the right direction. What I did on the system was to set the parameter ip_respond_to_echo_broadcast to zero with the ndd -set command. I also added the entry into the /etc/rc2.d/S72inetsvc file. Command: ndd -set /dev/ip ip_respond_to_echo_broadcast 0 This solved the problem until I replace my old Ascend Pipeline router. Thanks to: "Jonathan V Hays" "Steve Armijo" "Arzola, Nelson" Casper Dik Steve Maher Al Hopper "Preston, Mark" Tim Wort Again thanks to all, /mdb The original problem: > Sun Managers, > > I keep getting the following on several of the systems in my > network. I have commented out the echo lines in both the > /etc/inet/services and in /etc/inet/inetd.conf but I can not > seem to stop the DOS on my network. The below output goes on > for several hours and has basically put the network at a crawl. > > The systems are running Solaris 8 and 9 different updates. > > Anyone else run into this problem, and what else can be done to > prevent echo replys from the systems inside the network. > > TIA, > > /mdb > > -----snoop output ----- > www.recomp.tv -> xxx.xxx.xxx.255 UDP D=23920 S=5192 LEN=9 > www.recomp.tv -> xxx.xxx.xxx.0 UDP D=35903 S=56090 LEN=9 > www.recomp.tv -> xxx.xxx.xxx.255 ICMP Echo request (ID: 0 Sequence number: 0) > maya -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > navajo -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > dosmanos -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > apache -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > azteca -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > printsrv -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > yakii -> www.recomp.tv ICMP Echo reply (ID: 0 Sequence number: 0) > www.recomp.tv -> xxx.xxx.xxx.0 UDP D=50403 S=35197 LEN=9 > www.recomp.tv -> xxx.xxx.xxx.255 UDP D=52890 S=62425 LEN=9 > =============================================================== Martin D. Baldenegro (RRNC) Cwiz Resources, Inc. Re-Thinking Knowledge Westminster, CO. 80020-8723 Direct Phone #: 303-478-4790 E-mail: mailto:mdb at cwiz.com =============================================================== " Do Not Squander Time, For That Is The Stuff Life Is Made Of " _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From 0sqt7xekzs at earthlink.net Sun Apr 13 17:11:15 2003 From: 0sqt7xekzs at earthlink.net (Pasquale Franklin) Date: Sun, 13 Apr 03 21:11:15 GMT Subject: Purchase, Diet Pills, and other prescriptions with no Prescription Summaries Message-ID: <9v--1t9-1m0-1oxr4f@lnnb.f.e3js> An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030413/5d5f496e/attachment.html From osamaomar at link.net Sun Apr 13 05:59:49 2003 From: osamaomar at link.net (Osama Omar) Date: Sun, 13 Apr 2003 11:59:49 +0200 Subject: SUMMARY gcc compliling Message-ID: <000201c301a3$6c656ff0$c79e939d@onshore.pgs.com> Hi Many thanks to all who replied, I have installed a pre-compiled version of GCC and it worked finely Special thanks to : Rafael Angarita Roy S. Rapoport neil quiogue Joon Martin Hansen Regards Osama Omar Ali PGS Data Processing Middle East IT Dept. (002)(02) 2761353 (002)(02) 2761353 (002) (010) 6008263 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levent at kaa.desy.de Sun Apr 13 12:37:27 2003 From: levent at kaa.desy.de (Levent Demiroers) Date: Sun, 13 Apr 2003 18:37:27 +0200 (MEST) Subject: SUMMARY/NEW QUESITON: DNS cache server, sendmail Message-ID: Hi Managers, first, let me thank Sebastien DAUBIGNE Marc Sheldon James Zhao Timothy Lorenc Tim Chipman for their answers. I decided to try out djbdns (http://cr.yp.to/djbdns.html ) which turned out to be a easy installation. After following the five (six?) steps of the installation manual, the service was up and running! But.... The main reason I embarked on this whole quest was the difficulties I had with sendmails (using ver 8.11.6+Sun) Anti-Spam policies. When I tun on the strict host checking, a lot of emails are rejected with the error message "Domain of sender address blabla does not resolve", even though a lookup through > getent hosts blabla or > dnsip blabla gives a result. Any hints on this issue? I'd appreciate it. Cheers, Levent --------------------------------------------------------------------- Levent Demiroers kaa.desy.de/people/Demiroers.html _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sunsol19712003 at yahoo.es Sun Apr 13 15:15:07 2003 From: sunsol19712003 at yahoo.es (=?iso-8859-1?q?J=20Marquez?=) Date: Sun, 13 Apr 2003 21:15:07 +0200 (CEST) Subject: SUMMARY: SCSI CARD FOR ULTRA 5 Message-ID: <20030413191507.94593.qmail@web20702.mail.yahoo.com> Dear SunManagers, My problem is alredy solved: I needed to seat the SCSI card properly (it was in the right place -PCi riser board- but it needed to be pushed completely), besides a boot -r was needed. Thanks to: Olson, Ross A. Michael LehmannOsama Omar Robert Geoff Reed Hichael Morton Again thanks to all The original problem:Date: Sat, 12 Apr 2003 19:32:48 +0200 (CEST) From: =?iso-8859-1?q?Sun=20Sol?= Subject: SCSI CARD FOR ULTRA 5 To: sunmanagers at sunmanagers.orgHi gurus, I am very new with Hardware stuff and I got a problem related: I've just bought a SCSI card for my Ultra 5, but I am not sure how I gotta plug this into my machine. I tried to connect it to the "PCI Riser board" but when I boot the machine I could not see the SCSI card with the "prtdiag -v" command. The SCSI card is LSI SYMBIOS. It puzzles me that when I connect to the PCI riser board, the SCSI card has 2 interfaces without being used. Should I connect them to somewhere else in the machine? The card did not bring with any cable.. Thanks in advance and regards. --------------------------------- Yahoo! Messenger Nueva versión: Super Webcam, voz, caritas animadas, y más #161;Gratis! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From gprince at sealy.com.au Sun Apr 13 19:56:09 2003 From: gprince at sealy.com.au (Glenn Prince) Date: Mon, 14 Apr 2003 10:56:09 +1100 Subject: SUMMARY: iMS 5.2 Collection Problem In-Reply-To: <3E962BC3.30103@fnal.gov> Message-ID: <001901c30218$419b6700$f804a8c0@citrixnet.sealy.com.au> I think my problem is solved with regards to users unable to pickup POP3 mail. Rebuilding the mailboxes and applying the latest hotfixes has fixed the problem as far as I can see. Thanks to: William D Hathaway Al Lilianstrom For their advice. Glenn Prince _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jnaneshkalbag at indiatimes.com Sun Apr 13 23:10:54 2003 From: jnaneshkalbag at indiatimes.com (jnaneshkalbag) Date: Mon, 14 Apr 2003 08:40:54 +0530 Subject: SUMMARY:process info from netstat Message-ID: <200304140238.IAA27693@WS0005.indiatimes.com> Here is the question I posted. ----------------------------- > How can i get information regarding the processes that are ESTABLISHED ( out put from netstat) at this port 80. > > netstat | grep 80 > ------- > testmach23.80 121.22.37.86.39189 8760 0 32872 0 ESTABLISHED > testmach23.80 121.22.37.86.59311 8760 0 32872 0 ESTABLISHED > testmach23.80 121.22.37.86.39255 8760 0 32872 0 ESTABLISHED > > I am trying to get the process name that are clinging on to port 80. Is there any other way we can get the processes at this port. ------------------------------------------ Answers: ------- 1> #!/bin/sh PFILES=/usr/proc/bin/pfiles PROCS=`ps -ef|awk 'NR>1{print $2}'` for i in $PROCS do if [ -n "`$PFILES $i 2>/dev/null|grep S_IFSOCK`" ] then ps -f -p $i| awk 'NR>1{print}' $PFILES $i 2>/dev/null|awk '/S_IFSOCK/{print $1, $2, $5}/sockname:/{print}' fi done 2> Get lsof from http://freshmeat.net/projects/lsof/ and compile it, or get a prebuilt version from http://www.sunfreeware.com/ that is appropriate for whatever Solaris version you are running. You can then do lsof -I tcp:80 and it will list the processes on port 80. ------------------------------------- Many thanks to yura pismerov for the script in answer(1) and Rick anderson/Alexie for answer (2). Get Your Private, Free E-mail from Indiatimes at http://email.indiatimes.com Buy The Best In BOOKS at http://www.bestsellers.indiatimes.com Bid for for Air Tickets @ Re.1 on Air Sahara Flights. Just log on to http://airsahara.indiatimes.com and Bid Now! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From lists at internyc.net Mon Apr 14 01:06:18 2003 From: lists at internyc.net (fil) Date: Mon, 14 Apr 2003 01:06:18 -0400 Subject: [SUMMARY] weird sendmail problem on Solaris 9 Message-ID: <3E9A41CA.2050606@internyc.net> Thanks to everyone who responded: charlotte Crist J. Clark mat alan comp.mail.sendmail users The solution came from comp.mail.sendmail users. The solution is to add: O ResolverOptions=WorkAroundBrokenAAAA option to your sendmail.cf file or define(`confBIND_OPTS', `WorkAroundBrokenAAAA')dnl to your sendmail.mc file. And this is why (according to sendmail README file in sendmail source subdirectory): "When attempting to canonify a hostname, some broken name servers will return SERVFAIL (a temporary failure) on T_AAAA (IPv6) lookups. If you want to excuse this behavior, include WorkAroundBrokenAAAA in ResolverOptions. However, instead, we recommend catching the problem and reporting it to the name server administrator so we can rid the world of broken name servers." Thanks fil -- Fikus Development Corp. 42-06A Bell Blvd. #242 Bayside, NY 11361 Tel: 347-495-3649 Fax: 347-412-5378 --- original message --- I have a weird Sendmail problems and I have run out of ideas on how to fix it. Server is Netra X1 running Sendmail 8.12.9 and bind 9.2.2. DNS works 100% and all e-mail works except for one domain (1tv.tv). Any time I try to send e-mail to that doaminS my mail gets queued with following error: Running /var/spool/mqueue/h3A5t8VW020803 (sequence 1 of 1) 1tv.tv: Name server timeout ... Transient parse error -- message queued for future delivery They sit in the queue for 5 days and then get thrown out. Here is the weird part DNS works 100% and evry thing resolves fine, mine and recipients address. nslookup >> set type=any >> 1tv.tv Server: 127.0.0.1 Address: 127.0.0.1#53 Non-authoritative answer: 1tv.tv mail exchanger = 10 mxpool.postoffice.net. Name: 1tv.tv Address: 64.239.27.18 1tv.tv nameserver = ns2-hosts.nic.tv. 1tv.tv nameserver = ns1-hosts.nic.tv. Authoritative answers can be found from: 1tv.tv nameserver = ns2-hosts.nic.tv. 1tv.tv nameserver = ns1-hosts.nic.tv. ----------------------------------------------- nslookup mxpool.postoffice.net Server: 127.0.0.1 Address: 127.0.0.1#53 Non-authoritative answer: Name: mxpool.postoffice.net Address: 165.212.65.113 ----------------------------------------------- telnet mxpool.postoffice.net 25 Trying 165.212.65.113... Connected to mxpool.postoffice.net. Escape character is '^]'. 220 esmail02.eservices.usa.net ESMTP USA.NET-SMTA vC8.MAIN.1.11G; Thu, 10 Apr 2003 06:20:47 GMT QUIT 221 Goodbye Connection to mxpool.postoffice.net closed by foreign host. ------------------------------------------------- Also if I do Sendmail address test it fails with the same error /usr/lib/sendmail -bt ADDRESS TEST MODE (ruleset 3 NOT automatically invoked) Enter
>> 3 test at 1tv.tv canonify input: test @ 1tv . tv Canonify2 input: test < @ 1tv . tv > 1tv.tv: Name server timeout Canonify2 returns: test < @ 1tv . tv > canonify returns: test < @ 1tv . tv > == Ruleset 3 (3) status 75 And for the biggest kick of all, if I send e-mail to the same address from any other of my 5 mail servers it works fine. The only difference from the machines is that the rest of the servers are Linux and all the mail and dns software is exactly the same. Any ideas are welcomed. Thanks fil _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rj45 at slacknet.com Mon Apr 14 05:27:15 2003 From: rj45 at slacknet.com (RJ45) Date: Mon, 14 Apr 2003 03:27:15 -0600 (MDT) Subject: SUMMARY: ULTRASPARC real CPU Speed In-Reply-To: <3E9A6AF3.68D937B7@st.com> Message-ID: Generally it's not possible to overclock a 360MHz ultrasparcII to 450MHz, if you force the jumper on the ultra60 motherboard to be 450MHz ultra60 won't even boot. Thanks for the many dozen of peopel who replyed to my question Rick _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From likun at asiainfo.com Mon Apr 14 07:31:41 2003 From: likun at asiainfo.com (Kun Li) Date: Mon, 14 Apr 2003 19:31:41 +0800 Subject: [summary]How many dns queries can a 400Mhz cpu support Message-ID: <08603A641E6B2542B9C9AB7DEE2EE8368C5785@aints2.asiainfo.com> hello, here is the summary. Several suggestions, from a complete waste to 1300 queries/sec is a lot. Christophe suggested getting a few v100s instead, they are much cheaper and just suitable for dns business, to horizontal scaling. Peter kindly reminded me upgrading to bind9 before buying new hardware, and not to use a precompiled package, but compile it with gcc and use optimizing flags. Tim Chipman recommend alternates: djbdns/tinydns related tools, which is much better, you can get it from http://cr.yp.to/djbdns.html, also the performance , if you want to dig into it : http://lwn.net/2001/features/djbdns.php3 http://www.geocrawler.com/archives/3/514/2000/1/50/3160453/ http://forums.devshed.com/archive/36/2002/06/1/36632 http://djbdns.qmail.jp/performance/dnsspeed.html http://cr.yp.to/djbdns/blurb/server.html thanks all. although my question haven't been answered directly yet. My conclusion is it depends. One thing I noticed that may effect the performance of a dns server is , response time . for example, if a name query get 2s to be travelling around the internet, and then back to the client , compared to a query only get 20ms to get answered, the former server must hold much more sessions than the latter , so under the same query load, the former server may consume much more , maybe ten time, cpu power than the latter . Since I live in a country that the latency to the root dns server is some high, so I believe I need more resources than others to process the same load. likun followed is my original question: ============================ Hi, everyone, I really wonder how many queries per second a dns server can process with what kind of hardware configuration. I have got a dns server , SUN E450 with two 400MHz CPU, 4G mem, no other application, dedicated for dns service, Runing bind 8.3.3 Now, it handles almost 1300 queries per second, and have used up 80% of process power of one cpu , since bind8 is single-processed and single-threaded , it can't use the two cpu at the same time, that means the server has almost used up all cpu power this machine can provide for dns service. So I had to face two choice: upgrade to bind9 or upgrade the hardware, most probably the CPU . But I recall a lot of people said that dns service needs very few cpu power , so Why a 400MHz cpu in my server can only support up to 1300 queries per second? Does anyone has got any idea, in generally , how much load could be supported with a workstation like this, or with what kind of cpu power can you handle how many load ? Any inputs are welcome _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Shawn.Russell at GE.com Fri Apr 11 16:24:40 2003 From: Shawn.Russell at GE.com (Russell, Shawn (Indsys, FW, IT-E-Business Non GE)) Date: Fri, 11 Apr 2003 15:24:40 -0500 Subject: SUMMARY FW: HP JetDirect Message-ID: <9D669FA3F4331943A871E3F6FD0F1E53010692D7@FTWMLVEM01.e2k.ad.ge.com> Thanks to Will Enestvedt Carl Gobbo Aaron Facey Simone Cristina Sperandio They all made suggestions on coping /etc/lp/interfaces. I did not try this method. I went with the script provided by HP. Many Thanks. I was able to find a script that is bundled with V.10.34 Below are the directions that came with it: Transferqueue Instructions TRANSFER.UNX 26/04/2002 Purpose of "transferqueue" script: To transfer all JetAdmin created queues from one workstation to another into JetDirect Printer Installer while retaining the same set of configuration options that were configured through the JetAdmin Utility. It also converts Jetadmin created queues to JetDirect Printer Installer in the same system. NOTE: This script will ONLY transfer queues from HP-UX 10.x/11.x to HP-UX 10.x/11.x or Solaris 2.x to Solaris 2.x *** Do not transfer JetAdmin created queues from HP-UX to Solaris or from Solaris to HP-UX. These queues will not function properly and may become corrupted. 1. This script transfers queues from one workstation to another as noted above. The script will not update the queues with new fixes available in the latest Jetadmin modelscripts. 2. When transferring Jetadmin created queues to the destination workstation, the destination must have JetDirect Printer Installer version installed. 3. In case of a failure see /opt/hpnpl/tmp/transferqueue.log for more information. Transferring JetAdmin created queues to JetDirect Printer Installer in the same system ------------------------------------------------------------------- 1. Invoke transferqueue with "local" to convert local queues to JetDirect Printer Installer. 2. Select a queue or all to convert the queues to JetDirect Printer Installer. 3. Try printing from the converted queues Transferring JetAdmin created queues from HP-UX 10.x/11.x to HP-UX 10.x/11.x -------------------------------------------------------------------- 1. Login to the workstation (Source) as 'root' user from which the queues are to be transferred. 2. Change directories to /etc/lp/interface and remove old queues by executing "rm *.old" and remove the files which are not used as queues. 3. Invoke transferqueue with "backup" to tar all the queues created by JetAdmin/JetDirect Printer Installer for Unix. The tared file will be placed in /tmp/queues.tar. 4. Transfer the file '/tmp/queues.tar' to the '/tmp' directory of destination workstation (to which the queues are to be transferred) by executing the following commands. ftp enter 'root' as a user enter 'root' user password cd /tmp bin put /tmp/queues.tar bye 5. Login to the destination workstation as root user. 6. Install JetDirect Printer Installer (E.10.18 or later) if it is not already installed. 7. Make sure that queue names on the source workstation do not conflict with the queuenames on the destination workstation. If same queuenames exist on both workstations, then the one on the destination workstation will be overwritten by the source workstation's queue. 8. If you are using a host table - check to see if the proper /etc/host entries exists on the destination workstation for the queues you are transfering. 9. If you are using bootp - make sure the /etc/bootptab entries are available on the destination workstation. 10. Execute the script '/opt/hpnpl/bin/transferqueue'. This will create and install the queues on destination workstation with the same name as in the source workstation with same set of options. 11. Set the default queue (if it is necessary) by executing 'lpadmin -d. Transferring JetAdmin created queues from Solaris 2.x to Solaris 2.x ------------------------------------------------------------ 1. Login to the workstation (Source) as 'root' user from which the queues are to be transferred. 2. Change directory to /etc/lp/interfaces and remove old queues by executing "rm *.old" and remove the files which are not used as queues. 3. Invoke transferqueue with "backup" to tar all the queues created by JetAdmin/JetDirect Printer Installer for Unix. The tared file will be placed in /tmp/queues.tar. 4. Transfer the file '/tmp/queues.tar' to the '/tmp' directory of destination workstation (to which the queues are to be transferred) by executing the following commands. ftp enter 'root' as a user enter 'root' user password cd /tmp bin put /tmp/queues.tar bye 5. Login to the destination workstation as root user. 6. Install JetDirect Printer Installer (E.10.18 or later) if it is not already installed. 7. Make sure that queue names on the source workstation do not conflict with the queuenames on the destination workstation. If same queue name exists on both workstations, then the one on the destination workstation will be overwritten by the source workstation's queue. 8. If you are using a host table - check to see if the proper /etc/host entries exists on the destination workstation for the queues you are transfering. 9. If you are using bootp - make sure the /etc/bootptab entries are available on the destination workstation. 10. Execute the script '/opt/hpnpl/bin/transferqueue'. This will create and install the queues on destination workstation with the same name as in the source workstation with same set of options. 11. Set the default queue (if it necessary) by executing 'lpadmin -d . -----Original Message----- From: Russell, Shawn (Indsys, FW, IT-E-Business Non GE) Sent: Friday, April 11, 2003 2:50 PM To: sunmanagers at sunmanagers.org Subject: HP JetDirect I am configuring a series 100+ printers on 4 SunFire 480R servers. We are doing this in a Solaris 8, rev 10/02, environment via the HP JetDirect Print Installer. I would to know what files besides /etc/printers.conf is modified and does the JetDirect software keep an internal listing of all printers. I would like to be able to just configure one machine and copy the conf files over to the others. Thanks, Shawn Russell Unix Administrator shawn.russell at ge.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From TCarter at memc.com Fri Apr 11 16:34:07 2003 From: TCarter at memc.com (Thomas Carter) Date: Fri, 11 Apr 2003 15:34:07 -0500 Subject: SUMMARY: Help with DHCP/BOOTP Message-ID: I applied patch 109077-11 which, according to the README, is actually a new version of the DHCP server. After installing it and running in debug mode, I found the HP JetDirect box was showing a Client ID of 01 + the MAC address. changing the entry to 01080009A2AD65 caused it to work correctly. Thanks, Thomas Carter MEMC Southwest ----- Forwarded by Thomas Carter/MIS/SW/MEMC on 04/11/2003 03:31 PM ----- Thomas Carter 04/10/2003 05:18 PM To: sunmanagers at sunmanagers.org cc: Subject: Help with DHCP/BOOTP I'm trying to set up in.dhcpd on Solaris 8 to serve primarily as a BOOTP server for a number printers we have scattered around. Our router forwards all BOOTP requests to this server. I've set up the networks and addresses, and made sure the service is started and enabled. The server is set for Manual BOOTP Compatibility, but it doesn't seem to answer the BOOTP requests. When I run in.dhcpd in debug mode, it logs the following (IPs changed to protect the innocent): Packet received from relay agent: xxx.xxx.xxx.xxx Datagram received on network device: hme0 BOOTP client: 080009A2AD65 is looking for a configuration. But that client is defined in a configuration file: 123_123_123_123 -------------------------- 080009A2AD65 11 123.123.123.222 111.111.111.111 0 123.123.123.123 In DHCP manager, I defined this client ID as reserved, permanent, and only for BOOTP clients. What do I need to do to make in.dhcpd answer these requests? Am I missing something? Thanks, Thomas Carter MEMC Southwest _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From choogend at library.umass.edu Mon Apr 14 10:29:12 2003 From: choogend at library.umass.edu (Chris Hoogendyk) Date: Mon, 14 Apr 2003 10:29:12 -0400 Subject: PRE-SUMMARY: Configuring D1000 from Command Line Message-ID: <3E9AC5B8.3040704@library.umass.edu> I have receieved several replies with step by step for my situation. Hope to have it done while it is still Monday morning here. Will send summary & credits this afternoon or tomorrow. Thankyou, --------------- Chris Hoogendyk -- O__ ---- Network Specialist & Unix Systems Administrator c/ /'_ --- Library Information Systems & Technology Services (*) \(*) -- W.E.B. Du Bois Library ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Shawn.Russell at GE.com Mon Apr 14 10:35:39 2003 From: Shawn.Russell at GE.com (Russell, Shawn (Indsys, FW, IT-E-Business Non GE)) Date: Mon, 14 Apr 2003 09:35:39 -0500 Subject: SUMMARY HP JetDirect Message-ID: <9D669FA3F4331943A871E3F6FD0F1E53010692E0@FTWMLVEM01.e2k.ad.ge.com> Thanks to Will Enestvedt Carl Gobbo Aaron Facey Simone Cristina Sperandio They all made suggestions on coping /etc/lp/interfaces. I did not try this method. I went with the script provided by HP. Many Thanks. I was able to find a script that is bundled with V.10.34 Below are the directions that came with it: Transferqueue Instructions TRANSFER.UNX 26/04/2002 Purpose of "transferqueue" script: To transfer all JetAdmin created queues from one workstation to another into JetDirect Printer Installer while retaining the same set of configuration options that were configured through the JetAdmin Utility. It also converts Jetadmin created queues to JetDirect Printer Installer in the same system. NOTE: This script will ONLY transfer queues from HP-UX 10.x/11.x to HP-UX 10.x/11.x or Solaris 2.x to Solaris 2.x *** Do not transfer JetAdmin created queues from HP-UX to Solaris or from Solaris to HP-UX. These queues will not function properly and may become corrupted. 1. This script transfers queues from one workstation to another as noted above. The script will not update the queues with new fixes available in the latest Jetadmin modelscripts. 2. When transferring Jetadmin created queues to the destination workstation, the destination must have JetDirect Printer Installer version installed. 3. In case of a failure see /opt/hpnpl/tmp/transferqueue.log for more information. Transferring JetAdmin created queues to JetDirect Printer Installer in the same system ------------------------------------------------------------------- 1. Invoke transferqueue with "local" to convert local queues to JetDirect Printer Installer. 2. Select a queue or all to convert the queues to JetDirect Printer Installer. 3. Try printing from the converted queues Transferring JetAdmin created queues from HP-UX 10.x/11.x to HP-UX 10.x/11.x -------------------------------------------------------------------- 1. Login to the workstation (Source) as 'root' user from which the queues are to be transferred. 2. Change directories to /etc/lp/interface and remove old queues by executing "rm *.old" and remove the files which are not used as queues. 3. Invoke transferqueue with "backup" to tar all the queues created by JetAdmin/JetDirect Printer Installer for Unix. The tared file will be placed in /tmp/queues.tar. 4. Transfer the file '/tmp/queues.tar' to the '/tmp' directory of destination workstation (to which the queues are to be transferred) by executing the following commands. ftp enter 'root' as a user enter 'root' user password cd /tmp bin put /tmp/queues.tar bye 5. Login to the destination workstation as root user. 6. Install JetDirect Printer Installer (E.10.18 or later) if it is not already installed. 7. Make sure that queue names on the source workstation do not conflict with the queuenames on the destination workstation. If same queuenames exist on both workstations, then the one on the destination workstation will be overwritten by the source workstation's queue. 8. If you are using a host table - check to see if the proper /etc/host entries exists on the destination workstation for the queues you are transfering. 9. If you are using bootp - make sure the /etc/bootptab entries are available on the destination workstation. 10. Execute the script '/opt/hpnpl/bin/transferqueue'. This will create and install the queues on destination workstation with the same name as in the source workstation with same set of options. 11. Set the default queue (if it is necessary) by executing 'lpadmin -d. Transferring JetAdmin created queues from Solaris 2.x to Solaris 2.x ------------------------------------------------------------ 1. Login to the workstation (Source) as 'root' user from which the queues are to be transferred. 2. Change directory to /etc/lp/interfaces and remove old queues by executing "rm *.old" and remove the files which are not used as queues. 3. Invoke transferqueue with "backup" to tar all the queues created by JetAdmin/JetDirect Printer Installer for Unix. The tared file will be placed in /tmp/queues.tar. 4. Transfer the file '/tmp/queues.tar' to the '/tmp' directory of destination workstation (to which the queues are to be transferred) by executing the following commands. ftp enter 'root' as a user enter 'root' user password cd /tmp bin put /tmp/queues.tar bye 5. Login to the destination workstation as root user. 6. Install JetDirect Printer Installer (E.10.18 or later) if it is not already installed. 7. Make sure that queue names on the source workstation do not conflict with the queuenames on the destination workstation. If same queue name exists on both workstations, then the one on the destination workstation will be overwritten by the source workstation's queue. 8. If you are using a host table - check to see if the proper /etc/host entries exists on the destination workstation for the queues you are transfering. 9. If you are using bootp - make sure the /etc/bootptab entries are available on the destination workstation. 10. Execute the script '/opt/hpnpl/bin/transferqueue'. This will create and install the queues on destination workstation with the same name as in the source workstation with same set of options. 11. Set the default queue (if it necessary) by executing 'lpadmin -d . -----Original Message----- From: Russell, Shawn (Indsys, FW, IT-E-Business Non GE) Sent: Friday, April 11, 2003 2:50 PM To: sunmanagers at sunmanagers.org Subject: HP JetDirect I am configuring a series 100+ printers on 4 SunFire 480R servers. We are doing this in a Solaris 8, rev 10/02, environment via the HP JetDirect Print Installer. I would to know what files besides /etc/printers.conf is modified and does the JetDirect software keep an internal listing of all printers. I would like to be able to just configure one machine and copy the conf files over to the others. Thanks, Shawn Russell Unix Administrator shawn.russell at ge.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From anthony at fccl.co.uk Mon Apr 14 11:54:29 2003 From: anthony at fccl.co.uk (Anthony Firmin) Date: Mon, 14 Apr 2003 16:54:29 +0100 Subject: Summary: Finding out kernel parameter settings? In-Reply-To: Message-ID: adb -k does allow you to review all the parameters. sysdef -i will show you some, but not all, of the parameters and their values. Thanks to all who responded. Anthony _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ecorrado at hera.rider.edu Mon Apr 14 12:43:18 2003 From: ecorrado at hera.rider.edu (Edward M. Corrado) Date: Mon, 14 Apr 2003 12:43:18 -0400 (EDT) Subject: Summary: OpenSSH problem In-Reply-To: Message-ID: Thanks to some quick responses, I now have OpenSSH working on my machine. As it turns out, in the last couple of days, a new OpenSSH package came out on http://sunfreeware.com. I deleted all the keys and uninstalled OpenSSH version 3.5.1p1, and then installed the new OpenSSH package (3.6.1p1) and a newer version of OpenSSL (0.9.7a) from SunFreeware. Some other suggestions included: Mike (of Mike's List) suggested: try ssh -l and see if this helps...try ip instead of hostname machine.somewhere.com etc. He also mentioned that more Sun packages can be found at: ftp.patriots.net. Brian Pardy suggested deleting any keys in $HOME/.ssh. While that didn't help, it did lead me to attempt to connect using Protocol 1, which worked, but I still couldn't get Protocol 2 to work before upgrading to the just released version. I'm not sure if there was a bug, some other conflict between the first version of OpenSSH (3.5.1p1) and something on my system, or just that something became corrupted. Anyway, it is now working. Thanks to everyone that responded. Original post is below. Edward On Mon, 14 Apr 2003, Edward M. Corrado wrote: > Hello Sun Managers, > > I am in the process of configuring a Netra server running Solaris 8. Among > other things, I want to install OpenSSH on this machine. In order to do > this, I decided to download the packages from http://sunfreeware.com > (as I have done for other servers). However, even though I believe I have > followed all of the directions correctly, I am getting an error when I > try to connect. The error is: > > 49 (1)hera:~> ssh ecorrado at caddis > ssh_rsa_verify: RSA_verify failed: error:0407006A:rsa > routines:RSA_padding_check_PKCS1_type_1:block type is not 01 > key_verify failed for server_host_key > > This happens from more then one machine (in fact, it also happens from the > same machine that I am trying to install OpenSSH on). I tried recreating > the keys and a Google search didn't turn up any fixes besides one > suggestion to upgrade the client, but even the client on the same machine > that was installed with the same package is having the problem, so I don't > see how that could be the issue here. The OpenSSH version is the most > recent on sunfreeware (3.5p1). I have installed the patch that provides > /dev/random. > > Does anyone have any suggestions? I will summarize to the list. > > Edward Corrado _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From choogend at library.umass.edu Mon Apr 14 15:55:51 2003 From: choogend at library.umass.edu (Chris Hoogendyk) Date: Mon, 14 Apr 2003 15:55:51 -0400 Subject: SUMMARY: Configuring D1000 from command line Message-ID: <3E9B1247.5040704@library.umass.edu> Success! Thanks to: Eric Ray Darrn Dunham Sean Berry David Harrington John Tackman As John said, "It's actually quite simple once you 'get it' " It was the sum of all five that gave me confidence and comfort, but I ended up following Eric's instructions that were taylored to my circumstances. All the physical setup was done and working. So, I was just sitting in my terminal session back at my desk doing software stuff. Eric had recommended labeling one disk with 20Mb for slice 7 (for metadbs, just in case), then the rest for slice 0. Use prtvtoc /dev/rdsk/c3t0d0s0 | fmthard -s - /dev/rdsk/c3t1d0s0 and etc through all disks on the D1000 to obtain identical labels. The off-the-shelf configuration appeared useable and similar, so I went with that. It had 80Mb for 0 and 1 with the remainder (about 33Gb) for slice 6. I used format partition print to get a slightly more readable view for myself than the prtvtoc output. then I decided to go with that. I used metastat metastat d0 metadb etc to get a view of what the existing setup was. the root disk was mirrored and there is an older external array that is raided. there were adequate replicas of the state database already. I'll straighten it out to be more optimal later. I chose d20 for the mirror and d21 and d22 for the stripes that would make the mirror. then metainit d21 1 6 c2t0d0s6 c2t1d0s6 c2t2d0s6 c2t3d0s6 c2t4d0s6 c2t5d0s6 -i 32k (I got the -i 32k from the link that Darren sent for a page in the Solstice DiskSuite User's Guide at docs.sun.com. my application is for scanned image storage, so the larger size seemed better.) metainit d22 1 6 c3t0d0s6 c3t1d0s6 c3t2d0s6 c3t3d0s6 c3t4d0s6 c3t5d0s6 -i 32k metainit d20 -m d21 metattach d20 d22 which completed the DiskSuite process. then newfs /dev/md/rdsk/d20 which took a couple of hours. then cd / mkdir array2 mount /dev/md/rdsk/d20 /array2 then vi /etc/vfstab since I already had a line for an array1 that had been created using the graphical DiskSuite tool, I just did a yy and p to copy that line. then changed the 1's to 2's. so, it's like this: /dev/md/dsk/d10 /dev/md/rdsk/d10 /array1 ufs 3 yes - /dev/md/dsk/d20 /dev/md/rdsk/d20 /array2 ufs 3 yes - Finally, since I'm dealing with windows users and samba, go into samba and set up a new mount point. Also adjust permissions for /array to give proper access for the group that bought it. Done!! Thank you very much. --------------- Chris Hoogendyk -- O__ ---- Network Specialist & Unix Systems Administrator c/ /'_ --- Library Information Systems & Technology Services (*) \(*) -- W.E.B. Du Bois Library ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- -----Original Message----- From: Chris Hoogendyk [mailto:choogend at library.umass.edu] Sent: Sunday, April 13, 2003 18:15 To: Sun Managers Subject: configuring D1000 from command line I feel a bit foolish asking for help, but I'm under a very tight deadline. I'm working on a system that I really should rebuild, but I cannot take the time now, so I have to work with it as is. It is an E250 running Solaris 7. It was configured almost 5 years ago with inadequate /var space, and some things have gotten broken. In particular, I can't use the graphical disksuite tool, so I'm stuck back at the command line. I installed a new dual channel differential ultra SCSI PCI card and hooked it up to a new D1000 filled with twelve 36G drives. I put the card in the top PCI slot which has a different PCI bus, so it should give me the best throughput. I used both SCSI channels, hooking up the D1000 as though it were two hosts each taking half of the D1000. Both halves terminated. I did a "boot -r". If I look at /dev/rdsk now, I have six targets (0-5) on c2 each d0 with s0 through s7. I also have six targets (0-5) on c3 each d0 with s0 through s7. Seems to have worked just fine. I want to do raid 10. In other words create 6 mirrors, strip them, and then mount all that as a single directory. I already have a mirrored root disk, two other internal disks and a 6x18G external array done as a raid 5, all using the graphical disksuite tool several years ago. There are probably several disksuite database files distributed through those. What I need is a real succinct step by step for command line configuration. format drives first? how to partition for this situation? metadb ... metattach several times metainit ... newfs ... mount ... configure so it will come up at reboot I have found the disksuite documenation online at docs.sun.com, but it is a lot of reading and I have to have this running tomorrow morning. I am going through it, and I will use it for backup and filling in. TIA --------------- Chris Hoogendyk -- O__ ---- Network Specialist & Unix Systems Administrator c/ /'_ --- Library Information Systems & Technology Services (*) \(*) -- W.E.B. Du Bois Library ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bennyanderson at ireland.com Mon Apr 14 19:48:26 2003 From: bennyanderson at ireland.com (Benny Anderson) Date: Mon, 14 Apr 2003 15:48:26 -0800 Subject: SUMMARY: StorEdge T3 password reset Message-ID: <20030414234826.3144.qmail@ireland.com> Hey thanks a million to all the following: Zaigui Wang Moises Nicolas ceferino santander III Steve Armijo Jim Ennis Problem: I am currently working on a StorEdge T3 Disk Array which was obtained through acquisition, and I do not know the current root password. Is there anyway to reset this password at the EPROM level, or through tftpboot. I am not concerned with saving any info on the current configuration. Solution: The T3 serial connection kit is special and you might have to order one from Sun (about $40 - USD) if you did not get one with the T3. 1. Stop all I/O going to the T3(s). 2. Connect a console cable from the master unit to a video terminal and power off all units (both trays in a partner group). 3. Power up the master unit (leave alternate master off) and interrupt the boot process by hitting RETURN when you see the following message: "hit the RETURN key within X seconds to cancel..." NOTE: If you leave the alternate master on, the T3 Operating System (OS) will continue to boot. Eventually, the OS will assume there is a problem with the master (that we interrupted), disable it, and make itself the master. 4. At the "T300-EP" prompt, use the 'set passwd' command to view (and memorize...) the password. 5. Type 'reset' to reboot, and immediately power on the alternate master (if present). 6. If you choose, you can change the password after it boots with the 'passwd' command. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mbruntel at att.com Tue Apr 15 08:50:17 2003 From: mbruntel at att.com (Bruntel, Mitchell L, SOLCM) Date: Tue, 15 Apr 2003 08:50:17 -0400 Subject: A1000/E3500 Duh answered- thanks...(summary) Message-ID: <5F2B8267D7B55B47BDE2868D227BF50C062BAFF7@OCCLUST01EVS1.ugd.att.com> Thanks for the quick sanity check all: 1) To those who inquired, yes, I have the correct SCSI controller ( a differential scsi controller and cables...) 2) Yes, I do know what happens when you ASS/U/ME something, and was HOPING that my duh (as I hypothized) was correct). Bottom line: yes, As I thought, it IS plug and pray especially on a new install. It was suggested that technically, the boot -r command (at ok prompt) will * "recreate" the /dev and /devices structure, so newly added devices will be discovered and added to the os. and that 3) find/get a copy of of RAID Manager --(ver 6.22) Thanks again, happy holidays to those who celebrate this week! mitch -----Original Message----- From: Hichael Morton [mailto:mh1272 at yahoo.com] Sent: Tuesday, April 15, 2003 8:40 AM To: Bruntel, Mitchell L, SOLCM Subject: Re: A1000/E3500 Duh question? "assume" you do what it stands for, don't you! you will have to do a reconfiguration boot for any new hardware to be reconizaed. you will need the correct scsi controller. you mentioned the cables but not the controller. the onboard controller in the e3500 will not work with the a1000. you will need a differential scsi controller. the sun engineer handbook is online at sunsolve.sun.com. the information on the scsi controller (and other hardware) can be found in the se handbook. --- "Bruntel, Mitchell L, SOLCM" wrote: > This is likely a DUH RTFM question, but (it's SO big and so much > stuff).. > > Here's the dumb question. > Presuming I use the RIGHT (scsi) cables on both sides. AND > if I install a A1000 on a "brand new" (newly turned on/formatted) > E3500 > (and freshly installed OS) > > Is it correct to presume that it should see the new disk array as > simply additional SCSI, and > "AUTOMaGICALLY" (my favorite computer word!) > adds it into the configuration, and would thus > negate most of the configuration issues in installing this on a new > system? > > (and I can then at my leisure, do the "hard" stuff, like deciding > where and > how > to mount my items > > Sorry for seeming so stupid, but I've been up working all night on > this... > Mitch > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Jim.Coleman at corporate.ge.com Tue Apr 15 10:14:21 2003 From: Jim.Coleman at corporate.ge.com (Coleman, Jim (CORP)) Date: Tue, 15 Apr 2003 10:14:21 -0400 Subject: SUMMARY - OFF TOPIC POST - HP/UX Mailing List Message-ID: My thanks to: 'Boss' Shawn Evans Sebastien Daubigne Steve Hunter Gaziz Nugmanov Frank Velazquez Jim Zimmerman Responses, in no particular order, were: http://www.dutchworks.nl. Subscriptions go to majordomo at dutchworks.nl. http://hpux.ee.ualberta.ca/hppd/discuss.html (subscriptions to majordomo at cv.ruu.nl http://www.interex.nl/ (subscribe on line). http://isc.faqs.org/faqs/hp/hpux-faq/section-28.html (submissions to hpux-admin at dutchworks.nl) http://marc.theaimsgroup.com lists resources for *multiple* operating systems! -------------------------------------------- Jim Coleman Unix Administrator GE Corporate Global Computer Operations 214 Seventh Street, Parkersburg, WV 26101 304 424-5409, DC: 8*572-5409, Fax: 304 424-5499 Jim.Coleman at corporate.ge.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sun_consultant at hotmail.com Tue Apr 15 15:09:16 2003 From: sun_consultant at hotmail.com (Sun Consultant) Date: Tue, 15 Apr 2003 12:09:16 -0700 Subject: SUMMARY: GBIC Slot on Sun 4500? References: Message-ID: Thanks for all the replies---The questions was: > Hi on the Sun E4500 I/O board, there are slots for SBUS adapters, plus there > are two GBIC slots...Can these GBIC slots be used for connection to an A5200. The answer is yes, these GBIC slots support the Storedge A5200.. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From showie at uoguelph.ca Tue Apr 15 16:23:07 2003 From: showie at uoguelph.ca (Steve Howie) Date: Tue, 15 Apr 2003 16:23:07 -0400 Subject: SUMMARY: can't read changes made to filesystem on XP 512 made by another system Message-ID: <3E9C6A2B.1010304@uoguelph.ca> Many thanks to those who replied - Steve Mickler Bertrand Hutin Carl Ma Casper Dik Jay Lessert Darren Dunham The consensus is that this is not a good thing at all when using UFS file systems (which I am), since a file system mounted R-O is assumed to never have changes made to it. Directory inconsistencies can result, causing system panics if you're not careful. The suggested alternatives included: - Mount the file systems R/W to only one host, then export them to the other host using NFS and perform any updates from there. - Making use of a multi-mount capable filesystem such as Sun SAM-FS Veritas Cluster Filesystem ADIC StorNext Filesystem {Thanks to Darren Dunham) Once again thanks to those who responded Steve Howie University of Guelph ------------ original message --------------------- Howdy, We have a SunFire V880 with a number of file systems mounted on an XP512 SAN via a JNI card going through a Brocade switch. These file systems consist of relatively static data, and are mounted Read-Write should any changes need to be made. System "A" An E450 system also accesses these same file systems using a JNI card going through the Brocade switch and has them mounted Read-Only. System "B". Everything is working fine and the file systems are visible to both systems. However, it seems than changes made to System "A" do not show up as visible on System "B". I touched a file on A and there was no sign of it on B. Am I correct in assuming that this is just a caching issue on the XP512? It has a very large cache indeed accordng to our SAN guy. Any pointers appreciated, _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnb at mcmaster.ca Tue Apr 15 16:29:42 2003 From: johnb at mcmaster.ca (john benjamins) Date: 15 Apr 2003 16:29:42 -0400 Subject: SUMMARY: adding internal 73GB disks to V880 In-Reply-To: <1050427188.15008.114.camel@ockham> References: <1050427188.15008.114.camel@ockham> Message-ID: <1050438581.15008.151.camel@ockham> Hi, boy that was quick. got a flurry of answers that fell into one of two categories. a few people said that my original plan should be fine. however, the other group all said basically the same thing, which is best summed up by a FAQ on Sun's own web site. the answer is to be found at http://www.sun.com/servers/entry/880/faqs.html#3, which says the following: 3. How can I expand the internal disk array from six to twelve disks? The system comes standard with a single disk backplane which will support the first six drives. To expand beyond six, to a maximum of twelve disks a 6-bay internal expansion FC-AL disk backplane with either six 73 GB or six 36.4 GB disks will need to be ordered and installed. so, what they were trying to sell me was the 6 drives, PLUS the backplane needed to support the additional drives. thank you to everyone who was so quick to respond. it's very much appreciated. cheers, -john -- john benjamins LTRC - Teaching Infrastructure Support 905-525-9140 x27492 Thode Library B117B On Tue, 2003-04-15 at 13:19, john benjamins wrote: > Hi, > > we are thinking of adding some additional disk to our V880 (4GB RAM, six > 73GB internal disk, Solaris 8). so i called the Sun sales number for a > quote, and they said i had to buy six 73GB or six 36GB disks to add > disks to the V880. but i only want 2 more 73 GB disks! can i not buy > two 73GB FC-AL disks (X6742A) and have that be a supported > configuration? has anyone else added additional 73GB disks to a V880? > am i missing something obvious here? > > thanks! -john _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sbala at citadon.com Tue Apr 15 17:08:51 2003 From: sbala at citadon.com (Srini Bala) Date: Tue, 15 Apr 2003 14:08:51 -0700 Subject: Summary : Ultra Sparc IIIi processors - comparison to Sparc III Message-ID: <0FEA43415756E44AB368AC58C1A4F4F3011B4C01@a4mail2k01.citadoncorp.com> Thanks to everyone who contributed (Special thanks to Jay Lessert) 1. Sparc IIIi processors will be comparable (thanks to 1 MB L2 cache) to Sparc III, for applications which do not require intensive computation processing. For computation intensive applications like Oracle, Sparc III's external cache will come in handy. For Weblogic type of applicaiton servers, IIIi could give 85-100% of the performance of III. Actual benchmarking is required to make final decisions. Sun might relase sun benchmark comparisons this week. 2. Cache is *fuly* used. *everything* the applications touch, executables and data, are read into cache before being read by the processor, and written to the cache before being written to main memory. So unless process footprint is < 8MB, you're using it all. -Srini -----Original Message----- From: Srini Bala Sent: Tuesday, April 08, 2003 6:19 PM To: sunmanagers at sunmanagers.org Subject: RE: Ultra Sparc IIIi processors - comparison to Sparc III All, A related question would be : How could you find out the actual usage of the external cache on your CPUs? We have 280Rs with 2x900MHz processors (with 8MB external cache each) and I want to find out how much of the external cache is really getting used. Is it possible? Thanks again BTW, I have gotten couple of feedbacks on the first question and will eventually summarize the results. -Srini -----Original Message----- From: Srini Bala Sent: Tuesday, April 08, 2003 3:02 PM To: sunmanagers at sunmanagers.org Subject: Ultra Sparc IIIi processors - comparison to Sparc III All, The newly released Sparc IIIi processors have 1MB L2 cache (integrated) but no other external cache, compared to 8MB of external cache available in Sparc III processors (which have 512K L2 cache). Any idea how would this affect/improve performance? Thanks in advance... -Srini _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bin.lee at china-core.com Tue Apr 15 20:57:27 2003 From: bin.lee at china-core.com (Lee,Bin) Date: Wed, 16 Apr 2003 08:57:27 +0800 Subject: SUMMARY: What's the meaning of the error message ? Message-ID: <20030416084933.B4E3.BIN.LEE@china-core.com> Many thanks to Broun,Bevan for his quick response. My problem is > Could you kindly help me to get meaning with below error message, and > what's solution for the problem? > > Apr 15 11:00:00 benz savecore: [ID 570001 auth.error] reboot after > panic: free:freeing free frag, dev:0x7600000004, blk:149, cg:2629, > ino:15311307, fs:/export/home The answer should be, > A corrupt file system(/export/home) cause system panic and reboot. To > umount the system and fsck resolve the problem. Thanks again for Broun and all of your kindly support. :-) Best Regards, Lee,Bin _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From pattysunmanagers at hotmail.com Tue Apr 15 23:19:26 2003 From: pattysunmanagers at hotmail.com (Patty) Date: Tue, 15 Apr 2003 23:19:26 -0400 Subject: SUMMARY: e15k scalability Message-ID: I got a couple of answers basically saying that there are no obvious concerns regarding vertical scability of e15k for Oracle processes. Many thanks to all who answered... Patty _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bda at aime.insa-tlse.fr Wed Apr 16 08:02:59 2003 From: bda at aime.insa-tlse.fr (Philippe Bourdeu d'Aguerre) Date: Wed, 16 Apr 2003 14:02:59 +0200 Subject: SUMMARY: Trouble with external SCSI on SUN Blade 2000 In-Reply-To: <200304151542.38634.bda@aime.insa-tlse.fr> References: <200304151542.38634.bda@aime.insa-tlse.fr> Message-ID: <200304161403.00034.bda@aime.insa-tlse.fr> Many thanks to those who give advices: Juergen Waiblinger Dominic Clarke Luc I. Suryo Frangois Legal I retain explanation from Juergen Waiblinger > Hello, > > i have seen the same things (with external cdrom and dat) using the wrong > cables. You need a hd68 to hd50 cable which has termination inside. They > are a little bit more expensive but than it should work. Keeep in mind, > that you have 8 wires not used. If you have the possibility to check the > thing with a windows box you will see that the dat is working as the > lowlevel scsi driver from MS won't recognize the problem (but runs in > problems when you put data on the tape) > If you nare using a correct cable i would assume the DAT to be faulty. > btw. as far as i know in France you can get that from transtec > (www.transtec.fr) it's called CB-WT2SCSI-1. But I will not try this solution because I had success with another solution: I extracted the DAT from his box and installed it inside the SUN. Internals SCSI connectors was the same and it seems to work smoothly. On Mardi 15 Avril 2003 15:42, Philippe Bourdeu d'Aguerre wrote: > Hello, > > I try to install an external DAT on a SUN Blade 2000. It's a SEAGATE > STD124000N, DAT DDS3 with SCSI 2 interface. I connect it through a > HD68-HD50 cable but it doesn't work: > - active terminator doesn't light on > - on "boot -r" I get error messages: > Apr 14 10:44:35 limoux scsi: [ID 107833 kern.warning] WARNING: > /pci at 8,700000/scsi at 6,1 (glm1): Apr 14 10:44:35 limoux Connected command > timeout for Target 0.0 > Apr 14 10:44:35 limoux genunix: [ID 408822 kern.info] NOTICE: glm1: fault > detected in device; service still available Apr 14 10:44:35 limoux genunix: > [ID 611667 kern.info] NOTICE: glm1: Connected command timeout for Target > 0.0 Apr 14 10:44:35 limoux glm: [ID 280919 kern.warning] WARNING: > ID[SUNWpd.glm.cmd_timeout.6017] Apr 14 10:44:35 limoux scsi: [ID 107833 > kern.warning] WARNING: /pci at 8,700000/scsi at 6,1 (glm1): Apr 14 10:44:35 > limoux timeout on bus reset interrupt > Apr 14 10:44:35 limoux genunix: [ID 408789 kern.warning] WARNING: glm1: > fault detected in device; service unavailable Apr 14 10:44:35 limoux > genunix: [ID 451854 kern.warning] WARNING: glm1: timeout on bus reset > interrupt > > - devices are not created > > Can somebody confirms that SunBlade 2000 supports SCSI 2 devices ? Is there > any necessary conf ? > > Thanks. -- Philippe BOURDEU d'AGUERRE AIME - Campus de l'INSA http://www.aime.insa-tlse.fr/ 135, av. de Rangueil Tel +33 561 559 885 31077 TOULOUSE Cedex 4 - FRANCE Fax +33 561 559 873 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From GSchoep at storm.L-3Com.com Wed Apr 16 09:21:00 2003 From: GSchoep at storm.L-3Com.com (Schoep, Grant @ STORM) Date: Wed, 16 Apr 2003 09:21:00 -0400 Subject: SUMMARY: Reboot of server, tape drive keeps on "incrementing" in in /dev/r mt Message-ID: Sorry for the late summary. I decided to take the weekend off! Ok, my main question was sometimes when I do a "boot -r" Solaris just decides to increment the device number of my tape drive. I just wanted to know how to reset this. Two ways, one with a reboot, one without. I haven't tried the without method however. # cd /dev/rmt # rm * At this point you could reboot(boot -r) or run the following. # drvconfig # tapes # devlinks # /usr/ucb/ucblinks -grant -----Original Message----- From: Luc I. Suryo [mailto:luc at suryo.com] Sent: Tuesday, April 08, 2003 11:09 AM To: Schoep, Grant @ STORM Subject: Re: Reboot of server, tape drive keeps on "incrementing" in in /dev/r mt > I think we have a tape drive going bad(its really acting flaky now > preventing good backups hence my last email). > > But almost everytime I do a boot -r (reconfigure), on the box, the OS > sometimes decideds to increment the device in /dev/rmt. Meaning. Just > yestereday it was /dev/rmt/5, now after a boot -r, it is /dev/rmt/6. Its > done this before which is why I was up to 5 to begine with. > > So. Other than the drive going bad, is there away I can make it NOT do this? > Or at least make it reset back down to 1? I was thinking I could probably > boot to single user mode, remove the contents of /dev/rmt, and then do a > touch reconfigure, or boot -r from Openboot to rebuild the tree from > scratch. > > Will that work? yes, done that maytime... > > Using Solaris 2.6(latest patches as of about 5 months ago) on an Ultra 2 > with about 20 disks hanging off it(our heavily abused machine) hmm 2.6 :( have you check if there is ANY st. patch available that you might not have installed yet??? i had major problem with 2.6 (tape issue only).. glad i'm running 2.8 now ;) -- Kind regards, Luc Suryo _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From pattyatt at hotmail.com Tue Apr 15 23:07:31 2003 From: pattyatt at hotmail.com (Patty Eyenga) Date: Tue, 15 Apr 2003 23:07:31 -0400 Subject: SUMMARY sun e15k scalability References: Message-ID: I got a couple of answers basically saying that there are no concerns regarding vertical scability of e15k for Oracle processes. Many thanks to all who answered... Patty ----- Original Message ----- From: "Patty" To: "Sun Managers" Sent: Monday, April 14, 2003 6:07 PM Subject: sun e15k scalability > are there any limitations or concerns for vertical scaling of e15k boxes as > opposed to other sun boxes? i'm thinking in particilar regarding Oracle > processes.... > > thanks > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Wed Apr 16 12:26:40 2003 From: martini at mrpeabody.llnl.gov (Dave Martini) Date: Wed, 16 Apr 2003 09:26:40 -0700 (PDT) Subject: SUMMARY: swap and /tmp don't match up Message-ID: <200304161626.h3GGQeP05063@raider.llnl.gov> Thanks to all who replied and most people pointed me to an entry that I had in my vfstab file that was limiting the size of /tmp swap - /tmp tmpfs - yes size=100m I removed this from the vfstab file and rebooted and now both swap entries look show the same size. swap 1908616 8 1908608 1% /var/run swap 2057640 149032 1908608 8% /tmp Dave Martini LLNL _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mikelist at sky.net Wed Apr 16 12:52:25 2003 From: mikelist at sky.net (Mike's List) Date: Wed, 16 Apr 2003 11:52:25 -0500 (CDT) Subject: SUMMARY: Attempted hacking??? Message-ID: Thanks for all the replies, most agreed that the three lines below indicate nimda/code red worms/virus, whatever, but mainly affecting Windows running IIS. No need to worry if running Solaris/Apache, just pay attention to the log file not being too full. - Mike ---------- original message ---------- My access log for apache is full of the three lines below, is someone attempting to hack my web server thinking it's a Windows machine??? Solaris running Apache... x.x.x.x - - [13/Apr/2003:21:14:23 -0500] "GET /default.ida?XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXX%u9090%u6858%ucbd3%u7801%u9090%u6858%ucbd3%u7801%u9090%u6858%ucbd3%u7801%u9090%u9090%u8190%u00c3%u0003%u8b00%u531b%u53ff%u0078%u0000%u00=a HTTP/1.0" 404 164 x.x.x.x - - [13/Apr/2003:22:38:34 -0500] "GET /scripts/..%255c%255c../winnt/system32/cmd.exe?/c+dir" 404 - x.x.x.x - - [13/Apr/2003:22:39:24 -0500] "GET /scripts/..%255c%255c../winnt/system32/cmd.exe?/c+dir" 404 - Thanks. - Mike _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Wed Apr 16 13:28:12 2003 From: martini at mrpeabody.llnl.gov (Dave Martini) Date: Wed, 16 Apr 2003 10:28:12 -0700 (PDT) Subject: SUMMARY: How to tell if machine is 64bit or 32bit? Message-ID: <200304161728.h3GHSCn05173@raider.llnl.gov> Thanks for all the quick responses and most pointed me to the isainfo command. /usr/bin/isainfo -b Tells you what mode 32-bit or 64-bit the machine is running in. /usr/bin/isainfo -v Tells you if you have the capability of running 32 or 64. Here's a link I found on the sun site that describes how to boot in 32 and 64-bit modes and other good info. http://soldc.sun.com/articles/64_bit_booting.html#Q6 Dave Martini LLNL _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Rainer.Heilke at atcoitek.com Wed Apr 16 15:52:58 2003 From: Rainer.Heilke at atcoitek.com (Heilke, Rainer) Date: Wed, 16 Apr 2003 13:52:58 -0600 Subject: SUMMARY: SEAM, pam_krb5.so.1 and failover Message-ID: > -----Original Message----- > From: Houle, Michael > Sent: Wednesday, April 16, 2003 1:52 PM > To: Heilke, Rainer > Subject: SUMMARY: SEAM, pam_krb5.so.1 and failover > > Sorry folks, > > I had a hole in my head. the library is contacting both KDC's, but > somehow it doesn't > seem to like the answer. If I get an answer later on, I can post > another summary. > > Thanks, > Mike. > > -----Original Message----- > From: Houle, Michael > Sent: Tuesday, April 15, 2003 3:17 PM > To: 'sunmanagers at sunmanagers.org' > Subject: SEAM, pam_krb5.so.1 and failover > > Hi Sun managers, > > Anyone heard of this problem with SEAM's pam_krb5.so.1 under Solaris > 8? > > pam_krb5.so.1 doesn't seem to query more than 1 KDC before giving up. > > We're doing some fail-over testing and we've noticed that when we > shutdown > our primary kdc, pam_krb5.so.1 will not allow users to login. Our > linux boxes work just > fine so we know that the secondary kdc has good data and works. > > If we temporarily reverse the order of our kdc lines, we can > authenticate to > our slave kdc just fine under solaris. > > Our krb5.conf is: > > [libdefaults] > clockskew = 5 > ticket_lifetime = 600 > default_realm = TEST.CA > default_tkt_enctypes = des-cbc-crc > default_tgs_enctypes = des-cbc-crc > > [realms] > TEST.CA = { > kdc = kerberos.test.ca:88 > kdc = kerberos-1.test.ca:88 > admin_server = kerberos.test.ca:749 > kpasswd_protocol = SET_CHANGE > default_domain = test.ca > } > > [domain_realm] > .test.ca = TEST.CA > test.ca = TEST.CA > > [appdefaults] > kinit = { > renewable = false > forwardable= true > } > > List steps to reproduce problem (if applicable): > 1 shutdown primary KDC (kerberos.test.ca) > 2 try to telnet to box and enter password. > > authentication fails. When the primary KDC is back online, > authentication > succeeds. > > Thanks, and I will post a summary as soon as I can. > Mike. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Ebru.Tan at posta.mam.gov.tr Thu Apr 17 08:49:48 2003 From: Ebru.Tan at posta.mam.gov.tr (Ebru.Tan at posta.mam.gov.tr) Date: Thu, 17 Apr 2003 15:49:48 +0300 Subject: SUMMARY: to extract files from tape Message-ID: thanks all who replied my question. my question was "I have DDS2 tapes. I want to read these tapes by using dds3 tape driver. It has data about sea. There is no any explanation about how I can extract. But there are some notes on them. For example, box1.tar.gz. I understand from this note, there is a file in tape which name is box1.tar.gz. and I tried to extract this tape by using tar command: tar xvf /dev/rmt/0n but I get this message: 'tar: checksum error' ". the answer is to use GNUtar command. to obtain this file, I download tar file for solaris2.8 from HYPERLINK "http://www.sunfreeware.com"www.sunfreeware.com I tried and it worked. thanks again. ------------------------ Ebru TAN E-Mail : ebru.tan at posta.mam.gov.tr TUBITAK Marmara Research Center Earth and Marine Sciences Research Institute Pbox:21 41470 Gebze/KOCAELI Is tel :+90 262 641 23 00 ext:4321 Is fax :+90 262 641 23 09 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.459 / Virus Database: 258 - Release Date: 25.02.2003 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Asif.Ahmad at marconiselenia.com Thu Apr 17 09:30:10 2003 From: Asif.Ahmad at marconiselenia.com (Asif.Ahmad at marconiselenia.com) Date: Thu, 17 Apr 2003 14:30:10 +0100 Subject: Summary: Async memory fault on Sparc 5 running SunOS 4.1.3_U1 Message-ID: Thanks Peter Stokes and Bill Voight. I had two suggestions to solve the problem: 1) Memory may be sensitive. Try another motherboard. 2) Reseat and switch the memory DIMMs over. If it persists it could be the motherboard. As I have tried various DIMMS in various machines and the problem occurs every few months. I have put new memory in two machines so I will do bit more testing and see what happens. If there are any more suggestions Please let me know. Regards Asif ------------------------------------------------------------------------------------------------------------------------------------------------------------- The Problem: Hello all I am running SunOS 4.1.3_U1 on Sun Sparc 5s. Time to time I get the Async Memory fault ; Panic on CPU 0 and machine reboots. I have install the jumbo patch 101508-15 and 101592-09 and replaced the memory SIMMS number of times in various machines but I am still getting the fault. Initially I use to get Watchdog error and I installed the patch 101508-15 and 101592-09 for that and since than I never had a watchdog reset. But async memory fault is still a big problem. Can some one get me out this jigsaw. Many thanks: The early messages for Async memory fault: vmunix: Async memory fault mfsr=0x81802860 mfar=0xbb75598 vmunix: panic on cpu 0: async memory fault vmunix: syncing file systems... VAC ENABLED ..... ...... ...... and machine reboots Asif _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Thu Apr 17 18:16:55 2003 From: martini at mrpeabody.llnl.gov (Dave Martini) Date: Thu, 17 Apr 2003 15:16:55 -0700 (PDT) Subject: SUMMARY: Does PC NetLink Work with Active Directory?? Message-ID: <200304172216.h3HMGtm06561@raider.llnl.gov> The answer that I received from the one person that replied: No, PC NetLink emulates a NT-style domain. It has no knowledge of 2000-style Active Directory. It looks like since I don't have a PDC it won't work as far as trying to join my UNIX domain to my PC domain. I was told (by a SUN tech) I could try to have the PC domain join the UNIX domain/PDC but have not tried doing this yet. Dave Martini LLNL Here's my original question: We have an PC domain that is an active directory domain. Our domain > name is epd but we don't have a PDC (Primary Domain Controller) as I understand > it (This is what my NT administrator has explained to me) All our PC clients arerunning windows 2000. Our main PC server is running Windows 2000 server. > > When I try the joindomain command as shown below it fails when I put the name > of the domain in. Does anyone know if PC NetLink version 2.0 works with Active Directory? Is it wanting the name of a PDC on this question or a domain? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zaigui at yahoo.com Fri Apr 18 02:44:08 2003 From: zaigui at yahoo.com (Zaigui Wang) Date: Thu, 17 Apr 2003 23:44:08 -0700 (PDT) Subject: Summary: Solaris package modification and revalidation In-Reply-To: <20030418000627.57017.qmail@web12302.mail.yahoo.com> Message-ID: <20030418064408.71882.qmail@web12308.mail.yahoo.com> Crist Clark provided the instant solution to my problem: when you modify a package, the checksum and file size need to be updated also in the pkgmap file. To get the file size, do "ls -l". to get the checksum, use "sum" command. Thanks Crist for the quick and easy solution! In addition, I also got some suggestions from Rob De Langhe, who suggested "see man pkgmk" and Ryan A. Krenzischek, who gave some more details: --- You'll need to re-run the find . -print | pkgproto > prototype To re-map the directories, files, perms, and check sums. --- I have not yet tested this out. But for information on how to create a package from scratch, I found this document on sunsolve: http://sunsolve.sun.com/private-cgi/retrieve.pl?doc=infodoc%2F18303&zone_32=pkgproto Zaigui --- Zaigui Wang wrote: > Managers, > > If a modify a package (for example, modify the > checkinstall script), what should I do to > rebuild/revalidate the package so that it is still > valid (like checksum and stuff)? > > Thanks. > > Zaigui > The New Yahoo! Search - Faster. Easier. Bingo > http://search.yahoo.com > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers The New Yahoo! Search - Faster. Easier. Bingo http://search.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bodenste at mailcenter.csap.af.mil Fri Apr 18 10:53:24 2003 From: bodenste at mailcenter.csap.af.mil (Cathy M. Bodenstedt) Date: Fri, 18 Apr 2003 09:53:24 -0500 Subject: SUMMARY: newfs on s2 Message-ID: <3EA01164.B8F637D@mailcenter.csap.af.mil> Consenses is NO, which I pretty much knew. Any recommendations on using the entire disk without going into format? -- Phone: 210-925-3244 DSN: 945-3244 [demime 0.99c.7 removed an attachment of type text/x-vcard which had a name of bodenste.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bodenste at mailcenter.csap.af.mil Fri Apr 18 12:09:40 2003 From: bodenste at mailcenter.csap.af.mil (Cathy M. Bodenstedt) Date: Fri, 18 Apr 2003 11:09:40 -0500 Subject: SUMMARY2: newfs on s2 Message-ID: <3EA02343.2EBBAAF8@mailcenter.csap.af.mil> Just in case this comes in handy for someone. The orginal posting was, "Can I create a partition on s2 with newfs. " I needed a script that would set up a second disk with one partition that was the size of the entire disk. It may not be pretty but, it works. #!/bin/ksh #################################################### #Creates /newdir partition. #################################################### #Variables ROOT_DEVICE=`df -k |head -2 |grep dev | awk '{print $1}'|cut -d "/" -f4 |cut -d s -f1` DISK=c0t1d0 DISK_RAW=/dev/rdsk/${DISK} DISK_SIZE=`/etc/prtvtoc ${DISK_RAW}s2 |grep "2 5 00 0" |awk '{print $5}'` DISK_LS=`/etc/prtvtoc ${DISK_RAW}s2 |grep "2 5 00 0" |awk '{print $6}'` TMP=/tmp/disk_vtoc echo "* First Sector Last" > $TMP echo "* Partition Tag Flags Sector Count Sector Mount Directory " >> $TMP echo "2 5 00 0 ${DISK_SIZE} ${DISK_LS}" >> $TMP echo "6 0 00 0 ${DISK_SIZE} ${DISK_LS}" >> $TMP /usr/sbin/fmthard -s $TMP ${DISK_RAW}s2 echo "" echo "#######################################" echo "Current Disk Layout for ${DISK_RAW}s2." echo "#######################################" /etc/prtvtoc ${DISK_RAW}s2 echo "" echo "" echo "" newfs ${DISK}s6 if [ ! -d /newdir] ;then mkdir /newdir fi echo "" echo "" echo "#################################" echo "Creating entry in /etc/vfstab." echo "for /newdir" echo "#################################" echo "/dev/dsk/${DISK}s6 /dev/rdsk/${DISK}s6 /newdir ufs 2 yes logging" >> /etc/vfstab mount /newdir df -k |grep /newdir -- Phone: 210-925-3244 DSN: 945-3244 [demime 0.99c.7 removed an attachment of type text/x-vcard which had a name of bodenste.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mohamed at your-site.com Mon Apr 21 12:34:09 2003 From: mohamed at your-site.com (Mohamed Lrhazi) Date: Mon, 21 Apr 2003 12:34:09 -0400 Subject: Summary: How to RAID1+0 with Solaris 9 SVM References: <200304191706.KAA06860@redwood.taos.com> Message-ID: <3EA41D81.60008@your-site.com> Thanks for all who responded to me, especially Mike, mikelist at sky.net, who pointed out that he asked the same question and posted a summary last month. More info will be found in his summary. Sorry for missing that and resending the question. In short: Q: How to set up RAID1+0 using Solaris Volume Manager A: Just set up RAID 0+1, using same number of stripes and stripes of same size, the software will set up RAID1+0, and won't tell you it did! There is no way to ensure what you built was RAID1+0 and not 0+1, except from unplugging disks. dirk.boenning at ccf.de said he tested it himself. This behaviour is not documented anywhere, apart from the few sentences in SVM user guide that I pasted in my original post. alexei at soemail.rutgers.edu got me excited suggesting that A5000 have hardware RAID cards built in.... It turns out that is not correct, A5x00 do not have hardware RAID. One response, rich.teer at rite-group.com, claimed that what I configured was really RAID 0+1... I am not willing to unplug my disks right now in order to verify , I just hope he is wrong :) Thank you all, Mohamed~ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From meg991 at yahoo.com Mon Apr 21 19:00:04 2003 From: meg991 at yahoo.com (Meg Wall) Date: Mon, 21 Apr 2003 16:00:04 -0700 (PDT) Subject: SUMMARY: Jumpstart problem In-Reply-To: <001701c30814$3a18fc70$c3dc1e43@drake> Message-ID: <20030421230004.31874.qmail@web13506.mail.yahoo.com> > > Hi Managers, > > > > Thanks to the people who replied me, they are > > > > Jonathan V Hays > > Luc I. Suryo > > Cathy M. Bodenstedt > > Anton > > Aman Pannu > > > > None of the replies "hit the nail", but all of them helped me > > to check over everything, from /etc/bootparams to boot > > directory, netmask etc.. After I changed every hostname to ip > > address in /etc/bootparams file, the jumpstart started to > > work. I don't know why, the hostname used to be working in > > /etc/bootparams from other clients. > > > > Thanks!! > > > > Meg > > --- Meg Wall wrote: > > > Hi Managers, > > > > > > My jumpstart client won't talk to the server, can't figure > > out where > > > the problem is. > > > > > > Solaris 8, sparc., all servers on the same subnet, boot server, > > > profile server and configuration server, image server is the same > > > > server. All of the servers have two interface, one public ip, one > > > > internal ip (10.x.x.x). The jumpstart server hme1 is the internal > > > > interface, the client hme0 is the internal. > > > > > > I want the client talks to server through the internal interface. > > > > Every time, I get to this point and it hangs there: > > > > > > Requesting Internet address for 0:0:e:x:x:x > > > SunOS Release 5.8 Version Generic _xxxxxx-xx 64 bit > > > Copyright 1983-2001 Sun Microsystems, Inc. All rights reserved. > > > Configuring /dev and /devices Using RPC Bootparams for network > > > configuration information. Skipping interface hme1 > > > configured interface hme0 > > > / (it spins forever here) > > > > > > I did check documentation, believe I did profile right. The > problem > > > doesn't seem to get to profile part yet. My sysidcfg file, > > the part > > > I'm not sure, I wrote: network_interface=hme0 > > {netmask=255.255.252.0} > > > Should I write hme1 here? > > > > > > I tried both, none of them work, the same result. What could I > miss > > > here? I believe I add client right, -e, -s, -c, -p, all paths are > > > > right. > > > > > > Thanks, I will summarize. > > > > > > Meg -Sorry I forgot to write SUMMARY in the subject the first time. The New Yahoo! Search - Faster. Easier. Bingo http://search.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sridharapk at rediffmail.com Tue Apr 22 00:40:27 2003 From: sridharapk at rediffmail.com (sridhara pk) Date: 22 Apr 2003 04:40:27 -0000 Subject: SUMMARY: /dev/null cannot create. Message-ID: <20030422044027.14873.qmail@webmail7.rediffmail.com> An embedded and charset-unspecified text was scrubbed... Name: not available Url: http://www.sunmanagers.org/pipermail/summaries/attachments/20030422/ec37e41c/attachment.ksh From unixbeginer at hotmail.com Tue Apr 22 01:46:00 2003 From: unixbeginer at hotmail.com (beginer unix) Date: Tue, 22 Apr 2003 05:46:00 +0000 Subject: SUMMARY: problems backing up swap partition Message-ID: Hi all, Thanks for all the replies. Everybody wondered as to why I should be backing up a swap partition and explained what a swap partition is. I am aware of the fact that swap partition need not be backed up. But, we have a client, who backs up swap partition and insists on knowing the reason for the failure. One interesting thing I've noticed on the swap partition is that apart from a couple of regular files, we have 2 "door files". I want to know, if these would cause problem in backing up the partition Regards _________________________________________________________________ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From suganm at absa.co.za Tue Apr 22 04:44:03 2003 From: suganm at absa.co.za (Sugan Moodley) Date: Tue, 22 Apr 2003 10:44:03 +0200 Subject: SUMMARY: Strange environment variables In-Reply-To: <20030417121126.GA13423@alfa.absa.co.za> References: <20030401163240.13924.66066.Mailman@sunmgrs.cs> <20030417121126.GA13423@alfa.absa.co.za> Message-ID: <20030422084403.GA24582@alfa.absa.co.za> On Thu, Apr 17, 2003 at 02:11:26PM +0200, I wrote: >Greetings, > >I have a strange problem with environment variables. On a Sun E450 running >Solaris 7 Generic_106541-16 certain variables seem to exist but I cannot >find out where it is loaded from. Since the system has been up for over 230 >days software may have been removed or added that initially set these >variables. > >The variables in question are: > >CAIHPAD10=/CA/agents/data/hpa >AGENTWORKS_DIR=/CA > >I've checked /etc/profile ~/.profile but this is the only useful portion: > ># *---------------------------------------------- * ># Modifications made by setupNSM for ># Unicenter NSM Agent Technology ># *---------------------------------------------- * >if [ "$AGENTWORKS_DIR" = '' ] >then > AGENTWORKS_DIR="/ca" > export AGENTWORKS_DIR >fi > >if [ -f "$AGENTWORKS_DIR"/scripts/envset ] >then > . "$AGENTWORKS_DIR/scripts/envset" >fi > ># *---------------------------------------------- * ># End of Modifications made by setupNSM for ># Unicenter NSM Agent Technology ># *---------------------------------------------- * > >Somehow AGENTWORKS_DIR is being set somewhere but none of the usual places >contain this string. /bin/login, telnetd nor /bin/sh don't load these >variables. Creating a new user on the system and typing /usr/bin/env after >login shows these variables again. > >Any ideas on how to find where these variables are set? For the time being >forcing the setting in /etc/profile is the current solution but there must >be another explanation. > >thanks in advance. -------------------------------- Thank you for all your replies. I've decided to patch the system and schedule a reboot sometime soon which would destroy 241 days of uptime but then I have never really understood what all the hype was about a box being up for so long anyways :) A regular reboot (at most twice in 6 months) should be healthy for the box. The erroneous variables AGENTWORKS_DIR as well as CAIHPAD10 contain the wrong substring path: "/CA" which it should contain "/ca". These variables are in the INETD environment as mentioned by Darren Dunham but I am not sure how this is gonna help since the app is not spawned from there. However I should bounce the box after patching anyways since it's due for it. The "$AGENTWORKS_DIR/scripts/envset" script does not contain any setting related to the variables above and a grep on /etc/default/* comes up with nothing. Just a note to some people that the IF statement in the /etc/profile above will only set a variable if the variable is initially blank but it is not blank when /etc/profile is run therefore the inital incorrect value is preserved. Thank you all for your time and patience: Darren Dunham Pascal Grostabussiat Steve Sandau Ric Anderson Rich Kulawiec Frangois Legal Michael Schulte Crist Clark Johan Hartzenberg -------------------------------- On Thu, Apr 17, 2003 at 08:06:51AM -0700, Darren Dunham wrote: >Can you see if these variables are part of the inetd environment? > >ps -ef | grep inetd >/usr/ucb/ps eww -------------------------------- On Thu, Apr 17, 2003 at 02:23:17PM +0200, Johan Hartzenberg wrote: >Both those variables are used by TNG. They need to be defined for many TNG >agents and services binaries to run correctly. > >Both are normally set from /etc/profile, and they get added there by the >installation of TNG automatically. -------------------------------- On Thu, Apr 17, 2003 at 08:57:00AM -0400, Steve Sandau wrote: >It looks like the snippet you provide will set the variable >AGENTWORKS_DIR if it isn't already set. > >/etc/skel has system default .profile and .chhrc files. Beyond that, it >might be worth poking around in /etc/default... -------------------------------- On Thu, Apr 17, 2003 at 06:12:19AM -0700, Ric Anderson wrote: >Seems pretty obvious - the /etc/profile fragment >you included sets AGENTWORKS_DIR=/ca if >AGENTWORKS_DIR is not already set. Then, if the >file $AGENTWORKS_DIR/scripts/envset exists, it is >sourced by the current shell, and that could be the >place where CAIHPAD10 comes from. -------------------------------- On Thu, Apr 17, 2003 at 09:22:39AM -0400, Rich Kulawiec wrote: >Your article does not make it precisely clear what file contains this fragment: > >> # *---------------------------------------------- * >> # Modifications made by setupNSM for >> # Unicenter NSM Agent Technology >> and so on > >But if I understand you to say that this is in /etc/profile, then all users >(including new ones you've just created) will pick up the variables set there >when they login. -------------------------------- On Thu, Apr 17, 2003 at 03:29:09PM +0200, Frangois Legal wrote some HTML ONLY email that reads: You can check the files in /etc/default -------------------------------- On Thu, Apr 17, 2003 at 09:52:45AM -0500, Michael Schulte wrote: >> The variables in question are: >> >> CAIHPAD10=/CA/agents/data/hpa >> AGENTWORKS_DIR=/CA >> >> I've checked /etc/profile ~/.profile but this is the only useful portion: >> >> # *---------------------------------------------- * >> # Modifications made by setupNSM for >> # Unicenter NSM Agent Technology >> # *---------------------------------------------- * >> if [ "$AGENTWORKS_DIR" = '' ] >> then >> AGENTWORKS_DIR="/ca" >> export AGENTWORKS_DIR > >These lines set AGENTWORKS_DIR except the case is wrong. > > >> fi >> >> if [ -f "$AGENTWORKS_DIR"/scripts/envset ] >> then >> . "$AGENTWORKS_DIR/scripts/envset" > >This line executes the "script" envset, which probably sets CAIHPAD10. >Notice that this variable's value is a subdirectory of AGENTWORKS_DIR. > > >> fi >> >> # *---------------------------------------------- * >> # End of Modifications made by setupNSM for >> # Unicenter NSM Agent Technology >> # *---------------------------------------------- * -------------------------------- On Thu, Apr 17, 2003 at 11:36:23AM -0700, Crist Clark wrote: >Uh, well, it sure looks like that profile script is settinmg AGENTWORKS_DIR to >"/ca". Does the file it then loads, "/ca/scripts/envset," exist? Sugan Moodley Unix Systems Administrator - Midrange Support 2nd Floor ABSA Towers South, 160 Main Street, Johannesburg, 2001, South Africa PO Box 7735, Johannesburg, 2000, South Africa If little green men land in your back yard, hide any little green women you've got in the house. -- Mike Harding, "The Armchair Anarchist's Almanac" ______________________________________________ "The information contained in this communication is confidential and may be legally privileged. It is intended solely for the use of the individual or entity to whom it is addressed and others authorised to receive it. If you are not the intended recipient you are hereby notified that any disclosure, copying, distribution or taking action in reliance of the contents of this information is strictly prohibited and may be unlawful. Absa is liable neither for the proper, complete transmission of the information contained in this communication, nor for any delay in its receipt, nor for the assurance that it is virus-free." _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From whoha at med.umich.edu Tue Apr 22 11:21:18 2003 From: whoha at med.umich.edu (Edward Stankevich) Date: Tue, 22 Apr 2003 11:21:18 -0400 Subject: SUMMARY: Problem configuring HME0 Message-ID: Thanks to all of the following and maybe a few I missed ... Chris Grier,Alexander Sarreiter,Peter Stokes, Abhilash Mana,Chris Pinnock, Ben Garrett, ed rolison, Mike Demarco, Ryan Bishop,Joon Martin Hansen, Charles Gagnon, Martynas Buozis , John Mitchell, Vinnie German, Steven Haywood The consensus was "eeprom local-mac-address?=true" And to Martynas Buozis Thanks for the VCS info this is most likely the cause of some llt problems I've been having! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rj45 at slacknet.com Tue Apr 22 13:45:23 2003 From: rj45 at slacknet.com (RJ45) Date: Tue, 22 Apr 2003 11:45:23 -0600 (MDT) Subject: SUMMARY: updating form solaris MU CDs Message-ID: it's possible to do a upgrade installing Solaris from maintenance update CDs, the software will detect if a older solaris version is installed and will ask for upgrade/full install. thanks to Suresh Rajagopalan, Hargrave Mark E, Jon Godfrey _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mikelist at sky.net Tue Apr 22 15:00:25 2003 From: mikelist at sky.net (Mike's List) Date: Tue, 22 Apr 2003 14:00:25 -0500 (CDT) Subject: SUMMARY OT: Load Balancing Hardware Message-ID: Thank you goes to everyone that responded (really too many to list)... 1) f5 big ip (www.f5.com) --by far the most popular 2) cisco local directors (www.cisco.com) --second most 3) foundry networks (www.foundrynetworks.com) --third A few suggested Cisco Arrowpoint (formerly Arrowpoint which Cisco purchased, and I think the hardware might be EOL in favors of the local directors but not certain). The F5 BIG IP is the most preferred and everyone agreed that BIG IP is easy to manage/use and does all that I was looking for. - Mike ---------- original message ---------- Any recommendations for a hardware load balancing solution? ie. something easy to configure and manage and ability to manage the distribution. Example: one server has dual processors another with one processor, hence, 2/3 of processing requests go to server1 while 1/3 go to server2. ...and with additional servers added/removed, ability to modify. I'm looking for/prefer hardware solution not software. Thanks. - Mike _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From cradle at umd.edu Tue Apr 22 15:07:02 2003 From: cradle at umd.edu (David Eisner) Date: Tue, 22 Apr 2003 15:07:02 -0400 (EDT) Subject: SUMMARY: A1000 Solaris 2.6 to Solaris 8 Upgrade Question In-Reply-To: Message-ID: Thanks to: Graham Wood Deborah Cooker Darren Dunham David Foster Simon Burr Geof Lane Mark Hargrave Matthew Stier William Enestvedt There were five votes for scenario B -- install Solaris 8 and RM 6.22.1, and four for scenario A -- upgrade RM first. Two of the latter votes, however, were from people who had done an upgrade install of Solaris, not a clean install, which is what I'd be doing. The B voters thought there shouldn't be much problem, or had gone through the process and had had no problems -- one said it was "easy peasy". Also, RM is needed only to manage the array, not for data access, so even if RM is broken, you can get your data. There were a few caveats that don't apply to us, but merit mention. If you're upgrading with an A3000, and you have a 2 MB DacStore, you need to rebuild your LUNs when going to RM 6.22.x. (This is mentioned in the upgrade doc.) If you upgrade to 6.22.x on Solaris 2.6, pay careful attention to Sun Doc 22888: if your Solaris isn't patched, you may not be able to boot the system after upgrading RM. It was also suggested that I might want to either install Solaris onto a new disk, keeping the old one, or do a ufsdump of the boot disk, just in case something goes wrong. I think I'm going to go with method B. If anything goes heinously wrong, I'll post here. Thanks! Original Question: On Mon, 21 Apr 2003, David Eisner wrote: > We have two A1000's attached to an Ultra 30 running Solaris 2.6, with > RAID Manager 6.1.1. > > I need to "upgrade" the box to Solaris 8, while (hopefully) preserving > the data -- I'd prefer not to have to restore 200 GB of data from tape. > I put "upgrade" in quotes because it will really be a clean install after > the disk is repartitioned. None of the OS is on either array, and there's > a second internal disk where I can save data I'll need after I upgrade > the OS. > > According to Infodoc 43483, I need to upgrade to at least RM 6.22 (will > probably go with RM 6.22.1) for use with Solaris 8. > > Since this is hardware RAID, it seems like I shouldn't have much problem. > But the RAID Manager upgrade guide says things like this: > > Note Follow the procedures in this document exactly. > > It is possible, even likely, that you will encounter instructions within > this procedure that can be done better, faster, or even omitted > altogether. Resist the temptation to improve upon the procedures in this > document; if you alter the procedure, you risk an escalation due to an > upgrade gone wrong. > > so I'm a little worried. Further complicating matters is the fact that the > Solaris 2.6 currently installed isn't quite up to date, and patching it > may be difficult or impossible (for reasons I won't go into). > > Would it be better to A) upgrade to RM 6.22.1 first, then install Solaris 8 > and re-install RM 6.22.1; or B) just install Solaris 8 and RM 6.22.1. > > Obviously scenario A) only makes sense if the RM upgrade changes the state > of the A1000's, or if there's RAID Manager book-keeping info I'll need to > squirrel away on the second disk when I go from Solaris 2.6 to Solaris 8. > > Any advice is greatly appreciated. > > -David > ------------------------+--------------------------+ David Eisner | E-mail: cradle at umd.edu | CALCE EPSC | Phone: 301-405-5341 | University of Maryland | Fax: 301-314-9269 | ------------------------+--------------------------+ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ndo at woup.net Tue Apr 22 20:13:15 2003 From: ndo at woup.net (Nicolas Dorfsman) Date: 23 Apr 2003 02:13:15 +0200 Subject: SUMMARY: Who is doing chmod on homedirs ? Message-ID: <1051056795.3768.50.camel@trantor> A very late summary....no time before to write it. So, I wasn't the first guy to encounter this kind of magic chmod ! G.Hackett suggest: ----- You could write a wrapper script for chmod by moving chmod to chmod.sav and then put the folllowing lines in the file chmod #!/bin/sh who am i > /chmod_monitor exec $0.sav $@ Then make chmod executable. ------ ....this can't trap every inode modifications, but permits to exclude humans or scripts. Pavan Hotha wrote me about a similar issue...was Virus Scanner which scans samba shares and change permissions. SreenivasaRao Vadalasetty reports the same experience. I haven't any wintel machine with a virus scanner....but....hey ! what about my NFS shares ? So, I remove "root=xxx" in /etc/dfs/dfstab and reshare all. Problem solved ! One of my Linux box change permissions on remote shares ! I can't say which one (I've got some RH and Mdk) but they did ! Thanks for all their replies to: Pascal Grostabussiat Mr Rene Occelli joe.fletcher JULIAN, JOHN C G Hackett Pavan Hotha SreenivasaRao Vadalasetty _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From StephenB at datacom.co.nz Tue Apr 22 23:55:44 2003 From: StephenB at datacom.co.nz (Stephen Barnett (DSL AK)) Date: Wed, 23 Apr 2003 15:55:44 +1200 Subject: SUMMARY: Ultra-10 Model Message-ID: <04F92BA33955D51195B700508B4AB6E90D6AF3F8@dslak12.dnznet.co.nz> It turns out that all I needed was the CPU speed. Thanks to all those who replied. Cheers, Stephen W. Barnett Systems Consultant - Unix Division Datacom Systems Limited Auckland New Zealand _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jhartzen at csc.com Wed Apr 23 02:02:21 2003 From: jhartzen at csc.com (Johan Hartzenberg) Date: Wed, 23 Apr 2003 08:02:21 +0200 Subject: NON-SUMMARY: Mismatch between physical geometry and disk label contents Message-ID: Hi Gurus, Unfortunately, I did not get any solution. I had suggestions to just "Label" the disk, but my problem is not fixing a disk with an incorrect label, but rather finding such disks (LUNs) across many systems because these disks are currently "in use" and as a result, some portions of the disk is not available. Richard Hobbs mentioned having seen the same thing in the past when moving disks from one system to another. Using "format" to label a disk with an existing label does not re-read the disk geometry from the physical disk, it reads it from the existing label. The only way to really re-label a disk is by first corrupting the label, eg using the following procedure: First, corrupt the first sector on the disk c#t#d# dd if=/path/to/any/file of=/dev/rdsk/c#t#d#s2 count=1 bs=512 Next, use format to write a new label to the disk format select disk label? YES quit Then finally, encapsulate the disk, or partition and start using it as per normal. My Original question: > > Hi gurus, > > > > I was happily scripting a couple of weeks ago and came across something > > that produced a message similar to this (off the top of my head) > > > > Warning: The label contents does not match the geometry reported by the > > physical device. > > > > Now I can not remember the command I ran, but I want to use it in another > > script. I was busy writing a script to report on raw disk sizes as seen > by > > Solaris (not Veritas), and I was playing with commands like > > > > format, prtvtoc, fmthard > > > > I've been searching the web and sunsolve, but I can't find out what it is > I > > did. Makes me want to "script" to record everything I do... ! > > > > The situation I have is that sometimes when the storage administrators > > allocate a whole bunch of LUNs I do not check each individually, and > > sometimes I don't realise I am not seeing the full LUN (eg if the first > > sector of an expanded LUN was previously the first sector of a smaller > > LUN). I am trying to write a script which will tell me where I may be > > using LUNs without getting access to the whole full size of the LUN. > > > > > > > > moreover, I have tried to write something to compare the output of fdisk > -g > > with that of fdisk -G, but I always get the following message: > > > > # fdisk -g /dev/rdsk/c3t11d4s2 > > * Label geometry for device /dev/rdsk/c3t11d4s2 > > * PCYL NCYL ACYL BCYL NHEAD NSECT SECSIZ > > 39516 39516 2 0 15 96 512 > > # fdisk -G /dev/rdsk/c3t11d4s2 > > /dev/rdsk/c3t11d4s2: Cannot get physical disk geometry. > > > > > > > > devinfo appears totally useless, and on Solaris 2.6 I get for -g and -G > > both: > > > > Disk does not support fixed disk partition table > > > > > > Any ideas/suggestions ? > > > > Thanx, > > _Johan |---------+---------------------------> | | "Hobbs, Richard"| | | | | | | | | 17/04/2003 04:14| | | PM | | | | |---------+---------------------------> >-----------------------------------------------------------------------------------------------------------------| | | | To: Johan Hartzenberg/GIS/CSC at CSC | | cc: | | Subject: Re: Mismatch between physical geometry and disk label contents | >-----------------------------------------------------------------------------------------------------------------| Hello, How do you know the disk label is valid though?? What told you that it was valid? To produce the error, i took the 2GB disk out of my Ultra 1, and then took two 9GB disks out of an old veritas-managed disk array. I then put the two 9GB disks into my Ultra 1, and tried to install Solaris 9 onto them. These two disks were going to stay in my Ultra 1 permanently from now on. At some point in the install (fairly early on), i got the error you had... about the wrong physical geometry. I checked several things: - Whether that type of disk would actually work in my machine at all. - Whether the disk type was compatible with the bus i had plugged them onto. - Whether the disks were actually dead (by putting them in a working array). - Whether there were any master/slave jumpers on the disks. - Whether a "probe-scsi-all" command at the ok prompt could pick up the disks. - Whether a boot into single user-mode, followed by a 'format' would help. - ...and finally, i tried labelling them, and that worked! I don't know too much about it technically, i'm afraid, but there's my story. Hope it helps!! Try posting on my forum!! http://unixforum.co.uk Thanks, Richard. Quoting Johan Hartzenberg : > > > Hi, > > Thanx for your reply. The veritas stuff is not the problem, the problem is > that the disk label was valid, but the geometry was "wrong". > > What did you have to do to actually get the system to produce this message? > > So far I only got your reply, plus one from someone who didn't properly > read my question! > > Thanx, > _Johan > > > > > > |---------+---------------------------> > | | "Hobbs, Richard"| > | | | | @mongeese.co.uk>| > | | | > | | 17/04/2003 03:45| > | | PM | > | | | > |---------+---------------------------> > > > -----------------------------------------------------------------------------------------------------------------| > | > | > | To: Johan Hartzenberg > | > | cc: > | > | Subject: Re: Mismatch between physical geometry and disk label > contents | > > > -----------------------------------------------------------------------------------------------------------------| > > > > > Hello, > > I had that problem when i put two disks from a veritas-managed array into > an > Ultra 1. It said the disk was reporting geometry that didn't match the > physical > disk. > > I simply had to label it i think. Veritas doesn't label disks the same way > Solaris does, so this could be the problem. > > I'd be very interested to see the summary when yo figure it out though! > > Thanks, > Richard. > > > > _______________________________________________ > > sunmanagers mailing list > > sunmanagers at sunmanagers.org > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > > > -- > Richard Hobbs > hobbs at mongeese.co.uk > http://mongeese.co.uk | http://unixforum.co.uk > > "There's only one way of life, and that's your own" - The Levellers > > _____________________________________________________ > Send all your jokes to jokes at fishsponge.co.uk !! > To subscribe, email: jokes-subscribe at fishsponge.co.uk > > > > > -- Richard Hobbs hobbs at mongeese.co.uk http://mongeese.co.uk | http://unixforum.co.uk "There's only one way of life, and that's your own" - The Levellers _____________________________________________________ Send all your jokes to jokes at fishsponge.co.uk !! To subscribe, email: jokes-subscribe at fishsponge.co.uk _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sunsol19712003 at yahoo.es Wed Apr 23 03:49:05 2003 From: sunsol19712003 at yahoo.es (=?iso-8859-1?q?J=20Marquez?=) Date: Wed, 23 Apr 2003 09:49:05 +0200 (CEST) Subject: SUMMARY:CORE ANALYSIS Message-ID: <20030423074906.9277.qmail@web20709.mail.yahoo.com> The summary from the question below is: - Sun has a course on this, docs.sun.com has a book on the modular debugger - SolarisCAT http://wwws.sun.com/software/download/products/3e3af5aa.html - book PANIC! Unix System Crash Dump Analysis ISBN 0-13-149386-8 - run this for starters: #strings vmcore.# |head against the core files. Thanks to: Harold Reinstein, Alan Pae, Sun Systems Admin (Bill), John England,, Michael Peters and Jason Shatzkamer. Regards. Hi Gurus, I know this is a pretty broad question, but I would like to know if there is any document to analyse cores in Sun Machines. I can run adb commands or get the iscda output, but to be honest the output is not of big help to me, especially in "Bad traps" type of cores.I've been surfing Internet and Sun web but nothing good Ive found. Of course I am sure that with a document it won't analyse all the cases but I would like to have unless the most typical hints to get focused with an iscda output or russing adb commands. Ive also checked the book "Panic!..." but it's not too practical to me... Any help is welcome! Thanks in advance and regards. --------------------------------- Yahoo! Messenger Nueva versión: Super Webcam, voz, caritas animadas, y más #161;Gratis! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Jo.Ashmore at sis.securicor.co.uk Wed Apr 23 04:15:12 2003 From: Jo.Ashmore at sis.securicor.co.uk (Jo Ashmore) Date: Wed, 23 Apr 2003 09:15:12 +0100 Subject: SUMMARY: limiting the size of emails Message-ID: Thanks for all the replies, Consensus is that I should uncomment the O MaxMessageSize=nnnn line in sendmail.cf, setting nnnn to the size in bytes that we require the limit to be, and restart sendmail. Also, in response to: will MaxMessageSize work for outgoing emails? Yes, basically as far as sendmail is concerned it treats all emails as incoming when it receives them, whether from the local host for a remote host, or from a remote host for the local host or for relaying, and therefore runs its rules. The SIZE keyword was a bit more difficult, most said that if it wasnt specified or was 0 it would be infinity, so we're going to wait & see what happens. Thanks again, Jo -----Original Message----- From: Jo Ashmore [mailto:Jo.Ashmore at sis.securicor.co.uk] Sent: 22 April 2003 16:28 To: 'sunmanagers at sunmanagers.org' Subject: limiting the size of emails Hi all, I have been asked to limit the size of both incoming and outgoing emails - we are currently running a very basic vanilla sendmail on Solaris 8. Any ideas please? Jo Ashmore Unix Systems & Support Manager Securicor Information Systems Direct Dial 01249 665191 Mobile 07974 732668 ********************************************************************** The information contained in this e-mail message is intended only for the individuals named above. If you are not the intended recipient, you should be aware that any dissemination, distribution, forwarding or other duplication of this communication is strictly prohibited. The views expressed in this e-mail are those of the individual author and not necessarily those of Securicor Information Systems Limited. Prior to taking any action based upon this e-mail message you should seek appropriate confirmation of its authenticity. If you have received this e-mail in error, please immediately notify the sender by using the e-mail reply facility. ********************************************************************** _____________________________________________________________________ This message has been checked for all known viruses on behalf of Securicor Information Systems by the MessageLabs. For further information visit http://www.messagelabs.com or Email: mailsweeper.info at sis.securicor.co.uk _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers ********************************************************************** The information contained in this e-mail message is intended only for the individuals named above. If you are not the intended recipient, you should be aware that any dissemination, distribution, forwarding or other duplication of this communication is strictly prohibited. The views expressed in this e-mail are those of the individual author and not necessarily those of Securicor Information Systems Limited. Prior to taking any action based upon this e-mail message you should seek appropriate confirmation of its authenticity. If you have received this e-mail in error, please immediately notify the sender by using the e-mail reply facility. ********************************************************************** _____________________________________________________________________ This message has been checked for all known viruses on behalf of Securicor Information Systems by the MessageLabs. For further information visit http://www.messagelabs.com or Email: mailsweeper.info at sis.securicor.co.uk _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Gerard.Henry at cmi.univ-mrs.fr Wed Apr 23 05:36:30 2003 From: Gerard.Henry at cmi.univ-mrs.fr (Gerard Henry) Date: Wed, 23 Apr 2003 11:36:30 +0200 Subject: SUMMARY: help for a script Message-ID: <3EA65E9E.7090808@cmi.univ-mrs.fr> many thanks to all for your responses. sbhatia at siue.edu s.evans at TriCorInd.com alex at posixnap.net mh1272 at yahoo.com vogelke at pobox.com t at trey.net stephan.grund at isst.fhg.de dledger at ivdcs.demon.co.uk SDaubigne at bordeaux-bersol.sema.slb.com and specially for Sebastien DAUBIGNE, because i will use his command in my crontab: ps -eo pcpu,pid,args | awk '/netscape/ && $1 > 5 {print $2}' | xargs -n 10 kill -9 my first mail was not so clear, i want to automatically kill process such netscape because they are bugged, i don't want to do it by hand. Here is some contributions: ****************************************************************************** form Evans, Shawn: #!/bin/sh # PROC=netscape NUM=50 CPU=`prstat 1 1 | grep $PROC | awk '{print $9}' |awk -F. '{print $1}'` # if [ $CPU -ge $NUM ] ; then kill -9 `ps -ef | grep -i $PROC` fi from Alex J. Avriette: is there something wrong with ps -ef | grep netscape | awk '{ print $2 }' ****************************************************************************** from Karl Vogel: You might want to do it by hand using the script below; it looks for the processes on the command line and puts you in VI to pick the ones you want to kill. --------------------------------------------------------------------------- #!/bin/ksh # # $Id: skill-solaris.sh,v 1.3 2001/04/29 22:15:13 vogelke Exp $ # $Source: /src/opt/scripts/RCS/skill-solaris.sh,v $ # # NAME: # skill # # SYNOPSIS: # skill [-signal] [-dhinv] {tty user command pid regexp} # # DESCRIPTION: # "skill" is a program to blow away processes. You can specify # processes by tty, userid, command name, process id, or an # egrep-style regular expression. # # OPTIONS: # -d, --debug show debug output # -h, --help print this message # -i, --interact edit the process list before killing # -n, --nokill show processes but don't kill them # -v, --version print the version and exit # # AUTHOR: # Karl Vogel # Sumaria Systems, Inc. # Based on a script by Tom Christiansen -- tchrist at convex.com # # A similar script by Ric Anderson : # ftp://jaguar.cs.utah.edu/pub/skill/skill-3.7.tar.Z PATH=/bin:/usr/sbin:/usr/bin:/usr/local/bin : ${EDITOR=/bin/vi} # Set default editor if blank. export PATH EDITOR umask 022 tag=`basename $0` tmp=/tmp/skilltmp.$$ plist=/tmp/skillps.$$ trap "rm -f $tmp $plist" 0 trap "rm -f $tmp $plist; exit 1" ERR # ksh variants only # ======================== FUNCTIONS ============================= # # die: prints an optional argument to stderr and exits. # A common use for "die" is with a test: # test -f /etc/passwd || die "no passwd file" # This works in subshells and loops, but may not exit with # a code other than 0. die () { echo "$tag: error: $*" 1>&2 exit 1 } # usage: prints an optional string plus part of the comment # header (if any) to stderr, and exits with code 1. usage () { lines=`egrep -n '^# (NAME|AUTHOR)' $0 | sed -e 's/:.*//'` ( case "$#" in 0) ;; *) echo "usage error: $*"; echo ;; esac case "$lines" in "") ;; *) set `echo $lines | sed -e 's/ /,/'` sed -n ${1}p $0 | sed -e 's/^#//g' | egrep -v AUTHOR: ;; esac ) 1>&2 exit 1 } # version: prints the current version to stdout. version () { lsedscr='s/RCSfile: // s/.Date: // s/,v . .Revision: / v/ s/\$//g' lrevno='$RCSfile: skill-solaris.sh,v $ $Revision: 1.3 $' lrevdate='$Date: 2001/04/29 22:15:13 $' echo "$lrevno $lrevdate" | sed -e "$lsedscr" exit 0 } # ======================== MAIN PROGRAM ========================== # Defaults: ac_help= ac_prev= ac_invalid="invalid option; use -h or --help to show usage" argv= # Initialize some variables set by options. debug=n int=n dokill=y sig= for ac_option do # If the previous option needs an argument, assign it. case "$ac_prev" in "") ;; *) eval "$ac_prev=\$ac_option"; ac_prev=; continue ;; esac case "$ac_option" in -*=*) ac_optarg=`echo "$ac_option" | sed 's/[-_a-zA-Z0-9]*=//'` ;; *) ac_optarg= ;; esac # main switch case "$ac_option" in -d | -debug | --debug | --debu | --deb | --de) debug=y ;; -h | -help | --help | --hel | --he) usage ;; -i | -interact | --interac | --intera | --inter |\ --inte | --int ) int=y ;; -n | -nokill | --nokill | --nokil | --noki | --nok | --no) dokill=n ;; -v | -version | --version | --versio | --versi | --vers) version ;; -[0-9]*) sig="$sig $ac_option" ;; -di|-id) int=y; debug=y ;; -dn|-nd) dokill=n; debug=y ;; -in|-ni) dokill=n; int=y ;; -din|-dni|-nid|-ndi|-idn|-ind) dokill=n; int=y; debug=y ;; -*) die "$ac_option: $ac_invalid" ;; *) case "$argv" in "") argv="$ac_option" ;; *) argv="$argv $ac_option" ;; esac ;; esac done case "$ac_prev" in "") ;; *) die "missing arg to --`echo $ac_prev | sed 's/_/-/g'`" ;; esac # # Make sure we have something to look for. # case "$argv" in "") echo "I need something to look for."; usage ;; *) target="$argv" ;; esac # # The default signal strategy is to use TERM, INT, # and KILL on any processes. # case "$sig" in "") sig="-15 -1 -9" ;; *) ;; esac # # Get the processes to kill. Edit the list before showing # it so the -i, -d, and -n options can be used together. # Use a separate file for the process list, so we don't # include our own grep commands. # ps -ef | egrep -v " $$ " > $plist egrep "$target" $plist > $tmp test -s $tmp || die "No processes found." case "$int" in y) $EDITOR $tmp ;; *) ;; esac case "$debug" in y) echo "Signal(s): $sig Kill? $dokill" echo "Interactive? $int Target: $target" ;; *) ;; esac if test "$dokill" = "n" -o "$debug" = "y" then echo Not killing any of these: echo echo ' UID PID PPID C STIME TTY TIME CMD' cat $tmp case "$dokill" in n) exit 0 ;; *) ;; esac fi # # If we get this far, zap the process IDs. # pids=`awk '{print $2}' $tmp` echo "killing:" $pids | fmt -77 for each in $sig do echo kill $each $pids kill $each $pids sleep 1 done exit 0 from trey valenta: "/usr/ucb/ps aux" shows cpu in the 3rd col. Or: "/bin/ps -eo pid,cpu,comm" would get you the cpu usage in col2. But you can take what you have and do: /usr/bin/ps -eo pid,cpu,comm| awk '/netscape/{print $1,$2}' and get the pid and cpu percentage. ****************************************************************************** from Stephan Grund: Try something like: kill `prstat 1 1 | awk '/netscape/{if(substr($9,0,length($9)-1)>90)print $1}'` ****************************************************************************** from David Ledger: prstat 1 1 | awk '/netscape/ { split($9, tmp, "%"); print tmp[1] }' -- Gerard HENRY LATP UMR 6632 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From minoti.koul at wipro.co.in Wed Apr 23 07:12:37 2003 From: minoti.koul at wipro.co.in (Minoti Koul) Date: Wed, 23 Apr 2003 16:42:37 +0530 Subject: SUMMARY- script.(help) Message-ID: <1AB3C279D5C54942ADEDA401207936FF4482F8@punaomx1.wipro.co.in> Thanks to all who replied. Hichael Morton Ed Rolison Munish Sarin SANTHAKUMAR Siva Steven Hill I will go ahead with Steven's script. primarily because I could understand it df -k| awk '$5>70 { print "Aroogah! " $6 " is nearly full" }' As everyone recomended ..I will schedule it in cron. Thanks Once again. ======================= Regards, Minoti Koul Wipro Infotech,Pune 91-20-5535659 ext 25 ======================= -----Original Message----- From: Minoti Koul Sent: Wednesday, April 23, 2003 3:36 PM To: sunmanagers at sunmanagers.org Subject: script.(help) Hi SunManagers, One of the File systems on my Sun Machine keeps on getting full. My developers wanted that when ever the file system reaches say 70%, they should get a mail or some kind of an alert. I havent done scripting So i need your help. As I understand we need to pick up the %field from df -k and the do wall to all connected users. I wanted to broadcast something like " /ORACLE HAS REACHED 70% OF ITS CAPACITY,PLEASE TAKE CARE" Thanks, ======================= Regards, Minoti Koul Wipro Infotech,Pune 91-20-5535659 ext 25 ======================= DISCLAIMER: Information contained and transmitted by this E-MAIL is proprietary to Wipro Limited and is intended for use only by the individual or entity to which it is addressed, and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If this is a forwarded message, the content of this E-MAIL may not have been sent with the authority of the Company. If you are not the intended recipient, an agent of the intended recipient or a person responsible for delivering the information to the named recipient, you are notified that any use, distribution, transmission, printing, copying or dissemination of this information in any way or in any manner is strictly prohibited. If you have received this communication in error, please delete this mail & notify us immediately at mailadmin at wipro.co.in _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers DISCLAIMER: Information contained and transmitted by this E-MAIL is proprietary to Wipro Limited and is intended for use only by the individual or entity to which it is addressed, and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If this is a forwarded message, the content of this E-MAIL may not have been sent with the authority of the Company. If you are not the intended recipient, an agent of the intended recipient or a person responsible for delivering the information to the named recipient, you are notified that any use, distribution, transmission, printing, copying or dissemination of this information in any way or in any manner is strictly prohibited. If you have received this communication in error, please delete this mail & notify us immediately at mailadmin at wipro.co.in _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From HickeyM at bsci.com Wed Apr 23 07:54:50 2003 From: HickeyM at bsci.com (Hickey, Michael) Date: Wed, 23 Apr 2003 07:54:50 -0400 Subject: Summary: Awk script help Message-ID: <4059B48767FA8143B26D88957A902F8701039F77@natpr8.bscexc1.bsci.com> The overwhelming majority submitted the following: awk '$3>=15{print $3}' Thanks to all those who responded. Regards, Mike > -----Original Message----- > From: Hickey, Michael > Sent: Tuesday, April 22, 2003 11:05 AM > To: Sun Managers Mailing List (E-mail) > Subject: Awk script help > > Can someone push me in the right direction of the proper syntax to take a > file, > dr.out containing multiple rows and 3 fields as below, and print only the > lines that > have a number >= 15 in field 3? > > 04/12/03 20:56 15.3 > 04/12/03 21:56 15.3 > 04/12/03 22:56 15.3 > 04/12/03 23:57 18.5 > > Thanks in advance. > > Mike _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From daniel.bergman at seb.se Wed Apr 23 09:33:58 2003 From: daniel.bergman at seb.se (daniel.bergman at seb.se) Date: Wed, 23 Apr 2003 15:33:58 +0200 Subject: SUMMARY: Unable to add new metadatabase-devices. System will not accept them. Message-ID: <95FDD2234911D51180E5420045479348EFFAC8@smri023a.sebank.se> Resend with correct title. Daniel > -----Original Message----- > From: Jonathan V Hays [mailto:jhays at jtan.com] > Sent: den 23 april 2003 15:17 > To: daniel.bergman at seb.se > Subject: RE: SOLVED: Unable to add new metadatabase-devices. > System will > not a ccept them. > > > Please preface your Subject with "SUMMARY". > > Thanks. > > > -----Original Message----- > > From: sunmanagers-admin at sunmanagers.org > > [mailto:sunmanagers-admin at sunmanagers.org] On Behalf Of > > daniel.bergman at seb.se > > Sent: Wednesday, April 23, 2003 4:01 AM > > To: sunmanagers at sunmanagers.org > > Subject: SOLVED: Unable to add new metadatabase-devices. > > System will not a ccept them. > > > > > > Hi, > > > > Thanks for all the help I received! > > > > As Pal Baranyai and Preiszl Matthias pointed out I'd to add > > "forceload: drv/sd" to /etc/system to forceload SCSI-drivers, > > to load earlier actually, during system startup so that the > > SCSI-disks will be available when metadb loading. > > > > Regards, > > Daniel Bergman _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sun_consultant at hotmail.com Wed Apr 23 14:14:54 2003 From: sun_consultant at hotmail.com (Sun Consultant) Date: Wed, 23 Apr 2003 11:14:54 -0700 Subject: SUMMARY: Solaris 9 Installation on Sun Ultra 2 References: Message-ID: Thanks for the replies-- The problem turned out to be defective media. I reburned the Solaris 9 CDROM and it worked fine. ----- Original Message ----- From: "Sun Consultant" To: "Sun Managers" Sent: Tuesday, April 22, 2003 11:39 AM Subject: Solaris 9 Installation on Sun Ultra 2 > Hi I have a Sun Ultra 2, where I am having problems installing Solaris 9... > > The install starts, and in the middle gives the following error: > By the way, I am using a 18GB Disk with a big root partition, if it makes any > difference. > The latest PROM firmware is on the sytem. > > Thanks > > Error during installation: > Solaris 9 software installation partially failed > > Solaris 9 packages partially installed > > SUNWcsr > > . > > . > > . > > SUNWllcr > > Customizing system files > > - Mount points table (/etc/vfstab) > > ERROR: Could not open file (/a/etc/vfstab) > > ERROR: Could not create the file system mount table (/etc/vfstab) > > Could not create the file system mount table (/etc/vfstab) > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bkain1 at ford.com Wed Apr 23 14:55:32 2003 From: bkain1 at ford.com (Kain, Becki (B.)) Date: Wed, 23 Apr 2003 14:55:32 -0400 Subject: SUMMARY: ld.so.1: ./a.out: fatal: libstdc++.so.5 Message-ID: The problem was that I forgot to either put -R/usr/local/lib in my configure script or set my LD_LIBRARY_PATH to it. With thanks to: Ayhan Kanmaz henrik.mortensen Lars Hecking -----Original Message----- From: Kain, Becki (B.) [mailto:bkain1 at ford.com] Sent: Wednesday, April 23, 2003 2:24 PM To: 'sunmanagers at sunmanagers.org' Subject: ld.so.1: ./a.out: fatal: libstdc++.so.5 Unfortunatly, I found 3 questions like this in the archives, but no answers. I'm trying to compile nmap 3.25 on solaris 8 or 9 with gcc 3.2.2 precompiled from sunfreeware. The file libstdc++.so.5 is a link to libstdc++.so.5.0.2 which is sitting in the same directory and I'm compiling this as root with LDFLAGS set to -L/usr/local/lib so what am I doing wrong? And I promise to post the summary when I get it. thanks _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From aravind at csee.wvu.edu Wed Apr 23 17:26:46 2003 From: aravind at csee.wvu.edu (Aravind Vinnakota) Date: 23 Apr 2003 17:26:46 -0400 Subject: SUMMARY: boot problems with /var Message-ID: <1051133206.20056.29.camel@gandhi> First let me thank: Mike Bruno McClark Darren Dunham Tim Villa SreenivasaRao Vadalasetty SteinAxt Dominic Clarke German, Vinnie Williams, Mario Michael Schulte Mark Andrews AP for responding to my question. Many of them said the problem is /var needs to be mounted much before the other file partitions like /space is mounted but I am trying to create a var directory on /space and make a symbolic from /var to /space/var by which var is not accessible until /space is mounted. One way to overcome this problem is to make /space mount much before anything else is mounted but it may cause some problems. Some said make a partition for /var from /space so that /var can be mounted without any dependencies. But the way I did was changing the name of /space to /var. That way /var is instantly available and I can make use of the the disk space on /var for storing any other indexes by making symbolic links. Once again I want to thank all the people for their valuable suggestions. My original question was: I have a solaris 2.6 Ultra Enterprise 2 box with a couple of internal disks and an external disk set. As /var keeps growing, I decided to make use of the external disks. I raided and created file system and mounted the meta device on /space. Now I created a directory called var on /space and copied all the contents of /var onto /space/var in single user mode. Renamed /var to /var.old and made a soft link from /space/var to var on /. But for some reason when I reboot the machine after these changes I get the following error: INIT: Cannot create /var/adm/utmp or /var/adm/utmpx INIT: failed write of utmpx entry:" " INIT: failed write of utmpx entry:" " INIT: SINGLE USER MODE Type Ctrl-d to proceed with normal startup, (or give root password for system maintenance): I double checked to make sure that /space has rw permissions, I used the -p option to copy with permissions but still I am unable to understand why the system is unable to access the var link on /. Any idea where I am making a mistake? Any suggestions? I will summarize the solution. I am able to boot once I revert back to the inital settings without any fsck problems. Aravind _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From pln at razzle.Stanford.EDU Wed Apr 23 18:25:56 2003 From: pln at razzle.Stanford.EDU (Patrick L. Nolan) Date: Wed, 23 Apr 2003 15:25:56 -0700 (PDT) Subject: SUMMARY: All my mount points are busy Message-ID: <200304232225.h3NMPuW09306@razzle.Stanford.EDU> My original message is at the bottom. My problem is resolved, but I'm not sure I completely understand how I cured it. One key piece of information came from Casper Dik, of course. He pointed out that all my mount points are busy because they are being shared by NFS. Neither lsof nor fuser shows this usage. It is necessary to unshare before mounting or umounting. I didn't mention NFS in my original message. Even after unsharing the offending mount point, I couldn't mount it. The breakthrough came when I decided to fsck the partition, just to make sure the disk is really working. It looked like this: $: fsck /dev/dsk/c1t2d0s1 /dev/dsk/c1t2d0s1 is a mounted file system, ignored That's more or less the same baloney I had been seeing. For some reason I decided to umount it by device name, instead of mount point. I don't think I had ever done that before. $:umount /dev/dsk/c1t2d0s1 umount: warning: /dev/dsk/c1t2d0s1 not in mnttab Then I did the fsck again, and it worked! That umount must have accomplished something, in spite of the error message. After this, there was no problem. mount /glast/04 worked with no complaints. I have a hypothesis about what happened. I don't understand the details well enough to know if this is the truth. Suppose the file system got "mounted" at boot time, but there was no entry in mnttab for some reason. For quite a while, I was balked by the busyness of the mount point due to NFS. Even after that was cured, some programs were unable to deal with the file system by name, such as mount, umount, and df. (I don't see why mount should have suffered from this, though.) When I umount'ed the partition by device name, umount complained but did its job. Now mnttab was in agreement with reality, and things could proceed. Thanks also to Steve Maher, Marco Greene, John W. Ballard, Thomas M. Payerle, David Foster, Dan Astoorian, and jbarratt. Several pointed out the use of lsof and fuser, although I mentioned lsof in my message. Most mentioned the importance of mnttab. Original question: > I'm really confused. I can't mount one of my filesystems, and > all the mount points seem to be busy, when I know they aren't. > > This is on an Ultra 2 with Solaris 7. I just installed the new > kernel patch 106541-24 and rebooted. Apparently I installed > the patch without backing up the original files, so I can't back > it out. > > The first symptom I see is the dreaded message: > > $ /etc/mount /glast/04 > mount: /dev/dsk/c1t2d0s1 is already mounted, /glast/04 is busy, > or the allowable number of mount points has been exceeded > > According to the archives, the "number of mount points" part is > probably bogus. > > To check if it's already mounted, I ran df and looked at vfstab. > The string c1t2d0s1 doesn't appear in df, and it only appears > once in vfstab: > /dev/dsk/c1t2d0s1 /dev/rdsk/c1t2d0s1 /glast/04 ufs 4 yes logging > It's surrounded by a bunch of other similar entries. > > So maybe the problem is that the mount point is busy. That seems > awfully unlikely, but it's worth a shot. I did "rmdir /glast/04" and > it went away peacefully. I recreated it, and still mount fails. > > Now I began to wonder about the number of mount points. I decided to > unmount one of the other partitions to see if it freed up a slot. > I couldn't unmount any of them. I got this behavior: > $: /etc/umount /glast/11 > umount: /glast/11 busy > on nearly a dozen file systems. Some of these don't get used at all, > so this seems pretty unlikely. Then I ran lsof, and it tells me that > /glast/11 and its friends are not busy. I believe it, based on the > pattern of use on this computer. > > What's going on? * Patrick L. Nolan * * W. W. Hansen Experimental Physics Laboratory (HEPL) * * Stanford University * _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jhartzen at csc.com Thu Apr 24 02:29:04 2003 From: jhartzen at csc.com (Johan Hartzenberg) Date: Thu, 24 Apr 2003 08:29:04 +0200 Subject: SUMMARY NOW: Mismatch between physical geometry and disk label contents Message-ID: Hi there. That is exactly what I get, and the suggestion from McClarc works. Format should only update the disk label when you use the label command, and any other operations (print partition table, for example) will use the "current" label. format "Verify" and "Current" certainly does not pick up anything on the LUNs with "old" labels. I tried MCC's suggestion below. - Initially, Format sees disk #10 as "OPEN-E" - Inquire however reports it as "OPEN-E*2", whilst "Current" and "Verify" both print the contents of the Label. - And as mcc suggested, Type 0 detects the disk as an E*2 (27 Gb) Thanks _Johan P.S. In all fairness, Tommy Falsen hinted in the right direction too! # format Searching for disks...done AVAILABLE DISK SELECTIONS: 0. c0t0d0 /sbus at 54,0/SUNW,fas at 0,8800000/sd at 0,0 1. c1t0d0 /sbus at 55,0/SUNW,fas at 0,8800000/sd at 0,0 2. c2t14d33 /sbus at 54,0/fce at 1,400000/sd at e,21 3. c2t14d34 /sbus at 54,0/fce at 1,400000/sd at e,22 4. c2t14d42 /sbus at 54,0/fce at 1,400000/sd at e,2a 5. c4t16d33 /sbus at 54,0/fce at 1,401000/sd at 10,21 6. c4t16d34 /sbus at 54,0/fce at 1,401000/sd at 10,22 7. c4t16d42 /sbus at 54,0/fce at 1,401000/sd at 10,2a 8. c4t17d3 /sbus at 54,0/fce at 1,401000/sd at 11,3 9. c4t17d4 /sbus at 54,0/fce at 1,401000/sd at 11,4 10. c4t17d5 /sbus at 54,0/fce at 1,401000/sd at 11,5 Specify disk (enter its number): 10 selecting c4t17d5 [disk formatted] FORMAT MENU: disk - select a disk type - select (define) a disk type partition - select (define) a partition table current - describe the current disk format - format and analyze the disk repair - repair a defective sector label - write label to the disk analyze - surface analysis defect - defect list management backup - search for backup labels verify - read and display labels save - save new disk/partition definitions inquiry - show vendor, product and revision volname - set 8-character volume name ! - execute , then return quit format> inq Vendor: HITACHI Product: OPEN-E*2 -SUN Revision: 0117 format> ver Primary label contents: Volume name = < > ascii name = pcyl = 19759 ncyl = 19757 acyl = 2 nhead = 15 nsect = 96 Part Tag Flag Cylinders Size Blocks 0 unassigned wm 0 0 (0/0/0) 0 1 unassigned wm 0 0 (0/0/0) 0 2 backup wu 0 - 19756 13.57GB (19757/0/0) 28450080 3 - wu 0 - 0 0.70MB (1/0/0) 1440 4 - wu 1 - 19756 13.57GB (19756/0/0) 28448640 5 unassigned wm 0 0 (0/0/0) 0 6 unassigned wm 0 0 (0/0/0) 0 7 unassigned wm 0 0 (0/0/0) 0 format> cur Current Disk = c4t17d5 /sbus at 54,0/fce at 1,401000/sd at 11,5 format> ty AVAILABLE DRIVE TYPES: 0. Auto configure 1. Quantum ProDrive 80S 2. Quantum ProDrive 105S 3. CDC Wren IV 94171-344 4. SUN0104 5. SUN0207 6. SUN0327 7. SUN0340 8. SUN0424 9. SUN0535 10. SUN0669 11. SUN1.0G 12. SUN1.05 13. SUN1.3G 14. SUN2.1G 15. SUN2.9G 16. Zip 100 17. Zip 250 18. SUN9.0G 19. HITACHI-OPEN-E*15-SUN-0117 20. HITACHI-OPEN-E*10-0117 21. HITACHI-OPEN-E*3-0117 22. HITACHI-OPEN-E-SUN-0117 23. HITACHI-OPEN-E*5-0117 24. HITACHI-OPEN-E*2-0117 25. HITACHI-OPEN-E-0117 26. other Specify disk type (enter its number)[22]: 0 c4t17d5: configured with capacity of 27.13GB selecting c4t17d5 [disk formatted] |---------+---------------------------> | | mcclark | | | | | | 23/04/2003 01:49| | | PM | | | | |---------+---------------------------> >------------------------------------------------------------------------------------------------------------------| | | | To: Johan Hartzenberg | | cc: | | Subject: Re: NON-SUMMARY: Mismatch between physical geometry and disk label contents | >------------------------------------------------------------------------------------------------------------------| Johan, You might get some success using format, then 'type' and '0'. That should re-probe your disk geometry... I think it should ask you whether you want to relabel the disk or not at that point if it thinks the old geometry is not correct. I'm not sure about whether doing this will cause 'current' to then show you the modified geometry or not if you don't relabel (I don't have a disk I can try it on right now...). If you're slicing space off of large physical LUNs on a storage arrary, you might want to incorporate this as a first step in disk layout once the LUN is presented. Some storage arrays (like StorEdge 6960) don't write anything to the LUN at all when the space is allocated, and what can happen is that if an old LUN was de-allocated, and a new one created starting at the same space, Solaris may see the old label and presume that it is valid. -mcc Johan Hartzenberg wrote: >Hi Gurus, >Unfortunately, I did not get any solution. I had suggestions to just >"Label" the disk, but my problem is not fixing a disk with an incorrect >label, but rather finding such disks (LUNs) across many systems because >these disks are currently "in use" and as a result, some portions of the >disk is not available. > >Richard Hobbs mentioned having seen the same thing in the past when moving >disks from one system to another. > >Using "format" to label a disk with an existing label does not re-read the >disk geometry from the physical disk, it reads it from the existing label. >The only way to really re-label a disk is by first corrupting the label, eg >using the following procedure: > >First, corrupt the first sector on the disk c#t#d# > dd if=/path/to/any/file of=/dev/rdsk/c#t#d#s2 count=1 bs=512 > >Next, use format to write a new label to the disk > format > select disk > label? YES > quit > >Then finally, encapsulate the disk, or partition and start using it as per >normal. > > >My Original question: > > > >>>Hi gurus, >>> >>>I was happily scripting a couple of weeks ago and came across something >>>that produced a message similar to this (off the top of my head) >>> >>>Warning: The label contents does not match the geometry reported by the >>>physical device. >>> >>>Now I can not remember the command I ran, but I want to use it in >>> >>> >another > > >>>script. I was busy writing a script to report on raw disk sizes as >>> >>> >seen > > >>by >> >> >>>Solaris (not Veritas), and I was playing with commands like >>> >>>format, prtvtoc, fmthard >>> >>>I've been searching the web and sunsolve, but I can't find out what it >>> >>> >is > > >>I >> >> >>>did. Makes me want to "script" to record everything I do... ! >>> >>>The situation I have is that sometimes when the storage administrators >>>allocate a whole bunch of LUNs I do not check each individually, and >>>sometimes I don't realise I am not seeing the full LUN (eg if the first >>>sector of an expanded LUN was previously the first sector of a smaller >>>LUN). I am trying to write a script which will tell me where I may be >>>using LUNs without getting access to the whole full size of the LUN. >>> >>> >>> >>>moreover, I have tried to write something to compare the output of >>> >>> >fdisk > > >>-g >> >> >>>with that of fdisk -G, but I always get the following message: >>> >>># fdisk -g /dev/rdsk/c3t11d4s2 >>>* Label geometry for device /dev/rdsk/c3t11d4s2 >>>* PCYL NCYL ACYL BCYL NHEAD NSECT SECSIZ >>> 39516 39516 2 0 15 96 512 >>># fdisk -G /dev/rdsk/c3t11d4s2 >>>/dev/rdsk/c3t11d4s2: Cannot get physical disk geometry. >>> >>> >>> >>>devinfo appears totally useless, and on Solaris 2.6 I get for -g and -G >>>both: >>> >>> Disk does not support fixed disk partition table >>> >>> >>>Any ideas/suggestions ? >>> >>>Thanx, >>> _Johan _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sun at sun.twlight.net Thu Apr 24 10:54:07 2003 From: sun at sun.twlight.net (Jerry K) Date: Thu, 24 Apr 2003 09:54:07 -0500 Subject: [SUMMARY] Solaris 7 study CD-Rom from Sun In-Reply-To: <62F3FA70-7599-11D7-BB38-003065BDE51C@sun.twlight.net> Message-ID: <99AE11FC-7664-11D7-8B8D-003065BDE51C@sun.twlight.net> SUMMARY It does appear that Sun did produce a java based Solaris 7 sysadmin study cd, similar to the one produced for Solaris 2.6. No one on the list had one for sale but many people suggested that I search ebay. Thanks to the following for your replies. Sorry for anyone I left off. Jonathan Hays Mike Salehi Hichael Morton Kevin Smith ORIGINAL QUESTION ================= On Wednesday, April 23, 2003, at 09:39 AM, Jerry K wrote: > I currently have a Solaris 2.6 sysadmin certification study guide from > sun that was java based. The neat thing about it was even if I > couldn't get it to run on a certain platform, I could just export the > DISPLAY through X11 and run the program that way. > > I have (2) questions. > > Did Sun produce/update this product for Solaris 7? > > If so, I am hoping that someone on the list has one that they might be > willing to sell. I am not looking for any type of a windows > executable study guide. I am also not asking anyone to pirate any > software for me. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Christer.Eriksson at seb.co.uk Thu Apr 24 11:09:22 2003 From: Christer.Eriksson at seb.co.uk (Eriksson (London), Christer) Date: Thu, 24 Apr 2003 16:09:22 +0100 Subject: SUMMARY: Sunfire 480 network problem Message-ID: Thanks Ian Pease And David B Harrington For that fast answers I could confirm that it's was a network problem As it was a other SUN server on the same network lost it's connections as well but it didn't Write anything about it in /var/adm/messages But as we all now the network managers couldn't see anything that's was wrong and denied it It was any problem as usual Cheers Christer Chris; I had the same thing a while ago. There are some notes in the archives, including my problem statements. I never posted a summary (my BAD!). The issue turned out to be a bad port on the switch!! (which My network folks denied the possibility of this, but push them to check it out. Dave H I have found this is a Network Switch issue every time it happens here..... -----Original Message----- From: Eriksson (London), Christer [mailto:Christer.Eriksson at seb.co.uk] Sent: 24 April 2003 15:20 To: 'sunmanagers at sunmanagers.org' Subject: Sunfire 480 network problem Hi gurus I just was getting a very strange problem one of our production servers it kicked out everybody and complaind about someone was taking over it's ip lon014:root> tail /var/adm/messages Apr 24 14:49:40 lon014 ip: [ID 903730 kern.warning] WARNING: IP: Hardware addres s '00:03:ba:24:08:45' trying to be our address xxx.xxx.xxx.xxx! Apr 24 14:51:16 lon014 last message repeated 110670 times Apr 24 14:51:16 lon014 ip: [ID 903730 kern.warning] WARNING: IP: Hardware addres s '00:03:ba:24:08:45' trying to be our address xxx.xxx.xxx.xxx! Apr 24 14:54:11 lon014 last message repeated 381486 times lon014:root> ifconfig -a lo0: flags=1000849 mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 ce0: flags=1000843 mtu 1500 index 2 inet xxx.xxx.xxx.xxx netmask fffffff0 broadcast xxx.xxx.xxx.255 ether 0:3:ba:24:8:45 lon014:root> uname -a SunOS lon014 5.8 Generic_108528-17 sun4u sparc SUNW,Sun-Fire-480R and the strange thing is that it complaining over it's own mac address anyone that got clues Regards Christer __________________________ Christer Eriksson SEB London 2 Cannon Street, London. EC4M 6XX tel: +44 20 7246 4479 fax: +44 20 7489 1684 Cell: +44 7909 53 43 44 mailto: christer.eriksson at seb.co.uk __________________________ "There are only 10 types of people in the world: Those who understand binary, and those who don't" Confidentiality Notice The content of this e-mail is intended only for the confidential use of the person(s) to whom it is addressed. If the reader of this message is not such a person, you are hereby notified that you have received this communication in error and that reading it, copying it, or in any way disseminating its content to any other person, is strictly prohibited. If you have received this message in error, please notify the author by replying to this e-mail immediately. Skandinaviska Enskilda Banken AB - A Public Company (publ) - is incorporated in Stockholm Sweden with limited liability and is regulated by The Financial Services Authority for the conduct of designated investment business in the United Kingdom. It is registered in England and Wales under Number BR000979. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From patrick.oreilly at uni.edu Thu Apr 24 11:39:40 2003 From: patrick.oreilly at uni.edu (Patrick O'Reilly) Date: Thu, 24 Apr 2003 10:39:40 -0500 Subject: SUMMARY: Restore deleted files Message-ID: <007601c30a77$b81621b0$66f1a186@oreilly> Thanks go to all the respondents, who mostly said that I was SOL unless I had a backup (which I didn't). My favorite line stated that in UNIX, once a file is deleted, it is gone. Not so quick! I should say I am not a Sun sysadmin, but a meteorologist who runs a few Sun machines that only get used by me. So, yes I should have backups, and I do of other important scripts, but somehow forgot these. My loss...well, almost. Anyway....my original post: I recently deleted a directory of files that I would love to have back, months of work on a few scripts. Instead of typing rm *.gem *.gif I typed rm *.gem * gif and quickly hit enter without realizing there was no period between * and gif. Can I restore these files somehow? I am running patched Solaris 8 on a Ultra-10. ***************************** AND THE WINNERS ARE: Todd Boss and Ed Rolison.......with: - unrm, Lazarus; tools used to recover deleted unix files (part of TCT: the Coroner toolkit) - surf to http://www.fish.com/tct/help-recovering-file http://www.porcupine.org/tct. It's a lot of pain and effort though. ******************************* I went and got the TCT package and though it took overnight to run, worked like a charm! Wasn't that much of a pain, actually. I am once again the proud owner of the previously deleted scripts. A little work and time, but much less than re-writing all the scripts! I heartily endorse the TCT package. THANKS Todd and Ed! Patrick _______________________________________ Patrick O'Reilly Meteorological Decision Support Scientist The STORM Project - University of Northern Iowa patrick.oreilly at uni.edu ~ ph: 319-273-3789 http://www.uni.edu/storm "No trees were killed in the making of this e-mail...however, a large number of electrons were horribly inconvenienced." _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From crist.clark at globalstar.com Thu Apr 24 14:35:37 2003 From: crist.clark at globalstar.com (Crist Clark) Date: Thu, 24 Apr 2003 11:35:37 -0700 Subject: SUMMARY: Who Needs efdaemon? References: <3EA6D5C9.38845A16@globalstar.com> Message-ID: <3EA82E79.FDA450D5@globalstar.com> Casper Dik was kind enough to elaborate as to what the manpage for efdaemon(1M) means by "devices with FCode." It typically means machines with hot swappable PCI cards and system boards. And the daemon is only needed if you actually will be doing hot swapping. Original question: > > Can someone point me to some Sun documentation that indicates what > hardware/software (if any) actually requires efdaemon and or even the various > FCode packages? Or is this purely stuff for developers and not something > needed on a server or workstation? > > Where as I am in the "If I don't know what it is, I don't need it" camp, > I need to counter the some people with the ol' "if I don't know what it is, > install it, 'cause I might need it later" mentality. I need to justify > removing things rather than having to justify their presence. -- Crist J. Clark crist.clark at globalstar.com Globalstar Communications (408) 933-4387 The information contained in this e-mail message is confidential, intended only for the use of the individual or entity named above. If the reader of this e-mail is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any review, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this e-mail in error, please contact postmaster at globalstar.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From fhouston at east.isi.edu Thu Apr 24 18:42:44 2003 From: fhouston at east.isi.edu (Forrest Houston) Date: Thu, 24 Apr 2003 18:42:44 -0400 (Eastern Daylight Time) Subject: SUMMARY: Sol9 binaries Message-ID: Note to admin (if this hangs again) please delete this one. Note to everyone (if this gets through) please ignore the note to admin ;) Thanks everyone for the help. I've been trying for awhile now to send a summary message but the server keeps rejecting it (???) so I thought I would try a new message entirely instead of forwarding on the old one. Anyway sunfreeware.com seems to be the consensus and I'm d/l the package now (slow server though :( ). Some people pointed out the fact different versions of Solaris are usually compatible on the binary level, however the packages at sunsite must have some additional "checking" built in because whenever I tried to install one I got a "This version runs on Solaris _ only" type error. thanks again _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mattclark786 at yahoo.com Thu Apr 24 21:16:59 2003 From: mattclark786 at yahoo.com (david) Date: Thu, 24 Apr 2003 18:16:59 -0700 (PDT) Subject: summary: metadatabase not detect on storage array sunfire280r s1 array In-Reply-To: <20030425010547.16318.qmail@web20415.mail.yahoo.com> Message-ID: <20030425011659.94539.qmail@web20418.mail.yahoo.com> thanks all for help. it turned out /etc/system needed forceload drv/sd param. i found the answer on list previous posted thanks again though for responding. david > --- david wrote: > > Hi Guys > > we have brandnew SUNfire280r machine two internal > > optical drive mirrored already and new storage > array > > s1 with three 72gig drives(want to setup raid5 ). > We > > have two partitions on those three auxillary disk > > slice 7 c2t0d0s7 c2t0d0s7 c2t0d0s7 is being used > for > > metadatabase. We are using c2t0d0s0 c2t1d0s0 > > c2t2d0s0 > > for slice 0 for data . > > > > The problem is when I create metadatabase on > Storage > > array with metadb command the metadatabase is not > > detected after reboot. By the metadb command its > > puts > > in unknown state. Any idea what could be problem > > here > > thanks in advance for your help > > David > > platform: SunOS 5.8 Generic_108528-19 sun4u sparc > > SUNW,Sun-Fire-280R > > storage array s1 > > > > #metadb -i > > > > flags first blk block > count > > > > a m p luo 16 1034 > > > > > /dev/dsk/c1t0d0s7 > > > > a p luo 1050 1034 > > > > > /dev/dsk/c1t0d0s7 > > > > a p luo 16 1034 > > > > > /dev/dsk/c1t1d0s7 > > > > a p luo 1050 1034 > > > > > /dev/dsk/c1t1d0s7 > > > > M p unknown unknown > > > > > /dev/dsk/c2t0d0s7 > > > > M p unknown unknown > > > > > /dev/dsk/c2t1d0s7 > > > > M p u# format > > > > Searching for disks...done > > > > > ==================================================================================== > > > > > > > > AVAILABLE DISK SELECTIONS: > > > > 0. c1t0d0 > 107> > > > > > > > /pci at 8,600000/SUNW,qlc at 4/fp at 0,0/ssd at w21000020373e9703,0 > > > > 1. c1t1d0 > 107> > > > > > > > /pci at 8,600000/SUNW,qlc at 4/fp at 0,0/ssd at w21000020373e65d5,0 > > > > 2. c2t0d0 > 424> > > > > /pci at 8,700000/scsi at 6,1/sd at 0,0 > > > > 3. c2t1d0 > 424> > > > > /pci at 8,700000/scsi at 6,1/sd at 1,0 > > > > 4. c2t2d0 > 424> > > > > /pci at 8,700000/scsi at 6,1/sd at 2,0 > > > > > ====================================================================== > > > > #partition> print > > > > Current partition table (original): > > > > Total disk cylinders available: 14087 + 2 > (reserved > > cylinders) > > > > > > > > Part Tag Flag Cylinders Size > > > > > Blocks > > > > 0 unassigned wm 0 - 14075 68.30GB > > > > > (14076/0/0) 143237376 > > > > 1 unassigned wu 0 0 > > > > > (0/0/0) 0 > > > > 2 backup wu 0 - 14086 68.35GB > > > > > (14087/0/0) 143349312 > > > > 3 unassigned wu 0 0 > > > > > (0/0/0) 0 > > > > 4 unassigned wu 0 0 > > > > > (0/0/0) 0 > > > > 5 unassigned wu 0 0 > > > > > (0/0/0) 0 > > > > 6 unassigned wu 0 0 > > > > > (0/0/0) 0 > > > > 7 unassigned wm 14076 - 14078 14.91MB > > > > > (3/0/0) 30528 > > > > partions information for all three disks c2t0d0s7 > > c2t1d0s7 c2t2d0s7 > > The New Yahoo! Search - Faster. Easier. Bingo > > http://search.yahoo.com > > _______________________________________________ > > sunmanagers mailing list > > sunmanagers at sunmanagers.org > > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > The New Yahoo! Search - Faster. Easier. Bingo > http://search.yahoo.com The New Yahoo! Search - Faster. Easier. Bingo http://search.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rj45 at slacknet.com Fri Apr 25 05:18:29 2003 From: rj45 at slacknet.com (RJ45) Date: Fri, 25 Apr 2003 03:18:29 -0600 (MDT) Subject: SUMMARY: ultra 2 boot fail Message-ID: The solution was to install latest OBP+POST and the problem did not occour anymore. I noticed anyway the system power LED on the frontal panel did not work, I inspected further and I Saw on the motherboard the power led is missing. I found it inside and the LED was detached from motherboard because of a hit I think. acrually I bought the ultra2 in USA and they shipped it to me in Europe. it does not really look like a standard LED butit looks like smaller of a regular pin jumper it has a circular windows on the fase of its plastic surface with a symbol on it, the symbol is something like |--| --| |-- consider that the piece is very small as smal las a jumper. I think it is a LED because I Saw these kind of LEDs on many hard disks chips. Does the LED detach from the motherboard can be an ISSUE on proper functioning ? Rick _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jmcintosh at mlug.missouri.edu Fri Apr 25 10:01:52 2003 From: jmcintosh at mlug.missouri.edu (Jason McIntosh) Date: Fri, 25 Apr 2003 09:01:52 -0500 Subject: SUMMARY: V880 question on ethernet interfaces and services Message-ID: <7742D45B-7726-11D7-83A4-000A959CF864@mlug.missouri.edu> Ok, thanks to all who responded!!! There were more than a few and too many to list, but here's the general consensus of what I've received: With regards to the startup scripts, they're the usual /etc/rc*.d directories. The scripts are also as found on a linux system in /etc/init.d S##name is a start script, K##name is a stop script. For which runlevel corresponds to which "action" do a "man init. Additionally, I was told that /etc/rcS.d is the startup scripts directory. Further, a combination of rcS.d and rc2.d handle most of the startup and services. Specifically, S69inet and S72inetsvc for the network configuration. S30network.sh in /etc/rcS.d handles the network interfaces. Another warning - the network scripts in general don't support stop or restart parameters, as they're "designed to bring up the interfaces, not to bring them down." (thanks to Darren Dunham for a very specific set of information on this). Network interfaces are loaded through files in /etc such as /etc/hostname.ge0. the ge0 is the interface and the file contains the hostname. That hostname should be defined in /etc/hosts. This only counts for currently installed and configured network interfaces. But, the general network card interfaces for "plumbing" are as follows: ge# is the Gigabit Ethernet card. 100Mb cards are hme# Quad-fast-ethernet cards are qfe# SunBlade 150's (and apparantly the SunFire's) have eri# ports. To load an interface which hasn't be loaded at boot time, you use ifconfig with the plumb command. For a dhcp interface, you'd use the following: ifconfig eri0 plumb auto-dhcp The last part of the network config can be found in /etc/defaultrouter for setting the route information and in /etc/netmasks for setting the netmask configuration. For more information on the V880 the following link to the sun docs: http://docs.sun.com/db/coll/792.1 http://docs.sun.com/db/doc/816-2127-10 As for the different services running, I was told you can shut off pretty much anything in /etc/inetd.conf One person suggested commenting the following out: talk, finger, rsh, rquota, chargen, cachefs, kcms, bootps, comsat and anything labeled experimental in /etc/services can be shutdown in /etc/inetd.conf. If you're not running a dhcp server you can also comment out the dhcp stuff too. One suggestion was to stop using inetd and use xinetd instead (xinetd being more secure). And last, but not least, the SMC requirements. First, it seems no one has a clear idea of exactly what is required by the SMC (Mike Penny gave the only real clue) but one thing is the sadmind in /etc/inetd.conf. Most people apparently aren't using it, primarily because it's "a pig". But, no one seemed sure of what the SMC requires, how it's configured, etc. Thanks! Jason McIntosh Jason McIntosh Programmer/Analyst University of Missouri 573-884-3865 -----BEGIN PGP PUBLIC KEY BLOCK----- Version: GnuPG v1.2.1 (GNU/Linux) mQGiBD5AMRERBACcxAJ7hiB6udEDefnAksb49o6BDVC2bxdUTwkxP9jS0BmLqbQL egYYt09WjEJtn4eRuVdkku7A0fi/G8NIsXnE9oMKnWkqg2tjQ8q65D64Cass5zEU WG6j8qABpxpZNP9HGSTqm0yeYTR9f0dGaS6jZbxgme6hU0XDOGH3ug6/qwCg0j1h gYkkNl3jbPuwtpCrTIxSUYsD/0J18alKrswomFJfoRgjA7S9AezlV7YJoU3dCLSV 6D19SAVwmseTRltJm0S8e8Yf6Bq9l+1OdsJCHtT+HYBVuB0PmL8PDhJg6vAzIZlt 5c3hkfJrSSCssRSMBIr+8Hl2HBU4tKB79L2cI3Nrij+5DJTVzIa5QpvsFDAIO2Cp Ma/hA/9vOPY2PyoAWKb3JAQV8T2h7/rjzePxOv0WYI2/6THdgh2lLUP+GJ4FdH76 I/8d+qtiAzul/Zq2LkFvpejnu41dDZn+yhgsVTkz/xTRKWQX9sOud0QjvyN0nHD9 KqisFRgs1ByINQcWNK8KpwgLcBRLVS4EALDn5R6yL6AT6poT7bQ8SmFzb24gTWNJ bnRvc2ggKFByb2dyYW1tZXIvQW5hbHlzdCkgPE1jSW50b3NoSkBtaXNzb3VyaS5l ZHU+iF8EExECAB8FAj5AMREFCQHhM4AECwcDAgMVAgMDFgIBAh4BAheAAAoJEGZP +3FaGjd1P9IAoJQL5kaHEjG1TNVSt20bAXDx/DzmAJ4pVTDnpWUPNCxk+/kMBlGe bO97ObkCDQQ+QDEfEAgAjl6vRTDWrMTUfXyngnWAgU/3wRZmjcKONhCGcpqFOFR/ 2CiMeeJOnNXgSzrPxIfUJphlh00vBm1K/ngllg3MGFI9hOffuLuHXiw8e/Yc87uz YdtglWHeUz/9YQCe4ndKohtk7nZHUoxQd5OspJxYJH5J5cysSuH2V839NtNPJKBZ ai0VhyTFZKD3v9xTC8ZyMEO022bpkhWz1cs/9l5z4g2eg7mOwe+hJstMQFHk77Zr GbkPwi+gWwM/b2pxigz1xhQpHpR5HrO/4yM8zWYZHUbUGwxiuTJqKktFEVUKUFDD xopIpOAMOy0qVs+wXS5buJSduSgDxPmKq0xZFsJiZwADBwgAi66pXMCTolMQzn50 MAs9KRK6+3XphI+InmzfN+/OVknwzkAkGhRfisYI0DyN/26wSkn+zyoE87NBuUQt xtcNOpwwxS2WCqBx3PhtpVJ6yvaeFmSe3QC2lUf8418B2C6GR/e6IOPNhVW1cnhP IR0/yY8c8zQrJxEZNhBtj2SrkLY6Ps7j97lI8n+u21YC2/a5P3TPCa3x0w0m0APB zJrhXuNGwPcNtxqKZDz7m9KuROSijx0Xm1buQkGfDZqkqA1D8ljN2vdA+jx/v1eq H27iXk2iZ+i8bOyoiflniKh6nrd2UlVPvzmLscpiUf0rEGQuBV0Sq5AUUqO7MQM+ L8jCYYhMBBgRAgAMBQI+QDEfBQkB4TOAAAoJEGZP+3FaGjd1yhMAnjAwU37EIok3 WZu1BXhaD9y1ryjBAKC7fnv7AdrVrxdXkmdgdA5jxUFXOg== =ueqZ -----END PGP PUBLIC KEY BLOCK----- [demime 0.99c.7 removed an attachment of type application/pgp-signature which had a name of PGP.sig] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From r57a3smjf at qualityknitting.com Fri Apr 25 10:04:57 2003 From: r57a3smjf at qualityknitting.com (Bryce Parks) Date: Fri, 25 Apr 03 14:04:57 GMT Subject: Bulk Email Advertise to 28 Million People - $129 - g Message-ID: An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030425/c58883e6/attachment.html From edsbicapps58 at dlo.gsi.gov.uk Fri Apr 25 03:21:43 2003 From: edsbicapps58 at dlo.gsi.gov.uk (Procter, Paul Mr (EDS)) Date: Fri, 25 Apr 2003 08:21:43 +0100 Subject: SUMMARY: Overloading a CPU to failure Message-ID: Many thanks to all that offered assistance on my CPU problem, I have now tried both Sunvts and Seti and found them to be equally as good as each other ay CPU testing. Regards, Paul Procter The original issue was: ---------------------------------------------------------------------------- ---------------------------------------------------------------------------- Does anyone have a script that they use to test a cpu that may be a little on the 'worn out' side of things? We have been given a script that carries out a `wc -c /dev/mem &` but am not entirely sure that this is a good test. Many thanks in advance, I will summarize. Regards, Paul Procter _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers The original of this email has been scanned for viruses by the Government Secure Intranet (GSI) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs. This E-mail was sent from an E-mail address under the control of the MOD. Privileged / In Confidence Information may be contained in this message. If you have received this E-mail in Error, please destroy all copies in your possession or control and kindly notify the sender by reply e-mail. The addressee (or individual responsible for delivery of the message to such person) may not forward, discolse or copy this E-mail to any third party, without the prior consent of the MOD. Furthermore, MOD does not accept liability for any changes made to this E-mail after it was sent. Please advise immediately if you or your employer do not consent to Internet E-mail for messages of this kind. If you take offence to the contents of this E-mail, you can contact the System Management team at qcissmt at dlo.gsi.gov.uk The original of this email has been scanned for viruses by the Government Secure Intranet (GSI) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs. ........................................................ This E-Mail has been originated from the Internet. If you do not wish to receive E-mail from this address, please reply to this E-mail, and request that the sender remove your E-mail address from their system. QCIS SMT .............................................................. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From fhouston at east.isi.edu Thu Apr 24 16:13:33 2003 From: fhouston at east.isi.edu (Forrest Houston) Date: Thu, 24 Apr 2003 16:13:33 -0400 (Eastern Daylight Time) Subject: SUMMARY Re: Looking for binaries for Sol9 In-Reply-To: References: <007601c30a77$b81621b0$66f1a186@oreilly> Message-ID: sunfreeware seems to be the concensus on the binaries. Thanks for the quick responses :D On Thu, 24 Apr 2003, Forrest Houston wrote: > Hopefully I'm doing this right... I've only been lurking for a couple of > weeks. > > I'm cutting my teeth on Solaris 9 now and my usual "routine" has failed > me, sunsite.unc doesn't have gcc binaries for 9 yet. Anyone have any > hints/clues/suggestions as to other sites I can try for this (and other > packages)? > > While I have your attention ;) does anyone have a good security > reference/checklist for 9 as well? Especially something that outlines the > differences between 7/8 and 9. > > Thanks > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From fhouston at east.isi.edu Thu Apr 24 17:43:41 2003 From: fhouston at east.isi.edu (Forrest Houston) Date: Thu, 24 Apr 2003 17:43:41 -0400 (Eastern Daylight Time) Subject: SUMMARY Re: Looking for binaries for Sol9 In-Reply-To: References: <007601c30a77$b81621b0$66f1a186@oreilly> Message-ID: First time I tried this it got put into administrative hold (some sort of mail header problem) so let me give it another shot. Thanks everyone for all the help. I'm in the process of d/l it now. A couple of people have mentioned about the binary compatibility, and maybe it's just the packages on sunsite, but I was getting errors[1] when trying to install the packages which is why I had to look elsewhere. thanks again [1] 'This version requires Solaris X' type things On Thu, 24 Apr 2003, Forrest Houston wrote: > > sunfreeware seems to be the concensus on the binaries. Thanks for the > quick responses :D > > > > On Thu, 24 Apr 2003, Forrest Houston wrote: > > > Hopefully I'm doing this right... I've only been lurking for a couple of > > weeks. > > > > I'm cutting my teeth on Solaris 9 now and my usual "routine" has failed > > me, sunsite.unc doesn't have gcc binaries for 9 yet. Anyone have any > > hints/clues/suggestions as to other sites I can try for this (and other > > packages)? > > > > While I have your attention ;) does anyone have a good security > > reference/checklist for 9 as well? Especially something that outlines the > > differences between 7/8 and 9. > > > > Thanks > > _______________________________________________ > > sunmanagers mailing list > > sunmanagers at sunmanagers.org > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Rick.Brashear at ercgroup.com Tue Apr 15 11:01:50 2003 From: Rick.Brashear at ercgroup.com (Rick.Brashear at ercgroup.com) Date: Tue, 15 Apr 2003 10:01:50 -0500 Subject: SUMMARY: /dev/random for Solaris 8 Message-ID: <2857747EB2E7D311AC1A00508B6F9C7C0B1FC43B@kans020cercge.erc.gecapital.com> I wanted to report that an email came in after I had submitted this summary from Anrew J. Caines. I followed his advice and tested this script. It works great! This script allows applying the patch 112438-01 without rebooting. The script can be obtained at: http://halplant.com:88/software/Solaris/scripts/setup_random Thanks Andrew! > -----Original Message----- > From: Brashear, Richard G. (Rick) (GE ERC, Sun Systems, > Contractor) > Sent: Wednesday, April 09, 2003 10:49 AM > To: 'sunmanagers at sunmanagers.org' > Subject: SUMMARY: /dev/random for Solaris 8 > > Thanks for the overwhelming response! Great list! [Brashear, Rick (GE/ERC/ITS)] The original inquiry was: I have loaded the sunfreeware.com gnupg-1.2.1-sol8 package sparc and have had no success determining how to get the /dev/random - /dev/random devices for Solaris 8. Can anyone give me a clue? > The consensus was install patch 112438-01 and reboot. > > Other suggestions: > * SUNWski on the solaris easy access cds > * andirand (use google) > * prngd (use google) > > Download it from Andreas Maier site: > > http://www.cosy.sbg.ac.at/~andi/SUNrand/ > > Thanks (in order received) to: > Steven Haywood > Gary Chambers > Eric Trager > Randy Romero > Gavin Harris > Chistopher L. Barnard > Cathy M. Bodenstedt > Mario Williams > Rob De Langhe > Casper Dik > Lonnie Ratliff > Mike Salehi > Yakov Lerner > Hichael Morton > Sean Timmins > William Yodlowsky > Eric Forgette > Tim Thomas > Brian Pardy > Celso G. Lima > Kendall Lloyd > Luc I. Suryo > Preiszl Matthias > Suresh Vuthamaraju > Lars Hecking > and also to those that are still reponding.... > > g ERC > _______________________________ > Rick Brashear > Server Systems > Information Technology Department > Employers Reinsurance Corporation > 5200 Metcalf > Overland Park, Kansas 66201 > * 913 676-6418 > * rick.brashear at ercgroup.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnrams at cox.net Sat Apr 26 03:40:23 2003 From: johnrams at cox.net (John Rams) Date: Sat, 26 Apr 2003 00:40:23 -0700 Subject: Summary: syslog configuration Help Message-ID: <004e01c30bc7$188199d0$6501a8c0@svakahome> Many thanks to Mark Hargrave,Alan Angulo and Jay Lessert. Sorry for late summary. Alan mentoned the need to restart /etc/init.d/inetsvc for loghost to take effect after you change your /etc/hosts. Still not sure if i have to. But log are being redirected to the loghost. *************** 0) man syslog.conf Understand the "facility", "level", and "action" parts of syslog.conf 1) Edit /etc/hosts in all machines and add the "loghost" alias. example: 123.123.123.123 admin-machine loghost 2) Edit /etc/syslog.conf and tell the other 3 machines where to send the messages. example: auth.notice ifdef(`LOGHOST', /var/log/authlog, @loghost) 3) HUP the syslog daemon an all machines: kill -HUP `cat /etc/syslog.pid` ******************** *.err;kern.debug;daemon.notice;mail.crit;user.none@hostname Where "hostname" is the system you want to receive the messages. Don't forget to restart the syslogd daemon. *********** On the 3 systems you want to report from, put the following in the /etc/syslog.conf file: You configure syslog.con on the 3 other systems. See the syslog.conf(4) man page, look for the string "@server". > How can i do the same to receive checkpoint firewall logs? Configure syslog.conf on the firewall host. > -----Original Message----- > From: johnrams at cox.net [mailto:johnrams at cox.net] > Sent: Friday, April 18, 2003 3:49 PM > To: sunmanagers at sunmanagers.org > Subject: syslog configuration Help > > > How do i configure my syslog.conf on 1 system to be able to > receive all the system log messages of 3 other systems. How > can i do the same to receive checkpoint firewall logs? > > Thanks so much > John Rams -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.sunmanagers.org/pipermail/summaries/attachments/20030426/a67eb88b/attachment.html -------------- next part -------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bart at grafikon.com Sun Apr 27 13:55:57 2003 From: bart at grafikon.com (Bart Terryn) Date: Sun, 27 Apr 2003 19:55:57 +0200 Subject: SUMMARY: remove package from a broken kernel In-Reply-To: <20010916170528.E04E511CC5@ohno.mrbill.net> Message-ID: I know I'm late with it. But here is the 'winning' solution for my earlier question on this list. ----------Lucas Bunyan said: There is a sundocument that talks about removing a pathc whiel booted from cdrom. I am sure that you can adapt this for the pkgrm command. the document numnber is 50624. You do not need a sunsolve login id to get to this document. go to sunsolve.sun.com and search for this document. ----------------------------- And indeed the chgroot command described in this document allowed me to remove the package in question and return the (backup) system disks to a working state. (needed the copy /etc/mnttab trick from that doc as well. Thanks again to everybody that responded. Bart _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From nob-kami at jasmine.ocn.ne.jp Sun Apr 27 22:14:58 2003 From: nob-kami at jasmine.ocn.ne.jp (Nobuhira Kaminaga) Date: Mon, 28 Apr 2003 11:14:58 +0900 Subject: SUMMARY: snmp query by shell script References: <003901c304e5$46d08b40$1953f2aa@pc.japan.ml.com> Message-ID: <004401c30d2b$f7dc7180$1953f2aa@pc.japan.ml.com> Philippe BOURDEU d'AGUERRE of AIME - Campus de l'INSA suggested me net-snmp (http://www.sunfreeware.com/ or http://www.net-snmp.org/) use. This CUI tool can be called by shell script so this satisfies my requirement. You can also find W32 version (ucd-snmp) at net-snmp W3 site. Philippe, thank you for your information. Best Regards, Nobuhira Kaminaga ----- Original Message ----- From: "Nobuhira Kaminaga" To: Cc: "Nobuhira Kaminaga" ; "Nobuhira Kaminaga at office" Sent: Thursday, April 17, 2003 10:28 PM Subject: snmp query by shell script > Hello, > > What is the easiest way to query remote MIB value through snmp by shell > script ? > > Thank you, > > Nobuhira Kaminaga > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Przemyslaw_Bak at pgf.com.pl Mon Apr 28 03:30:34 2003 From: Przemyslaw_Bak at pgf.com.pl (Przemyslaw_Bak at pgf.com.pl) Date: Mon, 28 Apr 2003 09:30:34 +0200 Subject: SUMMARY: oracle 8.0.5 from Solaris 7 to Solaris 9 Message-ID: <20030428073034.GB1567@przemek.pgf.com.pl> Thanks to: pattysunmanagers at hotmail.com martinh at solid-state-logic.com krenzischek at Encompasserve.org matthew.darcy at hp.com In general: - oracle 8.0.5 is not supported on Solaris 9 (and is not supported at all) besides this: - it should be done without problem but it is better to do export/import przemol ----- Forwarded message from przemolicc at poczta.fm ----- > From: przemolicc at poczta.fm > To: sunmanagers at sunmanagers.org > Subject: oracle 8.0.5 from Solaris 7 to Solaris 9 > Date: Fri, 25 Apr 2003 14:01:15 +0200 > > Hello, > > we have oracle 8.0.5 running on solaris 7. I installed new sparc server > with solaris 9 and want to move oracle database from the old server to > the new server. Should I be prepared for any pitfalls ? Did anybody do > such movement ? > I don't want to export/import database. Just copy files. > > przemol ----- End forwarded message ----- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnrams at cox.net Mon Apr 28 02:48:29 2003 From: johnrams at cox.net (John Rams) Date: Sun, 27 Apr 2003 23:48:29 -0700 Subject: SUMMARY: sendmail 8.12.9 compilation problem install-docs cannot make Message-ID: <003e01c30d52$325f94b0$6501a8c0@svakahome> Thanks to Damien Curtain, Marcelo Maraboli, Dave Dunaway, Douglas Palmer, Al Hopper, Chris Pinnock, Roy Rapoport for helping. Solution seems (Special Thanks to Marcelo, Damien) to create directories cat8, cat1, cat5 under /usr/share/man To ignore building man pages, elect not to install the formatted manpages by setting define(`confDONT_INSTALL_CATMAN') in your site config, the standard manpages should still be installed. Regards John ----- Original Message ----- From: John Rams To: sunmanagers at sunmanagers.org Sent: Saturday, April 26, 2003 12:33 AM Subject: sendmail 8.12.9 compilation problem install-docs cannot make Managers: While compiling sendmail 8.12.9, i am getting this error. Looks like this seemed to be a problem from a long time. Unfortunately, didn't find a solution. # sh Build -c -f /usr/local/src/sendmail8.12.9/devtools/Site/site.config.m4 install nroff -man sendmail.8 > sendmail.0 || cp sendmail.0.dist sendmail.0 nroff -man aliases.5 > aliases.0 || cp aliases.0.dist aliases.0 nroff -man mailq.1 > mailq.0 || cp mailq.0.dist mailq.0 nroff -man newaliases.1 > newaliases.0 || cp newaliases.0.dist newaliases.0 ../../devtools/bin/install.sh -c -o bin -g bin -m 444 sendmail.0 /usr/share/man/cat8/sendmail.8 cp: cannot create /usr/share/man/cat8/sendmail.8: No such file or directory *** Error code 1 make: Fatal error: Command failed for target `install-docs' Current working directory /usr/local/src/sendmail-8.12.9/obj.SunOS.5.8.sun4/sendmail *** Error code 1 make: Fatal error: Command failed for target `all' My MANPATH is set to Made changes to config file to add these: (all others are default entries for Solaris 8) define(`confMANROOT', `/usr/share/man/cat') define(`confMANROOTMAN', `/usr/share/man/man') Read readme files, searched everywhere but didn't find any useful stuff. oops, just before i sent this to you all, i was able to compile using following constructs to bypass compiling mana pages. Still curious to find how to make & install man pages as well. (how would i compile only this part without building all again) define(`confNO_MAN_BUILD', `true') define(`confNO_MAN_INSTALL', `true') Please help. Thanks John _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Martin.Gabel at ffm3.siemens.de Mon Apr 28 04:03:34 2003 From: Martin.Gabel at ffm3.siemens.de (Gabel Martin) Date: Mon, 28 Apr 2003 10:03:34 +0200 Subject: SUMMARY: Creating cdrom Message-ID: <30888BF0456F6C41AD65E5415EC1A8130DC47C@ffmr222a.ffm3.siemens.de> Hi everybody, Thank you for for all who have replied. I got several good answers and suggestions. See my original message at bottom of this mail. My favorite came from Brendan Searle: Hi Martin, The ISO9660 data CDROM standard fits the "lowest common denominator" and is written for MS-DOS, VMS, Mainframes etc. with 8.3 filenames (ie. 8 character filenames with a 3 character extension - all upper case); I think VMS type version numbers were also supported, but typically ignored. Some Unix systems (as you've found) map 8.3 filenames to lower case and some leave them as upper case. Macs and Unix wanted long filenames from the start, so the "Rock Ridge" long filename extension was created. Rock Ridge is an extension to ISO9660, so 8.3 filenames still exist on the CD, but a mapping file is added that lists the long version of the filename for each 8.3 filename. A few things like symbolic links are also supported I think(?). Because Rock Ridge was created by the enemy, Microsoft chose to write a new standard for long filenames when Windows 95 came out sometime later. It's called "Joliet" and is different than "Rock Ridge". Just about all the Windows CD burning software writes long filenames as Joliet. I have about 6 different Windows applications (Nero, etc.) and none of them support Rock Ridge. You can create CD's with both Rock Ridge and Joliet, but you have to use the Unix "mkisofs" program; eg.: mkisofs -o /var/tmp/CDIMAGE -R -J -V "VOLNAME" This will create a CD "image" (/var/tmp/CDIMAGE) with both Rock Ridge (-R) and Joliet (-J) long filename extensions. You can then use your PC burning application to burn the binary image to a CD, or use the Unix utility "cdrecord"; eg: cdrecord -v fs=6m speed=8 dev=2,5,0 /var/tmp/CDIMAGE Trivia: ISO9660 was developed under the name "High Sierra Filesystem" (you'll still see references to HSFS under Solaris). High Sierra came from the location that the developers met (California?, Nevada?). As a bit of twisted humour, "Rock Ridge Filesystem" is named after the town in the movie Blazing Saddles. Regards, Brendan. Hi everyone, A colleague passed me this question yesterday. His job will be creating cdroms with appication parts for our customers. He sits in front of a Win 2K machine. He has several *.tar.Z files and - customer depending - license files. So every disk has to be created separately. He uses nero for burning cdroms. Now he wants to know the settings to make the cdrom accessable under a sol box. Every attempt failed so far. Means the cd was accessable under win and even linux but not under solaris. Another thing he reported was a strange behavior of the solaris (currently we're using 2.6 on an Ultra 10) - under win/linux all characters are displayed as upper case, in solaris as lower case. In the web you find a lot of "how to create a bootable cdrom", but that's not what I need. Do you have any suggestions regards may _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From cmap_sec at yahoo.co.uk Mon Apr 28 09:04:26 2003 From: cmap_sec at yahoo.co.uk (=?iso-8859-1?q?Elaine=20.?=) Date: Mon, 28 Apr 2003 14:04:26 +0100 (BST) Subject: SUMMARY: vnc placing windows Message-ID: <20030428130426.79759.qmail@web21009.mail.yahoo.com> hi I wrote about the requirement for me to place the window in vnc for oracle reports server to start.Trying to solve it with solaris/oracle or vnc.:) i've switched gui to CDE in vnc (starts twm by default) and thus cured the problem. Thanks Elaine __________________________________________________ Yahoo! Plus For a better Internet experience http://www.yahoo.co.uk/btoffer _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnrams at cox.net Sat Apr 26 04:00:03 2003 From: johnrams at cox.net (John Rams) Date: Sat, 26 Apr 2003 01:00:03 -0700 Subject: Summary: Login failure: /usr/lib/libc.so.1 too many open files Message-ID: <00f701c30bc9$da61cb40$6501a8c0@svakahome> Thanks to Casper Dick, Rick Andersonand Alan Bradley. Following are their answers. It was badly delayed to summarize. And my bad is a colleague in that remote site reinstalled OS from jumpstart without preserving the original config for evaluation and debugging! Casper's mail: The error you see is typical for systems that have been hacked with a specific rootkit. You should reinstall or ssave the disk for forensic purposes. It likely contains a few trojans. Ric's mail: Last box I saw with this symptom had been rooted. You'll probably find a secure shell (version 1) running on a high numbered port, along with a password capture program. run nmap against the box and then telnet to anything that nmap finds listening to see if the port in question responds with SSH-other stuff- If it does, format and reinstall, get the current patches in -AND- add the following lines to /etc/system so the next buffer overrun attack falls on the floor instead of getting in. * * Security fix - prevent execution on stack... set noexec_user_stack=1 set noexec_user_stack_log=1 Be sure you reboot after changing /etc/system... Alan's Mail: There is a posting in the archives of a similar problem someone had: http://www.sunmanagers.org/pipermail/sunmanagers/2002-August/015720.html There doesn't seem to be a resolution, but perhaps you could contact them and see if they did manage to resolve it. Regards John Rams > -----Original Message----- > From: sunmanagers-admin at sunmanagers.org > [mailto:sunmanagers-admin at sunmanagers.org] On Behalf Of > johnrams at cox.net > Sent: Monday, April 07, 2003 3:32 PM > To: sunmanagers at sunmanagers.org > Subject: Login failure: /usr/lib/libc.so.1 too many open files > > > Managers: > > On an Ultra 60, i am getting following error. How would i > logon to the system. FTP works with no problem. Tried to copy > the shared object by ftp, no luck. > > What can i do to resolve without having to reinstall? May be > booting from cdrom and copying contents of /usr file system. > > > $ telnet > Trying > Connected to . > Escape character is '^]'. > > SunOS 5.8 > > ld.so.1: login: fatal: /usr/lib/libc.so.1: Too many open files > Connection closed by foreign host. > > thanks > John _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From StephenB at datacom.co.nz Sun Apr 27 17:06:20 2003 From: StephenB at datacom.co.nz (Stephen Barnett (DSL AK)) Date: Mon, 28 Apr 2003 09:06:20 +1200 Subject: SUMMARY: Ultra-10 Model Message-ID: <04F92BA33955D51195B700508B4AB6E90D77B104@dslak12.dnznet.co.nz> Hi, I've been asked to clarify this. I had been asked to determine the "model" of Ultra-10 that a client has in order to determine what type of memory it uses. Cheers, Stephen -----Original Message----- From: Stephen Barnett (DSL AK) Sent: Wednesday, 23 April 2003 3:56 p.m. To: sunmanagers at sunmanagers.org Subject: SUMMARY: Ultra-10 Model It turns out that all I needed was the CPU speed. Thanks to all those who replied. Cheers, Stephen W. Barnett Systems Consultant - Unix Division Datacom Systems Limited Auckland New Zealand _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From wkwan at grow.net Mon Apr 28 12:05:12 2003 From: wkwan at grow.net (Kwan, William) Date: Mon, 28 Apr 2003 12:05:12 -0400 Subject: Summary: use DLT35/70 tape in a DLT40/80 drive Message-ID: <417F5E81F35DDB4CA208F5B00AFA8FE65A16@spruce.grow.net> Hopefully this is not too early to finalize the question. The original problem is trying to use DLT8000 (40/80) tapes that were written by a DLT7000(35/70) drive on a DLT8000 drive again. Tapes inserted in a DLT8000 drive will be ejected by the hardware. NO OS was involved. I wasn't able to send any command to the drive. I got suggestions from format, initialize to dd the tape. Mike Salehi suggested that a tape used in the lower density drives may need to be degaussed. On this HP page (http://h18000.www1.hp.com/products/quickspecs/11403_na/11403_na_Changes .HTML), it mentioned: "NOTE: To write to DLT Type IV data cartridges previously written by a DLT 20/40, DLT 35/70, and/or DLT 40/80, each data cartridge must first be degaussed before usage with a DLT VS 40/80 or DLT1 tape drive or device. For new, unused media, this step is not necessary." Since there's no button or anything on the tape drive to set the density, I think I need to find a big magnet. > -----Original Message----- > From: Kwan, William [mailto:wkwan at grow.net] > Sent: Monday, April 28, 2003 10:30 AM > To: sunmanagers at sunmanagers.org > Subject: use DLT35/70 tape in a DLT40/80 drive > > > Running out of tapes, want to try used DLT tapes in a DLT40/80 drive. > These tapes were used in a DLT35/70 drive before (they are 40/80 > tapes). It doesn't work... anyway to break this? > > ____________________________ > William Kwan > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From paul.coray at unibas.ch Mon Apr 28 12:09:14 2003 From: paul.coray at unibas.ch (Paul Coray) Date: Mon, 28 Apr 2003 18:09:14 +0200 Subject: SUMMARY: V120: Double serial connection freezes one host In-Reply-To: <3EACDECD.6060603@unibas.ch> References: <3EACDECD.6060603@unibas.ch> Message-ID: <3EAD522A.9050005@unibas.ch> A huge thanks to everybody who responded so quickly. In summary, the problem is resolved by changing or disabling the keyboard/serial break sequence. This can be done in two ways: - Using kbd(1)(non-persistent): # kbd -a alternate (changes break sequence to ~^b, that is, a tilde (~) followed by a control-B), or # kbd -a disable (disables keyboard and serial abort). - Editing /etc/default/kbd: Uncomment KEYBOARD_ABORT=disable or KEYBOARD_ABORT=alternate (effects same as above, but persistent). For immediate effect, do # kbd -i Alternatively, you can use Martin Schmitt's nobreak package, available at http://www.scsy.de/~mas/nobreak/ Rgds Paul > Sunmanagers, > > I have two SunFire V120, which operate side by side in a headless mode. > As I have no terminal concentrator, but I still need to have access to > the console of both servers, I planned to connect to each one's console, > doing a 'tip hardwire' on the other host, which would connect to the > console on serial a. This way I could even 'wake up' a host, sending a > 'poweron' to the LOM CLI. > > ----------------- ------------------ > | | | | > | serial a/lom --------- serial b | > | BOX A | | BOX B | > | serial b --------- serial a/lom | > | | | | > ----------------- ------------------ > > This seemed to work pretty well, but there is a big drawback: Each time > I (re)boot a host, e.g. sending an init 6 from box a to box b, the one I > use for terminal access (box a) freezes into ok-prompt. So I have to > wait for box b to be online, log in, connect to console of box a and > just type 'go' at the ok... > > This only happens, when both serial lines are connected as above. I > don't know if other systems would behave the same, as this is the first > time I did this kind of serial connect configuration. > > Obviously the host to boot sends a break signal to the connecting host, > causing the freeze. > Has anybody found out how to prevent this kind of behaviour? > > The two boxes have prettty much identical setup: > > # uname -a > SunOS ub-files 5.9 Generic_112233-04 sun4u sparc SUNW,UltraAX-i2 > > # prtconf -V > OBP 4.0.14 2002/07/03 09:17 > > # grep hardwire /etc/remote > hardwire:\ > :dv=/dev/term/b:br#9600:el=^C^S^Q^U^D:ie=%$:oe=^D: > > # eeprom > ttyb-rts-dtr-off=false > ttyb-ignore-cd=true > ttya-rts-dtr-off=false > ttya-ignore-cd=true > ttyb-mode=9600,8,n,1,- > ttya-mode=9600,8,n,1,- > output-device=ttya > input-device=ttya > > I need to get this configuration to production soon. Please help! Send > email and I will summarize to the list. > > Thanks for assistance. > > -- Mit freundlichen Gruessen Paul Coray Administrator Server und Netzwerk Oeffentliche Bibliothek der Universitaet Basel EDV-Abteilung Schoenbeinstrasse 18-20 CH-4056 Basel Tel: +41 61 267 05 13 Fax: +41 61 267 31 03 mailto:paul.coray at unibas.ch http://www.ub.unibas.ch _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mclausen at csit.fsu.edu Mon Apr 28 14:34:39 2003 From: mclausen at csit.fsu.edu (Matthew R. Clausen) Date: Mon, 28 Apr 2003 14:34:39 -0400 Subject: SUMMARY: StorEdge 3310 Question.... Message-ID: <3EAD743F.20008@csit.fsu.edu> The general consensus is that these come with trays, but are rendered unusable by Sun to act as drive trays. So I'm stuck buying the 73GB drives direct from Sun. Which after speaking with my sales rep. is the better way to go because using 3rd party drives in the 3310s voids the warrenty and the maintenance contracts (isnt that a pain!). Thanks to everyone that responded, and to those of you in the same boat as I am, we're stuck with it! :( ------------------------------------------------------------------------ Matthew R. Clausen mclausen at csit.fsu.edu Computational Science & Information Technology Florida State University voice=850-644-0193 400 Dirac Science Center Library pager=TBD Tallahassee, Florida 32306-4120 fax=850-644-0098 ------------------------------------------------------------------------ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Samier.Kesou at bfa.de Tue Apr 29 02:10:22 2003 From: Samier.Kesou at bfa.de (Samier.Kesou at bfa.de) Date: Tue, 29 Apr 2003 08:10:22 +0200 Subject: SUMMARY:trussing a slow process Message-ID: <20030429061527.EFEE31E67A@sunportal.sunmanagers.org> hi, thanks for the replies. joe fletcher, Bertrand Hutin, Vincent Detzel , Alan Pae LWP - light weight proccess ~ thread process. there for it can not be avoided. the overall understanding was that the process is waiting for some child process to get finished. This could be a slow disk or a slow network connection, or a slow tape drive. I went for the network and found out that there is a dial up link from my site to the backup client site. thanks for the replies, Samier. ----- Weitergeleitet von Samier Kesou/Extern/11/BfA-Berlin am 28.04.2003 16:33 ----- Samier.Kesou at bfa.de An: sunmanagers at sunmanagers.org Gesendet von: Kopie: sunmanagers-bounces at sunm Thema: trussing a slow process anagers.org 28.04.2003 15:55 Hi, i have an extrem slow backup process on a solaris 8 server. trussing it gives me multple of the following lines: ... lwp_cond_wait(0xFF08DFA0, 0xFF08DFB0, 0x00000000) (sleeping...) lwp_cond_wait(0xFF00FFA0, 0xFF00FFB0, 0x00000000) (sleeping...) ... man -k lwp gives me all the syscalls but i havent found the information what lwp does and why it seems to slow down my backup process can it be avoided? thanks in advance Samier. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jcayeirog at bankinter.es Tue Apr 29 06:42:24 2003 From: jcayeirog at bankinter.es (jcayeirog at bankinter.es) Date: Tue, 29 Apr 2003 12:42:24 +0200 Subject: SUMMARY: How delete a file? Message-ID: Thankx everybody for yours answers; specialy to: ed.rolison at itc.alston.com Tom Lojewski John Mitchell Bernard Sadlowsky but it4s wrong: rm "-I" rm "\-I" The only sentence that run OK is: rm -- -I The "--" means ignore all other command line switches. Lot of thanks for everybody Regards J.Cayeiro Remitente sunmanagers-bounces at sunmanagers.org Hi everybody, A user touch a file who name is "-I" , how can i delete this file? -rw-r--r-- 1 root other 0 Apr 29 12:15 -I Regards J.Cayeiro ____________________________________ "Este mensaje y los documentos que, en su caso, lleve anexos, pueden contener informacisn confidencial. Por ello, se informa a quien lo reciba por error que la informacisn contenida en el mismo es reservada y su uso no autorizado esta prohibido legalmente, por lo que en tal caso le rogamos que nos lo comunique por la misma vma o por telifono (901 113 113) , se abstenga de realizar copias del mensaje o remitirlo o entregarlo a otra persona y proceda a borrarlo de inmediato." _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers |--------------------------------------------------------------------------| | Destinatario sunmanagers at sunmanagers.org | | Copia | |--------------------------------------------------------------------------| Un saludo Javier ____________________________________ "Este mensaje y los documentos que, en su caso, lleve anexos, pueden contener informacisn confidencial. Por ello, se informa a quien lo reciba por error que la informacisn contenida en el mismo es reservada y su uso no autorizado esta prohibido legalmente, por lo que en tal caso le rogamos que nos lo comunique por la misma vma o por telifono (901 113 113) , se abstenga de realizar copias del mensaje o remitirlo o entregarlo a otra persona y proceda a borrarlo de inmediato." _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rachid.boukhari at st.com Tue Apr 29 07:38:09 2003 From: rachid.boukhari at st.com (rachid boukhari) Date: Tue, 29 Apr 2003 11:38:09 +0000 Subject: SUMMARY: Install over the network References: Message-ID: <3EAE6421.6EB3A8B1@st.com> Hello managers, Thanks to all who've replied (many good ideas and interesting weblinks) The method described below is right. I've made one mistake on the client machine; I've forgotten the blank between the "dush" and "install" "ok> boot net - install". Thanks to Steve Maher who've noticed this. Berst Regards, Rachid > > -----Original Message----- > From: rachid boukhari [mailto:rachid.boukhari at st.com] > Sent: Tuesday, April 22, 2003 10:58 AM > To: sunmanagers at sunmanagers.org > Subject: Install over the network > > Hi folks, > > I haven't resolve my problem yet concerning the install over the > network, I've tried many solutions but none of them was successful. > Here is again an abstract of what I 've done and the results I've > obtained. > Just to specify I 'm working on the same subnet so no need of boot > server. > > On the install server : > Insert CD 1/2 > cd /cdrom/cdrom0/Solaris8/Tools > ./setup_install_server /proj/Solaris_install > eject CD 1/2 > Insert CD 2/2 > cd /cdrom/cdrom0/Solaris8/Tools > ./add_to_install_server /proj/Solaris_install > Insert CD boot > cd /cdrom/cdrom0/s0 > ./modify_install_server /proj/Solaris_install /cdrom/cdrom0/s0/Boot > At the end of this command I 've got these 2 sentences: > Setup Complete > Kill:14580: no such process > > I'm not used to see this message before. > > I add the ethernet adress of the client machine to the /etc/ethers of > the server > > cd /proj/Solaris_install > ./add_install_client -e 8:0:24:9b:c8:95 -n sunhy:nis titan sun4u > > On the client machine(titan) at the ok prompt: > boot net -install > > Here is the message > Timeout waiting for ARP/RARP packet > 2a00 Warning:boot will not enable cache > Enter filename [Kernel/sparcv9/unix] > > What's the problem please ? > > I've tried to see with a another client machine installed with the same > method few months ago but always the same problem; the only thing I've > changed is the install server directory. > > Thanks a lot!! > > [demime 0.99c.7 removed an attachment of type text/x-vcard which had a name > of rachid.boukhari.vcf] > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers [demime 1.01b removed an attachment of type text/x-vcard which had a name of rachid.boukhari.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Paul_Wilkinson at utsglobal.com Tue Apr 29 15:43:20 2003 From: Paul_Wilkinson at utsglobal.com (Paul Wilkinson) Date: Tue, 29 Apr 2003 21:43:20 +0200 Subject: SUMMARY: SSH2 over PPPoE on Sol 2.7 Message-ID: <200304291944.h3TJiXB14600@mail.utsglobal.com> Thank's to Ben Garrett and Henri Servomaa who both have this working, Ben with a Linksys router between the Sun and the ADSL modem, and Henri by adjusting the MTU down to that suggested by their ISP. Since a Linksys router is not an option I further researched the MTU size and located an interesting test at : http://www.dslreports.com/faq/5793 This test method produced an MTU size of 538 in my case, changing all interfaces made no difference to the problem though. After applying greater pressure to the Telco who installed the ADSL line and modem I was told they effectively lower the priority of SSH and PPTP (perhaps others too) by placing HTTP and FTP packets in a high priority queue. I'm still waiting on a decision by the telco executives as to whether this policy could/should be changed. Paul. I asked : > Solaris 2.7 system configured with Free PPP 2.3.11 and SSH 2.3.0 > (non-commercial version) from www.ssh.com. > > For some time I've reliably used this system to connect to a remote > system using dial-up Internet access. > > Using PPP to connect to ISP, SSH to remote site and tunnel PPP > through the SSH session I establish a secure point to point link > between the two systems. > > With the availability of ADSL locally, I tried replacing the working > PPP -> ISP with RP-PPPoE 3.5 -> ADSL router via le0. > > Using PPPoE and ADSL, general Internet access is great. Performance > over the SSH session is extremely poor and unreliable, even without > the 2nd PPP tunnel. > > I note many references to setting the MTU size correctly and have > tried various suggested values without effect (le0 at 1500 and ppp0 > at 1492, le0 at 1492 and ppp0 at 1492). > > Anyone using SSH over a PPPoE link successfully ? > > Does Solaris 8 PPPD (PPPoE) play better with SSH ? > > Any help, advice and info welcome. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From luandea at alt.co.ke Wed Apr 30 08:26:13 2003 From: luandea at alt.co.ke (Andrew (alt.co.ke)) Date: Wed, 30 Apr 2003 15:26:13 +0300 Subject: SUMMARY: CDE desktop on VNC viewer Message-ID: <6238763931ED6642ACFCD1FD97D457B62A4541@NTSERVER> To summarize. I did the following In the user home directory there is a directory called .vnc This contains a startupscript (xstartup) which is run when you start vncserver as that user. If you put something like the following in xstartup it will start CDE. #!/bin/sh xrdb $HOME/.Xresources /usr/dt/bin/dtwm & Then I ran the vnc and it worked fine. Many thanks for all you help _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Wed Apr 30 11:35:30 2003 From: martini at mrpeabody.llnl.gov (Dave Martini) Date: Wed, 30 Apr 2003 08:35:30 -0700 (PDT) Subject: SUMMARY: dbm map "access": missing map file /etc/mail/access error message Message-ID: <200304301535.h3UFZUiY012671@raider.llnl.gov> Much thanks to Rich Kulawiec who helped me with this until it was solved. My Original question: I'm getting this error in my console window dbm map "access": missing map file /etc/mail/access I have solaris 8, sendmail 8.12.9. I have the access file in /etc/mail but there is nothing in it. What entries does it want in this file and why do I keep getting this error in my console window every few minutes? The Fix: I had the access file in /etc/mail which I created with a touch command but this wasn't the format that sendmail expected. My sendmail.cf expects the contents of access.db to be of format "dbm" because I have this line in my sendmail.cf file Kaccess dbm -T /etc/mail/access So, running this: /usr/sbin/makemap dbm /etc/mail/access < /etc/mail/access Created the access.db file in /etc/mail of the correct format, "dbm" for sendmail. Some sendmail.cf files could want the "hash" version instead: There may be a line in sendmail.cf that looks like this Kaccess hash -T /etc/mail/access So you would run this command to create the "hash" version of the access.db file. /usr/sbin/makemap hash /etc/mail/access < /etc/mail/access Dave Martini LLNL _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From wolf at uen.org Wed Apr 30 14:12:20 2003 From: wolf at uen.org (Wolfgang Schwurack) Date: Wed, 30 Apr 2003 12:12:20 -0600 Subject: SUMMARY: patch Berkeley DB References: <3EAFE550.8050006@uen.org> Message-ID: <3EB01204.1090100@uen.org> It would help if I would down load the correct files. I should have down loaded patch.4.1.24.1 and ran the patch command on that sorry must of had a brain fart on this Wolfgang Schwurack wrote: > Has any one appled a patch to Berkeley DB? I when on sleepycat web > site which gave the information on how to apply the patch but not sure > if it's working correctly. This is what is have > > Solaris 9 > Sunfire V120 > BerkeleyDB 4.1.24 > > command to apply patch from the root directory of the distribution. I > enter "patch -p0 < ../db-4.1.24.zip" and File to patch I enter > "btree/bt_open.c" > > I keep get this display on the screen, still running 20 min now, Is > this correct? > 15764 > 15764 > Looks like an ed script. > > > /opt/src/db-4.1.24 508 # patch -p0 < ../db-4.1.24.zip > Looks like an ed script. > File to patch:btree/bt_open.c > 15764 > 15764 > Looks like an ed script. > 15764 > 15764 > Looks like an ed script. > 15764 > 15764 > Looks like an ed script. > > -- -- Wolfgang Schwurack Unix System Administrator University of Utah/Utah Education Network Tel: (801) 587-9444 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bbrown at harper.cc.il.us Wed Apr 30 15:04:49 2003 From: bbrown at harper.cc.il.us (Bob Brown) Date: Wed, 30 Apr 2003 14:04:49 -0500 Subject: SUMMARY - Solaris 8 on Intel D845PT -- lan troubles! Message-ID: All fixed..I found something on sunsolve.... I needed to do "eeprom acpi-user-options=0x2". Upon reboot, ALL is well!!! original message: I'm trying to install solaris x86 8 2/02 on a system with an intel D845PT motherboard. It has built-in lan. The lan info that I've gotten seems to describe the LAN as: PCI:8086,3013 Intel i815/82559 I made some changes that I found documented in this list...changes to /boot/solaris/devicedb/master and /etc/driver_aliases The interface comes up as "iprb0" Once I do the ifconfig plumb etc, I can give it an IP and ping itself (works fine), but it can't see ANYTHING else on my network...even on the same subnet...even it's router (even after I add the default route). The H/W is fine...this pc came equipped with windows-xp, which talked to the network just fine. After changing O/S's, it wont talk to the network at all (except to it's own ethernet interface). Any ideas? thanks! -Bob bbrown at harpercollege.edu #### #### Bob Brown - KB9LFR Harper Community College ## ## ## Systems Administrator Palatine IL USA #### #### Saved by grace _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Jeremy.Russell at chickasaw.net Wed Apr 30 15:39:28 2003 From: Jeremy.Russell at chickasaw.net (Jeremy Russell) Date: Wed, 30 Apr 2003 14:39:28 -0500 Subject: SUMMARY: IP Multpathing Message-ID: <352B04312822444A962714393AED8A4D026847@ADAEVS01.int.chickasaw.net> Well, Thanks to all who replied. These are the resources I was pointed to for IPMP http://www.samag.com/documents/s=7933/sam0305a/0305a.htm http://www.sun.com/blueprints/ http://www.eng.auburn.edu/users/doug/howtos/multipathing.html http://www.sun.com/solutions/blueprints/0327-608/2011.pdf But, I ended up not using IPMP, I can't have to active IP's. So I wrote this little script to run. <------------------ cut here -------------------------> #!/bin/sh # This script will check the status of the nic, and fail over if there is a problem. PATH=/bin:/sbin:/usr/bin:/usr/sbin INTA="iprb0" INTB="iprb1" IP="10.64.16.1" HBIP="192.168.255.1" FAULT="$1" PING="ping -c 1 -t 1" while [ 1 = 1 ] do case $FAULT in 1) $PING $IP >>fault.log 1>&2 if [ $? = 1 ] then ifconfig $INTB $IP down ifconfig $INTB $HBIP netmask 255.255.255.252 -failover up ifconfig $INTA down ifconfig $INTA $IP netmask 255.255.0.0 failover up FAULT=2 fi ;; 2) $PING $IP >>fault.log 1>&2 if [ $? = 1 ] then ifconfig $INTA $IP down ifconfig $INTA $HBIP netmask 255.255.255.252 -failover up ifconfig $INTB down ifconfig $INTB $IP netmask 255.255.0.0 failover up FAULT=1 fi sleep 1 ;; reset) ifconfig $INTB down ifconfig $INTB $HBIP netmask 255.255.255.252 -failover up ifconfig $INTA down ifconfig $INTA $IP netmask 255.255.0.0 failover up ;; *) echo "usage: $0 { 1 | 2 | reset } ( where 1 or 2 is the STANDBY nic )" exit ;; esac done <----------------------- Cut again -------------------------------------> The only problem is this will only fail over on a complete interface failure, A link failure will don nothing. Does anyone have a take on what I might do to sense a link failure (some ifconfig option maybe?). Thanks for the help!! Jeremy Russell Network Administrator, CNI 580-272-2707 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Marco.Greene at nexinnovations.com Tue Apr 29 16:59:45 2003 From: Marco.Greene at nexinnovations.com (Greene, Marco) Date: Tue, 29 Apr 2003 16:59:45 -0400 Subject: SUMMARY: SDLT Tape Drive on Compaq MDR Message-ID: <62F2F0D25E95BE46B559C4E5347849800148D0A5@MISEXCVS01.corp.nexinnovations.com> I figured it out with a little pointer from HP (Compaq) Gold Support. If you don't have any SCSI maps defined all SCSI devices can be seen by all Fibre ports. As it turns out, I just had to increase my entries for number of luns in st.conf and rebuild the VERITAS sg driver. Mind you now everything is going through one fibre port, so balancing the tape drives is the next step. Tomorrow maybe. Regards, Marco Greene, BEng Infrastructure and Recovery Architect NexInnovations Inc. -----Original Message----- From: Greene, Marco Sent: April 29, 2003 2:23 PM To: 'List, Sun Managers' Subject: SDLT Tape Drive on Compaq MDR I currently have 4 SDLT tape drives connected through a Compaq MDR which works great! However, because I am backing up 120 servers we have decided to get a 5th tape drive. Now the MDR sees the tape drive fine. It is a 4 LVD - 2 FC with 2 SDLT drives on each scsi bus and 2 scsi bus' per fibre channel port. This means that the 5th drive needs to go over the second fibre port. I know from the host point of view that it sees the MDR port but it can not see the drive. There are no presentations on the MDR restricting host access. The server is running Solaris 8, 64-bit with Compaq SecurePath V2.1a. (SecurePath is required for the HSG80 drive.) [demime 1.01b removed an attachment of type application/ms-tnef which had a name of winmail.dat] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Wed Apr 30 11:34:23 2003 From: levins at westnet.com (Adam and Christine Levin) Date: Wed, 30 Apr 2003 11:34:23 -0400 (EDT) Subject: SUMMARY: confirmation of a theory (regarding A3500 disks) (fwd) Message-ID: Thanks to Morgan Jones and Sydney Weinstein . Both confirmed that I can do what I thought I could. The one thing to be aware of is that growfs will lock the filesystem while growing (which is mentioned in the growfs manpage, along with a way to alleviate the problem using the -s flag). Original question below. Thanks very much, -Adam We have a machine with an A3500 attached. It has four LUNs mounted: c4t5d0s0 c4t5d1s0 c5t5d0s0 c5t5d1s0 Each is around 300GB. We just added another A3500, because we need bigger disk volumes. What I'd like to do is unmount the existing volumes, create DiskSuite metadevices out of them, and then create concats with the four new volumes, so we'd have: d10 = c4t5d0s0 + c8t5d0s0 d20 = c4t5d1s0 + c8t5d1s0 d30 = c5t5d0s0 + c9t5d0s0 d40 = c5t5d1s0 + c9t5d1s0 Then, I'd just growfs the volumes to be 600GB each. I'm pretty sure this should work. The main questionable part is whether taking the curret /dev/dsk/c4t5d0s0 (etc) and making metadevices out of them will somehow disrupt the data. I know that when we mirror root disks, we just create d0 with /dev/dsk/c0t0d0s0, and nothing bad happens, so I think this will work. Can anybody confirm that? I'd rather do it this way than have to copy the data from the original LUNs to metadevices made out of the new LUNs in the second unit -- that could take quite a while. Thanks very much, -Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From peter at ashlyn.co.uk Wed Apr 30 13:22:19 2003 From: peter at ashlyn.co.uk (Peter Stokes) Date: Wed, 30 Apr 2003 18:22:19 +0100 Subject: [SUMMARY] NFS file limit on 2.6/8 Message-ID: Hi All Thanks to Jay Lessert Alan Scott (Anyone else I missed.....) They both suggested looking at the largefiles option for mount, it appears that ufs does this by default, but on 2.6 at least the nfs mount does not and you have to specify 'largefiles' as a mount option in vfstab, all is working fine now that is done. Peter Original Mail ---------------- Hi As the search facility is temporarily not operating and I am sure it is in there, can someone let me know if I can create a 30Gb+ file over NFS coming from a 2.6 client to 8 server, if so how? Thanks Peter --------------------------------- Peter Stokes Ashlyn Computer Services Ltd Tel: +44 (0)1636-627900 Fax: +44 (0)1636-627909 Mbl: +44 (0)7977-532320 Web: http://www.ashlyn.co.uk --------------------------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers