From netcomrade at bookexchange.net Fri Apr 1 14:46:10 2005 From: netcomrade at bookexchange.net (NetComrade) Date: Fri, 1 Apr 2005 14:46:10 -0500 (EST) Subject: SUMMARY: What to do about /tmp file system when it's not part of swap? In-Reply-To: Message-ID: It was confirmed that this configuration is valid. However, we do use /tmp heavily, and I changed it to tmpfs as was suggested fd - /dev/fd fd - no - /proc - /proc proc - no - /dev/dsk/c5t0d0s3 - - swap - no - /dev/dsk/c5t0d0s0 /dev/rdsk/c5t0d0s0 / ufs 1 no - /dev/dsk/c5t0d0s6 /dev/rdsk/c5t0d0s6 /usr ufs 1 no - /dev/dsk/c5t0d0s1 /dev/rdsk/c5t0d0s1 /var ufs 1 no - /dev/dsk/c5t0d0s5 /dev/rdsk/c5t0d0s5 /opt ufs 2 yes - thanks! On Fri, 1 Apr 2005, NetComrade wrote: > We accidentally made /tmp a file system when we setup the box. > > What would you advise to do about it? Just remove from vfstab? > The system came up with swap using the remainder of the disk: > > /dev/dsk/c5t0d0s0 1949551 247433 1643632 14% / > /dev/dsk/c5t0d0s6 4038702 918459 3079856 23% /usr > /proc 0 0 0 0% /proc > fd 0 0 0 0% /dev/fd > mnttab 0 0 0 0% /etc/mnttab > /dev/dsk/c5t0d0s1 4133838 105715 3986785 3% /var > swap 13932112 24 13932088 1% /var/run > /dev/dsk/c5t0d0s5 6053358 51817 5941008 1% /opt > /dev/dsk/c5t0d0s3 33038277 430437 32277458 2% /tmp > > Would like to figure out what's best before encapsulating the disk (we are > running an old version of vxvm that requires 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 bshah at citadon.com Fri Apr 1 14:48:21 2005 From: bshah at citadon.com (Bhavesh Shah) Date: Fri, 1 Apr 2005 11:48:21 -0800 Subject: SUMMARY: Apache 2.0 on Solaris 8 Message-ID: Hi, Well I was able to resolve this issue. It turned out to be openssl compilation issue. The solution is: Compile and install openssl with shared option. My original question was: Hi Gurus, I downloaded apache-2.0.53 and openssl-0.9.7f source from apache/openssl site and compiled on Solaris 8 and it went through without any errors. It works fine without ssl but when I start apache with ssl it gives me following error: # ./apachectl startssl Syntax error on line 251 of /opt/apache/conf/httpd.conf: Cannot load /opt/apache/modules/mod_ssl.so into server: ld.so.1: /opt/apache/bin/httpd: fatal: relocation error: file /opt/apache/modules/mod_ssl.so: symbol X509_INFO_free: referenced symbol not found I used following to compile and install Apache: ./configure --prefix=/opt/apache --enable-rewrite=shared -enable-proxy=shared --enable-ssl=shared enable-ssl --with-ssl=/usr/local/ssl --enable-mods-shared=all make && make install It went fine without any errors. Does anyone has faced this problem? Do I have to setup LD_LIBRARY_PATH? Will summarize. Regards B _______________________________________________ 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 sysadmin.linux at gmail.com Fri Apr 1 14:59:41 2005 From: sysadmin.linux at gmail.com (Linux Admin) Date: Fri, 1 Apr 2005 13:59:41 -0600 Subject: SUMMARY: Windows 2003 on SUN V880 Message-ID: <69693499050401115944b71f4@mail.gmail.com> Thanks to all the folks that wished me Happy April's Full. I do the same! I hope I did not cause to much of loss in productivity. Ilya _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Fri Apr 1 15:22:23 2005 From: levins at westnet.com (Adam Levin) Date: Fri, 1 Apr 2005 15:22:23 -0500 (EST) Subject: SUMMARY: additional info on quick V890 question Message-ID: Additional thanks to: joe_fletcher Adam Tomkinson Carl Marino RICHARD WINFIELD Michael Horton Geoff Lane "Bhaskara, Srikalyan (Sri), ALABS" Ilya Birman Bernd Schemmer A few tidbits: 1) I should try contacting Sun to let them remove the RSC password 2) I should send the machine back to the supplier and get one without a password 3) Someone speculates that admin/manager might be the user/pass combo Several suggested fixing this from a booted machine, but since we have not yet set up the machine, I need a console *first*. Several also mentioned that the power button trick does work, but it's quirky. You have to hit the button while the light is flashing quickly, which takes a few seconds after power on to happen, and you can miss it, so be careful. Finally, yes, I'm sure it's an 890 and not an 880, and yes, an 890 has a serial mgmt port, so labelled as part of the RSC card (and next to the net mgmt port). It turned out, after all, that the wrong cable was plugged into the TTY port, such that the RSC messages were still coming in on what we thought was the console. That's what happens when you're working remotely. Once we straightened out getting the correct cable into the TTY port, we were good to go at the OK prompt. Thanks again, all. -Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From asaenz at Found-Tech.com Fri Apr 1 16:20:29 2005 From: asaenz at Found-Tech.com (Al Saenz) Date: Fri, 1 Apr 2005 16:20:29 -0500 Subject: SUMMARY: System will not config after sys-unconfig Message-ID: My sisid files were renamed. >This is old, Solaris 8 sys-unconfig is broked. Get the patch Sys-unconfig doesn't have a problem. It is good to know this info just incase I run into a system that has the files but still wont config. Thank you Ryan, Michael, Alan, and Sunmanagers. al ***Solution used below from Ryan*** There are two scripts in /etc/rc2.d S30sysid.net -and- S71sysid.sys Most likely someone disabled these scripts. Something to the extent of renaming them so they don't begin with a 'S' for start. You can always run these manuall from the command line. /etc/init.d/sysid.net /etc/init.d/sysid.sys Regards, Ryan On Fri, 1 Apr 2005, Al Saenz wrote: > Well I have another problem. > > Machine: > Sun clone AX1105 mother board > Two onboard NICs > Solaris 8 > > I ran sys-unconfig on a system and now the system will not prompt for > configurations after a reboot. > It just come to the system login and I enter root and bang I am in at the # > prompt. > > I tried to type > > #sysidtool > sysidtool: not found > > > The document below says after I run sys-unconfig and reboot I will be prompted > for system config. > > http://docs.sun.com/app/docs/doc/806-5529/6jehkcs4b?a=view > "Because you unconfigured the system, you will be prompted for configuration > information by sysidtool programs when the system reboots." > Thank you for your assistance. > All this just to add DHCP client, good thing the system isn't in a operating > environment. > DHCP client should be able to be added w/o a sys-unconfig? > al > PS: I checked the archive and didn't find anything like my problem > _______________________________________________ > 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 Rich.Bonfoey at thenewstribune.com Fri Apr 1 14:12:12 2005 From: Rich.Bonfoey at thenewstribune.com (Rich Bonfoey) Date: Fri, 1 Apr 2005 11:12:12 -0800 Subject: SUMMARY:RE: TCP_keepalive Message-ID: Many thanks to Eric Voisard and Sam Nelson for sharing their expertise. Eric provided insight to another variable tcp_time_wait_interval, that it can have an affect on connections. Sam , very graciously provided a script that he uses to tune and sanitize the tcp stack Below is the original message With Eric and Sam's replays Thanks again for the info Rich -----Original Message----- From: Rich Bonfoey Sent: Thursday, March 31, 2005 9:55 AM To: Sun Managers (sunmanagers at sunmanagers.org) Subject: TCP_keepalive Greetings Have any of you, had much dealings with the tcp_keepalive ? On Solaris 8 and 9 the value is 7200000 or 2 hours. What are some of the pit falls if this value was lowered to about 10 minutes. I have read the RFC but was looking for people who have had real experience with tweaking this value. Part of the reason for tweaking, is we have a large number of PC and OS 10 users connecting to several different systems and they have been having performance issues. Our dba decided to bring in a consultant ( that's another story ) and they said this value should be 5 minutes. I am not quite buying it, as I have read that there are multiple opinion on the correct use of tcp_keepalive. Richard Bonfoey The News Tribune Information Systems Successfully Meeting the Business Needs of The News Tribune through Information Technology _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Nick at Pettefar.com Mon Apr 4 09:47:32 2005 From: Nick at Pettefar.com (Nick j Pettefar) Date: Mon, 4 Apr 2005 15:47:32 +0200 Subject: SUMMARY: Foolish Move In-Reply-To: <20031217152936.18A321E8B9@sunportal.sunmanagers.org> References: <20031217152936.18A321E8B9@sunportal.sunmanagers.org> Message-ID: <1112622452.4251457452031@webmail.eso.org> Thanks for the sage advice from those in the know. I did a foolish mv * /backupdir in the root directory and removed all the root files and directories. I could not do backupdir/bin/mv because it could not find the libraries. I had a machine (V240) without a CD drive. Many nice people told me to use usr/sbin/static/mv (and other statically linked utils in this directory) and this fixed the problem. Thanks! Other people should remember this too, judging by all the "You're hosed" type replies. Other people were on holiday, apparently. Nick One OS to fool them all, one browser to find them One email client to bring them all, and through security holes, blind them... Nick at Pettefar.com DoD 1069 MAG 73516 Bros 650 ZZR1100D R90s Z88s _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Sun Apr 3 16:33:53 2005 From: levins at westnet.com (Adam Levin) Date: Sun, 3 Apr 2005 16:33:53 -0400 (EDT) Subject: SUMMARY RE: problem jumpstarting a V890 (fwd) Message-ID: Thanks to Coy Hile and Jeffrey Donovan. I tried an installboot, but that didn't help. Jeffrey hit it, though: I could do an update install. After the flash install on the V890, I did a boot net and did an update install of 02/04 off my jumpstart image. That did it! Thanks! -Adam -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Adam Levin Sent: Sunday, April 03, 2005 10:12 AM To: Sun Managers Mailing List Subject: problem jumpstarting a V890 I'm having trouble properly jumpstarting a V890 using flash archives. We have flash archives of V240s and V210, but this is our first V890. The flash archives from the older machines are Solaris 8 07/03, patched. They have /usr/platform/SUNW,Sun-Fire-V890, which tells me that they should support the V890 platform. Note that we could not jumpstart with the unpatched 07/03 boot image, because it did not natively support the V890. We installed the 02/04 boot image, and then I was able to boot the V890 over the net. So, we lay down the flash image, but I get the boot: cannot open kernel/sparcv9/unix Enter filename [kernel/sparcv9/unix]: I'd love to be able to use the flash archive instead of rebuilding a system from scratch with Solaris 8 02/04. Is there any way to do that? The 07/03 flash archive isn't the *most recent* patch cluster, but it's within the last few months (enough so I at least see the V890 /usr/platform directory). Where is "kernel/sparcv9/unix"? Am I having a bootblk problem here? Thanks, -Adam _______________________________________________ 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 dskinner at blueprint.org Mon Apr 4 15:25:31 2005 From: dskinner at blueprint.org (Drew Skinner) Date: Mon, 04 Apr 2005 15:25:31 -0400 Subject: SUMMARY: Slightly different installation Message-ID: <425194AB.2040709@blueprint.org> Hi; Didn't receive too much to help here. Ryan suggested this site: http://www.lka.ch/projects/solcdburn/solcdburn.html and commented about UDF vs. ISO format. Someone else pointed out a Blueprint doc, but didn't mention it. I'm going to look a little deeper and toss together a web page once I have what I want working. All the best, Drew. -------- Original Message -------- Subject: Slightly different installation Date: Thu, 31 Mar 2005 10:58:25 -0500 From: Drew Skinner To: sunmanagers at sunmanagers.org Hi All; I wasn't actually quite sure what to call the subject, and I don't think this has been broached before, but here goes. I need to have a 'reference' system that will contain a production build of both Solaris and our software release. This system will be used to clone every system we ship. What I'd like to be able to do is ship a 'rescue' DVD with these systems. To do that I'd like to be able to burn a DVD like so: boot partition + mini root == jumpstart + flash archive Has anyone had experience with Solaris 10, and the type of custom environment I describe above? Note that we'll be including other software such as 64bit x86 MySQL, etc. The target 'reference' platform is a V20z with a 3511. If anyone is burning DVD's on Solaris in this manner I'd love to hear about it and your recommendations. Will summarize. Thanks, Drew. -- ("'-"-/").___..-""'-._ Drew Skinner '6_ 6 ) '-. ( ).'-.__.') Senior Systems Administrator (_Y_.)' ._ ) '._ '. "-..-' The Blueprint Initiative '--'_..-_/ /--'_..' ,' email:dskinner[[at]]blueprint.org (il),-" (li),' ((!.-' phone: 416.596.6251 ~~~~~~~~~~~~~~~~~~~~~~~ cell: 416.432.3500 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From alberto.dasilva at gmail.com Mon Apr 4 16:14:44 2005 From: alberto.dasilva at gmail.com (Alberto da Silva) Date: Mon, 4 Apr 2005 22:14:44 +0200 Subject: SUMMARY: SC service password - any ideas how to generate? In-Reply-To: <3c6dd26605040112211502fc67@mail.gmail.com> References: <3c6dd26605040112211502fc67@mail.gmail.com> Message-ID: <3c6dd266050404131427631ca6@mail.gmail.com> Question: Any ideas on how to generate the service password for the system controller? Answers: None Basically Sun can generate the password using an internal application https://sfservicepass.sfbay/mode-policy.html Why do you need the password? To re-enable components which are incorrectly Auto Diag'ed / Component Health Status disabled. I received info on a number of cases where AD had incorrectly CHS disabled additional components. eg. memory error disables memory + CPUs. In our case a network card loopback test on one card disabled 4 network cards - 2 onboard and 2 PCI. In some cases the component is changed, but the new component is still CHS disabled. To re-test the component you need to 'setchs -s ok -r "Testing" -c component' BUT you need a Sun supplied SC password to run setchs. If you have a server that has a system controller, and you are not on maintenance contract, then you are screwed. I suggest that if you are buying a new server, then before paying, insist on getting a permanent SC password. Alberto _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Francois.Bousquet at cgi.com Mon Apr 4 16:26:17 2005 From: Francois.Bousquet at cgi.com (Bousquet Francois) Date: Mon, 4 Apr 2005 16:26:17 -0400 Subject: SUMMARY: Sendmail on Solaris 9 Message-ID: <1F85129C88B3714185C912F3AF47605012A25C17@nc1981exc03.banquelaurentienne.ca> The problem was related to our ldap config (openldap libs) in /etc/nsswitch.conf. Removing it solve the problem. It seems like sendmail don't like openldap libs =) -----Original Message----- From: Bousquet Francois [mailto:Francois.Bousquet at cgi.com] Sent: January 13, 2005 3:48 PM To: Mailing List - SunManagers (E-mail) Subject: Sendmail on Solaris 9 I am trying to make mail command to send mail to an remote SMTP server on Solaris 9. Since the version of sendmail have change in Solaris 9, I need to configure sendmail to listen on localhost and to route the local clientmqueue. Here is the documentation I follow : http://sunsolve.sun.com/search/document.do?assetkey=1-25-74310-1 After following all the step, here is my error : # /etc/init.d/sendmail start 11085 Segmentation Fault 11084 Segmentation Fault Anyone have an idea of what I can do ? Syslogs are empty of error message.. I have no clue... Frangois _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From zhu_junca at yahoo.ca Mon Apr 4 16:48:11 2005 From: zhu_junca at yahoo.ca (Jun Zhu) Date: Mon, 4 Apr 2005 16:48:11 -0400 (EDT) Subject: summary: find the lost space + Message-ID: <20050404204811.50458.qmail@web51707.mail.yahoo.com> Hello, I got many responses. Lots of people recommend using lsof/pfiles, which failed in our environment. One possible reason is we are using vxfs filesystem and lsof doesn't recognize it. +++ oracle 5200 oracle cwd unknown file system type (vxfs), v_op: 0x783cdb 90 oracle 5200 oracle txt unknown file system type (vxfs), v_op: 0x783cdb 90 +++ Sherman Butler and Casper Dik suggests using: find /proc -links 0 -type f -ls It lists all the processes and can't tell which one is the stale one. Finally, I restart all the processes using that filesystem and the disk space is back.:-) Here I really appreciated the following people spending time to respond my question. Casper.Dik Petri Kallberg Sam Nelson Hall, Christian N. Allan West Alex Stade Chamkura, Balki Thomas M. Payerle Sherman Butler Allan West Thomas M. Payerle Brian Gerard Eric Lewandowski All best regards, zhu Original Question ------ I got many responses recommending lsof or fuser. I am sorry not to clarify my sitution in original mail. I don't know who/when the file was deleted and don't know the filename either. Just because of the weird space difference, I suspect some log files have been removed with processes up and running. I thought the /proc structure should contain the file node information and am writing a script to go through all the related process. In the mean time, I am open to other methods or opinions.:-) best regards, zhu ===== Hello everyone, One team member removed a big log file without killing the process, causing disk space un-released from process. We are running solaris 9 with volume manager 4.0. Is there anyway I can find that particular process without rebooting the whole system? thanks very much! zhu Here is the df output: Filesystem kbytes used avail capacity Mounted on /dev/vx/dsk/rootdg/a 1048576 946456 96334 91% /a #du -sk /a 1474 ______________________________________________________________________ Post your free ad now! http://personals.yahoo.ca _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bru1n at comcast.net Mon Apr 4 18:14:14 2005 From: bru1n at comcast.net (Marc Belanger) Date: Mon, 4 Apr 2005 18:14:14 -0400 Subject: SUMMARY: Reading material In-Reply-To: <4FF62919C3D35C4796D3211B226E71B47CCEF9@lonmailsp01.enterprise.corpad.timeinc.com> Message-ID: 1) Thanks to all who replied. I appreciate the effort. 2) I forgot about the list FAQ, and inappropriately posted here, instead of a more on-topic group, or list. My sincerest apologies. I'll summarize only because I said I would, maybe it could be added to the FAQ (or relevant source) and hopefully will close the thread. 3) Here's the list. I looked up ISBN's where I was able to find them. Almost all of these are in stock, and available at present. YMMV. Solaris Solutions for System Administrators (Wiley press) ISBN: 047143115X Solaris Systems Programming (Prentice Hall) ISBN: 0-201-75039-2 The Practice of System and Network Administration ISBN: 0201702711 Configuration and Capacity Planning for Solaris Servers ISBN: 0-13-349952-9 Unix Power Tools (O'Reilly) ISBN: 0-596-00330-7 Panic! Unix System Crash Dump Analysis ISBN: 0131493868 UNIX System Administration Handbook (3rd Edition) ISBN: 0130206016 Solaris Internals (2nd Edition) Coming to a bookstore soon! And the following websites were offered as well: http://www.sunmanagers.org http://www.sun.com/bigadmin http://www.sunhelp.org http://www.bookpool.org http://www.sun.com/blueprints http://www.piaffe.org/panic/ http://www.solarisinternals.com/si/index.php http://www.google.com Thanks again all. -Marc _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From eric at checkpoint.com Tue Apr 5 03:27:12 2005 From: eric at checkpoint.com (Eric Walg) Date: Tue, 5 Apr 2005 09:27:12 +0200 Subject: SUMMARY: tcsh for Solaris 2.7 In-Reply-To: <07D7946A4E98A048BD21C951EEDA8DBA045B04DA@MKEMLVEM02.e2k.ad.ge.com> Message-ID: <200504050727.j357REwx021786@michael2.checkpoint.com> Thank to: Anthony Richard Ronny Rahul Siva Kelly Gani And others for the help. Eric -----Original Message----- From: Chaturvedi, Rahul (GE Healthcare, non-ge) [mailto:Rahul.Chaturvedi at med.ge.com] Sent: Tuesday, April 05, 2005 8:57 AM To: Eric Walg Subject: RE: tcsh for Solaris 2.7 u need to change ur login shell in /etc/passwd for the user. -----Original Message----- From: Eric Walg [mailto:eric at checkpoint.com] Sent: Tuesday, April 05, 2005 12:13 PM To: Chaturvedi, Rahul (GE Healthcare, non-ge) Subject: RE: tcsh for Solaris 2.7 Thank you very much. I downloaded and installed tcsh and it works. Could you kindly inform me how I change my PATH, so that I automatically shall login with tcsh? Thanks, Eric -----Original Message----- From: Chaturvedi, Rahul (GE Healthcare, non-ge) [mailto:Rahul.Chaturvedi at med.ge.com] Sent: Tuesday, April 05, 2005 8:07 AM To: Eric Walg Subject: RE: tcsh for Solaris 2.7 try sunfreeware.com -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Eric Walg Sent: Tuesday, April 05, 2005 11:33 AM To: sunmanagers at sunmanagers.org Subject: tcsh for Solaris 2.7 Hi all, I just installed Solaris 2.7 by CD on a Ultra5 Sun Sparc. It appears that I have only sh, and I would like to install tcsh or bash. Does anyone have an idea what I could do? Thanks, Eric _______________________________________________ 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 dan.cave at gmail.com Tue Apr 5 10:10:22 2005 From: dan.cave at gmail.com (Dan Cave) Date: Tue, 5 Apr 2005 15:10:22 +0100 Subject: SUMMARY: V100 stuck at LOM prompt In-Reply-To: <2731bd3b050405044379e7f7f7@mail.gmail.com> References: <2731bd3b050405044379e7f7f7@mail.gmail.com> Message-ID: <2731bd3b050405071051f8f2d4@mail.gmail.com> Thanks to all of those who replied, Mike Imrick, Iain Miller, Mike Demarco, Doug Bell, Pablo Jejcic, Adam Tonkinson and Ryan Krenzischek, more specifically Mike Imrick and Doug Bell pointed me towards this sunsolve document after I got the "Console not shared" message from all permutations of "console -a/-f" http://sunsolve.sun.com/search/document.do?assetkey=1-25-78349-1 I ran option 2 which fixed the hung lom prompt.. :D It transpires that if one pugs the serial cable into the "serial port" you get to see the usual system at its booted state where nothing gets displayed on the output on the lom. Thanks all for their assistance and speedy replies.. my original post below. Rgds dan > > ( i have google'd for this) > I've been trying to get a V100 sunfire to start up as far as the OK > prompt, but whilst connected to the LOM (lights out management) port > it's just stuck at the lom. > > So far, i've tried "poweron, poweroff" physically powering off/on the > system and help doesn't give me any more than I tried. Also reset does > nothing. > > Does anyone know how to fix this or is this charactistic of a system > with a broke OBP? > > I will summarise, > > TIA. > > dan. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From gary.paveza at AIG.COM Tue Apr 5 10:43:02 2005 From: gary.paveza at AIG.COM (Paveza, Gary) Date: Tue, 5 Apr 2005 10:43:02 -0400 Subject: [summary] installing a locale Message-ID: <79CF2118BE3ED511AA110003476BCB940655AC0C@xwilmpf03.aig.com> Thanks to: Sergey Tsyganenko Adam Tomkinson Angelos Kanellopoulos They all pointed me to the following: http://developers.sun.com/dev/gadc/techtips/l10npkgadd.csh and http://developers.sun.com/dev/gadc/techtips/sol9_locale_pkgs/adding_locales. html#using-l10npkgadd --------------------------------------------------------- Gary Paveza, Jr. Senior Systems Administrator -CSA (302) 252-4831 - phone -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Paveza, Gary Sent: Tuesday, April 05, 2005 9:42 AM To: 'sunmanagers at sunmanagers.org' Subject: installing a locale I have been requested to install the German locale on one of my servers (SAP). I installed the German packages from the language CD, but it appears that it didn't install the locale. They want de_DE.ISO8859-1 . When I install via jumpstart and tell it locale de in the config file, I get it. But once a server is up and running, how do I install the locale? --------------------------------------------------------- Gary Paveza, Jr. Senior Systems Administrator -CSA (302) 252-4831 - phone _______________________________________________ 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 Rich.Bonfoey at thenewstribune.com Mon Apr 4 13:04:43 2005 From: Rich.Bonfoey at thenewstribune.com (Rich Bonfoey) Date: Mon, 4 Apr 2005 10:04:43 -0700 Subject: SUMMARY:RE: TCP_keepalive Message-ID: Apologies , fat finger does it again. Here are Eric and Sam's replies Eric's: We had to tweak around with the keepalive here. Typically, short keepalive values are good for web servers in example (lots of short and unreliable connections). Obvisouly long keepalive are good for long term connections such as applications having a connection to a database... We had to shorten this value and the "time wait" one because we had lots of open unused client sockets which were left open, consuming resources. But there is something: the keep alive parameter on the server will only be taken into account if "keep alive" is specified in the options when creating a TCP socket. For example, it will have no effect if a client application doesn't explicitely set the keep alive flag when connecting to a server... You might want to have a look at the "tcp_time_wait_interval" parameter as well... Sam's : I'm a sysadmin at an ISP in the UK - as such I deal with many busy machines in a production environment. In my experience you don't want to go lower than 60000ms (even on an extremely busy machine) with tcp_keepalive. I usually dump into /etc/rc2.d in this simple script on my boxes as a baseline to work with: #!/bin/sh # --------------------------------------------------------------- # # # S70stacktune 0.13 # 23/09/2004 Sam Nelson # Script to tune and sanitise the TCP stack on busy Solaris boxes. # UREVISION=`uname -r` fill () { awk '{leninput=length($($NF)); fill=63-leninput ; for (i=1; i< fill; i++) fillchar=fillchar"." ; printf $($NF) fillchar}' } setparams () { VAL=`/usr/sbin/ndd -set $1 $2 $3` printf "Value of $1 $2 is: " | fill printf " ${VAL} ($3)\n" } # Decrease the tcp time wait interval if [ ${UREVISION} = "5.6" ]; then printf "Decrease the tcp time wait interval\n" setparams /dev/tcp tcp_close_wait_interval 60000 else printf "Decrease the tcp time wait interval\n" setparams /dev/tcp tcp_time_wait_interval 60000 fi # Speed up the flushing of half-closed connection in state FIN_WAIT_2 printf "Flushing of half-closed connection in state FIN_WAIT_2\n" setparams /dev/tcp tcp_fin_wait_2_flush_interval 67500ms # Increase the receive and transmit window sizes printf "Increase the receive and transmit window sizes\n" setparams /dev/tcp tcp_xmit_hiwat 65535 setparams /dev/tcp tcp_recv_hiwat 65535 # decrease the retransmit interval printf "Decrease the retransmit interval\n" setparams /dev/tcp tcp_rexmit_interval_max 60000ms # increase number of half-open connections printf "Increase number of half-open connections\n" setparams /dev/tcp tcp_conn_req_max_q0 4096 # increase number of simultaneous connections printf "Increase number of simultaneous connections\n" setparams /dev/tcp tcp_conn_req_max_q 1024 # Decrease TCP connection abort interval printf "Decrease TCP connection abort interval\n" setparams /dev/tcp tcp_ip_abort_interval 60000 # Decrease TCP Keepalive Interval printf "Decrease TCP Keepalive Interval\n" setparams /dev/tcp tcp_keepalive_interval 60000 printf "Ok, tuning complete\n" ------------------------------------------------------- This usually does the trick. Hope this helps, and please summarise! Regards Sam -----Original Message----- From: Rich Bonfoey Sent: Friday, April 01, 2005 11:12 AM To: Rich Bonfoey; 'Sun Managers (sunmanagers at sunmanagers.org)' Subject: SUMMARY:RE: TCP_keepalive Many thanks to Eric Voisard and Sam Nelson for sharing their expertise. Eric provided insight to another variable tcp_time_wait_interval, that it can have an affect on connections. Sam , very graciously provided a script that he uses to tune and sanitize the tcp stack Below is the original message With Eric and Sam's replays Thanks again for the info Rich -----Original Message----- From: Rich Bonfoey Sent: Thursday, March 31, 2005 9:55 AM To: Sun Managers (sunmanagers at sunmanagers.org) Subject: TCP_keepalive Greetings Have any of you, had much dealings with the tcp_keepalive ? On Solaris 8 and 9 the value is 7200000 or 2 hours. What are some of the pit falls if this value was lowered to about 10 minutes. I have read the RFC but was looking for people who have had real experience with tweaking this value. Part of the reason for tweaking, is we have a large number of PC and OS 10 users connecting to several different systems and they have been having performance issues. Our dba decided to bring in a consultant ( that's another story ) and they said this value should be 5 minutes. I am not quite buying it, as I have read that there are multiple opinion on the correct use of tcp_keepalive. Richard Bonfoey The News Tribune Information Systems Successfully Meeting the Business Needs of The News Tribune through Information Technology _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From loverboi at gmail.com Tue Apr 5 20:43:33 2005 From: loverboi at gmail.com (g@ry) Date: Tue, 5 Apr 2005 20:43:33 -0400 Subject: Automount Message-ID: <57f6d16d05040517433307909f@mail.gmail.com> hi friends i have an NFS and NIS server... NIS clients are getting a "permission denied" doing a directory listing (ls) or changing directory (cd) to the shared NFS filesystem. I have checked /etc/auto_master and /etc/auto_home and restarted nfsd and mountd and automountd adn still getting permission denied... Any ideas ???????? Thanks ! G From loverboi at gmail.com Tue Apr 5 20:46:27 2005 From: loverboi at gmail.com (g@ry) Date: Tue, 5 Apr 2005 20:46:27 -0400 Subject: automount nfs nis Message-ID: <57f6d16d050405174617d23c89@mail.gmail.com> hi friends i have an NFS and NIS server... NIS clients are getting a "permission denied" doing a directory listing (ls) or changing directory (cd) to the shared NFS filesystem. I have checked /etc/auto_master and /etc/auto_home and restarted nfsd and mountd and automountd adn still getting permission denied... Any ideas ???????? Thanks ! G From ssefoen at yahoo.com Tue Apr 5 11:46:50 2005 From: ssefoen at yahoo.com (Z H) Date: Tue, 5 Apr 2005 08:46:50 -0700 (PDT) Subject: summary:problem with npasswd program Message-ID: <20050405154650.88911.qmail@web31601.mail.mud.yahoo.com> I only received one response from Christopher McNabb Chris suggested to use lsof to see whether there are checkpassword session already run in the system. I did not find any strayed checkpassword process in the system though. I have also run truss and truss does not seem to show any error message either. If any one has any additional suggestion, I will be happy to try. Thanks. ssefoen. Original question: o All Helpers: I have solaris 2.8 running npasswd version 2.05. Everything looks great except the checkpassword error message each time user runs checkpassword program. Here is the error message in messages log: Apr 1 09:16:28 alpha checkpassword[25938]: [ID 817389 user.error] Dictionary "/usr/lib/passwd/dictionaries/CRL-Words" already in use Apr 1 09:16:28 alpha checkpassword[25938]: [ID 817389 user.error] Dictionary "/usr/lib/passwd/dictionaries/CRL-Words" already in use Apr 1 09:16:28 alpha checkpassword[25938]: [ID 645434 user.error] Dictionary "/usr/lib/passwd/dictionaries/Congress" already in use Apr 1 09:16:28 alpha checkpassword[25938]: [ID 645434 user.error] Dictionary "/usr/lib/passwd/dictionaries/Congress" already in use Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ssefoen at yahoo.com Tue Apr 5 12:14:05 2005 From: ssefoen at yahoo.com (Z H) Date: Tue, 5 Apr 2005 09:14:05 -0700 (PDT) Subject: Summary: problem with npasswd program Message-ID: <20050405161405.3508.qmail@web31609.mail.mud.yahoo.com> After sending out the summary. I continue research further and I have found my own problem. The problem was from the configuration file. After commenting out the path to the dictoinaries, the checkpassword program will load the dictionaries only once instead of twice. I was using checkpassword -XD6 to see the detail process and found out the dictionaries were loaded twice. After commenting out the dictionaries path from configuration file, checkpassword program will only load the dictionaries once. ssefoen. __________________________________ Do you Yahoo!? Yahoo! Personals - Better first dates. More second dates. http://personals.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From larry at opsource.net Tue Apr 5 12:19:00 2005 From: larry at opsource.net (Larry Guest) Date: Tue, 5 Apr 2005 09:19:00 -0700 Subject: SUMMARY: Bad CPU's in a 220R Message-ID: <001601c539fb$2d191440$500118ac@LarrysIBM> Thanks to all who replied. I think everyone pretty much had the same answer which helped me solve the problem. I just needed to set the jumper on the mother board for the 450MHz CPU's. http://sunsolve.sun.com/data/806/806-1055/pdf/806-1055-12.pdf I now have a 220R running 2x450. Thanks !!!! Larry _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From johnjens at gmail.com Tue Apr 5 16:12:47 2005 From: johnjens at gmail.com (john jens) Date: Tue, 5 Apr 2005 16:12:47 -0400 Subject: SUMMARY: solaris 8 BIND stops resolving reverse DNS In-Reply-To: <2767c3fc050405121374d70010@mail.gmail.com> References: <2767c3fc050405121374d70010@mail.gmail.com> Message-ID: <2767c3fc05040513124cfa3f2f@mail.gmail.com> Many thanks to Richard Sullivan for pointing me in the right direction. I manually incremented the serial number to a much higher value on db.10.162.131 and while editing the file found a typo in one of the entries. I double-checked the BIND logging and indeed, the db file was not being loading. Once the hosts file was corrected and the db regenerated, named is able to resolve reverse lookups for all subnets. cheers, john jens > My internal DNS server has mysteriously stopped resolving reverse DNS > requests while I was on vacation (that'll teach me). I've rebuilt the > db files, via a customized hosts_to_network script, several times with > no effective change. Here's the kicker: the DNS server is multi-homed > and reverse lookups for the second subnet work on the secondary > interface. (clear as mud. see examples.) > > cheers, > john jens > > # nslookup foo > *** Can't find server name for address 10.162.131.21: Non-existent host/domain > *** Default servers are not available > # > # nslookup foo dns3 > Server: dns3.lab.priceinteractive.com > Address: 10.162.130.21 > > Name: foo.lab.priceinteractive.com > Address: 10.162.131.44 > # > # nslookup 10.162.131.44 > *** Can't find server name for address 10.162.131.21: Non-existent host/domain > *** Default servers are not available > # > # nslookup 10.162.131.44 dns3 > Server: dns3.lab.priceinteractive.com > Address: 10.162.130.21 > > *** dns3.lab.priceinteractive.com can't find 10.162.131.44: > Non-existent host/domain > # > # nslookup bar dns3 > Server: dns3.lab.priceinteractive.com > Address: 10.162.130.21 > > Name: bar.lab.priceinteractive.com > Address: 10.162.130.55 > # > # nslookup 10.162.130.55 > *** Can't find server name for address 10.162.131.21: Non-existent host/domain > *** Default servers are not available > # > # nslookup 10.162.130.55 dns3 > Server: dns3.lab.priceinteractive.com > Address: 10.162.130.21 > > Name: bar.lab.priceinteractive.com > Address: 10.162.130.10 > > === uname === > SunOS dns1 5.8 Generic_117350-11 sun4u sparc SUNW,Ultra-60 > > === /etc/hosts === > 127.0.0.1 localhost > 10.162.131.21 dns1 loghost > 10.162.130.21 dns3 > 10.162.131.20 vdns > > === ifconfig === > lo0: flags=1000849 mtu 8232 index 1 > inet 127.0.0.1 netmask ff000000 > hme0: flags=1000843 mtu 1500 index 2 > inet 10.162.131.21 netmask ffffff00 broadcast 10.162.131.255 > ether 8:0:20:c4:e8:84 > hme0:1: flags=1000843 mtu 1500 index 2 > inet 10.162.131.20 netmask ffffff00 broadcast 10.162.131.255 > qfe0: flags=1000843 mtu 1500 index 3 > inet 10.162.130.21 netmask ffffff00 broadcast 10.162.130.255 > ether 8:0:20:c4:e8:84 > > === netstat === > 10.162.131.21.53 *.* 0 0 24576 0 LISTEN > 10.162.130.21.53 *.* 0 0 24576 0 LISTEN > > === named.conf === > options { > directory "/usr/local/domain"; > }; > > logging { > channel my_file { > file "named_msgs.log" versions 3 size 20m; > severity dynamic; > }; > category default { my_file; }; > }; > > // Root Name Servers > zone "." { type hint; file "rootservers"; }; > > // Loopback > zone "0.0.127.IN-ADDR.ARPA" { type master; file "db.127.0.0"; }; > > // Lab Priceinteractive > zone "lab.priceinteractive.com" { type master; file "db.lab"; }; > zone "ohe.lab.priceinteractive.com" { type master; file "db.ohe"; }; > zone "dev.ohe.lab.priceinteractive.com" { type master; file "db.ohe.dev"; }; > zone "stg.ohe.lab.priceinteractive.com" { type master; file "db.ohe.stg"; }; > zone "dev.sp.lab.priceinteractive.com" { type master; file "db.sp.dev"; }; > zone "test.sp.lab.priceinteractive.com" { type master; file "db.sp.test"; }; > > // Reverse lookups for internal (10.x.x.x) networks > zone "128.162.10.IN-ADDR.ARPA" { type master; file "db.10.162.128"; }; > zone "130.162.10.IN-ADDR.ARPA" { type master; file "db.10.162.130"; }; > zone "131.162.10.IN-ADDR.ARPA" { type master; file "db.10.162.131"; }; > > === snoop === > ... > DNS: ----- DNS Header ----- > DNS: > DNS: Response ID = 63063 > DNS: RA (Recursion Available) > DNS: Response Code: 3 (Name Error) > DNS: Reply to 1 question(s) > DNS: Domain Name: 20.131.162.10.in-addr.arpa. > DNS: Class: 1 (Internet) > DNS: Type: 12 (Domain Name Pointer) > DNS: > DNS: 0 answer(s) > DNS: 1 name server resource(s) > DNS: Domain Name: 10.in-addr.arpa. > DNS: Class: 1 (Internet) > DNS: Type: 6 (Start Of a zone Authority) > DNS: TTL (Time To Live): 9375 > DNS: Start Of a zone Authority: > DNS: MNAME (Server name): prisoner.iana.org. > DNS: RNAME (Resposible mailbox): hostmaster.root-servers.org. > DNS: SeriDNS: Refresh: 1800 Retry: 900 Expire: 604800 > Minimum: 604800 > DNS: > DNS: 0 additional record(s) _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From john.christian at TheCReGroup.com Wed Apr 6 10:11:44 2005 From: john.christian at TheCReGroup.com (John Christian) Date: Wed, 6 Apr 2005 10:11:44 -0400 Subject: SUMMARY: verify physical host from terminal session Message-ID: <2AD31F5C087C144398DF520F2DB1B8BD0BF61D@magnum.TheCReGroup.com> THANKS TO: Michael Imrick, Rob McCauley, Pascal Grostabussiat, Graeme Burke, Iain Miller, the hatter, Bernd Schemmer, Anthony D'Atri, Michael Horton, Shane Gainer, Thomas Payerle, Lee Wood, Alan Pae, JV, Adam Tomkinson, Eugene Schmidt SUMMARY: The most elegant solution came from Michael Imrick who suggested /usr/sbin/locator -n which lights a very bright blue LED on the front of the host. Perfect! Shane Gainer suggested that a "bright ass blue light" can also be enabled from the ALOM (details below). Please note: The very bright blue LED and the bright ass blue LED are, in fact, the same LED. As it turns out, firing flares within the data center is frowned upon. Here are the alternatives: Visual Clues: enable bright blue locator light within Solaris (if h/w supports) /usr/sbin/locator -n enable bright blue locator light from ALOM Sc> setlocator on Sc> showlocator Locator LED is on very busy hard drive lights dd if=/dev/somedisk of=/dev/null format > analyze > [read | test] find / Audio Clues: Cat something to /dev/audio or run snoop -a (-a = listen to packets on /dev/audio) Physical Clues: If the network cables are labeled with switch/port information, login to the switch, check MAC and port tables, trace back to cable drop. If dual power supplies, pull one power cord out and check /var/adm/messages or local terminal console. -John Christian KEYWORDS: find physical hardware locator location flash flasher alert discover echo sound audio identify ________________________________ From: sunmanagers-bounces at sunmanagers.org on behalf of John Christian Sent: Tue 4/5/2005 5:07 PM To: sunmanagers at sunmanagers.org Subject: verify physical host from terminal session My colleague just installed Solaris 9 on ~20 hosts (440's and 240's) and forgot to label them before taking holiday.What's a quick way to verify that the terminal session I'm logged into corresponds to a specific machine in a rack? I'd like to invoke something that flashes the lights, sounds a beep, or shoots a flare. Yes, I can drag in my laptop and serial cables, but that's not very elegant. Snags: Many of the hosts are already in use, so destructive tests like pulling network cables or rebooting is out. Eject won't open the CD tray unless a disk is mounted. Since automountd is disabled, if I drop in a CD I'll have to try to mount on each server which is time consuming. echo [Ctrl+v][Ctrl+g] doesn't seem to ring any bells. At least, not that I can discern amongst three racks of running hosts. Hopes: I have all the IPs and passwords. Most hosts are dual power supplied. Is there a mechanism at the OS level that says "Warning: power supply A is now a doorstop" ? TIA, will summarize, -John _______________________________________________ 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 James.Swanson_CONTRACTOR at dtra.mil Wed Apr 6 11:31:59 2005 From: James.Swanson_CONTRACTOR at dtra.mil (Swanson, James CONTRACTOR) Date: Wed, 6 Apr 2005 11:31:59 -0400 Subject: Summary: Problems creating users in LDAP Message-ID: <300BD4FCBD290841BD61184DC67B03E103BB75B1@MAIL904.unet1.ad.dtra.mil> After much futzing, I figured something out. Creating the user using the Sun One console works. However, to have a home directory on a specific server, that home directory has to be created and the user has to be given rights to the folder based on their UID number. After creation of the user and directory, create the automount home directory in the Sun One console. This will fill the directory with the appropriate files, and the new user can fully use the home directory on the server. Writing a script or program to do this would probably be helpful. So would having Sun make this work through the console. As far as I can see, the whole point of the console is to alleviate the need for manually creating users and directories. My $0.02. Many thanks to those who helped. James Swanson Unix Sysadmin DTRA Phone: 703-767-4541 -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Swanson, James CONTRACTOR Sent: Monday, March 21, 2005 12:20 PM To: sunmanagers at sunmanagers.org Subject: Summary but not quite fixed: RE: Problems creating users in LDAP Thanks to Will Dowling, I'm not getting the home directory errors any more, and I can log on locally to our Sun boxes. Basically, I have to manually create the directories on each of the clients. But this brings up another issue. The users created by my predecessor can log in, and their /export/home/$username on the LDAP server is their home directory. I can't do that, I had to create home directories manually on each client box. Did I miss something? James Swanson Unix Sysadmin DTRA Phone: 703-767-4541 > > I've been trying to add myself to our Sun One 5.2 LDAP server. I'm > logged in as Admin, I added myself through the Sun One Server Console, > put in appropriate groups, added an automount directory, etc. Did all > that, hit okay , and tried to log in. Doesn't work. I looked up the > documentation from my predecessor, followed his instructions, and when > I get to the bottom of the document , he wrote, "And now what, the > user still can't logon and this procedure doesn't create a home > directory on the server." I'm having the same problem. What am I > missing? > Is there a secret command that I have to run to get the user to work? > Any help would be appreciated, as I'm still a beginner with this type > of system. Thanks. > > James Swanson > _______________________________________________ > 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 grgoffe at yahoo.com Wed Apr 6 12:55:38 2005 From: grgoffe at yahoo.com (George R Goffe) Date: Wed, 6 Apr 2005 09:55:38 -0700 (PDT) Subject: SUMMARY: UW Diff scsi vs UW scsi cables. Message-ID: <20050406165538.8791.qmail@web53103.mail.yahoo.com> Howdy, I got a LOT of responses to this one, my sincerest thanks to all who responded and to those who considered the problem as well. Basically, I was given what I was told was a DIFF SCSI card which turns out to be a "Dual Single-Ended Ultra/Wide SCSI (PCI)" card. What I needed a "Dual Differential Ultra/Wide SCSI (UDWIS/P)(HVD)" card which I just happen to have. This info is from a web site that Ken Rossman rossman at columbia.edu sent me. A GREAT SITE by the way. "http://sunsolve.sun.com/handbook_pub/" I hooked the card up to a multipack (12 drive version) but got some UGLY messages during boot which I had thought were due to trying to talk to a non-DIFF SCSI enclosure via a DIFF SCSI controller card. This origional card just might be bad. Thanks to all who responded: Ken Rossman Anthony D'Atri Darren Dunham Eric Lewandowski Jason Santos Thomas M. Payerle George... Origional post: Sorry to bother you with this baby question but I have an a1000 which has two diff scsi connections on the back and I have a two channel (low profile connectors) diff scsi (single ended) card in the system. I was told that just any uw scsi cable would work but I'm not seeing the drives with a uw scsi cable. Can you tell me if this should work or am I being fed a crock? ===== _/_/_/_/ _/_/_/_/ _/_/_/_/ _/_/_/_/ _/_/_/_/ _/_/_/_/ ----- _/ _/ _/ _/ _/ _/ _/ _/ _/ _/_/ _/_/_/_/ _/ _/ _/_/_/_/ _/ _/_/ _/_/_/_/ ----- _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/_/_/_/ _/_/_/_/ _/_/_/_/ _/ _/ _/_/_/_/ _/_/_/_/ ----- "It's not what you know that hurts you, It's what you know that ain't so." Will Rogers Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Chris.Bartram at usmint.treas.gov Wed Apr 6 14:58:46 2005 From: Chris.Bartram at usmint.treas.gov (Bartram, Chris (Contractor)) Date: Wed, 6 Apr 2005 14:58:46 -0400 Subject: SUMMARY: SNMP community string starting w/special a prob on Sol 8 ? Message-ID: <69181A9CDB58D411B9C3009027E867F00620FDAC@wdc2200.usmint.treas.gov> I only got one response, from Eric Lewandowski (thanks!), who confirms that he tried a similar setup at a previous site and found the sun-shipped SNMP software unable to cope with any special characters at the start of the community strings. He suggested switching to net-snmp, which would be nice if mgmt wasn't afraid of public domain software on production systems... -Chris Bartram Bartram, Chris (Contractor) <> wrote: > Our standard community strings start with a special character. It's > been ok so far on every other platform we've used... But on Sol 8 (on > an e15k domain fwiw) trying to enter the read/write-trap-community > string in snmpdx.acl generates > > Mar 29 13:31:56 solaris /usr/lib/snmp/snmpdx: [ID 702911 daemon.error] > syntax error in at line 20: the token < is not valid > Mar 29 15:15:10 solaris /usr/lib/snmp/snmpdx: [ID 702911 daemon.error] > syntax error in at line 20: the token $ is not valid > > (for variations of the string starting with a "<" and a "$" > respectively.) > > I determined the "in at line 20" is referring to the snmpdx.acl > file's line > 20. Trying to 'quote' or "quote" the strings yielded: > > Mar 29 15:16:33 solaris /usr/lib/snmp/snmpdx: [ID 702911 > daemon.error] syntax error at line 20: " > Mar 29 14:30:56 solaris /usr/lib/snmp/snmpdx: [ID 702911 > daemon.error] syntax error in at line 20: the token ' is not valid > > Don't seem to get complaints about the snmpd.conf file, though maybe > it's not getting that far... > > relevant portions of the snmpdx.acl are: > > acl = { > { > communities = %string1, %string2 <--- this is > "line 20" in the snmpdx.acl file > access = read-write > managers = * > } > } > > communities = { > %string1 read-only > %string2 read-write > } > > trap-community = %string1 > > trap-recipients = > { > host1,host2 > } > > Is is a bug? or is there a secret-handshake escape character I can > use to get it to accept a string starting with a special character? > (I tried "\" as a prefix as well.. no joy) > > And don't get me started on the idjit that picked those strings in > the first place! ;-) > > TIA - will summarize, > -Chris Bartram > _______________________________________________ > 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 craig.burtenshaw at amsa.gov.au Thu Apr 7 01:35:00 2005 From: craig.burtenshaw at amsa.gov.au (Burtenshaw, Craig) Date: Thu, 7 Apr 2005 15:35:00 +1000 Subject: SUMMARY: Samba Errors Message-ID: Hi, Original message, under answer. Thank you to Matthew Gillard, Kelly Ormsby, Alex Stade, Pablo Jejcic and Doug Hubbard for responding. I commented out the following entries from /etc/inetd.conf netbios-ssn stream tcp nowait /usr/local/samba/bin/smbd smbd netbios-ns dgram udp wait /usr/local/samba/bin/nmbs nmbd Then I kill -HUP The errors went away. Oh, and I'm running Samba as a Daemon (smbd -D) from now on too. Cheers Craig. Original message > I'm receiving the following errors in /var/adm/messages > > inetd[16689]: [ID 455825 daemon.error] getpwnam: > /usr/local/samba/bin/nmbd: No > such user > inetd[272]: [ID 858011 daemon.warning] nmbd: Hangup > > I have google'd, but the only responses were in Japanese. that any use or dissemination of this communication is prohibited. If you receive this transmission in error, please notify us immediately by telephone on +61 2 62795000 and delete all copies of this transmission together with any attachments. ********************************************************************** _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From yash at linux-delhi.org Thu Apr 7 06:48:40 2005 From: yash at linux-delhi.org (Yashpal Nagar) Date: Thu, 7 Apr 2005 16:18:40 +0530 Subject: SUMMARY: Sun pkg management In-Reply-To: <200504031412.47722.yash@linux-delhi.org> References: <200504031412.47722.yash@linux-delhi.org> Message-ID: <200504071618.41312.yash@linux-delhi.org> Thanks to Adam, Jacint & everyone those who replied. Lot of people replied that there are some areas where sun sucks over Linux. Last reply i got from Jacint, was interesting and found working. One may use pkgmap file in package to see what files it will install. for 2nd its not possible though there are some tools like http://www.blastwave.org which might work. Thanks Yash On Sunday 03 Apr 2005 2:12 pm, Yashpal Nagar wrote: > Hi Sun manager's! > > While learning Sun Solaris, i struck up at the following. > > 1. How do you check all files in a pkg with out installing a pkg. e.g i > have spooled a package SUNWcsr to /abc then how to check what are the files > in SUNWcsr ? aka in linux you can do "rpm -qpl pkg-name-i386.rpm". > > 2. Suppose if i want to install one package from CD 1/2 then if that > package requires some other dependencies in CD2, then how does it > curcumvent. Any tools liks apt-get, yum, etc :) I mean in general how to > meet dependencies problem? > > Regards, > Yash _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ssloh at singnet.com.sg Thu Apr 7 09:48:47 2005 From: ssloh at singnet.com.sg (ssloh) Date: Thu, 7 Apr 2005 21:48:47 +0800 Subject: SUMMARY: Resource Group offline in Sun Cluster References: <3C7D55BA53F0E54D9A0C5C3A738B490702253D5F@BGEEXC02.asiapacific.cpqcorp.net> <000c01c53aba$6b8abe30$020a0a0a@TEST> Message-ID: <007701c53b78$86976430$020a0a0a@TEST> Only two responses but GOOD enough, thanks Mazhar Khan who pointed me to use "RGOffload" resource type to accomplish my mission described Thanks Ramesh Edupalli who refer me to Sun Document ID : 78315 for resource group dependencies workaround for SunCluster 3.0 Yes, I am using SunCluster3.0 here, to be more specific 1. RG1 running on node1 2. RG2 running on node2 3. If RG1 failed on node1, it will failover to node2 4. Offline RG2 at node2 first prior to Online RG1 on node2 Cheers... ----- Original Message ----- From: "ssloh" To: Sent: Wednesday, April 06, 2005 11:03 PM Subject: Resource Group offline in Sun Cluster Hi folks, Would like to check how to accomplish below scenario RG1 running on node1, RG2 running on node2. Before failover RG1 to node2, RG2 need to be OFFLINE in node2 first. I could not find any pre-offline script attribute in RG. 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 alex at trdlnk.com Thu Apr 7 14:32:55 2005 From: alex at trdlnk.com (Alex Stade) Date: Thu, 07 Apr 2005 13:32:55 -0500 Subject: SUMMARY: controller enumeration Message-ID: <42557CD7.4030606@trdlnk.com> Thanks to everybody who replied so promptly, especially Darren Dunham who nailed it. Solaris builds and maintains a directory called '/dev/cfg' which contains the controller assignments. This happens whenever a new controller is seen. Device numbers are assigned on an incremental basis. It should be noted that messing with this stuff on a system which has volume managers or on controllers which hold the root disks, that re-assigning these controller names should be done with utmost caution. -Alex Original question was; I am trying to figure out how Solaris determines the controller ID for a particular SCSI controller. I am not having much luck. I have two 220R lab machines which I jumpstarted with Solaris 9 U7, both hosts are identical with respect to hardware as well as OS install. However, on one of the hosts the add-in SCSI controllers are called 'c1' and 'c2', whereas the other has 'c2' and 'c3'. I re-jumpstarted them and this time Solaris called them 'c1' and 'c2' on both hosts. I have seen this on both sparc and i386 hardware. Two identical PCs with add-in PCI SCSI controllers - one of the PCs end up having the internal IDE drive called 'c0d0' and the other PC ends up with 'c2d0'. So my question is two-fold. a) How does Solaris figure out what controller should recive which controller ID, e.g. 'c0', 'c1' etc.; and b) Is this something I can adjust (obviously with proper care and caution taken with respect to volume managers and boot devices) without having to jumpstart and keep my fingers crossed? -Alex _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bshah at citadon.com Thu Apr 7 14:35:28 2005 From: bshah at citadon.com (Bhavesh Shah) Date: Thu, 7 Apr 2005 11:35:28 -0700 Subject: SUMMARY: Can't delete folders..... Message-ID: Thanks a lot for your responses.... Despite the valuable suggestions from Gurus this mystery was not resolved At last I took the backup of /home and formatted it. Now things are OK. I assume it was a corrupted file system that might have caused this mystery. Thanks B -----Original Message----- From: Bhavesh Shah Sent: Tuesday, April 05, 2005 6:32 PM To: sunmanagers at sunmanagers.org Subject: Can't delete folders..... Hi Gurus, I have a strange issue on Solaris 8. I am trying to delete certain folders inside home directory it doesn't allow me to delete the same. I am able to some folders in the same directory but not all. Even as a root I can't delete those certain folders. I don't have /home automounted. I have given full permission to the folder still can't delete the same. Any Help/Pointer will be appreciated. Thanks B drwxr-xr-x 5 pnetops pnetops 512 Mar 4 14:33 development_7.0_200503040800 rm -rf development_7.0_200503040800 rm: Unable to remove directory development_7.0_200503040800/shared/tools/weblogic/8.1/domains/mydomain/ myserver/ldap/conf: File exists rm: Unable to remove directory development_7.0_200503040800/shared/tools/weblogic/8.1/domains/mydomain/ myserver/ldap: File exists rm: Unable to remove directory development_7.0_200503040800/shared/tools/weblogic/8.1/domains/mydomain/ myserver/projectnet/applications/CitadonCWWebApp/WEB-INF: File exists drwxrwxrwx 5 pnetops pnetops 512 Mar 28 15:53 development_6.7.0_200502010800 # rm -rf development_6.7.0_200502010800 rm: Unable to remove directory development_6.7.0_200502010800/scripts/reports: File exists rm: Unable to remove directory development_6.7.0_200502010800/scripts: File exists rm: Unable to remove directory development_6.7.0_200502010800/shared/tools/weblogic/8.1/domains/mydomai n/myserver/projectnet/serverclasses/citadon/bizcomponent/forms/entity: File exists # automount -v automount: /net mounted automount: /xfn mounted automount: no unmounts _______________________________________________ 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 gary.paveza at AIG.COM Thu Apr 7 14:55:20 2005 From: gary.paveza at AIG.COM (Paveza, Gary) Date: Thu, 7 Apr 2005 14:55:20 -0400 Subject: [summary]: jumpstart finish script Message-ID: <79CF2118BE3ED511AA110003476BCB940655B034@xwilmpf03.aig.com> I made three changes to the pkgadd. I added the -n for non-interactive, changed -d ${MNT} package to -d ${MNT}/package, and added all to the end of the line. I had done the -d change originally, so I think the problem was either the -n or the all. Either way, it now works. I'm getting an "exec failed. errno=2, but it appears to have installed correctly. I haven't been able to determine what this error is. I'm including my finish script below since a few people asked me to let them know how I ended up. It's not complete yet as I am still working out the patching and the JASS (just hadn't gotten to those items yet - doing it now). In the interest of being honest and open, one lesson I learned was that when testing this installation, it really wasn't necessary to use the SUNWXCall package :) While this is our standard for final installation, during testing, the SUNWXreq was all that was needed - and it sure sped up testing :) Thanks to everyone who had suggestions and helped me through this. #!/bin/sh # # @(#) Finish script for Jumpstarting Solaris 9 # @(#) $Revision 1.0 # @(#) Author: Gary Paveza # @(#) Created 2005/04/05 # @(#) Last Modified: 2005/04/05 BASE=/a MNT=${BASE}/mnt ADMIN_FILE=${BASE}/tmp/admin # Mount remote filesystem containing packages and patches mkdir ${MNT} >/dev/null 2>&1 mount -f nfs server:/jumpstart/Solaris9_config ${MNT} >/dev/null 2>&1 if [ $? != 0 ] then echo "Filesystem not mounted" exit 1 fi # Create admin script for package installation cat >${ADMIN_FILE} < ${BASE}/opt/jass-0.3.1/jass-execute.log 2>&1 cd # Copy patches echo "Copying 9_Recommended.zip file" cp ${MNT}/9_Recommended.zip ${BASE}/patches chown root:other ${BASE}/patches/9_Recommended.zip chmod 700 ${BASE}/patches/9_Recommended.zip unzip ${BASE}/patches/9_Recommended.zip > ${BASE}/patches/9_Recommended.log 2>&1 echo "Installing Recommended patch bundle" for patch in `cat ${BASE}/9_Recommended/patch_order` do cd ${BASE}/9_Recommended/$patch # echo yes | patchadd -u -d -R ${BASE} . echo "Will install patch $patch to ${BASE}" done # Setup root's environment echo "Setting up root's HOME (/root)" mkdir ${BASE}/root chmod 700 ${BASE}/root chown root:root ${BASE}/root cp ${MNT}/root.profile ${BASE}/root/.profile chown root:root ${BASE}/root/.profile chmod 400 ${BASE}/root/.profile cp ${MNT}/root.profile-EIS ${BASE}/root/.profile-EIS chown root:root ${BASE}/root/.profile-EIS chmod 400 ${BASE}/root/.profile-EIS cp ${BASE}/etc/passwd ${BASE}/etc/passwd.old sed s!Super-User:/:!Super-User:/root:! ${BASE}/etc/passwd > ${BASE}/etc/passwd.sed cp ${BASE}/etc/passwd.sed ${BASE}/etc/passwd # Set root's password ${BASE}/scripts/set_root_pw --------------------------------------------------------- Gary Paveza, Jr. Senior Systems Administrator -CSA (302) 252-4831 - phone -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Paveza, Gary Sent: Wednesday, April 06, 2005 2:24 PM To: 'sunmanagers at sunmanagers.org' Subject: jumpstart finish script I'm trying to use a finish script to install some packages via jumpstart (Solaris 9). I have the following in my finish script: #!/bin/sh BASE=/a MNT=${BASE}/mnt ADMIN_FILE=${BASE}/tmp/admin # Mount remote filesystem containing packages and patches echo "Mounting /jumpstart/Solaris9_config to ${MNT}" mkdir ${MNT} >/dev/null 2>&1 mount -f nfs :/jumpstart/Solaris9_config ${MNT} >/dev/null 2>&1 if [ $? != 0 ] then echo "Filesystem not mounted" exit 1 fi echo "Starting creation of admin script" # Create admin script for package installation cat >${ADMIN_FILE} < Well thanks to all that replied there were many!! A lot of replies going down the line of booting into single user mode and editing /etc/system & /etc/vfstab to remove the "md" parts and get it to boot from the underlying physical disks. This also failed with a different error: Copyright 1983-2003 Sun Microsystems, Inc. All rights reserved. Cannot assemble drivers for root /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/disk at w21000004cf8fc8ac,0:a Cannot mount root on /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/disk at w21000004cf8fc8ac,0:a fstype ufs As apposed to the original error of: Cannot mount root on /pseudo/md at 0:0,0,blk fstype ufs After much head scratching and time spent it was decided to re-install the O/S, this being on a single disk and still leaving us with the other half of the original root disk mirror with all data. This has been done & the server is now being patched. As to what the error was, what caused it and how it could have been fixed I'm still not sure. But thanks for your help - Julian. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From chris.wrigglesworth at atosorigin.com Fri Apr 8 06:47:41 2005 From: chris.wrigglesworth at atosorigin.com (WRIGGLESWORTH, Christopher) Date: Fri, 8 Apr 2005 11:47:41 +0100 Subject: SUMMARY: RBAC Role allows direct login. Will Summarize Message-ID: <2407A2B36F84D54683FE604C752B433604901195@notes4.sema.co.uk> Hi Managers Thanks to everyone for the response, especially Casper and Adam. Casper wrote: "You need to check /etc/pam.conf. For login, e.g., you must have: login account requisite /usr/lib/security/$ISA/pam_roles.so.1 as one of the lines. This is true for all account management definitions. And if you're using "SSH" you must make sure you have one which uses PAM correctly. Casper" You were right on the money Casper, the problem was OpenSSh and PAM. I installed the latest version of OpenSSH (openssh-4.0p1) and added: login account requisite /usr/lib/security/$ISA/pam_roles.so.1 to my /etc/pam.conf file and bingo jingo it worked. I then set UsePAM to yes in sshd_config. And it just worked first time. Once again thanks to you all for your fast replies. Regards Chris Wrigglesworth Unix Technical Specialist, Unix Technical Support (UK), Network Infrastructure Solutions, Atos Origin Chris.Wrigglesworth at atosorigin.com -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of WRIGGLESWORTH, Christopher Sent: Friday 08 April 2005 10:05 To: 'sunmanagers at sunmanagers.org' Subject: RBAC Role allows direct login. Will Summarize Hi Managers I have a Solaris 8 server that is behaving quite odd. I have set up an RBAC role to allow an operator to perform some admin tasks. As I understand it, an RBAC role should only allow a user to su to it, it should not allow anyone to login to the server directly as the RBAC role. However in testing I can always login to the server directly as the role. I have tried stopping and starting nscd but this has no effect. I have also searched sunsolve for any patches but I've found nothing (that doesn't mean there are no patches, just that I haven't found them :). Does anyone have any suggestions? Below are a few detail with the usual security adjustments, if any one would like more info let me know System details: SunOS HOSTNAME 5.8 Generic_117350-02 sun4u sparc SUNW,Sun-Fire-280R /etc/user_attr username::::type=normal;roles=rolename rolename::::type=role;profiles=profile name Thanks for your help. Chris Wrigglesworth Unix Technical Specialist, Unix Technical Support (UK), Network Infrastructure Solutions, Atos Origin * Chris.Wrigglesworth at atosorigin.com _______________________________________________________ This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Atos Origin group liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted. _______________________________________________________ _______________________________________________ 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 alex at trdlnk.com Fri Apr 8 08:14:54 2005 From: alex at trdlnk.com (Alex Stade) Date: Fri, 08 Apr 2005 07:14:54 -0500 Subject: SUMMARY2: controller enumeration Message-ID: <425675BE.8060107@trdlnk.com> Omitted by me, but pointed out by Vincent Cojot, it is also necessary to re-create /dev/dsk/cN* and /dev/rdsk/cN* at the same time as /dev/cfg/cN are re-created, lest Solaris will happily consult /dev/dsk/cN* and /dev/rdsk/cN* for hints on how to re-create /dev/cfg/cN. -Alex From Vincent; Alex, /dev/cfg/* is not enough. If you are renumbering your controller, you must -also- rename /dev/dsk/cX* and /dev/rdsk/cX* at the -same- time, otherwise, the /dev/cfg entries will be re-created in the prevous order. Aside from that, I've done controller re-numbering (to fix incorrectly ordered stuff due to human mistakes) on SF15k domains without problem. Cheers, Vincent _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From julian.grunnell at pipex.net Fri Apr 8 09:03:23 2005 From: julian.grunnell at pipex.net (Julian Grunnell) Date: Fri, 8 Apr 2005 14:03:23 +0100 Subject: V880 won't boot - Pt II - SUMMARY Message-ID: <002d01c53c3b$58ce2890$340211ac@internal.gxn.net> Easy really ... not enough sleep or something? Disksuite details backed up in /etc/lvm/md.tab. So just use metainit to create 1st half of the mirrors & test. When happy create the 2nd half of the mirror and attach to 1st half. Some metadevices were greater than 128 which threw me for a while. So need to edit /kernel/drv/md.conf and do a "boot -r". Other than that, its looking good. No replies ... but not to worry ;-) Thanks - J. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From william.m.dowling at nuim.ie Fri Apr 8 10:55:14 2005 From: william.m.dowling at nuim.ie (Will Dowling) Date: Fri, 08 Apr 2005 15:55:14 +0100 Subject: SUMMARY: excessive LDAP connections Message-ID: <42569b52.3c8.320e.678150048@may.ie> Hey all, Thanks to those who replied. I got a few conflicting reports on this. "The best solution, unless you are using xfn for something, is to remove the xfn line from /etc/auto_master, and stop/start autofs." - Rachel Polanskis "This is normal. It occurs on Linux also. Anytime you run a process that needs to resolve uids/guids to names will need to open a connection to LDAP. So running "ls -l" needs to contact LDAP." - Brian Miller "Make sure you have NSCD running on each client. Otherwise, every ls, id, etc command will do separate quieries for each and every file it looks at." - Chris_Sellers I tried all of these but none worked so I decided that Brian was right and it was normal. But today(after a busy week) it was solved, thanks to Robert Petkus : " Make sure that in /etc/nsswitch.conf you don't use ldap at all for services (services: files) then restart nscd." Works perfect. Cheers. Will > > Hey folks, > I'm running a Sun Fire v880 with Solaris 9 and Sun One > Directory Server5.2. LDAP is used to manage user accounts. > I just added four blade 2500's to > my network and configured LDAP correctly. > Now LDAP seems to be creating alot of connections. I > cannot remember if this is normal or not. > There always seems to be a fixed number of established > connections but then a ridiculous amount of close_wait > sessions > when I try and do some work on the workstations. > To give you an example this is when the workstation is > idle > > bash-2.05$ netstat | more > > TCP: IPv4 > Local Address Remote Address Swind Send-Q > Rwind Recv-Q State > -------------------- -------------------- ----- ------ > ----- ------ ------- > localhost.1022 server.nfsd 49640 0 49640 0 > ESTABLISHED > localhost.35077 localhost.35073 49152 0 49152 0 > ESTABLISHED > locahost.35073 localhost.35077 49152 0 49152 0 > ESTABLISHED > localhost.35078 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35078 49152 0 49152 0 > ESTABLISHED > localhost.35079 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35079 49152 0 49152 0 > ESTABLISHED > localhost.35080 localhost.35073 49152 0 49152 0 > ESTABLISHED > lcoalhost.35073 localhost.35080 49152 0 49152 0 > ESTABLISHED > localhost.35081 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35081 49152 0 49152 0 > ESTABLISHED > localhost.35084 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35084 49152 0 49152 0 > ESTABLISHED > > while this is when I do some work on it : > > bash-2.05$ netstat | more > > TCP: IPv4 > Local Address Remote Address Swind Send-Q > Rwind Recv-Q State > -------------------- -------------------- ----- ------ > ----- ------ ------- > localhost.1022 server.nfsd 49640 0 49640 0 > ESTABLISHED > localhost.35077 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073localhost.35077 49152 0 49152 0 > ESTABLISHED > localhost.35078 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35078 49152 0 49152 0 > ESTABLISHED > localhost.35079 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35079 49152 0 49152 0 > ESTABLISHED > localhost.35080 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35080 49152 0 49152 0 > ESTABLISHED > localhost.35081 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35081 49152 0 49152 0 > ESTABLISHED > localhost.35084 localhost.35073 49152 0 49152 0 > ESTABLISHED > localhost.35073 localhost.35084 49152 0 49152 0 > ESTABLISHED > localhost.35600 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35601 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35602 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35603 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35604 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35605 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35606 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35608 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35609 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35610 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35611 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35612 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35613 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35614 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35615 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35616 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35617 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35618 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35619 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35620 servere.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35621 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35622 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35623 server.ldap 49640 0 49640 0 > TIME_WAIT > localhost.35624 server.ldap 49640 0 49640 0 > TIME_WAIT > > Is this normal or do I have a problem? > _______________________________________________ > 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 ahoesch at smartsoft.de Fri Apr 8 07:35:10 2005 From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=) Date: Fri, 8 Apr 2005 13:35:10 +0200 Subject: Summary: Mirror down in SunFire 240R In-Reply-To: <76fd01e805040108351d9f1c0f@mail.gmail.com> Message-ID: <43D18566-A822-11D9-A09D-003065CCA582@smartsoft.de> Dear managers, thanks a lot for your very helpful responses: Daniel Nuno Adam Tomkinson "Thomas M. Payerle" "Dell, Mary" "Tom Grassia" "Michael Schneider/calispera.com" Steve Sandau "joe_fletcher" Sam Nelson "Harrington, David B (Contractor) (J6R)" Gary Chambers "NO UCE" I used format to verify that we have a bad disk AVAILABLE DISK SELECTIONS: 0. c1t0d0 /pci at 1c,600000/scsi at 2/sd at 0,0 1. c1t1d0 /pci at 1c,600000/scsi at 2/sd at 1,0 I broke the mirror > metadetach -f d0 d20 > metadetach -f d1 d21 > metadetach -f d3 d23 > > metaclear -f d23 1 1 c1t1d0s3 > metaclear -f d21 1 1 c1t1d0s1 > metaclear -f d20 1 1 c1t1d0s0 Since the machine was only half a year old this problem was resolved by Sun. They sent a technician that replaced the disk and most likely (I could not be present) did the following to rebuild the mirror. dd if=/dev/rdsk/c1t0d0s2 of=/dev/rdsk/c1t1d0s2 count=16 metadb -a -c 2 c1t1d0s7 metainit d20 1 1 c1t1d0s0 metainit d21 1 1 c1t1d0s1 metainit d23 1 1 c1t1d0s3 metattach d0 d20 metattach d1 d21 metattach d3 d23 What is the difference between dd if=/dev/rdsk/c1t0d0s2 of=/dev/rdsk/c1t1d0s2 count=16 and prtvtoc /dev/rdsk/c1t0d0s2 | fmthard -s - /dev/rdsk/c1t1d0s2 The second was recommended since dd was said to copy the whole disk. But as I understand it the above dd command only copies the partition table (partition s2). Is this correct or wrong? Thanks a lot! Regards, Andreas Original question: ==================================================================== From: Andreas Hvschler Date: Tue Mar 29, 2005 10:19:45 PM Europe/Berlin To: sunmanagers at sunmanagers.org Subject: Mirror down in SunFire 240R Dear managers, we have a production machine Sun Fire 240 with two mirrored 73 GByte SCSI-disks. The mirror was fine for a year. Now I get the following: bash-2.05# metastat d3: Mirror Submirror 0: d13 State: Okay Submirror 1: d23 State: Needs maintenance Pass: 1 Read option: roundrobin (default) Write option: parallel (default) Size: 10247232 blocks (4.9 GB) d13: Submirror of d3 State: Okay Size: 10247232 blocks (4.9 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t0d0s3 0 No Okay Yes d23: Submirror of d3 State: Unavailable Size: 10247232 blocks (4.9 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t1d0s3 0 No - Yes d1: Mirror Submirror 0: d11 State: Okay Submirror 1: d21 State: Needs maintenance Pass: 1 Read option: roundrobin (default) Write option: parallel (default) Size: 14337984 blocks (6.8 GB) d11: Submirror of d1 State: Okay Size: 14337984 blocks (6.8 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t0d0s1 0 No Okay Yes d21: Submirror of d1 State: Unavailable Size: 14337984 blocks (6.8 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t1d0s1 0 No - Yes d0: Mirror Submirror 0: d10 State: Okay Submirror 1: d20 State: Needs maintenance Pass: 1 Read option: roundrobin (default) Write option: parallel (default) Size: 114591936 blocks (54 GB) d10: Submirror of d0 State: Okay Size: 114591936 blocks (54 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t0d0s0 0 No Okay Yes d20: Submirror of d0 State: Unavailable Size: 114591936 blocks (54 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare c1t1d0s0 0 No - Yes Device Relocation Information: Device Reloc Device ID c1t1d0 Yes id1,sd at SSEAGATE_ST373307LSUN72G_3HZ6Y3DJ00007431G1EF c1t0d0 Yes id1,sd at SSEAGATE_ST373307LSUN72G_3HZ6Y1YR00007431G08H The system is up and running but very obviously there is something wrong with the second disk. I have never encountered such a situation and am not sure what to do now. I would really appreciate your hints before I do something wrong with a production machine. Thanks a lot in advance! Regards, Andreas _______________________________________________ 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 dave at primonics.com Mon Apr 11 09:43:38 2005 From: dave at primonics.com (Dave Hinse) Date: Mon, 11 Apr 2005 09:43:38 -0400 Subject: Solaris 8 to 9 upgrade and RAID Manager 6.22.01.14 Message-ID: <76E2D3FFBE34D411906400508BD899537EBDAE@beta.primonics.com> Currently running 2 servers (E450 and V880) on solaris 8. I need to upgrade to solaris 9 and was wondering if anyone has done this before and if they can give me some tips on what to do. I have been doing a lot of research but nothing comes close to what I need to do ! Can I upgrade from solaris 8 to 9 without any hickups ? Should I upgrade RAID Manager first then upgrade Solaris ? I was thinking of 2 scenarios to upgrade the systems: 1. Install solaris 9 and RAID Manager on a new disk, copy vfstab over I'm not sure if the currently configured disk array will survive such a transfer? Or 2.dd the boot drive to another drive, upgrade from sol8 to 9 and keep the same version of RAID Manager. I can't be down for more than 4 hours. Any comments or advise would be greatly appreciated Thanks -dave From john_lee at summithq.com Fri Apr 8 15:52:44 2005 From: john_lee at summithq.com (John Lee) Date: Fri, 8 Apr 2005 15:52:44 -0400 Subject: SUMMARY: Solaris doesn't recognise RAM upgrade. Message-ID: <7C1FA139E8CA6B4DA61A1B26C440274FE3B409@nyexchange02.summithq.com> Hi George, I read you post about your ugrading your Sun Fire 280R memory. So the Simpletech memory didn't fix the issue. You just put in original Sun memory? Thanks. John ----------------------------------------------------------------------------- ------------------- The views and opinions expressed in this email message are the sender's own, and do not necessarily represent the views and opinions of Summit Systems Inc. This message has been checked for viruses but the receiver should not depend upon this and should perform their own virus check. This message may contain confidential and/or privileged information and is intended for the named recipient(s) only. If you are not a named recipient, do not use, copy, transmit or otherwise disclose the contents of this message. Please inform the sender promptly of your receipt of this message. From hoogendyk at bio.umass.edu Fri Apr 8 17:17:07 2005 From: hoogendyk at bio.umass.edu (Chris Hoogendyk) Date: Fri, 08 Apr 2005 17:17:07 -0400 Subject: [Summary] keeping serial console open on Sun Blade 100 even with keyboard & mouse attached In-Reply-To: <4256A229.6070201@bio.umass.edu> References: <4256A229.6070201@bio.umass.edu> Message-ID: <4256F4D3.60900@bio.umass.edu> One reply - from Casper Dik # eeprom output-device=ttya # eeprom input-device=ttya Based on his information and a bit more searching, I found that this stuff is documented in the Open Boot Command Reference Manual -- docs.sun.com product categories solaris on sun hardware solaris 9 on sun hardware documentation set openboot 2.x command reference manual openboot 2.x quick reference openboot 3.x command reference manual openboot 3.x quick reference Anyway, it worked, but I'm still not quite where I want to be. I connected remotely through an ssh session and su'd to root. The eeprom commands can be issued without being at the ok prompt. Then I opened up another terminal session on my Mac, ran kermit and opened the serial port to the Sun Blade (I already had the cable in place), went back to the ssh session, and issued a reboot. The video monitor on the Sun Blade did not initially come on. The openboot output came through the serial port to my Mac desktop. At the end of the openboot, when it said it was initiating the desktop login on display :0, the video monitor on the Sun Blade came on and showed the graphical login. I logged in there and still had my serial connection to my Mac desktop. I opened a "console" window in the CDE environment on the Sun Blade, came back to my Mac desktop, ssh'd to the Sun Blade, and purposely typed a bad password into an su. I got the console error message on my serial connection, but not on the console window on the video monitor on the Sun Blade itself. So, it works. But the output is truly re-routed to ttya and not tee'd to the console window on the vga port. I can do something like 'tail -f /var/adm/messages' in the console window on the Sun Blade, but that won't catch everything. I could work on that; but, during the bootup process, you won't see anything unless you have something connected to the serial port watching. --------------- Chris Hoogendyk - O__ ---- Systems Administrator c/ /'_ --- Biology Department (*) \(*) -- 140 Morrill Science Center ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- Chris Hoogendyk wrote: > keeping serial console open on Sun Blade 100 even with keyboard & mouse > attached > > ---- > > I'm configuring a Sun Blade 100 to be a minor server. Took me a while to > discover that the serial console is deactivated if the system is booted > with a keyboard and mouse attached. It then re-routes all console output > to the vga port. > > Does anyone know if there is a way to keep the serial console port even > when the Sun Blade is booted with the keyboard and mouse attached? > Setting a boot prom variable or something? Seems like it could at least > tee the output or something. That way, I would have the serial console > as an emergency fallback for remote management & stuff. > > TIA > > > --------------- > > Chris Hoogendyk > > - > O__ ---- Systems Administrator > c/ /'_ --- Biology Department > (*) \(*) -- 140 Morrill Science Center > ~~~~~~~~~~ - 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 ghenry at cmi.univ-mrs.fr Sat Apr 9 06:08:23 2005 From: ghenry at cmi.univ-mrs.fr (=?ISO-8859-1?Q?G=E9rard_Henry?=) Date: Sat, 09 Apr 2005 12:08:23 +0200 Subject: SUMMARY: how to set correct path? In-Reply-To: <424D02A2.1040002@cmi.univ-mrs.fr> References: <41D95BC6.9050702@cmi.univ-mrs.fr> <424D02A2.1040002@cmi.univ-mrs.fr> Message-ID: <4257A997.3040909@cmi.univ-mrs.fr> sorry for this late summary, not in my lab during last week thanks to everybody answering my question: Eric Lewandowski alexei Ryan Krenzischeck David Foster Michael Schulte Matthew Stier Bill Williams first, i apologize because my question contains mistake: if /etc/default/login contains correct PATH, this PATH is correctly use ven if we login with ssh. My mystake was to ssh root at serv, then su - , and i forgot that is SUPATH being used but many interesting things in all responses (except d. foster which says: man ssh, thanks!) some people suggest to use openssh instead of sun ssh, it's not my choice. Finally, /etc/default/login seems a good choice to me, except for env var as MANPATH, so i need to have a mechanism to give correct MANPATH to allusers, so i continue to use something like: in ${HOME}/.cshrc source /modele/Cshrc and /modele/Cshrc contains everything user needs. Here is responses i received: ----------------------------------------------------------------------- Eric Lewandowski You may want to try setting UseLogin=yes in your sshd_config file and restart SSHD or if its running from inetd, kill -HUP . ----------------------------------------------------------------------- alexei For BASH login shell, place the system wide ENV variables, such as PATH, in /etc/profile For csh/tcsh login shells, place them in /etc/csh.cshrc It is generic for csh/tcsh on any Unix OS. If /etc/csh.cshrc doesn't exist, then just create it. I have all the ENV settings in /etc/csh.cshrc on my Solaris 8, 9; Linux Debian and Redhat systems. You may also check on /etc/csh.login, but it is only executed once at login, then the settings are overwritten by those in /etc/csh.cshrc ----------------------------------------------------------------------- Ryan Krenzischeck Because the /etc/default/login only gets sourced when /bin/login is used. You can set Use_login in the sshd_config to yes and restart sshd. There was some security issue a while back that I can't remember about using UseLogin. I'd check google out first. > thanks, but there is nothing about UseLogin in sshd_config according > to man ssd_config in solaris 9 (sun ssh) Hmmm, your right. Apparently sun feels that: strings sshd |grep -i uselogin uselogin %s line %d: "UseLogin" option is obsolete. Which means, go get the OpenSSH and OpenSSL packages. Sun really does some crazy crap to open sourced software. No wonder I de-install their versions and install vanilla versions. ----------------------------------------------------------------------- David Foster man ssh ----------------------------------------------------------------------- Michael Schulte You can also set the path in /etc/.login, which is executed first by all csh login shells (and it works for me when I login with ssh and have the login shell csh). ----------------------------------------------------------------------- Bill Williams SSHd has the default/beginning PATH compiled in as added security feature, and that's what you get when you login with SSH. (SSHd does NOT refer to the /etc/default stuff defined for the normal 'login'.) You will need to provide additional PATH stuff through a profile of some kind. (Possibly /etc/profile) For example: I have an account I use for development, testing, etc. that I have setup to use 'bash' as its default shell; it gets all the extras I want with profiles ~/.bash... ----------------------------------------------------------------------- Gerard Henry wrote: > hello all, > i want to set path for all users in /etc/default/login (solaris 9) > # ALTSHELL determines if the SHELL environment variable should be set > # > ALTSHELL=YES > > # PATH sets the initial shell PATH variable > # > PATH=/usr/bin:/usr/sbin:/usr/openwin/bin:/usr/openwin/demo:/opt/csw/mysql4/bin:/ > > usr/sfw/bin:/usr/ccs/bin:/usr/ucb:/opt/SUNWspro/bin:/opt/csw/bin:/usr/local/bin: > > /opt/sfw/bin > > > it's okay when user log in with dtlogin, but if i connect with ssh, PATH > have not the same value: > echo $PATH > /usr/bin:/bin:/usr/sbin:/sbin > > > is it a normal feature? what's the best way to set path, avoiding > something as source /modele/Cshrc in .cshrc? > > thanks in advance, > > gerard _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From benoit.audet at carra.gouv.qc.ca Mon Apr 11 08:30:23 2005 From: benoit.audet at carra.gouv.qc.ca (=?ISO-8859-1?Q?Beno=EEt=20Audet?=) Date: Mon, 11 Apr 2005 08:30:23 -0400 Subject: SUMMARY: Exiting a terminal session from the StorEdge 3510 firmware Message-ID: Managers, Many thanks to the ones who replied to my question. Unfortunately, it dosen't seem to have a "quit" option to the StorEdge 3510 firmware. I should have mention that this wasn't a "unable to quit a tip session" problem, since I've got many answers suggesting me to RTFM about tip! ;) Obviously, I should have been more specific. Special thanks goes to Andrew Williams, Tim Evans and Grzegorz Bakalarski for their replies. Here's an answer resuming the point: "So, the answer is there is no logout/quit function. You just need to disconnect (~. in tip, ctrl-] in telnet connection). That's it. If you set up password the system will always ask you for password whenever you try to change any setting and possibly at a beginning of session." Thanks again to all. Ben Audet My original posting was: <---- Cut here ----> Fellow managers, We have some StorEdge 3510 that we manage both by the way of IP addresses and serial connections. In both connexion types, we use the RAID firmware, revision 3.25. Accessind and using it is flawless and everything works perfectly. But, as soon as we want to "logout" of the interface, we face a problem: we didn't find any "quit" option. This is painfull, especially when doing a "tip" connexion, where we can't get out of the firmware interface. Browsing in the "docs.sun.com" and in the PDFs included with the StorEdge 3510's documentation CDs, we do have plently information about how to connect and initiate sessions to the firmware, but nothing about quitting it... Do some of you knows of a way to properly "quits" or "exits" from it? Thanks in advance and, of course, will summarise! <---- Cut here ----> _______________________________________________________________ Benoit Audet Conseiller - TELUS solutions d'affaires Commission administrative des rigimes de retraite et d'assurances (CARRA) Service de l'assistance et des technologies (SAT) Til : 418.644.5932 (CARRA) benoit.audet at carra.gouv.qc.ca benoit.audet at telussa.com _______________________________________________________________ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Francois.Bousquet at cgi.com Mon Apr 11 10:21:18 2005 From: Francois.Bousquet at cgi.com (Bousquet Francois) Date: Mon, 11 Apr 2005 10:21:18 -0400 Subject: SUMMARY: Command to show disks with non-root user ? Message-ID: <1F85129C88B3714185C912F3AF47605012A25C74@nc1981exc03.banquelaurentienne.ca> Thanks for the quick answer : iostat -En I am searching for a Solaris command that will list all the disks (like the format command) with which I can run with a non-root user. Anyone have an idea ? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From john at starinc.com Mon Apr 11 13:47:22 2005 From: john at starinc.com (John Malick) Date: Mon, 11 Apr 2005 13:47:22 -0400 Subject: SUMMARY: Very strange vmstat issue In-Reply-To: <0FBBDCE1D6F6DF49841621D706B7DE3113D05ECA@usvwoaahn02.abh.vw.com> References: <0FBBDCE1D6F6DF49841621D706B7DE3113D05ECA@usvwoaahn02.abh.vw.com> Message-ID: <425AB82A.8060900@starinc.com> Looks like the problem has been resolved. Why this situation exists I don't know. Two V480's same OS and presumably the same patch releases. Vmstat is whacked on one but not the other. patchadd -p reveals that the patch levels are the same on both machines. uname -a though reveals that 117350-xx is the kernel patch on one machine but 108528-29 is what uname -a reveals on the other. patchadd -p still shows 117350-xx on that buggy V480 but apparently something happened during the patch install which did not get installed properly. Anyway, I re-applied the 117350 patch along with the fix patch for vmstat (117000-05) and the problem has gone away. This would be the first time in 15 years that a showrev or patchadd command stated a patch was installed when apparently it was not. Thanks for the help pointing me to the 117000-05 patch. >> >>John Malick wrote: >> >>>Running Solaris 8 with latest recommended patches from 1 month >>>ago. >>> >>>Also running Veritas VM 3.5 and Veritas FileSystem 3.5. >>> >>>Vmstat's "b" column, for kernel threads blocked for >> >>resources keeps, >> >>>keeps climbing. >>> >>>Below is just one line of many: >>> >>>r b w swap free re mf pi po fr de >> >>sr s0 sd sd >> >>>1 38877704 0 4857296 1137056 262 1380 0 4 4 0 0 0 0 0 >>> 0 >>> >>>As you can see, the 'b' column is currently 38+ million and >> >>climbing. >> >>>What is more strange is that I would expect these numbers >> >>to manifest >> >>>themselves as a system that would be completely unusable. >> >>The fact is >> >>>that the system does not seem to be taking any kind of >> >>performance hit >> >>>and is running along great. >>> >>>Obviously this number is incorrect but I can't find >> >>anything indicating >> >>>why this number would be so high and growing. >>> >>>I have an identical system (V480) running the same version >> >>of Solaris >> >>>and Veritas and with exactly the same patch levels which is >> >>not showing >> >>>this strange behaviour. >>> >>>Strange? >>> >>>Any ideas? >>> >>>Thanks >>> >> >>-- >> >> >> >>John Malick >>Star Systems Engineering >>232-236 West Market St. >>York, PA. 17401 >> >>john at starinc.com www.starinc.com >>ph:717-854-5911 fx:717-852-9421 >>_______________________________________________ >>sunmanagers mailing list >>sunmanagers at sunmanagers.org >>http://www.sunmanagers.org/mailman/listinfo/sunmanagers >> > > > -----BEGIN PGP SIGNATURE----- > Version: PGP Personal Security 7.0.3 > > iQA/AwUBQlWMrFe1Bhkj9lZeEQJFYgCg2XqHDW1uJdglc3gjvYUNJZ3AF5gAoPjH > SzRwjtZnb05/g1AxoZ3+/V1U > =MkZy > -----END PGP SIGNATURE----- > > -- John Malick Star Systems Engineering 232-236 West Market St. York, PA. 17401 john at starinc.com www.starinc.com ph:717-854-5911 fx:717-852-9421 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From G.Bakalarski at icm.edu.pl Wed Apr 13 08:37:24 2005 From: G.Bakalarski at icm.edu.pl (Grzegorz Bakalarski) Date: Wed, 13 Apr 2005 14:37:24 +0200 Subject: SUMMARY: large LUNs & labels & inodes (plus additional info) Message-ID: <20050413123724.GP12807@delta.icm.edu.pl> Dear All, I've got to answers - thanks to Drew Skinner & Mark Day . Query & answers are attached at the end. Seems the trick given by Mark should work. However I went other way. I did not changed EFI labelling on temporary space (I'm copying out data from large array in order to replace all disks onto temporary space [which is loan array]). But I was more carefull creating new luns on my reinstalled array - simply created more smaller luns on array side - all were below some 500GB and it seems this is safe limit (I had to play little with setting of number of heads and cylinders in array) in order to be free in setting fragment size and inode size. Anyway I'm really very disappointed with the fack that for filesystems (ufs) larger than 1TBytes the inode size must be larger or equal 1Mbytes. >From my side this makes use of multiterabyte filesystems impractical ... Few additional notes from copying data: (note: for Solaris 9 SPARC, after applying patch cluster dated Apr/8/2005 uname -a = SunOS xyz 5.9 Generic_118558-05 sun4u sparc SUNW,Sun-Fire-V440 ) 1) cp -rp does not copy bad symlinks 2) /usr/sbin/tar - silently ignores (skips) files larger than 8GBytes I found a patch for last "feature" - 115336-03 (it's free patch) - which fixes it. Anyone using routinely tar on Solaris 9 machine should apply this patch (patch is dated Oct/25/2004) 3) /usr/sbin/tar has rarely used option "E" which states for Extended headers. At first look it changes granulity of access time saved in an archive, but this option is crutial for files larger than 8GBytes. Best Regards, Gb ============================= QUERY ================================= Dear All, I have the following quite urgent query (I have to start copying large amount of data immediatelly) ... I've got large SCSI array - some 3TBytes. First I logged to the array and created 2 LUNs - 1.5TB each (i.e. > 1TB ) and tried to label them with format. The partitions looked different from "normal old style solaris" ufs partitions - the first partition (0) started from 34th block and also there were partition number 8 called "reserved". I tried to modify them and I could not start from first block or delete this "reserved" 8th partition. Finally I gave up and created partition #6 containing all available blocks. Googling I learned that for devices > 1TB solaris makes EFI labels. Then I tried to make ufs filesystem, but I found out that for filesystems > 1TB the minimal number of bytes per inode can be 1 Mbytes (i.e mkfs -i 1048576 ... ). Because I have huge numeber of small file (some 50 milions of files from 1kBytes to 100Kbytes) that was not an option for me. So I logged to the array again and created 4 LUNs - 750GB each (i.e < 1TB). Then I labaled disks (LUNs) again. But that was strange: LUNs # 0 and 2 preserved somehow EFI labels and LUNs # 1 and 3 got old style labels. I tried to change the labels with format but I could not either force all EFI or all old style labels. Finnaly I gave up again. Interesingly when I tried to make filesystems on the LUNs, I got unexpected results. For EFI labeled LUNs I could make "-i 8192" i.e. 8KBytes per inode, but for old style labeled LUNs I could not make less than 12KBytes per inode (i.e. -i 12000 - option for mkfs).. So my question is if there is any tool or way to switch between "old style" and EFI labels for LUNs smaller than 1TB ??? And other question: may this labeling influence on filesystem performance ??? I tried to learn more on sunsolve but most of documents were protected (usually they were not) ... THanks for any help. Kind regards, Gb ============================== ANSWER 1 =============================== Drew Skinner Date: Thu, 07 Apr 2005 14:00:18 -0400 Hi; I've had exactly the same problem with T3's that were configured with 180GB disks. A 'complete' T3 would present itself as 1.2T which generated the EFI labels. As I didn't want that, I ended up having to destroy the LUNs on the T3's and recreate them. The trick I used to accomplish this was by turning on/off MPXIO on the Solaris host so as to alter the WWN's being presented back. I also found them to be persistent (and nasty to get rid of). I didn't see (or find) any tools to help here and I'm sure using MPXIO in this manner is unsupported :) -Drew ============================ ANSWER 2 ================================= Mark Day Date: Thu, 7 Apr 2005 11:00:36 -0700 (PDT) My notes from when I ran into a similar problem say: format -e will prompt for label type when labeling a disk that is incorrectly getting EFI automatically. Also you can use something like: # prtvtoc /dev/rdsk/cztydxs2 > /tmp/vtoc # fmthard -s /tmp/vtoc /dev/rdsk/cxtydzs2 To copy the label from a LUN with identical geometry to another LUN. Mark Day ====================================================================== _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From asaenz at Found-Tech.com Wed Apr 13 09:31:01 2005 From: asaenz at Found-Tech.com (Al Saenz) Date: Wed, 13 Apr 2005 09:31:01 -0400 Subject: Summary: mirroring and SQL Message-ID: Thank you Bernd and Sunmanagers for your responses. I messed up prior by making the mirror without stopping the dbms. I installed the OS, mirror, then apps/sql. My mirror seems to be a success. I did some queries on the database and all looks great. Thanks again. Al -----Original Message----- From: Bernd Schemmer [mailto:Bernd.Schemmer at gmx.de] Sent: Tuesday, April 12, 2005 5:43 PM To: Al Saenz Subject: Re: mirroring and SQL Al Saenz wrote: >Dear Sunmanagers, > >I am configuring a system that will run an application that uses SQL server >(same system) >Should I mirror the drive first and then install my app/sql or install my >app/sql then mirror? > >Thank you for your time > > > I always setup the mirrors (using SDS or Veritas) before I start installing any application. Saves time regards Bernd >Al >_______________________________________________ >sunmanagers mailing list >sunmanagers at sunmanagers.org >http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > > -- Bernd Schemmer, Frankfurt am Main, Germany http://home.arcor.de/bnsmb/index.html Mas temprano que tarde el mundio cambiara Fidel Castro _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From wong.raymond at pansar.com.my Thu Apr 14 02:48:03 2005 From: wong.raymond at pansar.com.my (Raymond Wong) Date: Thu, 14 Apr 2005 14:48:03 +0800 (MYT) Subject: CPU Memory error causes reboot Message-ID: <2724.129.200.9.207.20050414144803.1113461283.iaput@phantom> Hi, Our sun server self rebooted this morning. Search seems to indicate that it's an ecache problem, but our's involves 3 CPUs. Please help to analyze the log entries & shed light on the nature of the problem. Extract of /var/adm/messages ~~~~~~~ Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 868172 kern.warning] WARNING: [AFT1] WP event on CPU3, errID 0x0000046c.737b999f Apr 14 08:47:04 machine-name AFSR 0x00000000.00800002 AFAR 0x000001ff.f1500000 Apr 14 08:47:04 machine-name AFSR.PSYND 0x0002(Score 95) AFSR.ETS 0x00 Fault_PC 0x17fff6c Apr 14 08:47:04 machine-name UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0000 UDBL.ESYND 0x00 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 315841 kern.warning] WARNING: [AFT1] Uncorrectable Memory Error on CPU2 Data access at TL=0, errID 0x0000046c.890e8026 Apr 14 08:47:04 machine-name AFSR 0x00000000.80200000 AFAR 0x00000000.077c8568 Apr 14 08:47:04 machine-name AFSR.PSYND 0x0000(Score 05) AFSR.ETS 0x00 Fault_PC 0x1002533c Apr 14 08:47:04 machine-name UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0203 UDBL.ESYND 0x03 Apr 14 08:47:04 machine-name UDBL Syndrome 0x3 Memory Module 190x Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359128 kern.warning] WARNING: [AFT1] errID 0x0000046c.890e8026 Syndrome 0x3 indicates that this may not be a memory module problem Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 127160 kern.info] [AFT2] errID 0x0000046c.890e8026 PA=0x00000000.077c8568 Apr 14 08:47:04 machine-name E$tag 0x00000000.1cc000ef E$State: Exclusive E$parity 0x0e Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x00): 0x03021764.030238bc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x08): 0x0000004c.0300952c Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x10): 0x0302c2a4.00000060 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x18): 0x03025968.0303cfcc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x20): 0x0000008c.02f78564 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 989652 kern.info] [AFT2] E$Data (0x28): 0x02f78564.000020ec *Bad* PSYND=0x00ff Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x30): 0x02f78570.02f78570 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x38): 0x0000010c.0302e44c Apr 14 08:47:04 machine-name unix: [ID 321153 kern.notice] NOTICE: Scheduling clearing of error on page 0x00000000.077c8000 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 937065 kern.info] [AFT3] errID 0x0000046c.890e8026 Above Error detected by protected Kernel code Apr 14 08:47:04 machine-name that will try to clear error from system Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 387418 kern.warning] WARNING: [AFT1] Uncorrectable Memory Error on CPU2 Data access at TL=0, errID 0x0000046c.8cc25fd7 Apr 14 08:47:04 machine-name AFSR 0x00000000.80200000 AFAR 0x00000000.077c8568 Apr 14 08:47:04 machine-name AFSR.PSYND 0x0000(Score 05) AFSR.ETS 0x00 Fault_PC 0x1002533c Apr 14 08:47:04 machine-name UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0203 UDBL.ESYND 0x03 Apr 14 08:47:04 machine-name UDBL Syndrome 0x3 Memory Module 190x Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 752377 kern.warning] WARNING: [AFT1] errID 0x0000046c.8cc25fd7 Syndrome 0x3 indicates that this may not be a memory module problem Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 999817 kern.info] [AFT2] errID 0x0000046c.8cc25fd7 PA=0x00000000.077c8568 Apr 14 08:47:04 machine-name E$tag 0x00000000.1cc000ef E$State: Exclusive E$parity 0x0e Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x00): 0x03021764.030238bc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x08): 0x0000004c.0300952c Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x10): 0x0302c2a4.00000060 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x18): 0x03025968.0303cfcc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x20): 0x0000008c.02f78564 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 989652 kern.info] [AFT2] E$Data (0x28): 0x02f78564.000020ec *Bad* PSYND=0x00ff Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x30): 0x02f78570.02f78570 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x38): 0x0000010c.0302e44c Apr 14 08:47:04 machine-name unix: [ID 321153 kern.notice] NOTICE: Scheduling clearing of error on page 0x00000000.077c8000 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 676056 kern.info] [AFT3] errID 0x0000046c.8cc25fd7 Above Error detected by protected Kernel code Apr 14 08:47:04 machine-name that will try to clear error from system Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 612880 kern.warning] WARNING: [AFT1] Uncorrectable Memory Error on CPU0 Data access at TL=0, errID 0x0000046c.94ea7880 Apr 14 08:47:04 machine-name AFSR 0x00000000.00200000 AFAR 0x00000000.077c8568 Apr 14 08:47:04 machine-name AFSR.PSYND 0x0000(Score 05) AFSR.ETS 0x00 Fault_PC 0x1880064 Apr 14 08:47:04 machine-name UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0203 UDBL.ESYND 0x03 Apr 14 08:47:04 machine-name UDBL Syndrome 0x3 Memory Module 190x Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 307235 kern.warning] WARNING: [AFT1] errID 0x0000046c.94ea7880 Syndrome 0x3 indicates that this may not be a memory module problem Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 243430 kern.info] [AFT2] errID 0x0000046c.94ea7880 PA=0x00000000.077c8568 Apr 14 08:47:04 machine-name E$tag 0x00000000.1cc000ef E$State: Exclusive E$parity 0x0e Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x00): 0x03021764.030238bc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x08): 0x0000004c.0300952c Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x10): 0x0302c2a4.00000060 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x18): 0x03025968.0303cfcc Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x20): 0x0000008c.02f78564 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 989652 kern.info] [AFT2] E$Data (0x28): 0x02f78564.000020ec *Bad* PSYND=0x00ff Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x30): 0x02f78570.02f78570 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x38): 0x0000010c.0302e44c Apr 14 08:47:04 machine-name unix: [ID 321153 kern.notice] NOTICE: Scheduling clearing of error on page 0x00000000.077c8000 Apr 14 08:47:04 machine-name SUNW,UltraSPARC-II: [ID 370242 kern.info] [AFT3] errID 0x0000046c.94ea7880 Above Error is in User Mode Apr 14 08:47:04 machine-name and is fatal: will reboot Apr 14 08:47:04 machine-name unix: [ID 855177 kern.warning] WARNING: [AFT1] initiating reboot due to above error in pid 1470 (oracle) Apr 14 08:47:20 machine-name unix: [ID 221039 kern.notice] NOTICE: Previously reported error on page 0x00000000.077c8000 cleared Apr 14 08:49:04 machine-name pseudo: [ID 129642 kern.info] pseudo-device: tod0 Apr 14 08:49:04 machine-name genunix: [ID 936769 kern.info] tod0 is /pseudo/tod at 0 Apr 14 08:49:04 machine-name syslogd: going down on signal 15 Apr 14 08:49:33 machine-name genunix: [ID 672855 kern.notice] syncing file systems... Apr 14 08:49:33 machine-name genunix: [ID 904073 kern.notice] done ~~~~~~~ Thanks, Raymond Wong System & Network Engineer Pan Sarawak Company Sdn Bhd *************************************************************** CAUTION: This e-mail is confidential and may contain privileged information. If you are not the intended recipient, you must not disclose or use the information. If you have received this e-mail in error, please notify by returning the e-mail and delete the document. Any views expressed message are those of the individual sender, and may not necessarily reflect the views of the Company. *************************************************************** From ssefoen at yahoo.com Wed Apr 13 17:20:11 2005 From: ssefoen at yahoo.com (Z H) Date: Wed, 13 Apr 2005 14:20:11 -0700 (PDT) Subject: SUMMARY:how to search special character in vi Message-ID: <20050413212011.43881.qmail@web31603.mail.mud.yahoo.com> Thanks to Karyn Williams's reply. Since my goal is to remove all the special character in the file. Here is one of the solutions: :%s/[^ -~]/ /g I just globally search and replace whatever characters that is not in the range from space to ~ with space. And it works. Thanks, ssefoen Here is the original question: Hello to all helpers: I have a need to search extended ascii code 227 in a file using vi. Could anyone share the tip? Thanks. __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at raider.llnl.gov Wed Apr 13 19:32:32 2005 From: martini at raider.llnl.gov (Dave Martini 1) Date: Wed, 13 Apr 2005 16:32:32 -0700 (PDT) Subject: SUMMARY: What printers best work with Solaris? Message-ID: <200504132332.j3DNWWlH008789@raider.llnl.gov> I received many responses which I'll just include below to make it easier. Someone mentioned that the jetadmin utility isn't supported anymore running on Solaris. Is there an alternative GUI from HP that one can use from a Solaris box to configure HP printers and plotters and to create virtual queues that point to specific tryas on the HP printers?? Also, there were several suggestions to use cups. I'm not using cups I'm just using a Solaris machine as my print server currently. What will cups buy me if I setup a cups server and can all printers be configured with it? Is there a GUI that is used to configure trays and virtual queues for each printer or is that still handled via the manufacturers software like jetadmin (HP), centreware for unix (Xerox) , lexprt/markvision (Lexmark)?? And can a cups server be configured on a Solaris machine or is this linux only? Thank you. Dave. Below are responses I received from my original question: As long as you buy printers that do postscript, then you're all set. It's even better if they are network printers (USB support is a bit iffy, and it won't work at all with older versions of Solaris - you didn't specify what you use). As far as printing goes, we set up a print server using CUPS, then all the other machines learn about the printers from NIS or LDAP. All the machines use the lpr protocol to spool to the printer server, that then spool to the printer using IPP/Socket/LPR or whatever the printer likes. The printer server also acts as a Xprt server for those applications that know about the XPrint extension (like Mozilla). ---------- I can't speak for all printers, but I never had a problem with HP Laserjet printers. Basically, I've never had a problem with anything networked that had a HP Jet Direct card on it, or whatever those things are called See, they cause so little trouble, I even forget what they're called. --Tom --------- I have a Lexmark laser printer that works wonderfully under S10 and S9. Rich Teer, SCNA, SCSA, OpenSolaris CAB member -------- I would buy the printer I want, and then worry about how to make it work. In my experience I have always been happier with HP printers. Don't count on Jetadmin for HP though, they don't support it on Solaris anymore. But you can use CUPS instead, I haven't used it but I've seen a tutorial and it looks straightforward: CUPS Printing Software: http//www.cups.org http://www.sun.com/bigadmin/content/submitted/install_cups.html The Xerox config utility is great, we only have a very high end Xerox printer so I can't comment much on office Xerox printers. Dave Foster --------- Regularly get involved with printers (consultant) I have had good experience for most network reputable type printers like HP, Textronix/Xerox. Also Ricoh , Kyocera/Minolta is good. Lexmark -...??? Do not belive all you read. A few years ago did a LOT on QMS (worked for a reseller) Excellent but pricey. However, the best experience has still been HP,w ith their options. They are an excellent little printer company :-) If you require high volume, Oce is good. Note: High Volume. Make sure your printer has a network support. You do not wish to be dependant on a centally attached server with USB/Centronics etc attached. Make for bad config and muchos gremlins etc. For software, I have given up. It would apear there is an inverse aspect of printer quality vs web page layout. I go to google and do: +Solaris +printermodel + download site:support-url. :-) ---- Any Postscript capable printer will be easy to administer ---------------- We use HP postscript laserjet printers with jetdirect cards in them. Windows seems to deal with that OK, and the Solaris systems use the HPNPL package from HP to tweak a few things that make lp work with these printers. Nothing difficult about install or use. No experience with the lexmark gear on Solaris Ric Anderson (ric at arizona.edu) ---------------------- It has been my experience that HP Postscript printers work (and play) well with all the major OS players out there. The vast majority of our systems are Solaris, but we also have Apple (my desktop is a Power Mac G5), Windows (unfortunately), HP-UX, and Linux. None of these systems has any trouble printing to our HP printers. We have a Solaris print server, and all printers' queues reside on it. Gary Chambers --------------------------------- Lexmark is all we use. Period. -Alex ------------------ Jetadmin/HPPI have these options as well and surely HP makes better printers than Lexmark! well, just my 2 cents with the last statement. :-) HPPI does have tray options and queues, plus you can customize most of the HP scripts installed. -chris ------------------------- Just get a postscript enabled printer with a TCP/IP printerserver built in. --------------------------- Go with Lexmark. They're still very much supported. -- M. ------------------------- I have 500+ HP printers configured with HPNP on Solaris 8. Never a problem. I define Lexmark/Optra/Konica printers as REMOTE printers with no spooling, never a problem. The mid-level Konicas (7040+ series) are especially sweet, they understand most Postscript and lp options. Printing is not the drama it was back in the 90's. Back when I was a operator on Sequent/old Solaris 2.5.1, we had a guy who spent 5 hours a day chasing down printer BS. Now I just confine users to purchasing only HPs and I only spend 2 hours a week on printers, tops. YMMV JV711 ----------------------------------------- Lexmarks are good. You can't lose with a printer if it supports postscript. Pretty much everything from within Solaris prints postscript. I have a Brother Postscript capable laser printer and it works flawlessly. I print to it from Windows and from Solaris. I don't have Mac's but I'm sure they have a driver for that as well. As a matter of fact I think Brother provides MAC drivers. Good luck --------------------------------------------- hp postscript laserjet with built-in networking you print directly to the ip address with printmgr. you print directly to the port with cups. ------------------------------------------------ If you've used CUPS to set up printers on Mac OS X, perhaps you could consider its use on Solaris, and thereby get access to all the control it offers. I tried a commercialized version called EasyPrintPro but I wasn't able to get an eval copy in time for my project and went with something else (Xi-Text). Anyway, that would let you use any printer you like, 'cause CUPS has such wide printer support. We have lots and lots of HP printers. I use both jetadmin to print to them, as well as dumping text jobs into our Windows print servers' spoolers. I would prefer some more control over output, but you take what you can get. :7( -wde -- Will Enestvedt UNIX System Administrator _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From goldsun8 at yahoo.com.sg Thu Apr 14 05:21:57 2005 From: goldsun8 at yahoo.com.sg (Gold Sun) Date: Thu, 14 Apr 2005 17:21:57 +0800 (CST) Subject: Partial Summary: can't do "sudo -s" on hardened box In-Reply-To: 6667 Message-ID: <20050414092157.64732.qmail@web41506.mail.yahoo.com> Hi, Thanks to all who responded but looks like Bill's reply fits my problem here but I still haven't resolve my problem : Below is what I have in /etc/sudoers (line without the #) : # Defaults override compiled-in values Defaults mailto="root", mail_no_user, mail_no_host, mail_no_perms Defaults tty_tickets,!root_sudo,umask=0077,ignore_dot,timestamp_timeout=5 Defaults syslog=auth Cmnd_Alias SHELLS=/usr/bin/ksh, /usr/bin/csh, /usr/bin/sh, /usr/bin/rsh, /bin/ksh, /bin/csh, /bin/sh, /bin/rsh #-----------[ Summary Specification ]------------# # format is WHO WHERE=(ASWHO) COMMANDS # # all upper case WHO is probably a sudo User_Alias # all lower case WHO is probably a userid # % at beginning of WHO is a UNIX (or DCE) Group name # # all upper case WHERE is probably a sudo Host_Alias # all lower case WHERE is probably a system name # User privilege specification #ADMINS ALL=ALL root ALL=(ALL) ALL %smcadmin ALL=ALL %helpdesk ALL=ALL, But even if my 9gohpo id were to be part of helpdesk or smcadmin in /etc/group, it still doesnt allow my id to 'sudo -s' with message below: $ sudo -s Password: Sorry, user 9gohpo is not allowed to execute '/usr/bin/ksh' as root on sp01qtt02ist5s5. $ Something extracted from /var/adm/messages : Apr 14 16:05:28 sp01qtt02ist5s5 sudo: [ID 702911 auth.alert] 9gohpo : command not allowed ; TTY=pts/1 ; PWD=/home/9gohpo ; USER=root ; COMMAND=/usr/bin/ksh Apr 14 16:05:28 sp01qtt02ist5s5 sudo: [ID 702911 auth.alert] 9gohpo : command not allowed ; TTY=pts/1 ; PWD=/home/9gohpo ; USER=root ; COMMAND=/usr/bin/ksh Apr 14 16:05:28 sp01qtt02ist5s5 sudo: [ID 702911 auth.alert] 9gohpo : command not allowed ; TTY=pts/1 ; PWD=/home/9gohpo ; USER=root ; COMMAND=/usr/bin/ksh Apr 14 16:05:28 sp01qtt02ist5s5 sendmail[9649]: [ID 801593 mail.info] j3E85S19009649: from=9gohpo, size=213, class=0, nrcpts=1, msgid=<200504140805.j3E85S19009649 at sp01qtt02ist5s5.infra.sgdcw.mebs.ihost.com>, relay=root at localhost Apr 14 16:05:28 sp01qtt02ist5s5 sendmail[9649]: [ID 801593 mail.info] j3E85S19009649: from=9gohpo, size=213, class=0, nrcpts=1, msgid=<200504140805.j3E85S19009649 at sp01qtt02ist5s5.infra.sgdcw.mebs.ihost.com>, relay=root at localhost Apr 14 16:05:28 sp01qtt02ist5s5 sendmail[9649]: [ID 801593 mail.info] j3E85S19009649: from=9gohpo, size=213, class=0, nrcpts=1, msgid=<200504140805.j3E85S190096 Bill Teeple wrote: If you couldn't enter into SUDO mode, you would get an error message that you 'are not part of the Sudoers group, and that your actions would be reported' - so your predicament doesn't have anything to do with that. When you enter the 'sudo' command, it just doesn't like your password - check the /var/adm/messages file to see if any error messages are displayed (you may need root access to attain this). You may need to tweak your log settings: Q) Sudo is setup to log via syslog(3) but I'm not getting any log messages. A) Make sure you have an entry in your syslog.conf file to save the sudo messages (see the sample.syslog.conf file). The default log facility is local2 (changeable via configure). Don't forget to send a SIGHUP to your syslogd so that it re-reads its conf file. Also, remember that syslogd does *not* create log files, you need to create the file before syslogd will log to it (ie: touch /var/log/sudo). Note: the facility ("local2.debug") must be separated from the destination ("/var/adm/sudo.log" or "@loghost") by tabs, *not* spaces. This is a common error. Give that a shot and see what you come up with. Bill -----Original Message----- From: Gold Sun [mailto:goldsun8 at yahoo.com.sg] Sent: Wed 4/13/2005 12:16 AM To: sunmanagers at sunmanagers.org Cc: Subject: can't do "sudo -s" on hardened box Hi, I have a hardened Solaris box. I was told by a remote colleague that I can't do "sudo -s" because I have not included my id into sudo. How do I do that? Can I just edit /etc/sudoers ? The following is a screen dump login as: myuserid myuserid at 10.196.16.15's password: $ $ sudo -s Password: Sorry, try again. Password: Yahoo! Mobile - Download the latest ringtones, games, and more! Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From goldsun8 at yahoo.com.sg Thu Apr 14 23:39:19 2005 From: goldsun8 at yahoo.com.sg (Gold Sun) Date: Fri, 15 Apr 2005 11:39:19 +0800 (CST) Subject: Partial Summary: can't do "sudo -s" on hardened box In-Reply-To: 6667 Message-ID: <20050415033919.6755.qmail@web41511.mail.yahoo.com> Hi All, The problem is my colleague's id on the same server is OK (ie he could do 'sudo -s'). Yes, /usr/bin/ksh is in /etc/shells : # more shells /bin/sh /bin/csh /bin/ksh /usr/bin/sh /usr/bin/csh /usr/bin/ksh /bin/false /sbin/sh /usr/local/bin/bash I've also tried Frank's suggestion in /etc/sudoers : root ALL=(ALL) ALL 9gohpo ALL=(ALL) ALL %smcadmin ALL=ALL %helpdesk ALL=ALL, !SHELLS but I'm still getting the message : Sorry, user 9gohpo is not allowed to execute '/usr/bin/ksh' as root on sp01qtt02ist5s5 when doing 'sudo -s' Is there some daemon that I must restart for it to take effect? I've placed my id under both smcadmin & helpdesk groups in /etc/group Thanks G Sun ========================= is /usr/bin/ksh in /etc/shells? Could be something as silly as that. -Sal Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From kabinet at inf.u-szeged.hu Fri Apr 15 05:07:11 2005 From: kabinet at inf.u-szeged.hu (kabinet at inf.u-szeged.hu) Date: Fri, 15 Apr 2005 11:07:11 +0200 Subject: SUMMARY: system activity report Message-ID: <425F843F.3090007@inf.u-szeged.hu> For sar work correctly, you MUST uncomment the lines of the file /etc/rc2.d/S21perf _and_ configure the crontab of sys user. Thanks to Rodrigo Hernandez and Brian Miller for clearing this. Krisztian Here's my original question: -------------- I use sa1 to collect system usage, sys user's crontab is: 0 * * * 0-6 /usr/lib/sa/sa1 I use sa1 to collect system usage, sys user's crontab is: 0 * * * 0-6 /usr/lib/sa/sa1 20,40 8-21 * * 1-5 /usr/lib/sa/sa1 5 22 * * 1-5 /usr/lib/sa/sa2 -s 8:00 -e 21:01 -i 1200 -A When I reboot my machine, data collection stops until I remove /var/adm/sa/sa## and re-create it by su sys -c "/usr/lib/sa/sadc /var/adm/sa/sa`date +%d`" This way I lose collected data before reboot. How can I fix this? 20,40 8-21 * * 1-5 /usr/lib/sa/sa1 5 22 * * 1-5 /usr/lib/sa/sa2 -s 8:00 -e 21:01 -i 1200 -A When I reboot my machine, data collection stops until I remove /var/adm/sa/sa## and re-create it by su sys -c "/usr/lib/sa/sadc /var/adm/sa/sa`date +%d`" This way I lose collected data before reboot. How can I fix this? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ahoesch at smartsoft.de Fri Apr 15 06:21:05 2005 From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=) Date: Fri, 15 Apr 2005 12:21:05 +0200 Subject: Summary: ssh does not ask for a password In-Reply-To: Message-ID: <13253117-AD98-11D9-BAAD-003065CCA582@smartsoft.de> Dear managers, thanks a lot to all that responded. Most suggested to look into ~/.ssh/authorized_keys and ~/.rhosts for a corresponding entry. However, these files did not exist. What finally turned out made me very angry. Here is a recommendation: Do all maintenance on your own and never give away the root password to anyone, for example a technician from Sun. Somebody - I know for sure it was not me - removed my password from this machine. Thanks a lot! Regards, Andreas >> Dear managers, >> >> I just encountered weird behaviour of the ssh daemon on one of our >> servers. If I do >> >> ssh www..com >> >> I am logged in without being asked for a password. How can this be? It >> is a Solaris 9 Sparc machine. Any ideas? >> >> Thanks a lot! >> >> Regards, >> >> Andreas >> _______________________________________________ >> 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 goldsun8 at yahoo.com.sg Fri Apr 15 12:29:05 2005 From: goldsun8 at yahoo.com.sg (Gold Sun) Date: Sat, 16 Apr 2005 00:29:05 +0800 (CST) Subject: SUMMARY: can't do "sudo -s" on hardened box In-Reply-To: 6667 Message-ID: <20050415162905.21125.qmail@web41508.mail.yahoo.com> Hi, Thanks to all the helpful replies : just fyi I've been using "visudo" instead of just "vi /etc/sudoers" all these while & a couple of people suggested that using vi can corrupt sudoers file such that it will refuse to work. However, it's Bill Steeple who got the "spot-on" answer (which I've appended below). I thought "!SHELLS" in the /etc/sudoers file is just a comment/remark but it's not. Thanks, G Sun ===================================== Doesn't your sudoers file state... %helpdesk ALL=ALL, !SHELLS Anyone in the HELPDESK group cannot run anything in the SHELLS file and since you are using 'sudo -s' which calls SUDO and passes the SHELL command to it, you are basically stating that the user 9gohpo (who is in the group smcadmin and helpdesk) will not be able to run any of the shells listed in the /etc/shells file. remove your user account from the HELPDESK group and then try it again. This should remove the more restrictive settings you have for the HELPDESK group. Bill ---------------Original Message--------------- From: Gold Sun [mailto:goldsun8 at yahoo.com.sg] Sent: Thu 4/14/2005 8:39 PM To: sunmanagers at sunmanagers.org Cc: Subject: Partial Summary: can't do "sudo -s" on hardened box Hi All, The problem is my colleague's id on the same server is OK (ie he could do 'sudo -s'). Yes, /usr/bin/ksh is in /etc/shells : # more shells /bin/sh /bin/csh /bin/ksh /usr/bin/sh /usr/bin/csh /usr/bin/ksh /bin/false /sbin/sh /usr/local/bin/bash I've also tried Frank's suggestion in /etc/sudoers : root ALL=(ALL) ALL 9gohpo ALL=(ALL) ALL %smcadmin ALL=ALL %helpdesk ALL=ALL, !SHELLS but I'm still getting the message : Sorry, user 9gohpo is not allowed to execute '/usr/bin/ksh' as root on sp01qtt02ist5s5 when doing 'sudo -s' Is there some daemon that I must restart for it to take effect? I've placed my id under both smcadmin & helpdesk groups in /etc/group Thanks G Sun Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From goldsun8 at yahoo.com.sg Fri Apr 15 12:30:49 2005 From: goldsun8 at yahoo.com.sg (Gold Sun) Date: Sat, 16 Apr 2005 00:30:49 +0800 (CST) Subject: SUMMARY: can't do "sudo -s" on hardened box In-Reply-To: 6667 Message-ID: <20050415163049.76171.qmail@web41529.mail.yahoo.com> Hi, Thanks to all the helpful replies : just fyi I've been using "visudo" instead of just "vi /etc/sudoers" all these while & a couple of people suggested that using vi can corrupt sudoers file such that it will refuse to work. However, it's Bill Steeple who got the "spot-on" answer (which I've appended below). I thought "!SHELLS" in the /etc/sudoers file is just a comment/remark but it's not. Thanks, G Sun ===================================== Doesn't your sudoers file state... %helpdesk ALL=ALL, !SHELLS Anyone in the HELPDESK group cannot run anything in the SHELLS file and since you are using 'sudo -s' which calls SUDO and passes the SHELL command to it, you are basically stating that the user 9gohpo (who is in the group smcadmin and helpdesk) will not be able to run any of the shells listed in the /etc/shells file. remove your user account from the HELPDESK group and then try it again. This should remove the more restrictive settings you have for the HELPDESK group. Bill ---------------Original Message--------------- From: Gold Sun [mailto:goldsun8 at yahoo.com.sg] Sent: Thu 4/14/2005 8:39 PM To: sunmanagers at sunmanagers.org Cc: Subject: Partial Summary: can't do "sudo -s" on hardened box Hi All, The problem is my colleague's id on the same server is OK (ie he could do 'sudo -s'). Yes, /usr/bin/ksh is in /etc/shells : # more shells /bin/sh /bin/csh /bin/ksh /usr/bin/sh /usr/bin/csh /usr/bin/ksh /bin/false /sbin/sh /usr/local/bin/bash I've also tried Frank's suggestion in /etc/sudoers : root ALL=(ALL) ALL 9gohpo ALL=(ALL) ALL %smcadmin ALL=ALL %helpdesk ALL=ALL, !SHELLS but I'm still getting the message : Sorry, user 9gohpo is not allowed to execute '/usr/bin/ksh' as root on sp01qtt02ist5s5 when doing 'sudo -s' Is there some daemon that I must restart for it to take effect? I've placed my id under both smcadmin & helpdesk groups in /etc/group Thanks G Sun Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ahoesch at smartsoft.de Fri Apr 15 15:15:54 2005 From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=) Date: Fri, 15 Apr 2005 21:15:54 +0200 Subject: Summary: Problem with Sun Ray 1G In-Reply-To: <425FEE73.8000602@acntv.com> Message-ID: Dear managers, I found the following document. It told me what I needed. http://www.filibeto.org/sun/sunray-users/lib/sunsolve/22076.htm Basically /opt/SUNWut/bin/utxconfig -r 1280x1024 -s on -a /opt/SUNWut/sbin/utresadm -a -c default -t default 1280x1024 at 60 /opt/SUNWut/sbin/utresadm -a -c IEEE802.0003ba5d3b7a -t default 1600x1200 at 60d /opt/SUNWut/sbin/utresadm -o /opt/SUNWut/bin/utxconfig -l did the trick for me. Thanks a lot! Regards, Andreas >> Dear managers, >> we are using SRSS 2.x on Solaris 9. As long as I had a NEC TFT with >> 1280x1024 connected to my Sun Ray, everything was fine. I now >> replaced that with a MultiSync LCD 2080UX+ with a recommended >> resolution of 1600x1200. When I power up my Sun Ray I see the little >> graphic showing me the assigned IP-address, then the screen gets >> dark and the monitor on screen BIOS gives me a "OUT OF RANGE" >> message. It seems that the monitor accepts the (wrong) timing for a >> while but then switches away the signal. The log says something >> about 1600x1200, but obviously this resolution is switched back to >> 640x480 which the new monitor is not able to display. How can I fix >> that? >> I have telnet access to the server (280R with no monitor connected) >> and would like to set fixed resolutions for a given Sun Ray. >> Thanks a lot! >> Regards, >> Andreas >> ********************************************************************** >> ** *********************** >> Resetting servers... messages will be logged to >> /var/opt/SUNWut/log/messages. >> bash-2.05# tail -f /var/opt/SUNWut/log/messages >> Apr 15 17:24:10 sunfire utauthd: [ID 163784 user.info] main NOTICE: >> SmartCardConfigData: /etc/opt/SUNWut/smartcard/MondexUNU.cfg: 67 >> tokens processed >> Apr 15 17:24:10 sunfire utauthd: [ID 524863 user.info] main NOTICE: >> SmartCardConfigData: /etc/opt/SUNWut/smartcard/Cryptoflex.cfg: 144 >> tokens processed >> Apr 15 17:24:10 sunfire utauthd: [ID 651628 user.info] main NOTICE: >> SmartCardConfigData: /etc/opt/SUNWut/smartcard/UnknownCard.cfg: 63 >> tokens processed >> Apr 15 17:24:10 sunfire utauthd: [ID 723974 user.info] main NOTICE: >> Loaded module /opt/SUNWut/lib/modules/StartSession.m0 >> Apr 15 17:24:10 sunfire utauthd: [ID 612231 user.info] main NOTICE: >> Loaded module /opt/SUNWut/lib/modules/Authxlation.m1 >> Apr 15 17:24:11 sunfire utauthd: [ID 709793 user.info] main NOTICE: >> Loaded module /opt/SUNWut/lib/modules/ServerSelect.m2 >> Apr 15 17:24:11 sunfire utauthd: [ID 723977 user.info] main NOTICE: >> Loaded module /opt/SUNWut/lib/modules/StartSession.m3 >> Apr 15 17:24:11 sunfire utauthd: [ID 745953 user.info] main NOTICE: 4 >> authentication modules loaded. >> Apr 15 17:24:11 sunfire utauthd: [ID 826448 user.info] deviceManager0 >> NOTICE: DeviceManager.getDeviceManager: Initiate callback to >> utdevMgrd at localhost:7011 >> Apr 15 17:24:11 sunfire utauthd: [ID 914482 user.info] deviceManager0 >> NOTICE: DeviceManager.initiateCallback localhost:7010 established >> communication >> Apr 15 17:24:35 [192.168.1.40.2.2] 0x0.0x1f0 0:3:ba:5d:3b:7a >> Application: make_listener: software Version mismatch -- DHCP >> version (CoronaP7-2.0_37.b,REV=2002.12.19.07.46) booted >> (CoronaP7-MfgPkg_3.23,2.0,REV=2003.11.17.11.06) >> Apr 15 17:24:35 [192.168.1.40.2.2] 0x0.0x303 0:3:ba:5d:3b:7a >> Application: tftpopt: all read attempts failed, giving up >> Apr 15 17:24:35 [192.168.1.40.2.2] 0x0.0x303 0:3:ba:5d:3b:7a >> Application: make_listener: unable to get a new load, continuing >> with current version >> Apr 15 17:24:35 sunfire utauthd: [ID 121962 user.info] Worker0 >> NOTICE: whichServer pseudo.0003ba5d3b7a: >> Apr 15 17:24:35 sunfire utauthd: [ID 610736 user.info] Worker0 >> NOTICE: CLAIMED by StartSession.m3 NAME: pseudo.0003ba5d3b7a >> PARAMETERS: {_=1, >> clientRand=b/Ip/e9RqwTR9bnQMfXpef5237oGJ6qR2qgPBwmOTpy, >> rawId=0003ba5d3b7a, terminalIPA=192.168.1.40, startRes=1600x1200, >> state=disconnected, barrierLevel=200, initState=1, >> fw=MfgPkg_3.23,2.0,REV=2003.11.17.11.06,Boot:1.6; >> 2003.11.17-11:19:46-PST, pn=41458, rawType=pseudo, sn=0003ba5d3b7a, >> tokenSeq=1, MTU=1500, event=insert, id=0003ba5d3b7a, >> firstServer=c0a80101, cause=insert, hw=SunRayP7, type=pseudo, >> namespace=IEEE802} >> Apr 15 17:24:35 sunfire utauthd: [ID 140619 user.info] Worker0 >> NOTICE: CONNECT IEEE802.0003ba5d3b7a, pseudo.0003ba5d3b7a, all >> connections allowed >> Apr 15 17:24:35 sunfire utauthd: [ID 118787 user.info] Worker0 >> NOTICE: MTU = 1500 >> Apr 15 17:24:35 sunfire utauthd: [ID 839304 user.info] Worker0 >> NOTICE: SessionManager.getSessionManager: Initiate callback to >> utsessiond at localhost:7007 >> Apr 15 17:24:35 sunfire utauthd: [ID 218807 user.info] Worker0 >> NOTICE: SessionManager.initiateCallback localhost:7010 established >> communication >> Apr 15 17:24:35 sunfire utauthd: [ID 678911 user.info] Worker0 >> UNEXPECTED: AuthRecord::CreateClient:/opt/SUNWut/lib/utdtsession add >> :child error: openAdmin() failed: Internal system error >> Apr 15 17:24:35 sunfire utauthd: [ID 111219 user.info] Worker0 >> NOTICE: SESSION_OK pseudo.0003ba5d3b7a >> _______________________________________________ >> 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 Vgerman at us.nomura.com Fri Apr 15 17:16:09 2005 From: Vgerman at us.nomura.com (German, Vinnie) Date: Fri, 15 Apr 2005 17:16:09 -0400 Subject: SUMMARY: rsync asking for root password ???? Message-ID: Special thanks to: Matt Clausen [mclausen at csit.fsu.edu] Adam Levin [levins at westnet.com] dave at umiacs.umd.edu Matthew Stier [Matthew.Stier at us.fujitsu.com] Brewer, Dwight R. CONTRACTOR [Dwight.Brewer_CONTRACTOR at dtra.mil] g at ry [loverboi at gmail.com] William Enestvedt [William.Enestvedt at jwu.edu] Tim Chipman [chipman at ecopiabio.com] Thanks again guys. My posting is down below and here is the answer : Here's what you want to do: on Machine A (which is defined as the machine that will be sshing to a different machine) you need to run this command as root: ssh-keygen -t dsa -f ~/.ssh/id_dsa -N "" Which will create two files: id_dsa and id_dsa.pub Copy id_dsa.pub to Machine B and put it in the .ssh directory for root as authorized_keys and authorized_keys2 (some SSH installs use authorized_keys and others use authorized_keys2). Now if you SSH from Machine A to Machine B, it should not prompt you for a password at all and just log you right in. _____ From: German, Vinnie Sent: Friday, April 15, 2005 2:30 PM To: sunmanagers at sunmanagers.org Subject: rsync asking for root password ???? Hello, I'm rsycing data from ServerA to ServerB and data is being rsyced fine however Having the inconveniences of asking me for password every time data is synced Which is something I can not afford since I will be running this out of cron. I'm using ssh on a Solaris 8 environment. Does someone have an idea on which File will need to be modified to stop rsync from requesting root password? Thanks in advance, Vinnie. ----- This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please delete it and all copies from your system, destroy any hard copies and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Nomura Holding America Inc., Nomura Securities International, Inc, and their respective subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state the views of such entity. Unless otherwise stated, any pricing information in this message is indicative only, is subject to change and does not constitute an offer to deal at any price quoted. Any reference to the terms of executed transactions should be treated as preliminary only and subject to our formal written confirmation. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bshah at citadon.com Mon Apr 18 19:36:27 2005 From: bshah at citadon.com (Bhavesh Shah) Date: Mon, 18 Apr 2005 16:36:27 -0700 Subject: SUMMARY: Ghost Disk in E-3500 Message-ID: MANY THANKS to Eugene for his correct answer. It was wrongly connected from LB - UA changed it back to UA - LA The Solution is below: First, a fiber disk has two channels into it, channel A and B. The E3500 can dual host disks. As such you see the disk via port A and B, ie shows twice in format. Note w2200... = PORT B and w2100... = port A. OK - so you have two disk internal. The internal bays are made up as 4 top + 4 bottom. The top 4 bays drive from the top 2 ports ( A + B) on the FC-AL interface board . Ditto, the bottom 4 bays off the bottom two ports (A + B).FC-AL interface board is at rear right - next to power cord - can house up to 4 gbics. If you have 1 disk top, one disk bottom, it means you should move the one GBIC and FC-AL cable one port down, from second from top to 3rd from top on interface brd. It cab be drawn as: +-----+-----+-----+------------+-+ Top Bays - Port A 21000... +---+ +---+ +---+ +---+ +-+ | | | | | | | | | | | | | | | | | | | | | | | | +---+ +---+ +---+ +---+ +-+ +-----+-----+-----+------------+-+ Top Bays - Port B 22000... +-----+-----+-----+------------+-+ Bottom Bays - Port A 21000... +---+ +---+ +---+ +---+ +-+ | | | | | | | | | | | | | | | | | | | | | | | | +---+ +---+ +---+ +---+ +-+ +-----+-----+-----+------------+-+ Bottom Bays - Port B 22000... MY Original question was: Hi Gurus, I have a strange issue. I have Sun E-3500 with 2X9GB internal Fiber Channel Disks running Solaris 8 at co-location. But the problem is I can't see the second disk instead in format I see the root disk twice. The WWN of root disk is w21000020370e84aa whereas WWN of second disk (which is nothing but root disk only) shows WWN as w22000020370e84aa. Which means one disk is in Port A whereas second disk is in Port B leaving that both WWN no are same. I am wondering where the second disk gone? And why root disk is showing twice? I tried to delete partition 7 (Home) as was able to delete the same and eventually was deleted from root Disk 0. How come mounted partition can be deleted? Any help/pointer to figure out this GHOST Disk issue will be appreciated. I will summarize. Regards B format 0. c0t0d0 /sbus at 2,0/SUNW,socal at d,10000/sf at 0,0/ssd at w21000020370e84aa,0 1. c1t0d0 /sbus at 2,0/SUNW,socal at d,10000/sf at 1,0/ssd at w22000020370e84aa,0 Partition table Disk 0 Current partition table (original): Total disk cylinders available: 4924 + 2 (reserved cylinders) Part Tag Flag Cylinders Size Blocks 0 root wm 0 - 292 513.75MB (293/0/0) 1052163 1 swap wu 293 - 1461 2.00GB (1169/0/0) 4197879 2 backup wm 0 - 4923 8.43GB (4924/0/0) 17682084 3 var wm 1462 - 2046 1.00GB (585/0/0) 2100735 4 unassigned wm 2047 - 3169 1.92GB (1123/0/0) 4032693 5 usr wm 3170 - 4338 2.00GB (1169/0/0) 4197879 6 unassigned wm 0 0 (0/0/0) 0 7 home wm 4339 - 4923 1.00GB (585/0/0) 2100735 Disk 1 Current partition table (original): Total disk cylinders available: 4924 + 2 (reserved cylinders) Part Tag Flag Cylinders Size Blocks 0 root wm 0 - 292 513.75MB (293/0/0) 1052163 1 swap wu 293 - 1461 2.00GB (1169/0/0) 4197879 2 backup wm 0 - 4923 8.43GB (4924/0/0) 17682084 3 var wm 1462 - 2046 1.00GB (585/0/0) 2100735 4 unassigned wm 2047 - 3169 1.92GB (1123/0/0) 4032693 5 usr wm 3170 - 4338 2.00GB (1169/0/0) 4197879 6 unassigned wm 0 0 (0/0/0) 0 7 home wm 4339 - 4923 1.00GB (585/0/0) 2100735 _______________________________________________ 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 serafino at cshl.edu Tue Apr 19 10:19:26 2005 From: serafino at cshl.edu (Sal Serafino) Date: Tue, 19 Apr 2005 10:19:26 -0400 (EDT) Subject: SUMMARY: Semaphore Error Message-ID: <200504191419.j3JEJQL14905@phage.cshl.edu> Well, I only got one reply. Thanks to Eric Paul for suggesting that I raise the semmni from 100 to 256. That didn't do the trick, I'm afraid, but I've gotten much further along this time. Now, I was able to import the tablespace, but a separate process that builds indices on the tables in a different tablespace is doing the same thing. I've built a new SID and have been playing with the numbers. I am hoping that I have something bad someplace that is hidden and that a new instance will make it work properly. If this fails, I may have another post. Thanks to all, -Sal ------------- Begin Forwarded Message ------------- Date: Thu, 14 Apr 2005 15:21:56 -0400 (EDT) From: Sal Serafino Subject: Semaphore Error To: sunmanagers at sunmanagers.org Mime-Version: 1.0 Content-MD5: azQ73iTWcCzQjD3OE11x/w== Hi All- I have a situation where I am transporting an Oracle (9.0.1) tablespace between two identically outfitted and configured 450's running Solaris 8. My script sets up the transaction to use a huge rollback segment in order to offline drop the datafile and then to drop the tablespace including contents. Oracle will drop all objects from the database, but then hangs at the end. A trace of the processes spawned shows: 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) Err#11 EAGAIN 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) (sleeping...) 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) Err#11 EAGAIN 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) (sleeping...) 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) Err#11 EAGAIN 5944: semtimedop(2293762, 0xFFFFFFFF7FFF8FEC, 1, 0xFFFFFFFF7FFF8FD8) (sleeping...) I had to kill it to make it stop. I actually did get a "Tablespace Dropped." message on the Oracle side, and there are no objects owned by this user and no references to the tablespace anywhere. BUT... the tablespace still exists in dba_tablespaces and the datafile still exists in dba_data_files, so any attempt to import the transported tablespace afterwards dies. According to semop(2), The semtimedop() function will fail if: EAGAIN The timeout expired before the requested operation could be completed. The semtimedop() function will fail if one of the following is detected: EFAULT The timeout argument points to an illegal address. EINVAL The timeout argument specified a tv_sec or tv_nsec value less than 0, or a tv_nsec value greater than or equal to 1000 million. There are errors other than EAGAIN. My /etc/system file looks good: set msgsys:msginfo_msgmax=8192 set msgsys:msginfo_msgmnb=16384 set msgsys:msginfo_msgmni=1700 set msgsys:msginfo_msgtql=512 set semsys:seminfo_semmns=2048 set semsys:seminfo_semmnu=2048 set semsys:seminfo_semmsl=2048 set semsys:seminfo_semmni=100 set semsys:seminfo_semume=256 set shmsys:shminfo_shmmax=2147483647 set shmsys:shminfo_shmmin=1 set shmsys:shminfo_shmmni=100 set shmsys:shminfo_shmseg=10 set maxusers=256 set nproc=4096 set pt_cnt=256 set rlim_fd_cur=2048 set rlim_fd_max=8192 Ideas? Greatly appreciated, -Sal ------------- End Forwarded Message ------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From halla3 at corp.earthlink.net Tue Apr 19 10:37:28 2005 From: halla3 at corp.earthlink.net (Andrew Hall) Date: Tue, 19 Apr 2005 10:37:28 -0400 Subject: SUMMARY: solaris 10 and ssh In-Reply-To: <42545106.7050301@corp.earthlink.net> References: <42545106.7050301@corp.earthlink.net> Message-ID: <426517A8.7030509@corp.earthlink.net> This was caused by my finish script removing the setuid bit from /usr/lib/pt_chmod. Once I added the setuid bit, non root users can login. Drew Andrew Hall wrote: > Greetings, > > Looks like has bundled a version of ssh w/ solaris 10. With this > version I cannot login to the server as a non-root user. I get the > following error in messages: > > Apr 6 16:57:27 xxx sshd[736]: [ID 800047 auth.error] error: grantpt: > Permission denied > Apr 6 16:57:27 xxx sshd[736]: [ID 800047 auth.error] error: > session_pty_req: session 0 alloc failed > > and then my login just hangs and I have to use ~. to kill the ssh > session. Any idea what is causing this? > > Drew > _______________________________________________ > 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 ed.alexander at renesas.com Wed Apr 20 09:27:10 2005 From: ed.alexander at renesas.com (Ed Alexander) Date: Wed, 20 Apr 2005 09:27:10 -0400 Subject: SUMMARY: Solaris 8 Recommended Patches Broke LPRng Message-ID: <426658AE.5080202@renesas.com> Thanks to Dave Foster for his advice. Found the problem to be caused by patch 109320-13. Removing that patch fixed the problem. Haven't experienced any problems listed in the patch description, so am happy to live without it. Ed -------- Original Message -------- I recently built a V-240 with Solaris 8 (2/04) and an E-420 with Solaris 8 (4/01). We're set up to print using LPRng (v 3.7.9) on an NFS-mounted drive. Printing worked fine on both systems prior to installing the latest recommended patch cluster for Solaris 8. After the patch cluster install, only root or members of the wheel group could print using LPRng. Non-privileged users could print using /bin/lp, but not LPRng. With debug logging turned on, we see "permission denied" when the systems attempts a net_send_file to the remote print spooler. How can I find which patch broke printing? A search on Sunsolve didn't give me any clues. Thanks in advance. Ed _____________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers -- Ed Alexander Senior Systems Administrator Renesas Technology America Design Engineering Center - East ed.alexander at renesas.com (919) 767-7793 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From j3ff9ack at yahoo.com Wed Apr 20 09:49:32 2005 From: j3ff9ack at yahoo.com (Two Dog Flats) Date: Wed, 20 Apr 2005 06:49:32 -0700 (PDT) Subject: SUMMARY: WARNING: interrupt level 6 not serviced Message-ID: <20050420134932.80370.qmail@web51304.mail.yahoo.com> Hello all, Well, I have only received two responses. Rich Bishop replied that he sees these error messages occasionally with his v100 servers, but they've not caused any problems. He opened a ticket with Sun when it first occured, and they didn't really offer any help (beyond the network interface/driver explanation). Gaziz Nugmanov stated that as far as he can remember that similar warnings showed up on his systems several years ago and after talking to sun it turned out they are harmless, something related to scsi devices. The only other data I can provide seems to indicate that it might be SCSI disk driver issue. About 70% of the message times coincide with cron jobs that start disk activity (tripwire, backups, etc.). For now, I will monitor the servers and see if the frequency of the messages increases. If anyone has more info, please send it along and I will be more than happy to post a revised summary. Original Message: I have seen this error in the messages file on multiple machines after deploying the Recommended Patch Cluster dated March 29, 2005. I did not see any of these during our testing before rolling out these patches to 20+ production servers. All of these servers are either Netra T1 or SunFire V120 servers, Solaris 8, 512 MB memory, 2-36 GB SCSI drives and either qfe or GigaSwift ce quad Ethernet cards. I searched the archives and Googled for the error with the result that: A. Interrupt 6 typically refers to a network driver issue (but not always), and B. If anyone else has seen this lately, they haven't figured it out or posted a summary. I will be more than happy to summarize if someone can point me in the right direction to find out more about this warning, or if I can ignore the occasional unserviced interrupt. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rickv at memec.com Wed Apr 20 15:08:44 2005 From: rickv at memec.com (Rick von Richter) Date: Wed, 20 Apr 2005 12:08:44 -0700 Subject: SUMMARY: Java Desktop gnome terminal is slow as molasses In-Reply-To: <425C45EE.9060300@memec.com> References: <425C45EE.9060300@memec.com> Message-ID: <4266A8BC.9080301@memec.com> The basic consensus is that I'm . It appears to be a trade off between speed (dtterm/xterm) and eye candy and features (gnome term). Sigh.... I want my cake and eat it too. Original Question: ================== OS: Solaris 10 HW: Ultra 60 I am using the Java Desktop System and the default terminal is the Gnome terminal. I like this because of the multiple tabs in one window and other features. But, it is an incredibly slow window. E.g. if I do an 'ls -lAR' while I'm in the /etc directory on my local workstation, it takes the gnome terminal a whopping 78 seconds to list around 6300 files/directories. But, if I start a dtterm window and run the same command, it takes less than 1 (one) second. Another annoying problem is this. If I am logged onto another server and I list a directory that has a lot of files (or even do the same command above) I cannot immediate break out of the listing. I.e. the list is scrolling by and I hit Ctrl-C, my cursor is not returned to me immediately. It looks as though there is a lot of buffering going on and my keystroke (Ctrl-C) does not take affect until the buffer runs out. Again, if I run this in a dtterm window my Ctrl-C is immediate. Are there some settings or tweaks to do to the gnome terminal to make it behave like the dtterm window? TIA This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be assured to be secure or correct as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rickv at memec.com Wed Apr 20 15:27:47 2005 From: rickv at memec.com (Rick von Richter) Date: Wed, 20 Apr 2005 12:27:47 -0700 Subject: SUMMARY: V880 install can't see E3510 LUNs Message-ID: <4266AD33.70807@memec.com> Thanks for all the responses. Answer: ======= First, I made a mistake and we are using Solaris 9 08/03 release not 12/03. IAE, the install CDs do not contain the correct drivers to see the 3510. You must use a jumpstart image that has the drivers installed. Plus, you need to do a some work on the fabric so that things don't step on each other and continually appear after reboots. Thanks to Matt Murdock for pointing me at Sun's site again. I did originally look but I tried again and it's amazing what you can find if you just change a few search words. Here's the link http://docs-pdf.sun.com/817-1244-11/817-1244-11.pdf Original question: ================== We are trying to install Solaris 9 12/03 onto a V880 with an attached E3510. The LUNs have been configured on the E3510 and a regular Sol9 boot from HD shows the LUNs using the 'format' command (see below). The luxadm command also sees the SES. When we try to do a 'boot cdrom' from the Sol9 disk 1 CD, the install program starts but it only sees the internal drives. We want to install onto the E3510 LUNs. Any ideas? Output from format ================== bash-2.05# format Searching for disks...done AVAILABLE DISK SELECTIONS: 0. c1t0d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w500000e0108e7cd1,0 1. c1t1d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w210000008717dc49,0 2. c1t2d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w500000e0108e81c1,0 3. c1t3d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w500000e0108e78a1,0 4. c1t4d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w2100000c505a19da,0 5. c1t5d0 /pci at 8,600000/SUNW,qlc at 2/fp at 0,0/ssd at w2100000c505a19bf,0 6. c4t40d0 /pci at 9,600000/SUNW,qlc at 2/fp at 0,0/ssd at w216000c0ff883b25,0 7. c4t40d1 /pci at 9,600000/SUNW,qlc at 2/fp at 0,0/ssd at w216000c0ff883b25,1 8. c4t40d2 /pci at 9,600000/SUNW,qlc at 2/fp at 0,0/ssd at w216000c0ff883b25,2 Specify disk (enter its number): 6 selecting c4t40d0 [disk formatted] format> ve Primary label contents: Volume name = < > ascii name = pcyl = 26624 ncyl = 26622 acyl = 2 nhead = 64 nsect = 32 Part Tag Flag Cylinders Size Blocks 0 root wm 0 - 127 128.00MB (128/0/0) 262144 1 swap wu 128 - 255 128.00MB (128/0/0) 262144 2 backup wu 0 - 26621 26.00GB (26622/0/0) 54521856 3 unassigned wm 0 0 (0/0/0) 0 4 unassigned wm 0 0 (0/0/0) 0 5 unassigned wm 0 0 (0/0/0) 0 6 usr wm 256 - 26621 25.75GB (26366/0/0) 53997568 7 unassigned wm 0 0 (0/0/0) 0 Output from luxadm ================== bash-2.05# luxadm display -v /dev/es/ses0 SUNWGS INT FCBPL DISK STATUS SLOT DISKS (Node WWN) 0 On (O.K.) 500000e0108e7cd0 1 On (O.K.) 200000008717dc49 2 On (O.K.) 500000e0108e81c0 3 On (O.K.) 500000e0108e78a0 4 On (O.K.) 2000000c505a19da 5 On (O.K.) 2000000c505a19bf 6 On (Login failed) 7 On (Login failed) 8 On (Login failed) 9 On (Login failed) 10 On (Login failed) 11 On (Login failed) SUBSYSTEM STATUS FW Revision:922A Box ID:0 Node WWN:5080020000170560 Enclosure Name:FCloop SSC100's - 0=Base Bkpln, 1=Base LoopB, 2=Exp Bkpln, 3=Exp LoopB SSC100 #0: O.K.(922A/ 8D3C) SSC100 #1: O.K.(922A/ 8D3C) SSC100 #2: Not Installed SSC100 #3: Not Installed Temperature Sensors - 0 Base, 1 Expansion 0:26:C 1Not Installed Backplanes - A=Base, B=Expansion A: O.K. B: Not Installed Default Language is USA English, ASCII This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be assured to be secure or correct as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From DirkM at mxgroup.co.za Thu Apr 21 08:35:04 2005 From: DirkM at mxgroup.co.za (Dirk Moolman) Date: Thu, 21 Apr 2005 14:35:04 +0200 Subject: SUMMARY: NFS mount options Message-ID: <71CE21D77D53D0439D8BC729FA25DD9B5BF53F@mxmail1.mxgroup.co.za> Wow, I received so many replies on this one. First of all, I made a mistake, and added the noac option to the server instead of the client. So the "noac" option is available, I just used it on the wrong server. Thanks to: Pablo Jejcic, Dave Mitchell, Tim Chipman, Mike Demarco, Martin Wheatley, Bernd Schemmer, Brandon Haberfeld Second mistake, I found articles on the web saying that noac will improve my NFS performance, which I am told now is not true. Disabling the attribute caching will actually make my performance worse instead of improving it. Thanks to: Ben Garrett, Tim Chipman, Martin Wheatley, Brandon Haberfeld I will have to do some more digging, to find out what exactly is causing my NFS mounts to be slow. Thank you very much for all the feedback Dirk [ORIGINAL MESSAGE] I am hoping that you can shed some light on this for me. I am experiencing bad NFS performance. I do not have much experience with this. My support engineers told me to set the "noac" option in the /etc/vfstab file, to improve my NFS performance (after they also did some searching on the web). I tried Google, I tried the sysadmin guides, and cannot find examples of how this is done. I tried it myself by just putting in in the vfstab file, but received an error doing this. Is this option supported in Solaris 9 ? The line in /etc/vfstab (after I changed it): /dev/vx/dsk/testdg/intest /dev/vx/rdsk/testdg/intest /intest ufs - yes nolargefiles,suid,noac The error I received: mount /intest mount: /dev/vx/dsk/testdg/intest on /intest - WARNING unknown options "noac" Any help will be appreciated Dirk Moolman Database and Unix Administrator HEALTHCORP _______________________________________________ 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 CMcAdam at liaison.com Thu Apr 21 10:21:43 2005 From: CMcAdam at liaison.com (Clive McAdam) Date: Thu, 21 Apr 2005 10:21:43 -0400 Subject: [SUMMARY] UPDATE: bge0: Tx stall detected Message-ID: Looks like there's a patch for this. http://sunsolve.sun.com/search/document.do?assetkey=1-21-113680-05-1&searchc lause=113680-05 Thanks all. -Clive ---Original question below------------------------------------------------------------- I forgot to mention that both the server and the port on the switch are strapped 100Full no auto negotiate ---snip--- #> ndd /dev/bge0 link_speed link_duplex adv_autoneg_cap 100 1 0 ---end snip--- Anyone know what could be causing these in the messages file? ---snip--- Apr 20 13:35:47 servername bge: [ID 801725 kern.info] NOTICE: bge0: Tx stall detected, watchdog code 0x142b5 Apr 20 13:35:48 servername bge: [ID 801725 kern.info] NOTICE: bge0: automatic recovery function activated ---end snip--- SunOS servername 5.8 Generic_108528-27 sun4u sparc SUNW,Sun-Fire-V240 - Clive _______________________________________________ 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 Karl.Rossing at Federated.CA Fri Apr 22 15:29:47 2005 From: Karl.Rossing at Federated.CA (Karl.Rossing at Federated.CA) Date: Fri, 22 Apr 2005 14:29:47 -0500 Subject: SUMMARY: 3rd party Sun Maintenance providers Message-ID: My original query, summary below > I'm interested in find out your opinions on 3rd party sun maintenance > providers such as carefactor. > I'm trying to gauge what services we would end up loosing compared to > our Silver Sun Spectrum contracts. This is what i learned 1) Make sure you know what your getting into. 2) Make sure external attached devices are included. Sun includes this, but 3rd parties sometimes don't 3) 3rd party maintenance comes down to the competence of the providers presence in your area 4) You will loose your premium sunsolve access to patches 5) I'd expect that you would loose your Solaris upgrade media kits eg: 8 to 10 6) Consider moving your older Sun hardware to 3rd parties. Keep the newer hardware on SUN Spectrum maintenance. 7) You can always do your own maintenance with an onsite parts box. Email replies are below. Thanks Karl ==== Tobias N I know nothing about carefactor but akibia late or incomplete delivery less know how lot of trouble in front of back-delivery (in the last days they need 3 tries to get an v880 back, little car, less people to carry, no transportation material) We use akibia only for Hardware maintenance to get parts we need ==== Ric A We use IBM where I work to provide Sun maintenance. They were much more willing to work with us and support non-Sun internal bits (like RAM and Disks). Sun said they would only support Sun branded parts. ==== ssefoen I have very good experience with Akibia. ==== David F SunSolve just changed their terms so that people not on Spectrum contracts have less access to SunSolve than before. ==== Matt N I currently have a Platinum level service account with Marathon Int. They come with my highest recommendation. I have worked with CareFactor and Terix before and out of the 3x of them i would recommend Marathon Int. Going threw a service issue with Marathon Int. Is like walking into a first class hotel. ==== Mike S You do loose the access to constant info, essentially most these guys hire ex sun guys, but things change, you do loose access to the latest fixes. Essentially I think a competent in house staff will do as good a job as a 3rd party staff. ==== Iain M One quick point - we moved a pile of kit (not our decision - bean counters did it) and we jumped in two footed, then found out that they didn't support any externally attached devices. So all our tapes, disks and SCSI arrays suddenly became un-supported!!!! Moral of the story - make sure that you compare like with like before signing a year long contract :-) Any short term monetary gains may come back and bite you in the future!!!! ===== John C The 3rd party support companies are often the same groups that Sun outsources its field support to anyways, but now you can save 30-70% over the Sun-branded support by using them directly. In many cases, the same tech still comes on-site, it's just that he grabs the Acme Support ID badge out of his glove box instead of his Sun Support ID badge before he enters your lobby. For some hosts (or all) you might consider dropping support altogether and self-insure by purchasing some extra hardware components from eBay or refurbished hardware vendors. You could pick up a few extra disks or I/O cards that are likely to fail, or depending on the age of the equipment, you may be able to duplicate entire hosts pretty cheap. Keep the hardware on-site and use as needed. As for selecting a specific provider, look very closely at the organization before signing up. Some are primarily PC-focused vendors who figure they can add the Sun product line, sell the support, and be able to source the parts (usually grey market) on an as-needed basis. Do you require expert Solaris and software support, or do you usually call support when you've narrowed it down to a hardware issue? Many third party providers don't have strong (or any) Solaris expertise on staff. Overall, you need to check 'em out, visit their call center, see their warehouse, talk to some of their engineers directly, ask for (and check!) references with similar environments to yours. I'm not trying to toss FUD around. There are some good, solid third party maintenance providers out there and I'm a big fan of cutting costs. You may consider splitting your servers between different plans. Sun has historically priced support on older hardware at penalizing levels to prod folks along the buy-new-hardware path. Consider moving your older hosts to third party and keeping the new hosts on Sun official. ==== Matt M The biggest thing I think would be access to the contract area of Sunsolve which includes in-depth articles that non-support people have, and automatic updates of the o.s./apps that you have under the spectrum support. For myself, I am comfortable to install warranty parts on my own, so a third party would not benefit me. When out of warranty/support, there are plenty of third parties who sell sun parts reasonable inexpensive. The only benefit, I could see by going to a third party would be the labor to install parts during and after warranty. ==== James C Karl - I would recommend larger outfits like Terix, IBM or Fujitsu over Carefactor. The larger outfits have coverage, SLAs and economies of scale (that translate into better pricing for you) at least as good if not better than Sun. We have a Terix relationship and I can certainly make an introduction of you want me too. ==== _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ryan at vaxpower.org Fri Apr 22 23:37:58 2005 From: ryan at vaxpower.org (Ryan Krenzischek) Date: Fri, 22 Apr 2005 23:37:58 -0400 (EDT) Subject: SUMMARY: Jumpstart Locale Errors... In-Reply-To: References: Message-ID: Sorry for the delayed summary. Thanks to Geoffrey Walton who pointed out: In my case, and hopefully yours, this error is displayed when the locale in my sysidcfg file does not match the locale in the flash archive. When this is true I see this error at the start of the JS flash build/restore, and I am asked to insert the Solaris language CD at the end of the install process. If the locale's match neither of these events happen. Regards, Ryan On Thu, 7 Apr 2005, Ryan Krenzischek wrote: > Hello Felow Sun Managers! > > I'm a bit baffled. I'm seeing the following during a routine and successful > jumpstart installation: > > Beginning Flash archive processing > > Predeployment processing > /usr/bin/uudecode: ISO8859-1 to 646 conversion: Invalid argument > 16 blocks > /usr/bin/uudecode: ISO8859-1 to 646 conversion: Invalid argument > 16 blocks > /usr/bin/uudecode: ISO8859-1 to 646 conversion: Invalid argument > 16 blocks > > No local customization defined > > > > Is this something that can be safely ignored? Any ideas on how to fix it? > > Regards, > > Ryan > _______________________________________________ > 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 sunmanagers at freehome.ch Sat Apr 23 10:15:15 2005 From: sunmanagers at freehome.ch (sunmanagers at freehome.ch) Date: Sat, 23 Apr 2005 16:15:15 +0200 Subject: SUMMARY Solaris 8 Jumpstart Server Problem Message-ID: <20050423141047.61C608555A@helios.comcept.ch> Hi all I could solve the problem...! Specially thangs to Darren Dunham he put's me to the right directions it means the below description is absolutly except following: - I reinstalled all stuff it means I deleted all an installed all from scratch again and did not patch the original image! (DO NOT DO THIS UNTIL YOU MUST IT MEANS IN MY CASE IT WAS THE REASON THE JUMPSTART SERVER WAS NOT ANYMORE FUNCTION) - I made a mistake in the description with the position 'name_service=NONE' in sysidcf it means I described NONE but made some entries. Only use 'name_service=NONE'! - The entry for the client in /etc/ethers could be done manuell or by the 'add_install_client' command with the parameter '-e'. I did several test and I suggest to use not the manuell methode. Use the parameter '-e' and additional if neccessary add in the file the FQDN. The parameter '-e' does only enter the hostname and in some cases FQDN is required. - Do not use the parameter '-s' for 'add_install_client' because this parameter is only used if the Boot Server is on a seperate server. In my case all is on one server. - In the script client.profile it is not possible to deselect Package from core cluster and also to add some packages makes sometimes trouble. I suggest to do this with the finish script etc. Again after I did all from the scratch except the above mentioned notes etc. it works from beginning. Now I'm testing around with some scripts etc. Again many thanks to Darren Dunham! Andrea Trying to setup a jumpstart server but after net boot - install no error and also nothing happens. I setup the Jumpstart server as following: Sun Ultra 10 400 MHz 256 RAM 9 GB Disk Hostname jumpi-serv IP 192.168.1.127 Shell ksh Cluster April 05 Client Sun Ultra 1 Hostname jumpi-client IP 192.168.1.118 MAC 8:0:20:86:53:70 Because of some modifications on the OS I report some additional information: # pkginfo system SUNWadmr System & Network Administration Root system SUNWarc Archive Libraries system SUNWarcx Archive Libraries (64-bit) system SUNWbcp SunOS 4.x Binary Compatibility system SUNWbtool CCS tools bundled with SunOS system SUNWbtoox CCS libraries bundled with SunOS (64-bit) system SUNWbzip The bzip compression utility system SUNWcar Core Architecture, (Root) system SUNWcarx Core Architecture, (Root) (64-bit) system SUNWcg6 GX (cg6) Device Driver system SUNWcg6x GX (cg6) Device Driver (64-bit) system SUNWcsd Core Solaris Devices system SUNWcsl Core Solaris, (Shared Libs) system SUNWcslx Core Solaris Libraries (64-bit) system SUNWcsr Core Solaris, (Root) system SUNWcsu Core Solaris, (Usr) system SUNWcsxu Core Solaris (Usr) (64-bit) system SUNWdfb Dumb Frame Buffer Device Drivers system SUNWeridx Sun RIO 10/100 Mb Ethernet Drivers (64-bit) system SUNWesu Extended System Utilities system SUNWgzip The GNU Zip (gzip) compression utility system SUNWhea SunOS Header Files system SUNWhmd SunSwift SBus Adapter Drivers system SUNWhmdx SunSwift SBus Adapter Drivers (64-bit) system SUNWipc Interprocess Communications system SUNWipcx Interprocess Communications (64-bit) system SUNWkey Keyboard configuration tables system SUNWkvm Core Architecture, (Kvm) system SUNWkvmx Core Architecture (Kvm) (64-bit) system SUNWlibC Sun Workshop Compilers Bundled libC system SUNWlibCx Sun WorkShop Bundled 64-bit libC system SUNWlibm Sun WorkShop Bundled libm system SUNWlibms Sun WorkShop Bundled shared libm system SUNWlmsx Sun WorkShop Bundled 64-bit shared libm system SUNWloc System Localization system SUNWlocx System Localization (64-bit) system SUNWnamos Northern America OS Support system SUNWowbcp OpenWindows binary compatibility system SUNWpd PCI Drivers system SUNWpdx PCI Drivers (64-bit) system SUNWpl5u Perl 5.005_03 system SUNWrmodu Realmode Modules, (Usr) system SUNWscbcp SPARCompilers Binary Compatibility Libraries system SUNWscpr Source Compatibility, (Root) system SUNWscpu Source Compatibility, (Usr) system SUNWsprot Solaris Bundled tools system SUNWsprox Sun WorkShop Bundled 64-bit make library system SUNWswmt Install and Patch Utilities system SUNWtoo Programming Tools system SUNWtoox Programming Tools (64-bit) system SUNWxcu4 XCU4 Utilities Activating some deactivated startscripts! # cd /etc/rc2.d # mv NOK28nfs.server K28nfs.server # cd /etc/rc2.d # mv NOS71rpc S71rpc # touch /etc/inet/inetd.conf # chown root:sys /etc/inet/inetd.conf # chmod 666 /etc/inet/inetd.conf Activating inetd.conf because initially files was deleted! # vi /etc/inet/inetd.conf --------------- /etc/inet/inetd.conf --------------- tftp dgram udp wait root /usr/sbin/in.tftpd in.tftpd -s /tftpboot --------------- /etc/inet/inetd.conf --------------- Generating tftpboot dir! # mkdir -m 711 /tftpboot # chown root:root /tftpboot # cd /etc # ln -s /etc/inet/inetd.conf ./inet/inetd.conf Modifying inetd deamon because line was initially commented out! # vi /etc/init.d/inetsvc --------------- /etc/init.d/inetsvc --------------- /usr/sbin/inetd -s & --------------- /etc/init.d/inetsvc --------------- # /etc/init.d/inetsvc stop # /etc/init.d/inetsvc start Generating dir's for jumpstart server! # mkdir /install_server # mkdir /install_server/Solaris_8 # chown -R root:root /install_server # chmod -R 755 /install_server Configuring nfs mount's! # vi /etc/dfs/dfstab --------------- /etc/dfs/dfstab--------------- share -F nfs -o ro,anon=0 /install_server/Solaris_8 --------------- /etc/dfs/dfstab--------------- # /etc/init.d/rpc start # /etc/init.d/nfs.server start # ps -ef | grep nfs root 193 1 0 10:59:58 ? 0:00 /usr/lib/nfs/nfsd -a 16 root 191 1 0 10:59:58 ? 0:00 /usr/lib/nfs/mountd root 29948 1 0 10:50:01 ? 0:00 /usr/lib/nfs/nfsd -a 16 # shareall # share - /install_server/Solaris_8 ro,anon=0 "" # rpcinfo -s | grep mountd 100005 3,2,1 ticots,ticotsord,tcp,ticlts,udp mountd superuser Copy sources to jumpstart dir! # mkdir /cdrom # mount -F hsfs -o ro /dev/dsk/c0t2d0s0 /cdrom # cd /cdrom/Solaris_8/Tools # ./setup_install_server /install_server/Solaris_8 Verifying target directory... Calculating the required disk space for the Solaris_8 product Calculating space required for the installation boot image Copying the CD image to disk... Copying Install Boot Image hierarchy... Install Server setup complete # umount /cdrom # mount -F hsfs -o ro /dev/dsk/c0t2d0s0 /cdrom # cd /cdrom/Solaris_8/Tools # ./add_to_install_server /install_server/Solaris_8 # umount /cdrom # mount -F hsfs -o ro /dev/dsk/c0t2d0s0 /cdrom # cd /cdrom/components/sparc/Packages/ # cp -Rp SFWgcmn /install_server/Solaris_8/Solaris_8/Product/ # cp -Rp SFWgcc /install_server/Solaris_8/Solaris_8/Product/ # cp -Rp SFWgm4 /install_server/Solaris_8/Solaris_8/Product/ # cp -Rp SFWaconf /install_server/Solaris_8/Solaris_8/Product/ # cp -Rp SFWamake /install_server/Solaris_8/Solaris_8/Product/ # umount /cdrom # ls /install_server/Solaris_8/Solaris_8/ Docs EA Misc Patches Product Tools # rm -rf /install_server/Solaris_8/Solaris_8/MU # rm -rf /install_server/Solaris_8/Solaris_8/Patches/* Patching the Boot image from the jumpstart server! # mkdir /install_server/8_recommended # mv 8_Recommended.zip /install_server/8_recommended # cd /install_server/8_recommended # unzip 8_Recommended.zip # cd /install_server/8_recommended/8_Recommended # /usr/sbin/patchadd \ -C /install_server/Solaris_8/Solaris_8/Tools/Boot \ -M /install_server/8_recommended/8_Recommended \ patch_order Generating dir for check script and copy the script to this dir! # cd /install_server/Solaris_8/Solaris_8 # cd Misc/jumpstart_sample # mkdir /install_server/jumpstart_8 # mkdir /install_server/jumpstart_8/bin # chown -R root:root /install_server/jumpstart_8 # chmod -R 755 /install_server/jumpstart_8 # cp check /install_server/jumpstart_8/bin # chmod 755 /install_server/jumpstart_8/bin/check # chown -R root:root /install_server/jumpstart_8/bin Generating sysidcfg script with some entries! # touch /install_server/Solaris_8/sysidcfg # chmod 400 /install_server/Solaris_8/sysidcfg # chown root:root /install_server/Solaris_8/sysidcfg # vi /install_server/Solaris_8/sysidcfg --------------- /install_server/Solaris_8/sysidcfg --------------- timezone=MET name_service=NONE {domain_name=comcept.ch name_server=192.168.100.125 name_server=192.168.100.194} terminal=vt100 timeserver=localhost network_interface=primary {netmask=255.255.255.0 protocol_ipv6=no default_route=192.168.100.120} security_policy=NONE root_password=7jh5a9XLICBj2 --------------- /install_server/Solaris_8/sysidcfg --------------- Generating the client.profile script with some entries! # vi /install_server/Solaris_8/client.profile --------------- /install_server/Solaris_8/client.profile --------------- # profile keywords profile values # ----------------- ----------------- # # This profile keyword is required in every profile # install_type initial_install system_type standalone geo C_Europe # Following initial Cluster's are available for installation # # SUNWCreq (Core System Support cluster) # SUNWCuser (End-User cluster), # SUNWCprog (Developer cluster) # SUNWCall (Every OS package) cluster SUNWCprog # Partitioning will be done explixit # partitioning explicit # How the Client will pe partioned # filesys any 512 swap filesys any free / # How the Client will pe partioned for a "SUN4.2G" # #filesys rootdisk.s0 2480 / #filesys rootdisk.s1 1048 /var #filesys rootdisk.s2 all overlap #filesys rootdisk.s3 512 swap #filesys rootdisk.s4 10 /metadb1 #filesys rootdisk.s5 free /metadb2 # How the Client will pe partioned for a "SUN2.1G" # #filesys rootdisk.s0 472 / #filesys rootdisk.s1 1048 /var #filesys rootdisk.s2 all overlap #filesys rootdisk.s3 512 swap #filesys rootdisk.s4 10 /metadb1 #filesys rootdisk.s5 free /metadb2 # Deinstall SUN packages based on 32bit # package SUNWsndmr delete package SUNWsndmu delete package SUNWpcmci delete package SUNWpcmcu delete package SUNWpcmem delete package SUNWpcser delete package SUNWpcelx delete package SUNWpsdpr delete package SUNWauda delete package SUNWaudd delete package SUNWftpr delete package SUNWftpu delete package SUNWged delete package SUNWnisr delete package SUNWnisu delete package SUNWwsr2 delete package SUNWatfsr delete package SUNWatfsu delete package SUNWced delete package SUNWdtcor delete package SUNWfcip delete package SUNWfcp delete package SUNWfctl delete package SUNWi15cs delete package SUNWi1cs delete package SUNWluxop delete package SUNWmdi delete package SUNWnamow delete package SUNWses delete package SUNWsolnm delete package SUNWssad delete package SUNWudf delete package SUNWudfr delete package SUNWxwdv delete package SUNWxwmod delete package SUNWm64 delete package SUNWqfed delete package SUNWusb delete package SUNWuaud delete # Deinstall SUN packages based on 64bit # package SUNWauddx delete package SUNWcedx delete package SUNWfcipx delete package SUNWfcpx delete package SUNWfctlx delete package SUNWluxdx delete package SUNWluxox delete package SUNWpcmcx delete package SUNWqfedx delete package SUNWssadx delete package SUNWudfrx delete package SUNWmdix delete package SUNWsesx delete package SUNWtleux delete package SUNWuaudx delete package SUNWxwdvx delete package SUNWxwmox delete package SUNWusbx delete package SUNWm64x delete # Install SUN packages based on 32bit and 64bit # package SUNWtoox add package SUNWtoo add package SUNWlibCx add package SUNWlibC add package SUNWbcp add package SUNWowbcp add package SUNWscbcp add package SUNWscpr add package SUNWscpu add package SUNWipcx add package SUNWipc add package SUNWarcx add package SUNWarc add package SUNWgzip add package SUNWlibm add package SUNWbtoox add package SUNWbtool add package SUNWhea add package SUNWsprox add package SUNWsprot add package SFWgcmn add package SFWgcc add package SFWgm4 add package SFWaconf add package SFWamake add package SUNWmdr add package SUNWmdu add package SUNWmdx add --------------- /install_server/Solaris_8/client.profile --------------- Copy default rules to jumpstart dir and modify the rules that it matches the Jumpi-client (Ultra 1 sun4u with IP 192.168.100.118)! # cp /install_server/Solaris_8/Solaris_8/Misc/jumpstart_sample/rules \ /install_server/Solaris_8/rules # vi /install_server/Solaris_8/rules --------------- /install_server/Solaris_8/rules --------------- ############################################################################ # # RULE EXAMPLES # # The following rule matches only one system: # hostname jumpi-client.comcept.ch - client.profile - # The following rule matches any system that is on the 924.222.43.0 network # and has the sun4c kernel architecture: # Note: The backslash (\) is used to continue the rule to a new line. network 192.168.100.0 && karch sun4u - client.profile - # The following rule matches any sparc system with a c0t3d0 disk that is # between 400 to 600 MBytes and has Solaris 2.1 installed on it: #arch sparc && \ # disksize c0t3d0 400-600 && \ # installed c0t3d0s0 solaris_2.1 - upgrade - # # The following rule matches all x86 systems: #arch i386 x86-begin x86-class - # # The following rule matches any system: any - - client.profile - --------------- /install_server/Solaris_8/rules --------------- Verifying if rules are ok and generating rules.ok file! # cd /install_server/Solaris_8 # /install_server/jumpstart_8/bin/check \ -r rules \ -p /install_server/Solaris_8 Validating rules... Validating profile client.profile... The custom JumpStart configuration is ok. # /install_server/jumpstart_8/bin/check \ -p /install_server/Solaris_8 Validating rules... Validating profile client.profile... The custom JumpStart configuration is ok. # ls -la rules.* -rw-r--r-- 1 root other 157 Apr 22 18:56 rules.ok Make the entry for the jumpi-client in /etc/ethers and /etc/hosts! # vi /etc/ethers --------------- /etc/hosts --------------- 8:0:20:86:53:70 jumpi-client.comcept.ch --------------- /etc/hosts --------------- # vi /etc/hosts --------------- /etc/hosts --------------- 192.168.1.118 jumpi-client jumpi-client.comcept.ch --------------- /etc/hosts --------------- Use add_install_client script to add the client and generating /etc/bootparams as the Files in /tftpboot! # cd /install_server/Solaris_8/Solaris_8/Tools # /install_server/Solaris_8/Solaris_8/Tools/add_install_client \ -e 8:0:20:86:53:70 \ -i 192.168.1.118 \ -s jumpi-serv:/install_server/Solaris_8 \ -c jumpi-serv:/install_server/Solaris_8 \ -p jumpi-serv:/install_server/Solaris_8 \ jumpi-client.comcept.ch \ sun4u starting rarpd starting bootparamd starting nfsd's starting nfs mountd updating /etc/bootparams copying inetboot to /tftpboot # ps -ef | grep nfs root 20559 1 0 16:48:04 ? 0:00 /usr/lib/nfs/mountd root 20561 1 0 16:48:04 ? 0:00 /usr/lib/nfs/nfsd -a 16 # ps -ef | grep /usr/sbin/ root 20564 1 0 16:48:04 ? 0:00 /usr/sbin/in.rarpd -a root 20532 1 0 16:47:56 ? 0:00 /usr/sbin/rpcbind root 20566 1 0 16:48:04 ? 0:00 /usr/sbin/rpc.bootparamd root 20527 1 0 16:47:49 ? 0:00 /usr/sbin/nscd # ps -ef | grep inet root 26210 1 0 19:00:50 ? 0:00 /usr/sbin/inetd -s # cat /etc/bootparams --------------- /etc/hosts --------------- jumpi-client root=jumpi-serv:/install_server/Solaris_8/Solaris_8/Tools/Boot install=jumpi-serv:/install_server/Solaris_8 boottype=:in sysid_config=jumpi-serv:/install_server/Solaris_8 install_config=jumpi-serv:/install_server/Solaris_8 rootopts=:rsize=32768 --------------- /etc/hosts --------------- # ls -la /tftpboot/ total 346 drwx--x--x 2 root other 512 Apr 22 16:49 . drwxr-xr-x 21 root root 512 Apr 21 22:08 .. lrwxrwxrwx 1 root other 26 Apr 22 16:49 C0A86476 -> inetboot.SUN4U.Solaris_8-1 lrwxrwxrwx 1 root other 26 Apr 22 16:49 C0A86476.SUN4U -> inetboot.SUN4U.Solaris_8-1 -rwxr-xr-x 1 root other 159424 Apr 22 16:49 inetboot.SUN4U.Solaris_8-1 -rw-r--r-- 1 root other 317 Apr 22 16:49 rm.192.168.100.118 Goin to the client and connecting with serial console to ok prompt! Giving following Command! (both jumpi-server and jumpi-client are connected to the same switch) # boot net - install Rebooting with command: boot net - install Boot device: /sbus/ledma at e,8400010/le at e,8c00000 File and args: - install On the jumpstart server I see with snoop that there are some Broadcast requests from Jumpi-client to the jumpstart server (tftp read C0A86476 (octet) etc.) but there is Nothing going to the jumpi-client. Also on the client there is no other messages as The Boot device message. I tried also to use easier configuration as mentioned here it means for sysidcf and For client.profile. I controlled all logs local and on the log server no messages no Errors. In my eyes it looks all fine also with the files etc. but why it is not function? Please help......many thanks in advanced for your help! Andrea _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ahoesch at smartsoft.de Mon Apr 25 06:47:11 2005 From: ahoesch at smartsoft.de (=?ISO-8859-1?Q?Andreas_H=F6schler?=) Date: Mon, 25 Apr 2005 12:47:11 +0200 Subject: Summary: cron jobs no longer work Message-ID: <60E21FCC-B577-11D9-A0E8-003065CCA582@smartsoft.de> Dear managers, thanks to Bradley Alan "Lennon, Padraig" Jacint JUHASZ They suggested to do /etc/rc2.d/S75cron stop /etc/rc2.d/S75cron start and then do tail -f /var/cron/log to check what it is doing. My first impression was (no luck) but after repeating the stop/start cycle and looking closer I realized that the thing is ticking again. Thanks a lot! Regards, Andreas > Dear managers, > > we run out of disk space on one of our servers. After removing a big > file, things went to normal again, except the root cron jobs (crontab > -e), that are no longer executed. Do I need to restart a process that > might have died? I am stuck here. Where should I start to look for the > reason? > > Thanks a lot! > > Regards, > > Andreas > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > > *********************************************************************** > ************ > OMAM is a licensed Financial Services Provider > This communication is subject to specific restrictions and disclaimers. > Details are contained at the following link: > http://www.omam.co.za/docs/docredir.asp?docid=3460 > > *********************************************************************** > ************ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From solarislist at gmail.com Mon Apr 25 06:46:13 2005 From: solarislist at gmail.com (Solaris List) Date: Mon, 25 Apr 2005 13:46:13 +0300 Subject: SUMMARY: OS upgrade with VxVM Message-ID: Hello Managers; First of all, I am sorry for late reply. I finally succeed to upgrade Solaris with Veritas VxVM. Thanks all who replied. In general, All suggestios are composed in two different ways. - Unmount all staff about Vx , then reboot system "-s" and run ./upgrade_start ( in veritas cd ) then upgrade , then run upgrade_finish script - Unmount all staff about Vx, deport datadg(s) ( try import to understand if you will be able to import after upgrade or not ), protect the one of the rootdg mirrors for backup, remove VxVM package and patch about Veritas, then upgrade Solaris , then install VxVM packages ( install with pkgadd and select update ) and import datadg(s) I used 2nd way to upgrade. Finally, I finished my work regards _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sunman at lists.gaziz.ca Fri Apr 22 18:05:52 2005 From: sunman at lists.gaziz.ca (Gaziz Nugmanov) Date: Fri, 22 Apr 2005 18:05:52 -0400 Subject: SUMMARY and apology Re: Some bash/tty questions Message-ID: <367427380.20050422180552@lists.gaziz.ca> Hi, There is no general consensus on subj. Some people tend to create complex login scripts while others think that at 2 am scripts might fail and you end up troubleshooting environment issues so things should be kept very simple and generic. Another good point is that per FAQ this question shouldn't be sent to this list at all. I sincerely apologize. Here is the list of replies: ============= "Paveza, Gary" If you don't allow direct login to root, but rather su to root, then so something like this in your profile: USER=`who am i | awk '{print $1}'` Case ${USER} In User1) PS1="whatever";; User2) PS1="whatever";; Esac We do that here. Another admin here likes PS1 to show current directory, while I prefer just #. Set your HISTFILE=filename.${USER} so everyone has a unique history file. That also allows for limitd auditing as you can see what someone did. --------------------------------------------------------- Gary Paveza, Jr. Senior Systems Administrator -CSA (302) 252-xxxx - phone ================== Coy Hile Leave it at the defaults. I would argue that one should not modify root's environment to make it "friendly"; that leads to one being too comfortable as root and therefore doing too much as root unnecessarily. -- Coy Hile ================== Ric Anderson We do the following in linux's /root/.bash_profile and Solaris /.profile # # See if user specific code is to be run. Be very cautious of # system run state and make sure we are connected to a terminal. ID=/usr/bin/id SED=/bin/sed TTY=/usr/bin/tty WHO=/usr/bin/who ${TTY} -s ttyflag=$? if [ ${ttyflag} -eq 0 -a -x ${WHO} -a -x ${SED} ]; then rl=`${WHO} -r | ${SED} -e 's/^.*run-level[ ]*\([0-9]\).*$/\1/'` # Slackware uses 4 for startup with X11, RH uses 5 for same thing, # so remap those to rl=3 for our purposes. test \( ${rl} -eq 4 -o ${rl} -eq 5 \) && rl=3 case $rl in 2|3) if [ `${ID} | ${SED} -e 's/ .*$//'` = "uid=0(root)" ]; then ans="nopossiblewayjOsE" while [ ! -f "/private/rootrc/$ans" -a -n "$ans" ]; do echo -e "Enter your login name and press RETURN or just press RETURN:\c" read ans done test -n "$ans" -a -f "/private/rootrc/$ans" && \ . "/private/rootrc/$ans" fi ;; esac fi This lets a user press "enter" to get root's default setup, or by entering a file name (typically the user's login name) to execute a specific setup. As an example, my startup (/private/rootrc/ric) contains # This is an sh sourceable script that sets up whatever options # you want to have when running "su - root". It can exec your shell # as well. ENV=/home/ric/.kshrc export ENV I_AM_ROOT=y export I_AM_ROOT unset LD_LIBRARY_PATH # bastion of all evil EXINIT='set number|map g G|set redraw' export EXINIT export DISPLAY test -n "${OPENWINHOME}" && export OPENWINHOME test -x /usr/bin/ksh && exec /usr/bin/ksh -o vi test -x /bin/ksh && exec /bin/ksh -o vi echo >&2 "$0: could not find ksh - continuing current shell". #-end of /private/rootrc/ric. Cheers, Ric Anderson ===================== Rich Kulawiec On Thu, Apr 21, 2005 at 03:21:32PM -0400, Gaziz Nugmanov wrote: > These are not Solaris specific questions Then they don't belong on this list -- per the FAQ. When can I expect to see your public apology to everyone on the list for your deliberately rude behavior? ---Rsk ===================== Mats Oberg Hi, not to bash down on bash, but perhaps you should try zsh, it has the shared history thing built in. / Mats ==================== francisco and original question. > These are not Solaris specific questions but I hope SUMMARY will be > helpful for most of admins. > > 1/ There are several admins in our environment and a > question of bash environment (PS1, HISTSIZE, aliases, etc) sometime comes up. > > What would be the best practise to keep shared env variables > for root or other shared account to be generic enough but still custom to each user? May best idea > is to "resource" custom user's scripts and enjoy. Don't. As a policy, root should have a consistent environment, otherwise you end up troubleshooting wierd problems with why scripts work for you but not for Joe, or why things aren't working the way you expect when you get in at 2am and Jane got there first, logging in to console and then su'ing to root. But if you really need to, use 'su root' not 'su - root' and source $HOME/.bash_profile if it exists. > What is your preferred root PS1 variable on production servers? `# ` or at most `hostname# ` I like it simple b/c it needs to work even on hosed systems. > 2/ I am using "screen" multiplexor (sooooo happy > about this) and sometime I want to repeat the commands I just typed in > another window. > > Is there a way to sync bash history every let's say every 30 secs > between all sessions of the same user? Good question, but you may want to reconsider if that would really be a good idea to do always. I don't know an easy way, but you could put together something with the `history` builtin bash command to sync the current history with a file (which would pressumably also be written to by other shells). I don't think that would work as a background job, maybe you'd have to execute it as part of your prompt (ugly). > 3/ What is the best way to keep history of all typed commands and all > outputs to consoles/psedo terms? I liked the "screen" way of logging > but not all users want to learn new app (tons of CTRL-A shortcuts+). Try using script. It will keep all input/output in a generic file, or one you specify. No need to learn new ctrl-a things either. iirc, GNU script even comes with a timer, so you can record the time between input/ouput and then play it back, for something like demo purposes. -f http://www.blackant.net/ =================== _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From benoit.audet at carra.gouv.qc.ca Mon Apr 25 14:30:45 2005 From: benoit.audet at carra.gouv.qc.ca (=?ISO-8859-1?Q?Beno=EEt=20Audet?=) Date: Mon, 25 Apr 2005 14:30:45 -0400 Subject: SUMMARY: Oracle problem after a restore Message-ID: First of all, I would like to thank all of those who responded to my demand for help. Especially, Tommy Falsen David B Harrington, Bernd Schemmer ans Scott Mickey, who pointed me to interesting directions. Surprisingly, just before noon, I finally got from one of the DBAs that they did a change last week to the server: they added 3 more Oracle instances. The server asn't been rebootet before this time, of course, and everything went well until yesterday. We did a full restore, via NetBackup, on new disks, changing *nothing else than the swap partition*, which is now twice the size it was before the move. Looking closely (two more "sunmanagers"'s mails about swap from me), I finally saw that Oracle instances was taking much more swap than it was before the restore. I just added another swap partition *just to see* and TADAAAM! everything came up flawlessly! Strange thing I suspect is 1) Oracle would need a serious fine-tuning effort, and 2) it's strange that the server was running without problem since the DBAs added 3 more instances, but they did add 3 more instances on the server *without warn us* they did so! Well, that's almost it. Thanks again to all who replied! Ben Audet My original post was: Fellow managers, I know that this is probably more an Oracle related problem and that here isn't an Oracle mailing list, but since we don't know where to look, I try here anyway. We have a Sun Fire 280R with a StorEdge S1, runnung both Oracle 8.1.6 and 8.1.7.4. We had to perform a full restore on the server yesterday. Despite the facts that we had a "good" full backup, the restore has been applied successfully, and that the server came up very well after the operation, Oracle don't work properly. We have this message in the listener log file: TNS-12500: TNS:listener failed to start a dedicated server process TNS-12540: TNS:internal limit restriction exceeded TNS-12560: TNS:protocol adapter error TNS-00510: Internal limit restriction exceeded Solaris Error: 12: Not enough space Everywhere I look for an explanation, it points to the fact that the server is "missing" memory and/or swap. But the settings are the same as "before" we had to restore the server, except that the swap partition is twice the size (we replaced internal disks with bigger ones, so we had room for more swap). The actual server have 1 Gb. of RAM, plus 4 Go. of swap: oracledev# swap -l swapfile dev swaplo blocks free /dev/md/dsk/d2 85,2 16 8389632 8389632 oracledev# swap -s total: 597896k bytes allocated + 4140144k reserved = 4738040k used, 209280k available Here are the "/etc/system" values related to Oracle: set shmsys:shminfo_shmmax=4294967295 set shmsys:shminfo_shmmin=1 set shmsys:shminfo_shmmni=100 set shmsys:shminfo_shmseg=10 set semsys:seminfo_semmap=64 set semsys:seminfo_semmns=4096 set semsys:seminfo_semmni=4096 set semsys:seminfo_semmnu=4096 set semsys:seminfo_semume=64 set semsys:seminfo_semmsl=250 set semsys:seminfo_semopm=100 set semsys:seminfo_semvmx=32767 set rlim_fd_max=2048 set rlim_fd_cur=4096 Well, this is almost it. Unfortunately, the Oracle knowledge we have here is a little limited... ...so that's why I ask if some of you managers would have a clue on what could help me to fix the problem. Thanks in advance and, of course, will summarise. Ben Audet _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From foster at ncmir.ucsd.edu Mon Apr 25 19:39:45 2005 From: foster at ncmir.ucsd.edu (David Foster) Date: Mon, 25 Apr 2005 16:39:45 -0700 (PDT) Subject: SUMMARY: StarOffice 7.0 patch 116519-07 broke my installation Message-ID: <200504252339.j3PNdjbe012965@dim.ucsd.edu> SunOS <> 5.8 Generic_117000-03 sun4u sparc SUNW,Ultra-80 Solaris SunOS <> 5.9 Generic_118558-02 sun4u sparc SUNW,Ultra-60 Solaris Just a heads up that StarOffice 7.0 patch 116519-07 completely broke my StarOffice. Upon installation of the patch got many complaints about not being able to register components. After installation, upon startup StarOffice complained about missing components. Tried to run ./program/setup but it complained about missing program files. Anyone else install this successfully? Installation entails changing to patch directory and typing: umask 022 ./so-7-pp4-02-bin-solsparc.bin -patch Dave Foster =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= David Foster National Center for Microscopy and Imaging Research IT Manager, Programmer University of California, San Diego dfoster[at]ucsd[dot]edu Department of CRBS, Mail 0608 (858) 534-7968 http://ncmir.ucsd.edu/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable." -- George Bernard Shaw ------------- End Forwarded Message ------------- << All opinions expressed are mine, not the University's >> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= David Foster National Center for Microscopy and Imaging Research IT Manager, Programmer University of California, San Diego dfoster[at]ucsd[dot]edu Department of CRBS, Mail 0608 (858) 534-7968 http://ncmir.ucsd.edu/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable." -- George Bernard Shaw _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From G.Bakalarski at icm.edu.pl Tue Apr 26 07:27:55 2005 From: G.Bakalarski at icm.edu.pl (Grzegorz Bakalarski) Date: Tue, 26 Apr 2005 13:27:55 +0200 Subject: SUMMARY: su - ; \rm -rf * Message-ID: <20050426112755.GF20079@delta.icm.edu.pl> Dear All, Recovery finished more or less successfully. Time for summary ... I've got a lot of replies ... Too many to fully specify auhtors. At the end of this e-mail you'll find votes statistics. My short summary/commentary follows... There is no real firm way to prevent such disaster i.e. a way to block accidental recursive removing of whole filesystem by root. However Solaris10's rm is to forbid to remove recursively from / (i.e. /bin/rm -rf / will be refused). Most of you suggested aliasing rm to "rm -i". Many adviced to use sudo or RBAC ... Some agreed that creating file named like "-a or -i or -z" is safe and could be a good trick which may work in limited situations. Changing root home to /root is common practice but will not prevent against stupid admin. Someone suggested to write a kind of wrapper for rm (i.e. change rm to what_ever_you_like e.g. put rm in /usr/verysecure/bin/rm and in /usr/bin put a script called rm which would check for id=root if any of arguments is / or * or /usr or /var or ... and failing if it is true; in other case just execute /usr/verysecure/bin/rm). it looks clever but I'm not sure if is is secure and healthy to OS. Few of you wisely suggested that training and learning good habits is the only way to avoid such disaster, e.g admin should never use rm & *, admin before any rm should 3 times check pwd etc.. My additional note is that when you do backups on disk it should be umounted or mounted read_only after doing backup ... My note #2, for those who suggested cutting fingers, hands etc. You'll never knows what happens with you in a future ... Sometimes im maybe a white space between / and tmp/scratch (i.e "rm -r / tmp/scratch" instead of "rm -r /tmp/scratch"), other time you may mismatch controller numer of device when doing newfs ... We'll ask then about your fingers. Nevertheless: before rm always check pwd, never work in a hurry, never work very tired, good backup is your friend. Thanks again & best regards, GB ------------ In addition to my own comments I put a notes from Colin Bigam who gave most comprehensive reply. Thanks! From: Colin Bigam [...] There are several things wrong with your initial example. "su -" should (nearly) always be used instead of "su". \rm escapes any aliasing for the rm command, but doesn't specify where it is. Specify the full pathname for dangerous binaries, i.e. /bin/rm Also, your friend apparently didn't check their current path before issuing the command. I type 'pwd' more times than I can count, when I'm logged in as (or su - to) root. ALWAYS know what directory you're in when issuing commands that might modify files. This wouldn't matter if you replaced the * with a full pathname again, and not using 'tilde expansion.' So... #su - #pwd #/bin/rm -rf /home/username Or possibly; #su - #cd /home/username #pwd #/bin/rm -rf ./* Changing root's home directory can have some repurcussions (vipw fails), and is a debatable subject. Adding the '-z' file to each directory you want to save won't help much either. If you do "/bin/rm -rf *" in /, it won't recognise the -z in any subdirectories, such as /var or /usr. Also, it's dangerous to rely on "tricks" of any sort when you're playing with root. The only real trick is training. As root, use full pathnames, don't combine commands, don't use aliases, and don't set up conveniences that are going to cause harm when you go to work on another system that doesn't have them. Good luck with the restore. ------------------- ------------------------------------- ------------------- VOTES ------------------------------------- ------------------ ---------------------------------- forbid root access (or in other words "Don't let idiots have the root password.") +++ Use sudo or RBAC instead of su: ++++++ upgrade to Solaris 10 Solaris 10 has a built-in protection against this calamity: # rm -fr / rm of / is not allowed + Learn careful work + change root's home to /root ++++ allow root logins only on console + alias rm to /bin/rm -i +++++++ never issue "rm -rf * , use rm -rf absolute_path +++ train using "-f" only then it is really needed + cut out fingers anyone using " \rm -rf * " +++ alias su + create wrapper for the rm command which can just grep for a *, /* or whatever you want and deny the command + ------------------------------------------------------------------ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From debbie at icus.com Wed Apr 27 13:28:49 2005 From: debbie at icus.com (Debbie Tropiano) Date: Wed, 27 Apr 2005 12:28:49 -0500 (CDT) Subject: SUMMARY: Windows Storage Server 2003 serving NFS to Linux/Unix hosts Message-ID: <200504271728.j3RHSnU6010925@enterprise.icus.com> Hello - While there were a few positive responses, most people said that it was more trouble than it was worth, and that neither Dell nor MS were of much help. Since we already have NFS issues with Linux, Solaris and MAC systems sharing a network, I don't want to add any more, so plan to avoid Windows Storage Server 2003 as well as SFU (MS's Services for Unix). While I gather that SFU 3.0 is better than the previous version, it apparently still has problems (and I'm looking for something to make my life easier :-). Also, Craig Warnol suggested getting W2K3 Server rather than WSS2k3 since the only interface for WSS2k3 configuration is via a web interface. If we were going with Windows at all, that would be helpful to know. Thanks for all of the replies from Chris Ricker, John Madden, Craig Warnol, Ric Anderson, Mark Gosselin, Will H. Backman, Rami Aubourg-Kaires, Jerry B. Altzman, Erald Troja, Steve Sandau and Bob Finch (some of whom may be from the linuxmanagers list since I posted this message there as well). Debbie PS Will H. Backman sent me the following URL, but I'm wondering if "Best" just means that there's nothing much else out there. http://www.cioupdate.com/news/article.php/1576081 Original message: > We're considering acquiring a NAS device and the one my cohorts are > looking at runs Windows Storage Server 2003. It claims to serve NFS > and play happily in the Linux/Unix environment, but having been burnt > before I don't fully trust the marketing speak (especially when it > comes from Microsoft :-). > > So before we ask them to send us a system for evaluation, I'd like to see > what the "real" experts have to say. Does Windows Storage Server 2003 > *really* play nicely in a mostly Linux/Unix environment? Are there > difficulties to setting it up or running it that a Linux/Unix admin would > want to know beforehand? > > Any information would be appreciated, > Debbie > > PS The one we're looking at is the PV745 from Dell. Other projects have > this one, but don't run it with Linux/Unix. -- + 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 martini at raider.llnl.gov Wed Apr 27 14:01:42 2005 From: martini at raider.llnl.gov (Dave Martini 1) Date: Wed, 27 Apr 2005 11:01:42 -0700 (PDT) Subject: SUMMARY: rpcinfo -p output Message-ID: <200504271801.j3RI1gUY000668@raider.llnl.gov> The solution was to modify the services line in my /etc/nsswitch.conf file to look like this services: files nisplus [NOTFOUND=return] and then to restart nfs.client and now I see nlockmgr on udp as well as tcp. Dave. Below is my original quesiton I have a solaris 9 machine that shows the nlockmgr running on tcp when I do a #rpcinfo -p 100021 1 tcp 4045 nlockmgr 100021 2 tcp 4045 nlockmgr 100021 3 tcp 4045 nlockmgr 100021 4 tcp 4045 nlockmgr I have a different machine running solaris 2.7 that shows nlockmgr running on tcp AND udp. 100021 1 udp 4045 nlockmgr 100021 2 udp 4045 nlockmgr 100021 3 udp 4045 nlockmgr 100021 4 udp 4045 nlockmgr 100021 1 tcp 4045 nlockmgr 100021 2 tcp 4045 nlockmgr 100021 3 tcp 4045 nlockmgr 100021 4 tcp 4045 nlockmgr My question is how do I get the solaris 9 machine to run the nlockmgr on udp as well as tcp?? Thanks much! Dave Martini LLNL _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From carolyn at cs.usna.edu Wed Apr 27 14:07:37 2005 From: carolyn at cs.usna.edu (Carolyn Mayr {ITSD/CS}) Date: Wed, 27 Apr 2005 14:07:37 -0400 (EDT) Subject: SUMMARY: CDE login hangs on Solaris 10 hosts Message-ID: <200504271807.j3RI7bc3002986@rattler.cs.usna.edu> Sorry for the delay in the summary but I wanted to hear back from Sun tech supports escalation group. Thanks to the following sunmanagers sysadmins for their troubleshooting assistance: Murdock, Matt Alan Crosby Rick von Richter Steve Sandau Answer: It's a Tooltalk bug. This was first noticed in Solaris 9 but was not fixed in Solaris 10. By disabling IPv6 in Solaris 10, the CDE login problem was resolved. This bug affects CDE, dtmail, and anything else that requires Tooltalk. Refer to Document ID 6206423 (viewable for Sun Spectrum customers only). Sun is working on a Solaris 10 patch. My advice would be not to install IPv6 during the Solaris 10 OS installation until the patch is released. Carolyn Mayr =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Carolyn Mayr (IT Specialist/UNIX sysadm) Voice: 410-293-6808 (sec-6800) Computer Science Department, DivMath&Sci Email: carolyn at cs.usna.edu 572M Holloway Road, Michelson Hall, 9F FAX: 410-293-2686 U.S. Naval Academy WWW: http://www.cs.usna.edu Annapolis, MD 21402-5002 USNA: 410-293-1000 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ------------- Begin Forwarded Message ------------- Date: Mon, 25 Apr 2005 09:45:38 -0400 (EDT) From: Carolyn Mayr {ITSD/CS} Subject: CDE login hangs on Solaris 10 hosts To: sunmanagers at sunmanagers.org Cc: carolyn at cs.usna.edu I just upgraded 2 hosts named "slink" and "tangier" from Solaris 9 to Solaris 10. My home directory physically resides on tangier and my login name is "carolyn". Prior to the Solaris 10 upgrade I tarred my entire "carolyn" home directory from tangier to another machine, did a clean Solaris 10 install then tarred my "carolyn" directory back to tangier. Now I can't log in as "carolyn" into the slink CDE login screen. It just hangs. I tried removing my .dt, .dtlogin, .cshrc, .profile and that didn't work. I have no trouble logging into slink remotely via ssh. I have no trouble logging in via the CDE as carolyn on any other hosts. I have no trouble logging in via the slink CDE as any other user. I upgraded a 3rd machine to Solaris 10 and am having the same trouble logging into it as carolyn via the CDE. I originally saw a message on slink "/usr/dt/bin/ttsession Error: rpc.tdbserverd on tangier is not running "/usr/dt/bin/ttsession [2112]:[ID908584 daemon.error]_Tt_db_client::connectToDb():fcntl(F_SETFD): Bad file number" even though tangier's ttdb server was running. Per Sun tech support, I tried cleaning the Tooltalk database per SRDB 12729 on slink and tangier but still no go. I no longer see the ttdb error message or any other error messages and the .dt/startlog looks fine: --- Mon Apr 25 09:26:06 EDT 2005 --- /usr/dt/bin/Xsession starting... --- starting /usr/dt/bin/dthello & --- starting /usr/openwin/bin/speckeysd --- Xsession started by dtlogin --- starting /usr/dt/bin/dtsession_res -load -system --- sourcing //.dtprofile... --- sourcing /usr/dt/config/Xsession.d/0010.dtpaths... --- sourcing /usr/dt/config/Xsession.d/0015.sun.env... --- sourcing /usr/dt/config/Xsession.d/0020.dtims... --- sourcing /usr/dt/config/Xsession.d/0030.dttmpdir... --- sourcing /usr/dt/config/Xsession.d/0040.xmbind... --- starting /usr/dt/bin/dtsearchpath --- starting /usr/dt/bin/dtappgather & --- starting /usr/dt/bin/dsdm & --- session log file is /dev/null --- DTSOURCEPROFILE is 'true' (see //.dtprofile) --- execing /usr/dt/bin/dtsession with a /sbin/sh login shell ... --- starting desktop on /dev/pts/3 /usr/dt/bin/ttsession[21421]: starting So it seems it is something related to my home directory even though it was tarred from the original carolyn directory. What could possibly be the problem? =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Carolyn Mayr (IT Specialist/UNIX sysadm) Voice: 410-293-6808 (sec-6800) Computer Science Department, DivMath&Sci Email: carolyn at cs.usna.edu 572M Holloway Road, Michelson Hall, 9F FAX: 410-293-2686 U.S. Naval Academy WWW: http://www.cs.usna.edu Annapolis, MD 21402-5002 USNA: 410-293-1000 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From janoosch at freenet.de Thu Apr 28 06:08:33 2005 From: janoosch at freenet.de (Alex Duile) Date: Thu, 28 Apr 2005 12:08:33 +0200 Subject: SUMMARY: Unused diskspace is unavailable on mirrored rootdisk with Veritas Message-ID: A tad late but here is my promised summary about the problem i had... Thanks to the following guys who tried to help me: Jeff Woolsey Murdock, Matt Bhaskara, Srikalyan Marble, Jeff R Darren Dunham Bernd Schemmer After Veritas takes Controll about the harddisk my unassigned Slice 7 (about 35 GB) was not longer visible for later configurations. Here is what i have done: With # vxprint -ht i could discover that my Slice7 now named rootdisk7vol (That was the Info I had overread ;-)) The next Steps were: Stop the Veritas Volume: # vxvol stop rootdisk7vol Erase it: # vxassist remove volume rootdisk7vol After erasing it i could configure my two new volumes: # vxassist -b -g rootdg make test1 15g layout=mirror-concat # vxassist -b -g rootdg make test2 15g layout=mirror-concat At first i thougt i had to create the volumes without the mirror-layout because my rootdg was already mirrored. But then it creates only an unmirrored volume... this showed me the (mirrored) results: # vxprint -ht: v test1 - ENABLED SYNC 31457280 SELECT - fsgen pl test1-01 test1 ENABLED ACTIVE 31464192 CONCAT - RW sd rootdisk-03 test1-01 rootdisk 69644544 31464192 0 c1t0d0 ENA pl test1-02 test1 ENABLED ACTIVE 31464192 CONCAT - RW sd rootmirr-01 test1-02 rootmirr 0 31464192 0 c1t1d0 ENA v test2 - ENABLED SYNC 31457280 SELECT - fsgen pl test2-01 ENABLED ACTIVE 31464192 CONCAT - RW sd rootdisk-05 test2-01 rootdisk 101108736 31464192 0 c1t0d0 ENA pl test2-02 test2 ENABLED ACTIVE 31464192 CONCAT - RW sd rootmirr-05 test2-02 rootmirr 31464192 31464192 0 c1t1d0 ENA Creating the directories: # mkdir /test1 # mkdir /test2 Adding the following entries to /etc/vfstab: /dev/vx/dsk/test1 /dev/vx/rdsk/test1 /test1 ufs 1 yes - /dev/vx/dsk/test2 /dev/vx/rdsk/test2 /test2 ufs 1 yes - Creating new filesystem: # newfs /dev/vx/dsk/test1 # newfs /dev/vx/dsk/test2 Mouting all: # mountall /dev/vx/rdsk/test1: is logging. /dev/vx/rdsk/test2: is logging. Many Thanks to this nice "Sunmanagers-institution" Alex Duile _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From loris.serena at bt.com Thu Apr 28 10:19:05 2005 From: loris.serena at bt.com (Loris Serena) Date: Thu, 28 Apr 2005 15:19:05 +0100 Subject: SUMMARY: How to know when a patch has been applied. Message-ID: <90735D183F490247AC911C2C7E86E7E7024ACA00@exchange2.dublin.esat.ie> Probably the fastest SUMMARY in the history of the list ;-) ls -lrt /var/sadm/patch (Thanks to Jeff Marble) -----Original Message----- From: Loris Serena Sent: 28 April 2005 15:10 To: 'sunmanagers at sunmanagers.org' Subject: How to know when a patch has been applied. > Hello Guys, > > 1. > Is there any way to find out WHEN an OS patch has been applied? > > 2. > Is there a way to find what OS patches have been applied after a given > date? > > TIA > > Loris > > > > > > BT Communications Ireland Limited is a wholly owned subsidiary of BT Group plc Registered in Ireland, Registration No. 141524 Grand Canal Plaza, Upper Grand Canal Street, Dublin, Ireland This electronic message contains information (and may contain files) from BT Communications Ireland Limited which may be privileged or confidential. The information is intended to be for the sole use of the individual(s) or entity named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information and or files is prohibited. If you have received this electronic message in error, please notify us by telephone or email (to the numbers or address above) immediately. http://www.btireland.ie _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers BT Communications Ireland Limited is a wholly owned subsidiary of BT Group plc Registered in Ireland, Registration No. 141524 Grand Canal Plaza, Upper Grand Canal Street, Dublin, Ireland This electronic message contains information (and may contain files) from BT Communications Ireland Limited which may be privileged or confidential. The information is intended to be for the sole use of the individual(s) or entity named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information and or files is prohibited. If you have received this electronic message in error, please notify us by telephone or email (to the numbers or address above) immediately. http://www.btireland.ie _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ragnar.moller at ericsson.com Thu Apr 28 10:37:56 2005 From: ragnar.moller at ericsson.com (Ragnar Moller (MA/ESF)) Date: Thu, 28 Apr 2005 16:37:56 +0200 Subject: Summary - Solaris 10 at home Message-ID: Just for a summary of what I have so far: The best so far is quite a detailed work instruction for quad boot on http://www.sun.com/bigadmin. Search for "quad boot" and click on the third entry in the list. It is this http://onesearch.sun.com/search/clickthru?qt=quad+boot&url=http%3A%2F%2Fwww.s un.com%2Fbigadmin%2Fcontent%2Fsubmitted%2Fquad_boot_article.pdf&pathInfo=%2Fo nesearch%2Findex.jsp&hitNum=3&col=support-bigadmin It's really detailed and answered some of my hesitations when doing the intallations. Other recommendations are - Use a dedicated machine for Solaris - Use a dedicated disk for Solaris (This advise I also got from colleagues but I AM STUBBORN AS HELL). Plus some links where I can search for more (but it's along search) May be you can install grub without linux see : http://www.gnu.org/software/grub/grub.html but I used Ranissh boot manager usefull links : http://multiboot.solaris-x86.org/ http://solaris-x86.org/ Nothing so far about recovering partitions lost. Ragnar _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From DirkM at mxgroup.co.za Thu Apr 28 11:15:16 2005 From: DirkM at mxgroup.co.za (Dirk Moolman) Date: Thu, 28 Apr 2005 17:15:16 +0200 Subject: SUMMARY: C-license Message-ID: <71CE21D77D53D0439D8BC729FA25DD9B5BFBAE@mxmail1.mxgroup.co.za> Thank you very much everyone. I received many replies, the best one from Vincent Cojot (see below), thanks Vincent! (and to everyone else). It took me 30 minutes to find the answer from this mailing list, compared to WEEKS(!) with my support company (and still not making any progress with them). [Vincent] Forte (Up to 6.2) used node-locked FLEXlm licensing. Starting with 7.0, there is a serial number that activates the software but no node-locked stuff anymore which means you can move the software to a different machine. To check if you have a post 7.0 version, see if: /SUNWspro/bin/snit is there.. On my system (Sun Forte 10 aka SUN studio 10): root at tsaroth:[~][502]# /opt/SUN-Studio-10.0/SUNWspro/bin/snit Number of serial numbers found: 1 Total RTUs available: 500 License file used: /opt/SUN-Studio-10.0/SUNWspro/bin/../prod/bin/../lib/serial.dat Product Name: Sun[tm] Studio 10 software Serial Number (SN): SSS100-[snip] Is SN valid?: yes RTUs: 500 Is trial SN?: no If you have Forte 6.2 or below, you will have a file similar to: /etc/opt/licenses/licenses_combined which will most likely show node-locked licenses.. I hope this helps.. To get the version, run: # pkginfo -l SPROcc Example here with Forte 6.2: PKGINST: SPROcc NAME: Forte Developer 6 update 2 Compiler C CATEGORY: application ARCH: sparc VERSION: 6.2 BASEDIR: /opt VENDOR: Sun Microsystems, Inc. DESC: Forte Developer 6 update 2 Compiler C PSTAMP: 010516124340-22819-80cb2941 INSTDATE: Dec 08 2003 18:09 HOTLINE: Please contact your local service provider STATUS: completely installed FILES: 110 installed pathnames 41 shared pathnames 48 directories 24 executables 16068 blocks used (approx) [Original post] I am getting very frustrated with the support I am getting from the company that supports my SUN Solaris platform .... sigh We (my company) bought a c-license from them, and I am now trying to find out what type of license we have, an enterprise license (can be installed on many servers), or stand alone. They are telling me to run pkginfo and grep for Forte, but this doesn't tell me anything about the type of license. Now they tell me to search on Google. With this type of support I don't need support (sorry, just blowing off some steam). Can anyone here tell me how to check what type of license I have - if it is possible at all. I know some packages don't tell you, and that the paperwork is the only thing that tells you what type of license you have. Please, any help will be appreciated. TIA Dirk _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From JWalton at alldata.net Thu Apr 28 13:23:46 2005 From: JWalton at alldata.net (Walton, Joe) Date: Thu, 28 Apr 2005 13:23:46 -0400 Subject: SUMMARY: Segmentation Fault Message-ID: Thanks to all for your help. It looks like this issue was addressed by Sun here: http://sunsolve.sun.com/search/advsearch.do?collection=PATCH&type=collec tions&max=50&language=en&queryKey5=109873&toDocument=yes This was for a V880, but I'm sure there is one for the 12K too. Thanks again. Joe Walton Unix System Administrator Email: jwalton at alldata.net _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From DirkM at mxgroup.co.za Fri Apr 29 08:17:59 2005 From: DirkM at mxgroup.co.za (Dirk Moolman) Date: Fri, 29 Apr 2005 14:17:59 +0200 Subject: SUMMARY: Debugging process in Unix Message-ID: <71CE21D77D53D0439D8BC729FA25DD9B5BFCD2@mxmail1.mxgroup.co.za> Thanks everyone. Again I had lots of replies, and I will copy the the most complete answers below: [Ric Anderson] There are many truss -p pid shows what system calls a process is making. ps -fp pid shows general status pldd pid shows what dynamic libraries (.so) files a process is using. In addition, there are a number of tools in /usr/proc/bin that let you look at the actions of a process. If you have lsof installed, then lsof -p pid will show open files for a process. Some debuggers (e.g., gdb) can also be used with a pid to interactively debug a running process. [Jonathan Birchall] It all depends on how deep you wish to go and what version of Solaris you are using. Truss -wall -vall -fall -o -p will give you details of system calls etc. see Man truss You can also use the p commands see man pfiles , ptree , pmap etc. In Solaris 10 you can also now use dtrace for a detailed view of what the process is doing. Using a mixture of the about should give you a fairly detailed view of what is happening. You can also use adb (debugger).... which also pointed me to mdb (solaris 9) With Ric's help, using lsof, we managed to track down the problem. [ORIGINAL POST] > -----Original Message----- > From: sunmanagers-bounces at sunmanagers.org > [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Dirk Moolman > Sent: Friday, April 29, 2005 12:27 AM > To: sunmanagers at sunmanagers.org > Subject: Debugging process in Unix > > > What is the command to see what a process is doing in Unix ? I know > there is a command that you type in with the PID, and it show you > exactly what that process is doing. > _______________________________________________ > 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 goldsun8 at yahoo.com.sg Fri Apr 29 11:13:00 2005 From: goldsun8 at yahoo.com.sg (Gold Sun) Date: Fri, 29 Apr 2005 23:13:00 +0800 (CST) Subject: SUMMARY: patchadd/pkgadd failed for 114219-11 In-Reply-To: <20050428045124.48022.qmail@web41511.mail.yahoo.com> Message-ID: <20050429151300.14150.qmail@web41504.mail.yahoo.com> Hi, Received numerous replies & I've extracted the following few which basically give similar solutions. Apologies that I'm unable to name the lengthy list of contributors. Raymond Jackson I just hit this one yesterday. At some point the patchadd process does an su to the install user if one exists, otherwise to user nobody. If the patch files and all parent directories aren't readable by install/nobody, you get exactly the error you report. For me setting execute permission for all on /var/spool/patch fixed the problem. ================================================== You can workaround this in two ways, one is to make sure that the user "nobody" can read all patch files and execute a "pwd" in the patch directory or add an account "install" to /etc/passwd: install:x:0:1:installpatch braindamage:/:/bin/true Installpatch and patchadd use "nobody" as a fallback if it cannot find the "install" user. You indicated that you'd checked the permissions on the subdirectories for world readability, but did you check the permissions on the directories in the path above the patch directory? ========================================= I've gone insane trying to resolve this before. This may not be your problem, but I think you'll find you need world read all the way to root (/) wherever the patch is. Uh... http://www.science.uva.nl/pub/solaris/solaris2.html#q5.59 says If any of the patch files or if any part of the path leading up to the patch directory cannot be read by nobody, an error similar to the following will appear:... . =================================== Gold Sun wrote:Hi, Received several suggestions (I appreciate it) but it's still not resolved yet. I've inserted below replies to requests for more information : A)I've verified that all directories/file under the patch directory has world read, rerun patchadd/pkgadd & still got the same error. Then I issued "umask 022" from another suggestion & rerun again & still getting the same error B)return status code is 5 : one reply asks for "echo #?" to see what's the status code but it always give a blank output & so I issued "echo $?" which gave "5" as follows (note: pkgadd returns "3") : # patchadd -d /home/9mydir/patch/114219-11 Checking installed patches... Verifying sufficient filesystem capacity (dry run method)... Patch 114219-11 failed to install due to a failure produced by pkgadd. See /var/sadm/patch/114219-11/log for details Patchadd is terminating. # echo $? 5 C) Content of /var/sadm/patch/114219-11/log : (I've inserted in between **...** my remarks) : Package not patched: PKG=SUNWdtim Original package not installed. This appears to be an attempt to install the same architecture and version of a package which is already installed. This installation will attempt to overwrite this package. ** I believe this is because 114219-11 contains 2 ** ** packages, SUNWdtim & SUNWolrte, & SUNWolrte ** ** is already installed : ** ** # pkginfo | grep -i Wolrte ** **system SUNWolrte OPEN LOOK toolkits runtime environment** /home/9mydir/patch/114219-11/SUNWolrte/install/checkinstall: /home/9mydir/patch/ 114219-11/SUNWolrte/install/checkinstall: cannot open pkgadd: ERROR: checkinstall script did not complete successfully Dryrun complete. No changes were made to the system. Do I need to do "pkgrm" to remove a certain package which may be hindering this or is it this patch is not applicable as this box is missing some pre-requisite packages? Thanks G Sun Gold Sun wrote: Hi, I'm asked to patch a security vulnerability & I refer to url http://sunsolve.sun.com/search/document.do?assetkey=1-26-57769-1&searchclause= After downloading the normal unsigned patch from Sun website indicated in the above url for Solaris 9, the patch could not install (I did "sudo -s" to gain root privilege on our hardened box) : # pkginfo | grep -i tiff system SUNWTiff libtiff - library for reading and writing TIFF # # patchadd -d ./114219-11 Checking installed patches... Verifying sufficient filesystem capacity (dry run method)... Patch 114219-11 failed to install due to a failure produced by pkgadd. See /var/sadm/patch/114219-11/log for details Patchadd is terminating. # # # patchadd -d ./114219-11 Checking installed patches... Verifying sufficient filesystem capacity (dry run method)... Patch 114219-11 failed to install due to a failure produced by pkgadd. See /var/sadm/patch/114219-11/log for details Patchadd is terminating. # pkgadd -d ./114219-11 The following packages are available: 1 SUNWdtim Solaris CDE Image Viewer (sparc) 1.5,REV=10.2002.03.13 2 SUNWolrte OPEN LOOK toolkits runtime environment (sparc) 3.7.0,REV=1.2001.10.12 Select package(s) you wish to process (or 'all' to process all packages). (default: all) [?,??,q]: Processing package instance from Solaris CDE Image Viewer (sparc) 1.5,REV=10.2002.03.13 Copyright 2002 Sun Microsystems, Inc. All rights reserved. Use is subject to license terms. ## Executing checkinstall script. PaTcH_MsG 8 Version 1.5,REV=10.2002.03.13 of SUNWdtim is not installed on this system. checkinstall script suspends Installation of was suspended (administration). No changes were made to the system. There is 1 more package to be installed. Do you want to continue with installation [y,n,?] y Processing package instance from OPEN LOOK toolkits runtime environment (sparc) 3.7.0,REV=1.2001.10.12 Copyright 2001 Sun Microsystems, Inc. All rights reserved. This appears to be an attempt to install the same architecture and version of a package which is already installed. This installation will attempt to overwrite this package. ## Executing checkinstall script. /home/9mydir/patch/114219-11/SUNWolrte/install/checkinstall: /home/9mydir/patch/114219-11/SUNWolrte/install/checkinstall: cannot open pkgadd: ERROR: checkinstall script did not complete successfully Installation of failed. No changes were made to the system. I verify that the above patch is relevant to my system as follows : # pkginfo | grep -i tiff system SUNWTiff libtiff - library for reading and writing TIFF # # cd /usr/openwin/lib # ls *tif* libtiff.so libtiff.so.3 Hope I have not corrupted anything & I promise to summarize this one. Many thanks in advance G Sun Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers Yahoo! Mobile - Download the latest ringtones, games, and more! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Bernd.Schemmer at gmx.de Fri Apr 29 14:31:01 2005 From: Bernd.Schemmer at gmx.de (Bernd Schemmer) Date: Fri, 29 Apr 2005 18:31:01 +0000 Subject: Summary: Emulex Performance 4000 Message-ID: <42727D65.4060801@gmx.de> Summary: I got one reply from Petri Kallberg pointing me to the web page http://www.emulex.com/ts/legacy/cs/index.html There's the documentation for the Performance 4000 thanks Bernd --- -------- Original Message -------- Subject: Emulex Performance 4000 Date: Fri, 29 Apr 2005 16:51:59 +0000 From: Bernd Schemmer To: sunmanagers at sunmanagers.org Hi, I bought a used Emulex Performance 4000 to use it as terminal server for my Sun machines. Unfortunately there's no documentation and my google search wasn't successfull. Is anybody using this Terminalserver and can tell me how to configure it? TIA Bernd -- Bernd Schemmer, Frankfurt am Main, Germany http://home.arcor.de/bnsmb/index.html Mas temprano que tarde el mundio cambiara Fidel Castro _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers -- Bernd Schemmer, Frankfurt am Main, Germany http://home.arcor.de/bnsmb/index.html Mas temprano que tarde el mundio cambiara Fidel Castro _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rickv at memec.com Fri Apr 29 12:47:32 2005 From: rickv at memec.com (Rick von Richter) Date: Fri, 29 Apr 2005 09:47:32 -0700 Subject: SUMMARY: Serial console vs. RSC console Message-ID: <42726524.8070804@memec.com> Many thanks to those that responded and suffered thru my replies. 8-) First of all, I received many answers that there is NO difference between the RSC console and the serial console. The console is the console. Nuff said... I finally figured this one out.There were two different issues on two servers. Server #1 ========= Originally, I went into the OBP, ran nvedit and looked at the lines and did not see any problems. What I failed to do was look at ALL of the lines. I merely stopped at the first blank line thinking it was the last. In nvedit, I found that Ctrl-L will list all of the lines. I found this: {0} ok nvedit 0: devalias test2 test2 1: devalias vx-rootdisk1 /pci at 1f,4000/scsi at 3/disk at 0,0:a 2: probe-all install-console banner 3: cd /pci at 1f,4000/scsi at 5 4: 6 " scsi-initiator-id" integer-property 5: device-end 6: 5: device-end 6: 7: 8: 9: 10: 11: 12: 13: 14: 15: devalias vx-rootdisk2 /pci at 1f,4000/scsi at 3/disk at 1,0:a 16: setenv boot-device vx-rootdisk2 Apparently, some lines were erased but not deleted and I completely missed them. Server #2 ========= This one was a bit easier to find since use-nvramrc? was set to false. I found a /etc/rcS.d script that had this line in it: eeprom boot-device rootdisk-p rootmirror-d The reason why it was intermittent was because I didn't boot all of the time. I had set auto-boot? to false to troubleshoot. So, this would not hit until I actually did a boot. This is what you get when you inherit systems. Cheers, Rick Original question ================= We're running across some strange behavior. We are setting aliases and noticed something strange. If we set the aliases from a serial console they stick across reboots. If we set them from the RSC console then they revert. Is there a difference between the serial console and the RSC console (other than how you get in obviously)? This makes no sense to me because it's the same OBP. TIAWS This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be assured to be secure or correct as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ssefoen at yahoo.com Fri Apr 29 13:18:03 2005 From: ssefoen at yahoo.com (Z H) Date: Fri, 29 Apr 2005 10:18:03 -0700 (PDT) Subject: Summary: volume manager 3.5 upgrade In-Reply-To: 6667 Message-ID: <20050429171803.71476.qmail@web31607.mail.mud.yahoo.com> Thanks to Darren Dunham for his expertise. I used vxdg -C import disk_group to clear the old block and I see all the disks under vxdisk list. After the reboot, all is well and working beautifully. Thanks also to Vicky Mangat for her response. Have a great weekend! ssefoen. -------------- Original question: > To All Helpers: > > Platform: solaris 2.8 and VM 3.5 > > machine: Ultra2 (this is very old machine and not a > production server) > > Problem: the box was running sun volume manager 2.4 > and solaris 2.6. The root disk was encapsulated. > > boot the system from underlying partition, remove > the > vm 2.4 without export the disk group (this may be > the > problem). upgrade the system to 2.8 without > problem. > Install 3.5 MP1 without problem. Run vxinstall > after > the installation of vm 3.5 and encapsulate the root > disk again. > > Everything is fine. But the VM can not see any > other > old disk groups from vxdisk list, although it can > see > all the disks. > > I have used vxdg -f import option, but it failed. > "import failed: No valid disk found containing disk > group" > > I can used vxdisk -o alldgs list to see the old disk > group but the disk has no name on it. > > rpt root [5] vxdisk -o alldgs list > DEVICE TYPE DISK GROUP > STATUS > c0t0d0s2 sliced rootdisk rootdg > online > c0t1d0s2 sliced - - > error > c2t0d0s2 sliced - (ssa_app) > online > c2t0d1s2 sliced - (ssa_app) > online > c2t0d2s2 sliced - (ssa) > online > c2t0d3s2 sliced - (ssa) > online > > I also can use vxprinvutil dumpconfig to dump the > config from each disk. > > My question is that is there anyway that I can force > the vm 3.5 to see the disks created under vm 2.4? > > Thanks in advance > > ssefoen. > Tired of spam? Yahoo! Mail has the best spam > protection around > http://mail.yahoo.com > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From hoogendyk at bio.umass.edu Fri Apr 29 16:52:13 2005 From: hoogendyk at bio.umass.edu (Chris Hoogendyk) Date: Fri, 29 Apr 2005 16:52:13 -0400 Subject: [Summary] ufsdump, solaris 9 & RBAC not working correctly Message-ID: <42729E7D.2060206@bio.umass.edu> Didn't get any direct answers to my question, but I got some very good explanatory information from the one person (Bill R. Williams ) who replied, and that lead me to the answers I needed. Bottom line, bug 4961690 posted on Sun on March 30, 2005. No corresponding patch that I could find. Workaround - run ufsdump as root until patch is available. Since I'm using a remote tape drive for this server, and I don't want to have root logging in on the remote server, I did it this way: /usr/lib/fs/ufs/ufsdump 0cafNu cat1 - /dev/rdsk/c0t0d0s3 \ `/usr/lib/fs/ufs/fssnap -o raw,bs=/export/home/scratch,unlink / ` \ | su - backup -c 'ssh remoteserver "dd obs=64b of=/dev/rmt/0n"' of course, embedded in a larger script with all the other appropriate stuff like "fssnap -d /", etc. and key exchange setup for backup user. original message and exchanges below chronologically. --------------- Chris Hoogendyk - O__ ---- Systems Administrator c/ /'_ --- Biology Department (*) \(*) -- 140 Morrill Science Center ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- ---------------- Original Message Posted to List ---------------- Subject: ufsdump, solaris 9 & RBAC not working correctly Date: Thu, 28 Apr 2005 09:55:54 -0400 From: Chris Hoogendyk To: Sun Managers List I've found brief discussions on a couple of lists attributing the error Unable to create temporary directory in any of the directories listed below: /tmp/ /var/tmp/ / Please correct this problem and rerun the program. to a "bug" in ufsdump in Solaris 9. One person said he replace the Solaris 9 ufsdump binary with the Solaris 8 ufsdump binary and it worked without the error. the error seems to be cause by the creation of a directory with 0 permissions on /tmp and then an attempt to create a subdirectory under that. root can do it, but if you run ufsdump as non-root, it cannot, even though ufsdump is suid root. in one thread, Casper Dik said "The only thing ufsdump/ufsrestore use set-uid root for is to use rcmd(3) for remote tape style dumping. They do not run with euid == 0 when doing anything else." I don't really get that. I thought suid was suid. Anyway, I've created a role "backup" with the following specs: # grep backup /etc/passwd backup:x:7000:7000:Tape Backup:/u1/home/.backup:/bin/pfksh # grep backup /etc/user_attr backup::::type=role;profiles=Dump # grep Dump /etc/security/prof_attr Dump:::Tape Backup User: # grep Dump /etc/security/exec_attr Dump:suser:cmd:::/usr/lib/fs/ufs/ufsdump:euid=0;gid=sys Dump:suser:cmd:::/usr/lib/fs/ufs/fssnap:euid=0;gid=sys If I 'su - backup' and do a ufsdump, I still get the same error described above. Presumably, I'm not only running ufsdump as root as per the role, but ufsdump is also suid root. I know the role is working, because I got a permission denied on the fssnap before I had it set up, and now the fssnap works. In both cases, I'm using the full path to the binary and not the symlink. I really want to use RBAC rather than run this whole thing as root. Any ideas? --------------- Chris Hoogendyk - O__ ---- Systems Administrator c/ /'_ --- Biology Department (*) \(*) -- 140 Morrill Science Center ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers ---------------- First Response from Bill R. Williams ---------------- Subject: Re: ufsdump, solaris 9 & RBAC not working correctly Date: Thu, 28 Apr 2005 10:10:13 -0400 From: Bill R. Williams To: Chris Hoogendyk References: <4270EB6A.3060801 at bio.umass.edu> Hi Chris, I can't speak to the RBAC, but perhaps I can clear up one thing for you. Referring to suid program remarks you said: > I don't really get that. I thought suid was suid. It is; however, a suid program may drop/regain the suid privileges any time it wants to. Any well-behaved program designed to run with suid=root had BETTER drop priv and regain only when absolutely necessary. Or more specifically: A program (task, job, whatever) has a REAL uid and an EFFECTIVE uid which in most cases are the same -- the uid of the user running the program. A suid program run by some user, spawns with the euid of its owner and the ruid of the user which ran it. IOW: it is "effectively" running as the program owner. THE FIRST THING such a program is supposed to do is save euid and switch to ruid so that it is now running as the user. It should switch back to the euid ONLY for operations that require the privilege of the euid, and as soon as the operation is complete it should drop back to ruid. This is how things like Apache start as root then run as some other user -- apache, httpd, or nobody. Of course, if 'root' is running a program, it doesn't really matter what the suid is or whether the program "drops privilege" -- it's always running as root. Maybe this helps with the suid is suid thing. It is, but only as long as it wants to be. -- --------------------------------------------- Bill R. Williams ------------------------ ETSU Library Systems ---------------- My Reply to Bill R. Williams ---------------- Subject: Re: ufsdump, solaris 9 & RBAC not working correctly Date: Thu, 28 Apr 2005 12:36:26 -0400 From: Chris Hoogendyk To: Bill R. Williams References: <4270EB6A.3060801 at bio.umass.edu> <20050428141013.GA24653 at brwms.etsu.edu> Bill, Thanks for the reply. It's the only one I've gotten. Very informative. It may actually answer the question, although now I have to figure out a solution. The /etc/security/exec_attr specifications for the role I created are to run ufsdump with euid=0. Basically, then, this is no change from the application being suid in the first place. By Casper Dik's comments and your explanation, ufsdump is doing most of its processing as user/role "backup" rather than as root. Since one of the things it does is to create a directory in /tmp with 0 privileges and then try to create a subdirectory in that directory, one must assume that either ufsdump is intended to be run as root or that this is a bug. However, if it was intended only to be run as root, then it wouldn't be suid root. So, it must be a bug. I suppose my next task is to try to find what it is doing and add to the role exec_attr so that whatever it is runs as root. I assume RBAC is low level enough that it will work even from inside a binary. Ok, actually found a bug report 4961690 posted March 30, 2005. Can't find a corresponding patch. --------------- Chris Hoogendyk - O__ ---- Systems Administrator c/ /'_ --- Biology Department (*) \(*) -- 140 Morrill Science Center ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- ---------------- Bill's Reply Back to Me ---------------- Subject: Re: ufsdump, solaris 9 & RBAC not working correctly Date: Thu, 28 Apr 2005 16:05:41 -0400 From: Bill R. Williams To: Chris Hoogendyk References: <4270EB6A.3060801 at bio.umass.edu> <20050428141013.GA24653 at brwms.etsu.edu> <4271110A.8060409 at bio.umass.edu> Chris, In the not really help but FWIW department... I'm guessing that in most cases 'ufsdump' is run as root; therefore, bug-reports resulting from non-root use are not as plentiful. I think that usually 'ufsdump' is being run to backup a filesystem, (for instance by i-node) for the purposes of getting a "pure" representation of the state of the filesystem, and you simply cannot get a "pure" filesystem backup if you are not 'root'. If Casper is correct about ufsdump being suid=root in order to deal with the tape drive, and only the tape drive -- not the disk devices -- then the suid=root bit on it will not provide privilege to read/write anything, anytime, anywhere. Actually, no matter what you use you can only backup what you have permission to read, and if you want to include attributes in your backup you must be able to read them, too. (Hint, hint.) IOW: No matter what you are using, you must have some user set up in some way to read all aspects you want to preserve of the specified filesystem. (I'm not a RBAC guru, so I'll just leave that as that. :-) That isn't to say you can't get a "fine for what I need" backup using something besides 'ufsdump' and as a user other than root. I do this sort of thing quite often: sometimes I use GNU tar (gtar) and sometimes I use cpio. (I pick the utility based on the task at hand.) BTW: any of them can use a fssnap device if you mount it ro somewhere. I tend to use 'cpio' for "almost" pure backups because: * It allows me fine-grained control over the files which get backed up. This would do a 'cpio' backup, but omit any first level ./tmp/* files while keeping the ./tmp entry. cd /FILESYSTEM find . -fstype ufs -depth -print \ | grep -v '^./tmp/' \ | cpio -o >SOMEWHEREnotin/FILESYSTEM/SOMEFILE.cpio As you can imagine you could have more 'grep's, etc. in there to filter out all non-essential junk stuff. In quick & dirty mode, I "find . -depth -print >/tmp/file.list" and then edit that file to just feed to cpio: cpio -o /tmp/backup.cpio And, you *could* add '| gzip' after the cpio instead of '>SOMEFILE...' to compress "on the fly". * It is a reasonably portable format which means I *could* retrieve a file from it on another system -- possibly even a Linux system. * Sun uses 'cpio' for their stuff. For example... In Solaris9 the part of a flash archive (flar) that contains the backup is a 'cpio' container! IOW: cpio is available at jumpstart/boot time. Actually, I frequently use the 'cpio' technique to backup certain data through an ssh pipe from one Solaris9 system to another. Often the user running the backup (source) is root, but the target user (on the receiving system) is NEVER root! My biggest tip for any backup (ufsdump, cpio, gtar, ...) is when writing to anything NOT a local file is to specify the blocksize explicitly. (In the case of a tape drive give it the biggest blocksize the drive will allow). And RECORD the blocksize used; it'll save you a lot of aggravation later. (Some utilities want the same blocksize coming in as they "know" they wrote going out.) But, you may already know all that and wasted your time reading this whole thing. Or you may be able to point out some flaws in my methods which will prevent me from shooting myself in the foot. -- --------------------------------------------- Bill R. Williams ------------------------ ETSU Library Systems _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jjah at sol.ccsf.cc.ca.us Sat Apr 30 03:39:30 2005 From: jjah at sol.ccsf.cc.ca.us (Joe R. Jah) Date: Sat, 30 Apr 2005 00:39:30 -0700 (PDT) Subject: SUMMARY: File system full, but not according to du;( In-Reply-To: Message-ID: Wow! That was quick;) Many thanks to: Mike Salehi "Mangat, Vicky" "Sjolshagen, Thomas (Tru64 UNIX Eng.)" Rick Rezinas Scott M. Sorrentino Michael Schulte Matthew Stier Iain Miller Original question: > Since early this morning one of my partitions have become full: > > # df /dptweb > Filesystem 1K-blocks Used Avail Capacity Mounted on > /dev/amir0d 25019863 23770030 -1161 100% /dptweb > > However: > > # du -s /dptweb > 6680278 /dptweb > > How can I resolve the problem. Michael's response is the most comprehensive, and doesn't need to be summarized;) On Fri, 29 Apr 2005, Michael Schulte wrote: > Date: Fri, 29 Apr 2005 13:58:05 -0500 (CDT) > From: Michael Schulte > To: jjah at sol.ccsf.cc.ca.us > Subject: Re: File system full, but not according to du;( > > The easy solution: reboot > The hard solution: find the process that is holding a deleted file open, and > kill that process. > > The problem is (very probably): when a file is deleted, the file space is > NOT recovered until all the processes that have the file open have > closed it (or died). Many system processes open a file when they start > and never close it; if someone deletes that file, the space is still > allocated. > > You notice the problem because df and du count the space differently. > Df counts space from the inodes (which gives all allocated space); > du follows the directory structure, so it only get file space accessible > by name. > > --- > Michael Schulte Specialist in Computer Science > Math/CS Dept. (MC-61) (314) 516 5239 > Univ. of Missouri-St. Louis schulte at cs.umsl.edu > One University Boulevard http://www.cs.umsl.edu/~schulte > St. Louis, MO 63121 USA Regards, Joe -- _/ _/_/_/ _/ ____________ __o _/ _/ _/ _/ ______________ _-\<,_ _/ _/ _/_/_/ _/ _/ ......(_)/ (_) _/_/ oe _/ _/. _/_/ ah jjah at sol.ccsf.cc.ca.us _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers