From G.Bakalarski at icm.edu.pl Thu Apr 1 08:37:26 2004 From: G.Bakalarski at icm.edu.pl (Grzegorz Bakalarski) Date: Thu, 1 Apr 2004 15:37:26 +0200 Subject: SUMMARY: [V880 Solaris 9 ) prtdiag problem Message-ID: <20040401133726.GB18002@burza.icm.edu.pl> Dear All Sorry for late summary. But I wanted to check one advice (i.e. turn off the system fully any start with full test mode - then restart again). This din't help. As a workaround following other advice I copied /usr/platform/SUNW,Sun-Fire-880/lib/libprtdiag_psr.so.1 from other unpatched machine and this cured a problems. Today I found also official SUN statement on that: Patch-ID# 113574-03 Keywords: libpsvc sunw,sun-fire-880 psvc_sema_rdwr picld Synopsis: SunOS 5.9: SUNW,Sun-Fire-880 libpsvc Patch Date: May/23/2003 ******************************************************* Patch 113574-03 has been re-instated on Mar. 25, 2004. Patch 113574-08 was withdrawn. Reason: Patch 113574-08 causes a segmentation fault when "prtdiag" is run. Please reference bugID 5016636 for more info. Recommendation: Please backout patch 113574-08 and install patch 113574-03, if necessary. ******************************************************* I think this solves problem at the moment. Cheers, GB -----------------Original Query-------------------------- Date: Mon, 15 Mar 2004 14:21:50 +0100 From: Grzegorz Bakalarski To: sunmanagers at sunmanagers.org Dear ALL, Recently I installed 113573-04 & 113574-08 patches (libpsvcpolicy & libprtdiag_psr). Now I get Segmentation Fault whenever I try to run prtdiag. Is there any way (other than patchrm) to get this command working? Machine is: SUN FIRE V880 - 6x900MHz Ultra III sparc 12GB uname -a: SunOS 5.9 Generic_112233-11 sun4u sparc SUNW,Sun-Fire-880 patched to recent (Feb 27th/2004) recommended patches. GB PS: I've also intalled on the same day patch 113277-19 (sd & ssd). I don't think this is related but since that day I have been noticing larger max system load (i.e above some 16 vs under some 8) _______________________________________________ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From benoit.audet at carra.gouv.qc.ca Thu Apr 1 15:56:18 2004 From: benoit.audet at carra.gouv.qc.ca (=?ISO-8859-1?Q?Beno=EEt=20Audet?=) Date: Thu, 01 Apr 2004 15:56:18 -0500 Subject: SUMMARY : Sun Blade 100 - input / output problem... Message-ID: I finally rafined my "search keywords" on the Sun Microsystems' website and found the anwser to my problem. In fact, the "Stop" "N" combination was ok for "non USB" keyboard systems... ...and that was my problem, since Sun Blade 100 uses USB keyboard and mouse... Here's the Sun's awnser on this, located at: http://docs.sun.com/source/817-4189-10/oboot.html "To Use a Stop-N Equivalent 1. After turning on the power to your system, wait until the front panel power button LED begins to blink and you hear an audible beep. 2. Quickly press the front panel power button twice (similar to the way you would double-click a mouse)." Thanks anyway for those of you that will may be awnsers me before seing my "SUMMARY"! ;-) The original post was: Hi managers, We have a Sun Blade 100 here to do some tests and to get the ability to save a job log, I wanted to hook up the workstation to a laptop, using a serial cable setting. Normally, when not using monitor, keyboard and mouse, Sun workstations/servers are redirecting input/output to the serial port. So I did connect the cable I normally use, with a null modem module and a DB25-to-DB9 adapter. I tested this setting on another Sun Blade 100 and it works (with the same laptop). My problem is that it was not working on this particular Sun Blade 100. So, I went to the OBP and these settings was present: ok printenv output-device output-device = screen ok printenv input-device input-device = keyboard I browsed a bit to see if there's something about a known problem using serial console with Sun Blade 100 models (since there's only one DB9 serial connector, etc., I though maybe there's something that may be done...), and I found someone suggesting to but "ttya" values for both "output-device" and "input-device" in the OBP. So I did it, but guess what? It's not working. But the big surprise arrised when I tried to plug back the USB keyboard and mouse, and the monitor: I no longer have the hability to interract with the machine! I knew an old technique to reset the OBP values with SPARCstations, which consist to maintain the "Stop" and "N" keys on the keyboard and you're back in business... ...I of course tried it with the Sun Blade 100, but I doesn't seem to do anything. Anyone have an idea of what I could do, look at, or... ...??? Thanks in advance and, of course, will summarise! _______________________________________________________________ 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 : 418.528.1281 (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 BLucas at accela.com Sat Apr 3 12:14:08 2004 From: BLucas at accela.com (Brian Lucas) Date: Sat, 3 Apr 2004 10:14:08 -0700 Subject: SUMMARY: pkginfo file is corrupt or missing Message-ID: <1678E37D857493458F8E9EE62C7A67441DAD4D@slc-mail.accela.com> OK, I figured it out. I just replaced the pkginfo file in SUNWlibm with a good one from another server and everything works well. -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Brian Lucas Sent: Saturday, April 03, 2004 9:56 AM To: sunmanagers at sunmanagers.org Subject: UPDATE: pkginfo file is corrupt or missing OK, I can see that the pkginfo file for the package SUNWlibm is corrupt. Is there an easy way to fix this? -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Brian Lucas Sent: Saturday, April 03, 2004 9:47 AM To: sunmanagers at sunmanagers.org Subject: pkginfo file is corrupt or missing Gurus, Am getting the error "pkginfo file is corrupt or missing" when trying to retrieve the list of packages installed using "pkginfo" It displays a series of packages but then exits with that error as seen below: (snipped for brevity) system SUNWlccom Localization common files system SUNWlcl Locale Conversion Library system SUNWlclx Locale Conversion Library (64-bit) system SUNWless The GNU pager (less) system SUNWlibC Sun Workshop Compilers Bundled libC system SUNWlibCf SunSoft WorkShop Bundled libC (cfront version) system SUNWlibCx Sun WorkShop Bundled 64-bit libC pkginfo: ERROR: pkginfo file is corrupt or missing I can see packages in the /var/sadm/pkg folder just fine. Is there a way to rebuild the pkginfo information? Thanks, Brian _______________________________________________ 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 mparamas at iupui.edu Wed Apr 7 19:08:49 2004 From: mparamas at iupui.edu (Paramasivam, Meenakshisundaram) Date: Wed, 7 Apr 2004 18:08:49 -0500 Subject: [SUMMARY] SCSI-IDE Ultra10 Message-ID: <244A4457EB02E942BCB26BB535AFF3C0E1AB39@iu-mssg-mbx03.exchange.iu.edu> The problem really was a disconnected power cable inside enclosure. I apologize for wasting valuable time and resources. I would like to thank especially Darren Dunham who communicated back and forth on this case. I placed the 250 GB IDE disk on Blade2000 (should also work on Ultra10) scsi chain to get the disk up quick. Solaris 8 (Generic_108528-12), OPB 4.2.4. Format sees it now: 4. c0t14d0 /pci at 8,700000/scsi at 6,1/sd at e,0 Part Tag Flag Cylinders Size Blocks 0 root wm 0 0 (0/0/0) 0 1 swap wu 0 0 (0/0/0) 0 2 backup wu 0 - 60562 232.88GB (60563/0/0) 488380032 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 0 - 30280 116.44GB (30281/0/0) 244185984 7 unassigned wm 30281 - 60562 116.44GB (30282/0/0) 244194048 Note: 232.88 GB size in machine = 250 GB size by vendor x 0.9313 where 0.9313 = (1000/1024 kb) (1000/1024 kb) (1000/1024 kb). format.dat was not needed. > -----Original Message----- > From: Paramasivam, Meenakshisundaram > Sent: Tuesday, April 06, 2004 6:14 PM > To: 'sunmanagers at sunmanagers.org' > Subject: SCSI-IDE Ultra10 > > Hi, > > I have WD2500PB disk and AEC-7726Q SCS-IDE bridge > connected to Ultra 10 running Solaris 8 (Generic_108528-17) > with OBP 3.19 > > probe-scsi-all: > can not find this particular drive > format: > can not find this particular drive > > I do not want to use on-board 24-bit IDE controller. I did boot -r. > > Question: > What will resolve the problem? (Looks like firmware, OS and patch updates are due). > > FYI: > I have WD2500JB working fine via SCSI-IDE bridge on a Blade1500 running Solaris 8 (Generic_108528-27) with OBP 4.9.5 (2003/09/15). > > Thanks in advance. > > Sundar _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From steeler_dude99 at yahoo.com Wed Apr 7 21:15:33 2004 From: steeler_dude99 at yahoo.com (Chuck) Date: Wed, 7 Apr 2004 18:15:33 -0700 (PDT) Subject: SUMMARY: exercising a sun In-Reply-To: Message-ID: <20040408011533.32608.qmail@web10412.mail.yahoo.com> There is a package called VTS, it's on the supplement CD and will hammer CPUs and memory as much and long as you want. Thanks to: kevin.a.sindhu at Mail.AC Jeremy.Loukinas at evenflo.com hike1272-sunhelp at yahoo.com jsims at MusicNet.com Yahoo! Small Business $15K Web Design Giveaway - Enter today _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From aad at verio.net Thu Apr 15 03:43:57 2004 From: aad at verio.net (Anthony Talltree) Date: Thu, 15 Apr 2004 07:43:57 -0000 Subject: Summary: On the topic of mice & keyboards... In-Reply-To: <43ADFEB6-8D66-11D8-8FA1-000A95A637EC@verio.net> References: <43ADFEB6-8D66-11D8-8FA1-000A95A637EC@verio.net> Message-ID: > Have any of you tried to adapt a Type 5 keyboard and/or mouse to USB? > I've been allocated a Mac G5 as a desktop machine, and loathe Apple's > keyboard and mouse -- control/capslock are switched on the keyboard, > eg., and the mouse lacks two buttons. I'm currently using a USB Type > 6 mouse + keyboard, which function perfectly, but the stiff keys and > the shape of the mouse are killing my hands. I see various sorts of > adapter dealies out there, but they're moderately expensive and I'd > like to see a success story before I spend the bucks. Thanks. Thanks as usual the the horde of folks with badly misconfigured autoresponders who think I care that they're away from their offices. I've received pointers to two devices that should work: o Lightwave Communications product called "Chameleon". Some web sites offer the product for sale in the $150 range, though it appears that this company was acquired by Lantronix, and I wonder about availability. One respondent is using this product, albeit on a Sun. Another mentioned a similar, perhaps even OEM'd product from UltraSpec: http://www.ultraspec.com/pcatalog.asp?pID=5188 o Raritan has a product pn ASUSB that appears to be intended for use with some wacky KVM switch, but which should also work for my application. It's available in the $80 range, eg here: http://www.cdw.com/shop/products/default.aspx?EDC=323624 I'll probably give one of the latter a try. A number of folks misunderstood what I was after, suggesting various other USB keyboards and USB-PS/2 adapters, which wouldn't help because they all have broken layouts, except for the Happy Hacking models, which are two minimalist for comfortable desktop usage on a Mac, and ludicrously expensive. Thanks to all who responded. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From andy.ford at telindus.co.uk Sun Apr 18 16:05:13 2004 From: andy.ford at telindus.co.uk (Andy Ford) Date: Sun, 18 Apr 2004 20:05:13 -0000 Subject: SUMMARY: ifconfig configuration persistency Message-ID: <1082322266.1474.9.camel@home> Thanks everyone for responding - a sharp, concise, accurate response as ever ;0) I created an entry in the following... /etc/inet/hosts /etc/netmasks /etc/hostname.eri0 /etc/hostname.eri1 Andy _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From adave at vantage.com Mon Apr 19 10:51:42 2004 From: adave at vantage.com (Anjan Dave) Date: Mon, 19 Apr 2004 14:51:42 -0000 Subject: SUMMARY: mouting / filesystem Message-ID: <4BAFBB6B9CC46F41B2AD7D9F4BBAF785098143@vt-pe2550-001.vantage.vantage.com> This list rocks! Most suggested remounting it, and that worked. Thanks, Anjan -----Original Message----- From: Anjan Dave Sent: Mon 4/19/2004 10:27 AM To: sunmanagers at sunmanagers.org Cc: Subject: mouting / filesystem Folks, I made a mistake of mounting the wrong partition as / file system, and now the system boots up only in read-only for root, and i can't modify the vfstab file anymore. Single-user mode doesn't go forward after typing the password, hangs there. Is there any other alternative to be able to modify the /etc/vfstab? (unfortunately, there's no CDROM drive, but i can try to put one, and boot from Solaris CD). Thanks, Anjan _______________________________________________ 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 Wed Apr 21 14:07:58 2004 From: bshah at citadon.com (Bhavesh Shah) Date: Wed, 21 Apr 2004 18:07:58 -0000 Subject: SUMMARY: E 4500 Panic Message-ID: <0FEA43415756E44AB368AC58C1A4F4F3021B7480@a4mail2k01.citadoncorp.com> Hi, I received only one response in this regards and very much thankful to Tom Davis for that. The Problem seems to be a faulty Hardware (CPU1) in this case and needs to be replaced. Thanks Tom My Original Question was: Hi Gurus, We have Sun E 4500 Running Solaris 8. Recently Server got rebooted with following error: ------------------------------------------------------------------------ ------------------------------------------------------------------------ ------------------------------------------------------------------------ ------------ Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 125304 kern.warning] WARNING: [AFT1] EDP event on CPU1 Instruction access at TL=0, errID 0x00 3d4b09.64085e0d Apr 20 14:58:36 sccpdb01 AFSR 0x00000000.00400002 AFAR 0x00000000.fabed1c8 Apr 20 14:58:36 sccpdb01 AFSR.PSYND 0x0002(Score 95) AFSR.ETS 0x00 Fault_PC 0x13e51b0 Apr 20 14:58:36 sccpdb01 UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0000 UDBL.ESYND 0x00 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 364788 kern.info] [AFT2] errID 0x003d4b09.64085e0d PA=0x00000000.fabed1c8 Apr 20 14:58:36 sccpdb01 E$tag 0x00000000.08401f57 E$State: Shared E$parity 0x04 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x00): 0xc6526000.8938e01f Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 989652 kern.info] [AFT2] E$Data (0x08): 0x81800004.01008000 *Bad* PSYND=0x0002 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x10): 0x01000000.01000000 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x18): 0x88f8e064.2e800002 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x20): 0x09200000.88592064 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x28): 0x8620c004.8600e064 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x30): 0x8608e0ff.c62aa001 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 359263 kern.info] [AFT2] E$Data (0x38): 0x1080001e.c64a6002 Apr 20 14:58:36 sccpdb01 SUNW,UltraSPARC-II: [ID 623616 kern.info] [AFT2] errID 0x003d4b09.64085e0d AFAR was derived from E$Tag Apr 20 14:58:36 sccpdb01 unix: [ID 321153 kern.notice] NOTICE: Scheduling clearing of error on page 0x00000000.fabec000 Apr 20 14:58:46 sccpdb01 unix: [ID 221039 kern.notice] NOTICE: Previously reported error on page 0x00000000.fabec000 cleared Apr 20 14:58:46 sccpdb01 SUNW,UltraSPARC-II: [ID 788098 kern.info] [AFT3] errID 0x003d4b09.64085e0d Above Error is in User Mode Apr 20 14:58:46 sccpdb01 and is fatal: will reboot ------------------------------------------------------------------------ ------------------------------------------------------------------------ ------------------------------------------------------------------------ ------------- Does anyone know what that error means? CPU, Memory or Board problem? Any help in this will be greatly appreciated. Thanks 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 chocobofrank at hotmail.com Thu Apr 22 12:26:19 2004 From: chocobofrank at hotmail.com (Frank Cheong) Date: Thu, 22 Apr 2004 16:26:19 -0000 Subject: SUMMARY backup "Device busy" References: <200404221329.i3MDTPJ11560@babbage.cs.umsl.edu> Message-ID: Yeah, thx very much ! I have overlooked...... and killed the tar process (not running for days) and everything is working again now. Frank ----- Original Message ----- From: "Michael Schulte" To: Sent: Thursday, April 22, 2004 9:29 PM Subject: Re: backup "Device busy" > > Recently, my tape drive failed to response with "Device Busy" whenever I tried > > to touch on the tape drive by just any command include "mt rewind" or "tar > > clvf /dev/rmt0" > > > > I have switched the tape drive off and then on again but problem stilll exist. > > > > I was wondering if there is any command to reset the tape related modules on > > the OS level without restarting the machine. > > You might also look for running processes that reference the drive. > I have had this problem if a backup fails in starting (because I forgot the > tape!); the tar process was still running in the background, and had > taken control of the device and wouldn't let go without a kill. > > --- > 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 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From antonia at fib.upc.es Fri Apr 23 06:00:48 2004 From: antonia at fib.upc.es (Antonia Gomez) Date: Fri, 23 Apr 2004 10:00:48 -0000 Subject: SUMMARY "problems sunscreen" Message-ID: <4088EC6D.6010805@fib.upc.es> Thanks! The solution is: ssadm with command authuser -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Antonia Gomez Gonzalez FIB (Laboratori de Calcul) UPC Barcelona ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ahoesch at smartsoft.de Fri Apr 30 15:03:44 2004 From: ahoesch at smartsoft.de (Andreas Hoeschler) Date: Fri, 30 Apr 2004 19:03:44 -0000 Subject: Summary: Getting Sun Ray to work In-Reply-To: <0EBFC22FDE33C84B97D546599FFC449101C9119A@STL-EXBE-I03.corp.erac.com> Message-ID: <09A9A753-9AD9-11D8-B0C8-000393CA0072@smartsoft.de> Dear managers, special thanks to Jarett Stevens Lebar, Rusell Loukins, Jeremy Tim, Chipman I finally got the Sun Ray to work with th followin approach cd /opt/SUNWut/sbin ./utadm -D 192.168.1.0 ./utadm -r ./utadm -A 192.168.1.0 ./utadm -c ./utrestart pntadm -P 192.168.1.0 Using ./utadm -a eri0 was not a good idea as it turned out. After getting the Sun Ray working one can set the resolution with ./utsettings ./utxconfig -r 1024x768 -s off Works great. Unfortunately the setting of the resolution does not survive a system reboot. But I can live with that for a while... Regards, Andreas _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Andrew_Rotramel at cch-lis.com Thu Apr 29 11:36:51 2004 From: Andrew_Rotramel at cch-lis.com (Andrew_Rotramel at cch-lis.com) Date: Thu, 29 Apr 2004 15:36:51 -0000 Subject: SUMMARY: Read OK, Write Errors Message-ID: Turned out it was the switch. I connected the array directly to the server, copied 13 GB, and got no errors. Thanks to Tim Chipman for his suggestion that the disks in the arrays may have bad sectors. Thanks to Alan Pae for his suggestion that I talk to the vendor, which I had already done. It was the vendor who gave confirmed that my idea of removing the switch from the equation was a good thing to test. Andrew Andrew Rotramel 04/28/2004 04:19 PM To: sunmanagers at sunmanagers.org cc: Subject: Read OK, Write Errors I am having trouble writing to, but not reading from, an array. It is a Raid 5 array conntected to an E450 running Solaris 2.6. The connection goes from the server, through an Antares P-036 fiber card, through a copper cable, to a Gadzooks 1 GB fiber switch, through a copper cable, to an Infortrend 2200 array controller with internal SCSI disks. Copying TO the array from another array gives me this error about once per minute. "SCSI transport failed: reason 'aborted' The data seems to copy fine, but with that error. Copying TO one array partition to another array partition on the same array gives me the same error. (Solaris sees an array partition as a hard drive). Copying FROM the array gives me no error. Since the copy FROM is OK, I am guessing that the HBA, switch and cables are OK, and the problem is with the array itself. Any suggestions on how to troubleshoot this? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ahoesch at smartsoft.de Fri Apr 30 15:42:48 2004 From: ahoesch at smartsoft.de (Andreas Hoeschler) Date: Fri, 30 Apr 2004 19:42:48 -0000 Subject: Summary: SunPCI III on SunBlade 100 In-Reply-To: <1083351951.3140.31.camel@mach2> Message-ID: <7FAA66E6-9ADE-11D8-B0C8-000393CA0072@smartsoft.de> Dear managers, > use m64config Yeapp! That was the command. Thanks a lot!!! Regards, Andreas >> How can I alter the resolution and the color depth on a SunBlade 100? >> I >> used to know that but I currently don't remember the procedure. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From cbar44 at tsg.cbot.com Thu Apr 8 13:17:35 2004 From: cbar44 at tsg.cbot.com (Christopher L. Barnard) Date: Thu, 08 Apr 2004 17:17:35 -0000 Subject: SUMMARY: 9 character passwords invalid? In-Reply-To: References: Message-ID: I asked: > With the advent of Solaris 9 and the end of the 8 character limitation on > passwords, I have started to gradually increase the length of the passwords. > I am doing this slowly to make sure nothing breaks. > > Well, I think something has. > > On the few solaris 9 servers that have 9 character root passwords, it is > impossible to go to single user mode. When prompted for the root password > to enter single user mode, the 9 character password is not recognized. > The first 8 characters of that 9 digit password were also not recognized. > It was necessary to bring the box back up to multi-user mode, change the > root password to an 8 character password and then go into single-user > mode. The server accepted the new 8 character password just fine. > > btw, on these servers I have changed the default algorithm from DES > (__unix__) to blowfish. However changing the password, still in blowfish, > to 8 characters resolved the problem so I do not believe its the algorithm. > > So has anyone else seen this behavior where on a Solaris 9 server a root > password of more than 8 characters is not honored when entering > single-user or maintenance mode? TIA. The solution: Sun Alert number 57463. We actually solved the problem last night, since that Alert specifies a patch needed and that was one of the patches we installed. A one-sentence summary of SA 57463: install 112874-06 or higher and 113482-02 or higher. Thanks to: Hendrik Visage Darren Dunham Sun Microsystems Support +-----------------------------------------------------------------------+ | Christopher L. Barnard O When I was a boy I was told that | | cbarnard at tsg.cbot.com / \ anybody could become president. | | (312) 347-4901 O---O Now I'm beginning to believe it. | | http://www.cs.uchicago.edu/~cbarnard --Clarence Darrow | +----------PGP public key available via finger or PGP keyserver---------+ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Alexandre.Dery at sct.gouv.qc.ca Thu Apr 8 15:58:04 2004 From: Alexandre.Dery at sct.gouv.qc.ca (Alexandre =?iso-8859-1?Q?D=E9ry=2Fdgt=2Fsct?=) Date: Thu, 08 Apr 2004 19:58:04 -0000 Subject: SUMMARY : Sol9 - VxVM 3.5 : Log plex problem after disk replacement Message-ID: Hi, Many thanks to Matt Wanke for his response : Original problem was that one of my log plex was "STALE" : ---- v dsa1vol01-L01 fsgen ENABLED 71114752 - ACTIVE - - pl dsa1vol01-P01 dsa1vol01-L01 DETACHED LOGONLY - STALE - - sd dsa1_temp2-01 dsa1vol01-P01 DETACHED 594 LOG RECOVER - - ---- And I couldn't attach it to my volume (with vxplex att volume plex) Solution : Delete and recreate the log plex, with these commands : ---- vxplex -g DSA1dg dis dsa1vol01-P01 vxedit -g DSA1dg -rf rm dsa1vol01-P01 vxassist -g DSA1dg addlog dsa1vol01-L01 nlog=1 logtype=drl vxedit rename dsa1vol01-L01-01 dsa1vol01-P01 ---- Side note : The reason I got this problem is because I have 2 servers connected to a A5200 array. I did the disk change using the first server, and everything was fine. Then I imported the diskgroup/volume on the second server just to make sure it worked there : it didn't work because the second server's path to the replaced disk was wrong, so every plex/subdisk on the disk went it "stale" state. So I switched back the diskgroup to the first server. Then, the data plex was recovered, but the log plex stayed in "stale" state, and I fixed it using the above commands. Then I did a "reboot -- -r" on my second server, so that all the device links were ok. And then I was able to import successfully my diskgroup to my second server. Thanks, Alexandre Diry Technicien, SCSA, CCNA Direction des services d'inforoute 1500-E, rue Jean-Talon Nord 1er itage Sainte-Foy (Quibec) G1N 4T6 Tiliphone: (418) 646-9273 Tilicopieur: (418) 528-0418 [demime 1.01b removed an attachment of type image/gif] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bshah at citadon.com Thu Apr 8 18:12:54 2004 From: bshah at citadon.com (Bhavesh Shah) Date: Thu, 08 Apr 2004 22:12:54 -0000 Subject: SUMMARY: NFS Security in Solaris 8 Message-ID: <0FEA43415756E44AB368AC58C1A4F4F3021B7431@a4mail2k01.citadoncorp.com> Thanks to all who replied: Darren Dunham Anatoliy Lisovskiy Driddick Mark Cain Special Thanks to Darren Dunham which pointed me to the right direction towards solving the problem quickly. Solution was: Reverse lookup was not setup correctly in DNS. Once that done everything started working with FQDN. My Original Question was Hi Gurus, I have a NFS Security question: I am running Solaris 8. Following is my /etc/dfs/dfstab entry: /usr/sbin/share -F nfs -o rw=host1:host2,anon=0,ro /dir even then host1 or host2 can not have write access to /dir if I try to touch a file to host1 or host2 it gives an obvious error :"touch: test cannot create" but if I replace host1 and host2 with ip addresses everything works fine. I can ping to host1 and host2 from NFS Server box. Was wondering how that can happen? I changed nsswitch.conf to search for dns first and then files but no success. I guess there is some configuration problem on NFS Server. Any Help will be greatly appreciated. Thanks in advance B _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From alan at motorola.com Thu Apr 8 19:03:09 2004 From: alan at motorola.com (McIntosh Alan-R54071) Date: Thu, 08 Apr 2004 23:03:09 -0000 Subject: SUMMARY: NIS map transfer problems Message-ID: <7FD24C15A06DD511BF9E00D0B73E9952109C7C15@az33exm05.corp.mot.com> Thanks for the input from Jerry Kemp, Richard Felkins, Darren Dunham and about a dozen folks lucky enough to be out of the office! As it turns out, this domain(which is prehistoric) utilizes a wrapper script for ypxfr to pull maps from the appropriate master(there are two). When an update of some support scripts was installed on each of the slaves several weeks back, it wiped out the wrapper script and replaced it with a new copy of ypxfr. After reinstalling the wrapper script, everyone is happy again(especially me!). - Alan -----Original Message----- From: McIntosh Alan-R54071 Sent: Thursday, April 08, 2004 1:43 PM To: Sunmanagers (sunmanagers at sunmanagers.org) Subject: NIS map transfer problems All, Recently, NIS map transfers have been failing for various slaves within our environment. When a make is attempted on the passwd map from the master, the following error occurs: root at master :> ./make passwd updated passwd yppush listener: Garbage transaction status (value 3) from ypxfr on slave8. yppush listener: Garbage transaction status (value 3) from ypxfr on slave11. yppush listener: Garbage transaction status (value 3) from ypxfr on slave15. yppush listener: Garbage transaction status (value 3) from ypxfr on slave8. yppush listener: Garbage transaction status (value 3) from ypxfr on slave11. yppush listener: Garbage transaction status (value 3) from ypxfr on slave15. pushed passwd Not all maps are affected by this problem, the passwd map is the only one I have noticed as a matter of fact, probably because it is the most affected to users. When attempting to pull the map to the host from the slave, the following error rears it's ugly head: root at slave8 :> /usr/lib/netsvc/yp/ypxfr -f passwd.byuid (info) Can't get order number from ypserv at master. Reason: can't bind to a server which serves domain. The master is running Solaris 2.6 and the slave is running Solaris 7, but I doubt that matters much, since it has been fine for four or five years, and most other maps still update. Thanks in advance for any help!! Later, Alan _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From carlos_sevillano at ureach.com Tue Apr 13 13:10:48 2004 From: carlos_sevillano at ureach.com (Carlos Sevillano) Date: Tue, 13 Apr 2004 17:10:48 -0000 Subject: SUMMARY: X11 Forwarding with OpenSSH Message-ID: <200404131704.NAA15016@www22.ureach.com> Thanks to those who replied. I seem to understand the nature of the problem and are working to implement the scripts and suggestions: Susan Hanna" From Susan: http://www.tldp.org/HOWTO/Remote-X-Apps-7.html X Applications from Another User-id >From Alek: The Problem: When you login you get an $DISPLAY of "localhost:10.0" or something like that. This is an identifier for a specific ssh tunnel. If you do a 'su - user' then you also change the enviroment var $DISPLAY. You have to set it to the old value, or you cannot connect to your local x-server. The next Problem is the X-Auth-Cookie. If you change your user (with enviroment [su - user]) your default x-Auth will change! There a two ways to fix this Problem: 1.) Anytime you change a user, do it WITHOUT enviroment!!!! This will save your original $DISPLAY and X-AUTH-Cookie. 2.) If you need to change the enviroment to the new user then write a script (place the code in .bash_profile/.profile/.kshrc) that print you origin $DISPLAY var into a file. Also copy your origin x-AUTH. After 'su -' or something simillar wrap your new $DISPLAY to the old one which is saved in the file. (try this first, if it doesn't work - continue). - Temporary copy your origin X-Auth-Cookie to the one of the new user. DON'T FORGET TO SAVE THE ORIGIN! And copy the origin back on logout. Original Posting was: Users unable to do X11 forwarding once the switch to another user via PPAK, su, or sudo. ________________________________________________ Get your own "800" number Voicemail, fax, email, and a lot more http://www.ureach.com/reg/tag _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bshah at citadon.com Tue Apr 13 15:24:38 2004 From: bshah at citadon.com (Bhavesh Shah) Date: Tue, 13 Apr 2004 19:24:38 -0000 Subject: SUMMARY: Find to exclude dirs/subdirs Message-ID: <0FEA43415756E44AB368AC58C1A4F4F3021B7443@a4mail2k01.citadoncorp.com> I got two replies and very thankful to them for their valuable suggestions: Russell It might be easier to create the list of directories with find, and then process it with sed to remove the ones you don't want. If the list isn't all that big, and you don't need to do this very often, it may be worthwhile fixing it in an editor. Oscar would use the find command to list the directories and then pipe the output into grep or xargs and filter the directories there. You would issue a mkdir command only on the ones that satisfied the criteria But I figured out the solution: To Exclude DIR1, DIR2 and DIR3 from list simply use: find / -type d \( -name DIR1 -o -name DIR2 -o -name DIR3 \) -prune -o -type d -print My Original Question was: Hi Gurus, I have a shell script which creates a Dir structure to newly created Sun Box from a File which has Dir List. I would like to create a New File with a complete Dir structure excluding certain Directories (Only Dirs not files) If I use find . -type d will give me a recursive list of Dirs in a current Dir which is fine. find . -type d ! -name '[0-9]*' will exclude Dirs which starts with numbers but to exclude mutilple dirs find . -type d ! -name '[0-9]*' -o -type d ! -name . -o -type d -name NT doesn't work neither find . -type d ! \( -name '[0-9]*' -o -name NT -o -name .snapshot \) -print I tried -prune switch as well but it lists files as well besides directories. Also tried ls -d * / but it gives dir list in current Dir only not the recursively My question is: How can I exclude multiple directories with find command? I am running Solaris 8 Do I need any other tool ? or which is a best way to create a dir list recursively in a current dir without files listing? Any Help will be greatly Appreciated. Thanks in advance B _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From aballestas at lycos.com Tue Apr 13 18:17:03 2004 From: aballestas at lycos.com (Andres Ballestas) Date: Tue, 13 Apr 2004 22:17:03 -0000 Subject: SUMMARY : Blade-Cluster Message-ID: Hello, Thanks all people sent your comments, the resume. For free clustering I think you'll need to use linux rather than Solaris. For cheap(ish) clustering look at RSF-1 from www.high-availability.com. Then again you could always buy a couple of second hand VAX machines. ;-) by Joe Fletcher It's more difficult to do useful work without some form of sharedstorage between the two, but you might check out FreeHA. by Darren Dunham Un grid engine 5.3 is free. I will test it in the future by Gerard Henry OpenMOSIX cluster freeware, and see their documents. by Sundar You could do a cold standy type cluster. You could write a script that watches for a certain event then trigger an app shutdown on server A and a start on B. You won't be able to use like Sun cluster or Veritas cluster. Sun has a PDF on creating a Beowolf cluster .... www.sun.com/blueprints by Jeremy Loukinas Try http://www.fstha.com by Reggie Beavers However, OpenBSD on Sparc should be able to run CARP failover, if you're doing web or something it might work.. by Paul Greidanus I choose a few options, I will send my comments. --- Andres-B. ____________________________________________________________ Find what you are looking for with the Lycos Yellow Pages http://r.lycos.com/r/yp_emailfooter/http://yellowpages.lycos.com/default.asp?SRC=lycos10 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From bshah at citadon.com Fri Apr 23 15:22:34 2004 From: bshah at citadon.com (Bhavesh Shah) Date: Fri, 23 Apr 2004 19:22:34 -0000 Subject: SUMMARY: E 4500 Hardware Failure Message-ID: <0FEA43415756E44AB368AC58C1A4F4F301F9DD9B@a4mail2k01.citadoncorp.com> Well got one reply besides out of office mails. Grateful to Anjan Dave for his quick and accurate analysis. Its Memory in 4 Banks that went bad at the same time - Strange Will replace the Memory and see how it goes. Thanks Anjan My Original Question Was: Hi Gurus, Sorry for troubling once again but we had one more E 4500 Hardware failure. Would appreciate it if some can guide me what exactly is failed as this Server is located at remote location. 1) Board 0 and 2 2) Memory on Board 0 and 2 - Bank J3201 and J3101 or all memories on board 0 and 2 3) CPUs on board 0 and 2 4) All Boards( 0 & 2) CPUs and Mem Thanks in advance B Below is the prtdiag output ========================= CPUs ========================= Run Ecache CPU CPU Brd CPU Module MHz MB Impl. Mask --- --- ------- ----- ------ ------ ---- 0 0 0 400 8.0 US-II 10.0 0 1 1 400 8.0 US-II 10.0 2 4 0 400 8.0 US-II 10.0 2 5 1 400 8.0 US-II 10.0 4 8 0 400 8.0 US-II 10.0 4 9 1 400 8.0 US-II 10.0 ========================= Memory ========================= Intrlv. Intrlv. Brd Bank MB Status Condition Speed Factor With --- ----- ---- ------- ---------- ----- ------- ------- 0 0 1024 Failed Unknown 60ns 1-way 0 1 1024 Failed Unknown 60ns 1-way 2 0 1024 Failed Unknown 60ns 1-way 2 1 1024 Failed Unknown 60ns 1-way 4 0 1024 Active OK 60ns 2-way A 4 1 1024 Active OK 60ns 2-way A ========================= IO Cards ========================= Bus Freq Brd Type MHz Slot Name Model --- ---- ---- ---------- ---------------------------- -------------------- 1 SBus 25 0 QLGC,isp/sd (block) QLGC,ISP1000U 1 SBus 25 1 SUNW,qfe SUNW,sbus-qfe 1 SBus 25 1 SUNW,qfe SUNW,sbus-qfe 1 SBus 25 1 SUNW,qfe SUNW,sbus-qfe 1 SBus 25 1 SUNW,qfe SUNW,sbus-qfe 1 SBus 25 2 QLGC,isp/sd (block) QLGC,ISP1000U 1 SBus 25 3 SUNW,hme 1 SBus 25 3 SUNW,fas/sd (block) 1 SBus 25 13 SUNW,socal/sf (scsi-3) 501-3060 Detached Boards =============== Slot State Type Info ---- --------- ------ ----------------------------------------- 3 disabled disk Disk 0: Target: 10 Disk 1: Target: 11 Failed Field Replaceable Units (FRU) in System: ============================================== ac unavailable on CPU Board #0 PROM fault string: fail- mem-bank0: J3200 mem-bank1: J3201 Failed Field Replaceable Unit is CPU board 0 ac unavailable on CPU Board #2 PROM fault string: fail- mem-bank0: J3200 mem-bank1: J3101 Failed Field Replaceable Unit is CPU board 2 Detected System Faults ====================== PROM detected failure Detected Fri Apr 23 00:36:18 2004 Board 2 fault: PROM detected Fault: fail- mem-bank0: J3200 mem-bank1: J3101 Detected Fri Apr 23 00:36:18 2004 Board 0 fault: PROM detected Fault: fail- mem-bank0: J3200 mem-bank1: J3201 Detected Fri Apr 23 00:36:18 2004 Most recent AC Power Failure: ============================= Mon Oct 6 19:20:18 2003 ========================= Environmental Status ========================= Keyswitch position is in Normal Mode System Power Status: Redundant System LED Status: GREEN YELLOW GREEN WARNING ON ON BLINKING Fans: ----- Unit Status ---- ------ Rack OK Key OK AC OK System Temperatures (Celsius): ------------------------------ Brd State Current Min Max Trend --- ------- ------- --- --- ----- 0 OK 35 35 35 stable 1 OK 35 35 35 stable 2 OK 37 37 38 stable 4 OK 40 38 41 stable CLK OK 24 24 24 stable Power Supplies: --------------- Supply Status --------- ------ 0 OK 1 OK 2 OK PPS OK System 3.3v OK System 5.0v OK Peripheral 5.0v OK Peripheral 12v OK Auxilary 5.0v OK Peripheral 5.0v precharge OK Peripheral 12v precharge OK System 3.3v precharge OK System 5.0v precharge OK AC Power OK ========================= HW Revisions ========================= ASIC Revisions: --------------- Brd FHC AC SBus0 SBus1 PCI0 PCI1 FEPS Board Type Attributes --- --- -- ----- ----- ---- ---- ---- ---------- ---------- 0 1 5 CPU 100MHz Capable 1 1 5 1 1 22 Dual-SBus-SOC+ 100MHz Capable 2 1 5 CPU 100MHz Capable 3 Unknown 100MHz Capable 4 1 5 CPU 100MHz Capable System Board PROM revisions: ---------------------------- Board 0: OBP 3.2.30 2002/10/25 14:03 POST 3.9.30 2002/10/25 14:04 Board 1: FCODE 1.8.29 2001/06/18 17:26 iPOST 3.4.29 2001/06/18 17:49 Board 2: OBP 3.2.30 2002/10/25 14:03 POST 3.9.30 2002/10/25 14:04 Board 4: OBP 3.2.30 2002/10/25 14:03 POST 3.9.30 2002/10/25 14:04 # _______________________________________________ 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 adam at 68e.com Mon Apr 26 14:39:15 2004 From: adam at 68e.com (Adam Mazza) Date: Mon, 26 Apr 2004 18:39:15 -0000 Subject: SUMMARY: mirroring with Jumpstart on Solaris 9 04/04 In-Reply-To: References: Message-ID: Sorry for the late summary, but just got an answer from Sun. This is a bug, you currently cannot use the 'free' or 'auto' keyword along with the mirror keyword. Regards, Adam Mazza On Mon, 5 Apr 2004, Adam Mazza wrote: > Hello, > > I'm trying to mirror two disks during a jumpstart setup with Solaris 9 > 04/04. The check script seems to choke on the "free" option when I use it > for slice size. Here's an example: > > metadb c0t0d0s5 > metadb c0t1d0s5 > partitioning explicit > filesys mirror c0t0d0s0 c0t1d0s0 3072 / logging > filesys mirror c0t0d0s1 c0t1d0s1 1024 swap > filesys mirror c0t0d0s3 c0t1d0s3 512 /var logging > filesys mirror c0t0d0s4 c0t1d0s4 3072 /export/home logging > filesys mirror c0t0d0s6 c0t1d0s6 3072 /usr logging > filesys mirror c0t0d0s7 c0t1d0s7 free /opt logging > > I checked on docs.sun.com, and didn't see mention of 'free' for the size > variable anymore. Is this no longer supported? > > Regards, > > Adam Mazza > _______________________________________________ > 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 andy.ford at telindus.co.uk Wed Apr 28 04:33:18 2004 From: andy.ford at telindus.co.uk (Andy Ford) Date: Wed, 28 Apr 2004 08:33:18 -0000 Subject: SUMMARY: UNIX fifo file (off the wall) In-Reply-To: <1083079595.2807.14.camel@evo@telindus.co.uk> References: <1083079595.2807.14.camel@evo@telindus.co.uk> Message-ID: <1083018362.2398.8.camel@evo@telindus.co.uk> Thanks to all that responded. Suggestions were to use mkfifo or mknod. Both do exactly the same with (with a few parameters!!) I tested the fifo as follows stage 1: mkfifo -m 0666 /tmp/file stage 2: perl program #!/usr/bin/perl -w $count=0; while(1) { open(FIFO, ">/tmp/file") or die "Couldn't open log for writing: $!\n"; print FIFO $count."\n"; close FIFO; $count++; } stage 3: tailed the file result: the file stayed 0 in size and displayed the increasing count when tailed - perfect Thanks Andy On Tue, 2004-04-27 at 16:26, Andy Ford wrote: > Is there such a thing on Solaris as a fifo file? > i.e. a file that will take only one entry so when its tailed, the > content will scroll up the screen, but the file will not grow in size. > > Thanks > > Andy -- perl -e 'print qq^;@) [###]^^qq^z\.MY{eLQ9^' in:control developer, Telindus, RG27 9HY DDI: +44 1256 709211, GSM: +44 7810 636652 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From choogend at library.umass.edu Wed Apr 28 14:11:37 2004 From: choogend at library.umass.edu (Chris Hoogendyk) Date: Wed, 28 Apr 2004 18:11:37 -0000 Subject: SUMMARY: why so many ports open on Solaris In-Reply-To: <408F0C9F.4030603@library.umass.edu> References: <408F0C9F.4030603@library.umass.edu> Message-ID: <408FF35F.8050405@library.umass.edu> boatload of replies. thanks to everyone. it looks like I have some serious work to do. I'll do a very brief summary, but the only way to do justice to the richness of the replies is to include a number of them at the end of this, after my original message. key item that one person gave me was that the install option you choose is important. I had installed the entire distribution, figuring that the programming tools would be there. I should have installed the minimal system for servers. next is to clean out rc2.d and rc3.d in addition to inetd.conf. lots of things started from there. of course, I had done a 'kill -HUP' of the inetd process, but it is also necessary to restart the system or kill processes that have already been started. in this case, I had actually restarted the system. tools: -- lsof widely recommended. get it from sunfreeware or from purdue. 'lsof -i' gives ports and processes. can pipe to grep. http://www.sunfreeware.com -- jass, Sun's security tightening tool. get it from Sun. it's a script. can read it. can modify it. can just run it. http://www.sun.com/security -- someone pointed to a setup_rc script that removes all the stuff you don't want. run it again after doing patches, because patches can put startup scripts back in. I had already encountered this with sendmail. I get rid of it, do recommended patches, then have to get rid of it again. best to completely uninstall stuff you don't want so a startup script won't find it anyway. lots more detail in the replies, and a chuckle or two. Thanks again to everyone. --------------- Chris Hoogendyk - O__ ---- Network Specialist & Unix Systems Administrator c/ /'_ --- Library Information Systems & Technology Services (*) \(*) -- W.E.B. Du Bois Library ~~~~~~~~~~ - University of Massachusetts, Amherst --------------- -------- My Original Question -------- Subject: why so many ports open on Solaris Date: Tue, 27 Apr 2004 21:45:03 -0400 From: Chris Hoogendyk To: Sun Managers Why does Solaris (e.g. 8) have so many ports open even when I've gone through inetd.conf and commented out virtually everything? I've got several web guides to securing Solaris. I've seen the SysAdmin Magazine articles on locking down Solaris. I've done all that stuff. But I still have ports open whose purposes and sources I don't understand. Does anyone know where there is a guide or discussion of the absolute minimum necessary and what you lose or don't lose by shutting down everything else? I don't want to use a port blocking mechanism. I use tcpwrappers to regulate access to ports that I do want open. It seems I should find the source of excess ports and actually shut down the processes that are opening them. I presume a lot of them come from rc2.d or rc3.d. I'm getting hammered by some folks who think I should only have about 2 ports open. TIA -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Tue, 27 Apr 2004 21:55:15 -0400 From: Chris To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> Couple tips for you. It takes forever to manually go figure out all the processes that are running with listening ports. To save a TON of time, check www.sun.com/security and get the "JASS" script they have there for free. Take a moment to read through it and then run it. It will lock the box down but good. Actually just a heads up, it will leave NO means to connect to the box over the network and lock out root login from anything but the console. That is the default. You can customize your own 'profile' to chose what it leaves running or not once you get used to how the script works. Another tip, www.sunfreeware.com , get 'lsof' it's there as a sun package. lsof 'lists open files' including network connections. You can find out what user/process owns any open files or network sockets on a running system. It's handy for what you're doing, it's also handy to figure out what process is hanging on to a mounted filesystem when you try to unmount it (CD or floppy in particular). For general box security, there's a mildly useful utility called ASET, check into that as well, it handles locking down the ridiculously wide open file permissions on a system. It is a script as is JASS. -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 10:02:21 +0100 From: Simon Crowther To: choogend at library.umass.edu Chris, You have to consider that the Solaris target audience is very broad, from workstation users through developers to large server environments. Some of these users will not have a great Sys Admin background or knowledge, and these users especially will want a more no hassle approach to installations where products and services are installed and running that might be integral to a 3rd party application, With so many 3rd party apps out there having differing dependancies, its no wonder there is an "all lights on" approach... Solaris does address this to a degree, by having different install options, packages are clustered in the following fashion: Core install End User System Support Developer System Support Entire Distribution Entire Distribution + OEM The core install is considered A minimum package set required which is supported by SUN (this may have changed now, since the popularity of the Sun Blueprints Minimisation Document which describes hardening techniques and further package removal) The Entire Distribution + OEM installs a great deal of product and services. The considerations for what should be running and what should not are dependant on the intended end use of the machine. For instance, a back-end server that runs a database which serves a web site may only have SSH and Oracle related daemons listening. The folk you speak of are right in principal, as you should attempt to configure your servers to serve only the services that make up it's intended use. some people achieve this by placing a host based firewall on the server or by setting TCP Wrappers and editing inetd.conf (which is similar to installing a host based firewall) and others will go for a "Defense in depth" approach... So the big Qn is HOW? This has been covered by many Docs and articles out on the web, but limiting factors are so often time and/or experience. A good starting point is Suns Blueprints which can be found here: http://www.sun.com/solutions/blueprints/browsesubject.html In particular....(this one is solaris 9) http://www.sun.com/blueprints/1102/816-5241.pdf Other examples of minimisation work can be found here: http://www.spitzner.net/ also there are many varied documents here: http://www.securityfocus.com/infocus/unix Good resources to be found here: http://www.stokely.com/unix.sysadm.resources/faqs3.sun.html#perf.tun and a good step by step document here: http://www.filibeto.org/sun/lib/security/hardening_solaris_v0.86.pdf It will take time for you to develop safe and solid techniques, but the more you put in, the more you will get out ;-) Hope this helps, Simon Crowther. -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Tue, 27 Apr 2004 21:27:38 -0700 From: Ric Anderson To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> Depends on the use of the machine. rpcbind services (like ttdbserver) run on workstations, but are not needed on servers. Make darn sure you have * Security fix - prevent execution on stack... set noexec_user_stack=1 set noexec_user_stack_log=1 in /etc/system, and you rebooted since you put those lines there; that will stop most of the crap (if you are running on Sparc hardware). The Intel lovers have no hardware equivalent protection, as the pentium and lower chips don't differentiate between stack read and stack execute on a per-page basis. Itaniums might have fixed that, but I don't know for sure. Sort of normal open ports are 22(ssh), 25 (smtp), 111 (RPC), 4045 (lockd), and 3277x (rpc services, like statd and dtlogin). If a windowing server is running, port 6000 (X11) will show up also. This is about as far as I trim my machines. I could, with more work, turn off sendmail, and run it from cron to make sure no outbound messages get queued up for any length of time, and kill off dtlogin. However, since all my boxes are either NFS clients (to mount home dirs) or NFS servers (or both), I can't get rid of rpcbind, statd, and lockd. In a non-NFS, non-console windowing world you could hack the startup scripts to eliminate those boxes, but you'll then have to deal with patch installs unding your work, or failing because you touched those scripts in some cases, so approach with caution. Cheers, Ric Anderson (ric at opus1.com) -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 09:56:33 +0100 From: Simon Burr To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> I tend to just comment out all of inetd before sending it the HUP. You have two options; one is to install IP-Filter which provides router-like ACLs on a per network interface. That will guarentee that even if a port is open, no one can reach it; this assumes that IP-Filter is configured correctly tho; you can get IP-Filter from http://coombs.anu.edu.au/~avalon/ I've got a couple of scripts which I run on servers which lock them down quite nicely. The first job I do is remove a gaggle of packages which I don't need or replace with others - a good example of this is removing the sendmail packages (replaced by PostFix) as sendmail has a habit of being re-enabled after patch clusters have been applied. The other job is to then disable certain startup scripts in /etc/rc2.d and /etc/rc3.d; personally I do this by prepending "no." to the start of the file names. The scripts are: ## Remove certain packages cat > /tmp/pkgrm-admin < Reply-To: ssandau at bath.tmac.com To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> If you never run CDE or any other GUI, you can have like 2 ports open. I do this on an Oracle server on Sol 8. CDE, Gnome and so on open up (need?) many ports to start with. In addition many other optional services run out of scripts in /etc/rc2.d. I can't give you a list, but I have tracked many down in the past by reading the script and looking at the man page for the particular binary. I think that KDE, Gnome and others open lots of ports on Linux as well. Really minimal ports open is related to the window manager, not the OS. My opinion anyway... ;) SteveS -------- Original Message -------- Subject: RE: why so many ports open on Solaris Date: Wed, 28 Apr 2004 08:31:30 -0400 From: William Enestvedt To: Chris Hoogendyk Well, not everyuthing listening on a port gets started via inetd, if I recall correctly. (See the instructions for installing TCP Wrappers: the explanation of the two methods of installation might shed more light on this than I can.) Also, you restarted inetd after changing its conf file,right? :7) SANS publishes a book about securing Solaris that's quite good; if you read through it, it explains why certain services are being disabled -- but I must confess that it wants you to accet their assurances pretty blindly. I have taken to disabling a lot of the things in /etc/rc2.d and rc3.d, but I try to read the man pages to figure out whether I can get by without them (like picld, which I'd love to shut off but which I *think* is required by Solaris) before I kill them. Suns "JASS Toolkit" for securing Jumpstarting Solaris systems contains scripts for securing various services and ports. You could probably glean a lot from reading the supporting paper on the Sun Blueprints site. I think many Linux distributions use xinetd to start more services/deamons/processes than Solaris does, which is why they can rely on keeping more things disabled by default (feeling safe that the right stuff will get launched when it tickles xinetd). But I could be wrong. -wde -- Will Enestvedt UNIX System Administrator Johnson & Wales University -- Providence, RI William.Enestvedt at jwu.edu -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 10:58:57 -0400 From: Andrew J Caines Reply-To: Andrew J Caines Organization: H.A.L. Plant To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> Chris, You should have exactly as many ports open as are used by the network services you want offered by the system, no more. inetd is just one handler of network services. Solaris comes with a bucketload of other server which start by default on install, or after patching which puts the start scripts back. You are expected to manually turn them off, or better not install the software in the first place. See the setup_rc script[1], which makes the process of removing all unwanted startup scripts. Run it after install and patching. Consider removing the packages containing the software you don't use. Since you didn't mention any details, you need to find out what's listening on those ports. I suggest using "lsof -i" and looking for processes in a LISTEN state on each port. You can look for the process listening on a particular port by specifying it, eg. # lsof -i :22 COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME sshd 257 root 3u IPv6 0x30001e54638 0t0 TCP *:ssh (LISTEN) sshd 257 root 4u IPv4 0x30001e547b8 0t0 TCP *:ssh (LISTEN) A good reference is Alex Noordergraaf's Sun Blueprints, "Minimizing the Solaris Operating Environment for Security"[2] and "Solaris Operating Environment Minimization for Security: A Simple, Reproducible and Secure Application Installation Methodolgy"[3]. Other Blueprints will probably be of interest to you, too. [1] http://halplant.com:88/software/Solaris/scripts/setup_rc [2] http://www.sun.com/blueprints/1102/816-5241.pdf [3] http://www.sun.com/blueprints/1100/minimize-updt1.pdf -Andrew- _______________________________________________________________________ | -Andrew J. Caines- Unix Systems Engineer A.J.Caines at halplant.com -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 10:34:05 +0200 From: Gandalf el gris To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> Hi Chris By default Sun Soalris come with a lot of open services. If you want to close these services you can use a Security tool like JASS or Titan, or make it by your self with a guide, a very good book about that is syngress Hard Proffig Sun Solaris. With JASS you can harden your sistem clossing almost all open ports, or securizing them. JASS is a Sun developed software and is the tool that SUN use to harden their systems. I hope this can help you. Cheers MArcos -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 10:25:14 -0400 (EDT) From: Mark Montague To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> You can install a copy of lsof. It doesn't come standard with Solaris 8, but you can get it from ftp://vic.cc.purdue.edu/pub/tools/unix/lsof Running "lsof -i" will tell you what processes are using which ports. This will tell you which /etc/init.d scripts to disable. If you are not actually using a port, you should not have it open, in my opinion. A common mistake is to leave a port open because you might need it. Turn off the service, and if you ever wind up needing it, turn it on (permanently) then. Mark Montague LS&A Information Technology The University of Michigan markmont at umich.edu -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 13:19:04 -0400 From: Rich Kulawiec To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> On Tue, Apr 27, 2004 at 09:45:03PM -0400, Chris Hoogendyk wrote: > Why does Solaris (e.g. 8) have so many ports open even when I've gone > through inetd.conf and commented out virtually everything? The glib answer: Because Sun ships systems that way in order to minimize support costs to them: otherwise they'd be fielding an endless stream of "Why doesn't FOO work?" calls. Compare/contrast with OpenBSD, which ships with darn near everything turned off by default. The more useful answer: Because while inetd "listens on behalf of other daemons" and thus opens those ports that those daemons provide services on, some daemons and other processes do their own listening: thus any ports that they choose to open are, uh, open. > Does anyone know where there is a guide or discussion of the absolute > minimum necessary and what you lose or don't lose by shutting down > everything else? I don't want to use a port blocking mechanism. I use > tcpwrappers to regulate access to ports that I do want open. It seems I > should find the source of excess ports and actually shut down the > processes that are opening them. I presume a lot of them come from rc2.d > or rc3.d. The best answer to this is "it depends", because which ones you can turn off without disabling a vital service depends on which services are vital to you. I can offer three bits of guidance: 1. Get lsof, as mentioned in the Sun-Manager's FAQ, because running lsof will enable you to figure who has which port(s) open. 2. Resist the temptation to disable everything at once. Again, this depends on what you're doing with your system, but even when I *know* that eventually I will probabbly end of turning off lots of things, I've found it better to take things one step at a time, and make sure -- after each change -- that everything I think should still be working IS still working. 3. Things that I find that I can often disable without screwing things up (and these are from Solaris 9, so salt to taste): nfs.client nfs.server lp keymap sendmail volmgt autofs init.snmpdx init.dmi picld skipkey ---Rsk -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Tue, 27 Apr 2004 23:34:48 -0600 From: Colin Bigam Reply-To: colin at west.gecems.com To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> Hi Chris; First of all, if services are shut down in inetd, then you'll probably find about five remaining ports open. Sendmail(port 25) is one you can shut down in Solaris 8, and still mail out stuff from that machine. nfs.client can safely be shut down if the machine won't be NFS mounting anything. The remaining few are probably RPC-related ports. It's close to impossible to shut down RPC entirely, so you'll have to look at deregistering them. Getting this far will eliminate nearly all of the open ports. As for a guide, Sun has a whitepaper on hardening Solaris/Sparc. Look that up, and you'll get quite a few interesting bits of info. Colin -- Colin Bigam Senior Unix Analyst, GEITS colin at west.gecems.com (403) 699-4584 -------- Original Message -------- Subject: RE: why so many ports open on Solaris Date: Tue, 27 Apr 2004 23:25:40 -0400 From: Roetman, Paul To: Chris Hoogendyk Sun put out this doc: Minimizing the Solaris Operating Environment for Security 816-5241.pdf Which has some quite good reading! Cheers Paul -------- Original Message -------- Subject: RE:why so many ports open on Solaris Date: Wed, 28 Apr 2004 14:39:42 +0200 From: Pavic, Aleksander To: choogend at library.umass.edu Hi, rpc Services are not handled with /etc/inetd.conf. If you really want to disable everything and open just the things you need, you have to disable the S71rpc script in /etc/rc2.d. But think about your needs, some services need rpc (like nis,nfs,) There are probably some other services that are not controled by rpc or inetd.conf. Then you have to disable the startscript for this service. To find out the startscript for a service thats called "lala" you can mostly find all scripts with 'find /etc/rc?.d | xargs grep -i lala'. HTH Aleks -------- Original Message -------- Subject: RE: why so many ports open on Solaris Date: Wed, 28 Apr 2004 08:40:15 -0400 From: Brent Mcdaniel To: Chris Hoogendyk Chris, We tie our boxes down to only ssh and whatever app is running on it, i.e. Weblogics, database, etc.... So if you have commented out almost everything in /etc/inetd.conf and HUP'd it, then the only other place would be in /etc/rc2.d and /etc/rc3.d If you want to give me a list from a "netstat -an | grep LISTEN" and "netstat -an | grep Idle", I'd be happy to tell you what ports those are and how to stop that process. Brent I n t e r c o n t i n e n t a l E x c h a n g e _____________________________________________ Brent McDaniel | http://www.intcx.com | Senior Systems Administrator cell -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 09:31:56 -0400 From: Matt Clausen To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> A lot of the inetd processes will hang around even after you restart the inetd server (either by a kill -HUP to force it to reread its configuration file or killing it all together and restarting it). If you reset the box you may find that a lot of the open ports will disappear. You can also use tools like nmap to scan these ports and it will often give you some clues as to what the ports that are open are. -------- Original Message -------- Subject: Solaris network ports open Date: Wed, 28 Apr 2004 08:31:54 -0400 From: Schernau, Ed To: 'choogend at library.umass.edu' Just install ipfilter, then they won't see any ports open. I routinely do it here, to mask my machines from prying eyes. Set up a policy to drop all but the stuff you know about. Ed Schernau Systems Management Specialist, ECC Citizens Bank, East Providence Operations Center 401.282.1262 ed.schernau at citizensbank.com -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 00:47:36 -0500 From: Kelly Setzer To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> In my crankier moods, I dream about just typing 'killall' and pronouncing the system "secure". Kelly < ;-) > -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Wed, 28 Apr 2004 00:58:37 -0500 (EST) From: J. Oquendo To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> You more than likely have some of the RPC services open. Grab yourself a copy of lsof from Sunfreeware.com if you don't have it and lsof|grep -i listening to see what exactly is accessing what port using what. Another thing you may want to do to really restrict the machine itself is looking into using ACL's if you have users, and running Titan on the machine. Titan is available for free via www.fish.com and is a pretty nifty tool. TCP Wrappers if you ask me are rather obsolete I haven't used them since about 1998 or so. Currently on my personal machine I have it modified by Titan which resolves almost 95% of the problems, I've got most known patches I need, and I have a modified version of Pitbull running on ths machine. (www.argus-systems.com) Although Pitbull is not free, it is worth picking up if you have a budget. Other tools I used are for deception. Modified DTK (Deception Tool Kit), Port Sentry. I used to run Snort to maintain awareness of who was doing what but too many false positives, and a high load on the system made me chuck it. =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ J. Oquendo -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Tue, 27 Apr 2004 19:03:06 -0700 From: Roy S. Rapoport To: Chris Hoogendyk References: <408F0C9F.4030603 at library.umass.edu> *TWO* ports? These sort of naive fools are what makes security so hard! An open port is an invitation to hacking, which is why I prefer to keep my systems with all network ports closed, superglue gumming up the serial and network interfaces, power disconnected, and the machine itself embedded in half a ton of concrete. That's the only way to be sure! Sorry :) You likely need exactly as many ports open as services you're offering, no more and no less. This likely means 1 (remote access) + whatever public services you're offering. For servers, this is quite easy -- if you've got a web server, you really should only have, say, 22 (for ssh) and 80 open. For desktops it gets a bit ugly because every full-featured desktop system out there seems to rely on network ports for some of its communication. Regardless, there are two sources for open ports on Solaris (well, and other systems): inetd will spawn ports if it's configured to do so; and server processes will always be listening on a given port. You *can* -- and *should* -- run through every process running on the machine, familiarize yourself with it, and know what it does. You *can* -- and *should* -- then go and check out JASS, the Jumpstart Architecture and Security Scripts, AKA the Solaris Security Toolkit. JASS, when integrated with Jumpstart, will result in systems that come out of the jumpstart process nicely tight. JASS is also a really nice architecture to manage Jumpstart, by the way. Hope this helps, -roy -------- Original Message -------- Subject: Re: why so many ports open on Solaris Date: Tue, 27 Apr 2004 20:57:19 -0500 (CDT) From: Mike's List To: Chris Hoogendyk So list the ports so other can see what it is and tells you where it's coming from. Yes, some ports are open with some services are enable, in /etc/rc2.d and /etc/rc3.d. ie. if you don't need /etc/rc3.d/S16boot.server, stop the process and rename the file so it won't start. www.sun.com/bigadmin --start here and search. http://www.spitzner.net/ http://www.fish.com/titan/ http://www.yassp.org/ - Mike _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Aleksander.Pavic at telekom.de Thu Apr 8 05:18:16 2004 From: Aleksander.Pavic at telekom.de (Pavic, Aleksander) Date: Thu, 08 Apr 2004 09:18:16 -0000 Subject: SUMMARY: Where to get SUNWGlib Message-ID: <5D1219B98DCF4A4080B6D27B3B1FE0D40626ED@E0-S2K-2.ads.tnetpro.de> Hi all, the solution was to unpack the netscape zip file and install SUNWGlib which is a part of netscape. Very BIG thanks to Matthew!!!!!!!! greetz Aleks -----Urspr|ngliche Nachricht----- Von: Matthew Stier [mailto:Matthew.Stier at fnc.fujitsu.com] Gesendet: Mittwoch, 7. April 2004 21:24 An: Pavic, Aleksander Betreff: Re: Where to get SUNWGlib Sun's Netscape 7 install package. I use the SUNWGlib and SUNWGtk packages to run Mozilla and FireFox on my Solaris 8 workstations. http://wwws.sun.com/software/solaris/netscape/getnetscape70.html Pavic, Aleksander wrote: >Hi all, >i need SUNWGlib as Package. > >Is there anyone who know where to get this? > >greetz >Aleks >_______________________________________________ >sunmanagers mailing list >sunmanagers at sunmanagers.org >http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > -- Matthew Lee Stier * Fujitsu Network Communications Unix Systems Administrator | Two Blue Hill Plaza Ph: 845-731-2097 Fx: 845-731-2011 | Sixth Floor Matthew.Stier at fnc.fujitsu.com * Pearl River, NY 10965 [demime 1.01b removed an attachment of type text/x-vcard which had a name of Matthew.Stier.vcf] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From adam at 68e.com Tue Apr 27 17:09:11 2004 From: adam at 68e.com (Adam Mazza) Date: Tue, 27 Apr 2004 21:09:11 -0000 Subject: SUMMARY: mirroring with Jumpstart on Solaris 9 04/04 Message-ID: Sorry for the late summary, but just got an answer from Sun. This is a bug, you currently cannot use the 'free' or 'auto' keyword along with the mirror keyword. Regards, Adam Mazza On Mon, 5 Apr 2004, Adam Mazza wrote: > Hello, > > I'm trying to mirror two disks during a jumpstart setup with Solaris 9 > 04/04. The check script seems to choke on the "free" option when I use it > for slice size. Here's an example: > > metadb c0t0d0s5 > metadb c0t1d0s5 > partitioning explicit > filesys mirror c0t0d0s0 c0t1d0s0 3072 / logging > filesys mirror c0t0d0s1 c0t1d0s1 1024 swap > filesys mirror c0t0d0s3 c0t1d0s3 512 /var logging > filesys mirror c0t0d0s4 c0t1d0s4 3072 /export/home logging > filesys mirror c0t0d0s6 c0t1d0s6 3072 /usr logging > filesys mirror c0t0d0s7 c0t1d0s7 free /opt logging > > I checked on docs.sun.com, and didn't see mention of 'free' for the size > variable anymore. Is this no longer supported? > > Regards, > > Adam Mazza > _______________________________________________ > 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 vikk at comcast.net Thu Apr 29 10:05:52 2004 From: vikk at comcast.net (vikk at comcast.net) Date: Thu, 29 Apr 2004 14:05:52 -0000 Subject: Short Disk Read error Message-ID: <042920041405.21254.40910BBA0009F6FF000053062200734076FF949496@comcast.net> Hi, I just purchased an Ultra 5 with 128 Megs of Ram and a 8 Gig Hard Drive. I am trying to install Solaris 8 on it and when I boot from cdrom via the command: ok> boot cdrom I get a error message saying "Short Disk Read" and than it says "The File just loaded does not appear to be executable". I am completely new to solaris hence I do not know what to do next. I was trying to install a new CD-Drive as that might be the issue. Can someone please suggest how and what should I do? Thanks From debbie at icus.com Tue Apr 13 17:49:32 2004 From: debbie at icus.com (Debbie Tropiano) Date: Tue, 13 Apr 2004 21:49:32 -0000 Subject: SUMMARY: CDE dtlogin - do not log failed login attempts In-Reply-To: <200404082229.i38MTYwM045686@enterprise.icus.com> from "Debbie Tropiano" at Apr 08, 2004 05:29:34 PM Message-ID: <200404132142.i3DLgTuD090161@enterprise.icus.com> Hello - According to Sun, the /var/adm/loginlog file is for remote login attempts only (rlogin or telnet), but not for local console login attempts (CDE, GNOME or failsafe). There are two options for logging failed login attempts from the system console: turn on BSM or add a debug option to the dtlogin service_name for the Authentication Services Module in /etc/pam.conf along with adding user.debug to /etc/syslog.conf. Neither one is a great option, but at least I could get it to work. BTW I got no replies to my message, other learning that several people are out of the office. Debbie Original message: > I did not see an answer or summary for this question, > but am having an issue with this now. Is it possible > to get CDE to properly log failed login attempts? > > I can get failed login attempts via remote logins to > be logged, but not directy at the CDE console. > > Thanks in advance for any help, > Debbie > > Forwarded message: > > From: "Salum, Felipe" > > To: "'sunmanagers at sunmanagers.org'" > > Subject: CDE dtlogin - do not log failed login attempts > > Date: Thu, 8 Nov 2001 07:20:52 -0500 > > > > Hi people. > > > > I have a doubt. > > > > Users that log in locally using CDE do not have their failed login attempts > > written in /var/adm/loginlog or whatever file that failed login attempts are > > written in a envinronment running hp, sun, aix ?? > > What do I need to do to make this log working ? > > > > My conclusion is that dtlogin is not secure, if it doesnt log anything the > > user can try various passwords to crack an account. > > > > Thanks in advance, > > > > Felipe Salum, SCSA > > Security and Systems Administration > > EDS Brazil > > > > _______________________________________________ > > sunmanagers mailing list > > sunmanagers at sunmanagers.org > > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > > > -- > + 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 + > -- + 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 epaul at profitlogic.com Wed Apr 14 11:59:45 2004 From: epaul at profitlogic.com (Eric Paul) Date: Wed, 14 Apr 2004 15:59:45 -0000 Subject: SUMMARY: sar not working in Solaris 9 on V880 or V240 Message-ID: Many thanks to everyone who replied. On Darren Dunham's suggestion, I trussed the running cron pid and found that it was having pam troubles. After some googling, I was able to come across some sunforum posts saying that somewhere along the line, Sun changed how PAM behaves in Sol 9. I added these lines to pam.conf: cron account required pam_projects.so.1 cron account required /usr/lib/security/pam_unix.so.1 And restarted cron. NOTE: You MUST restart cron. I made the changes to pam.conf and did not restart cron, and was still seeing it not working. So after all this, it turned out to be not a SAR problem, not even really a CRON problem, but a PAM problem :P Many jeers to: Pohl, Stefan [Stefan.Pohl at Dresdner-Bank.com] Kulkarni Pawan-W19278 [Pawz at motorola.com] Obst, Thomas [Thomas.Obst at t-mobile.de] Ian Pease [I.Pease at esteem.co.uk] Broderick, Sean [Sean.Broderick at irl.xerox.com] for not having properly configured auto-responders! May a million spambots chew on their emails!! -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Eric Paul Sent: Wednesday, April 14, 2004 10:01 AM To: sunmanagers at sunmanagers.org Subject: UPDATE: sar not working in Solaris 9 on V880 or V240 Thanks to everyone who has responded so far. The suggestions have been good sanity checks, but so far no one has figured out the problem. Here is some additional information. - If I "su - sys" and manually run /usr/lib/sa/sa1, it works fine. So it appears to be a problem in cron. - If I try adding a truss to the cronjob, I get no results back. If I tell truss to output to a file "/bin/truss -f -o /tmp/trace /usr/lib/sa/sa1" the file never gets created. Ditto without the -f. Ditto if I don't use -o and just use >>. - I received many suggestions saying that cron wasn't running for a number of reasons. /var/cron/log shows my jobs being executed. The user can run cron jobs. I have tried changing the password in /etc/shadow to all manner of things, no difference. - Syslog shows nothing. No mails are being generated, not to root, not to sys. - I received one suggestion saying there was a problem with openssh on sol9, and that I should only change the sys crontab from the console. I dragged my laptop over and plugged into the serial port. No dice. - I received a suggestion telling me about a problem with the Virtual SCSI Device Driver on V240s. I can't see that I am loading that. # pkginfo |grep SCSI system SUNWfcp Sun FCP SCSI Device Driver system SUNWfcpx Sun FCP SCSI Device Driver (64-bit) system SUNWses SCSI Enclosure Services Device Driver system SUNWsesx SCSI Enclosure Services Device Driver (64-bit) If anyone has further suggestions, I am open to anything at this point. Thanks! Eric -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Eric Paul Sent: Tuesday, April 13, 2004 4:56 PM To: sunmanagers at sunmanagers.org Subject: sar not working in Solaris 9 on V880 or V240 Ok, I have a problem that is stumping myself and the other admin here. We just got some new servers in, a couple of V880s and some V240s. We installed Solaris 9 on them (12/03 on the V240, 06/03 on the V880 when we saw problems on the 240) For some reason, sar is not working as it does everywhere else. I am using the same crontab entries that I have working on a V120 running Sol 9, and still no dice. It starts up fine, but the incremental updates don't seem to be happening. Here are the relevant files and info: # cat /var/spool/cron/crontabs/sys #ident "@(#)sys 1.5 92/07/14 SMI" /* SVr4.0 1.2 */ # # The sys crontab should be used to do performance collection. See cron # and performance manual pages for details on startup. # # 0 * * * 0-6 /usr/lib/sa/sa1 # 20,40 8-17 * * 1-5 /usr/lib/sa/sa1 # 5 18 * * 1-5 /usr/lib/sa/sa2 -s 8:00 -e 18:01 -i 1200 -A 00,05,10,15,20,25,30,35,40,45,50,55 * * * * /usr/lib/sa/sa1 01 00 * * * /usr/lib/sa/sa1 59 23 * * * /usr/lib/sa/sa1 01 00 * * * find /var/adm/sa -name "sa*" -type f -ctime +15 -exec rm -f {} \; # cat /etc/rc2.d/S21perf #!/sbin/sh # # Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T. # All rights reserved. # # THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF AT&T # The copyright notice above does not evidence any # actual or intended publication of such source code. # # Copyright (c) 1997 by Sun Microsystems, Inc. # All rights reserved. # #ident "@(#)perf.sh 1.7 97/12/08 SMI" # Uncomment the following lines to enable system activity data gathering. # You will also need to uncomment the sa entries in the system crontab # /var/spool/cron/crontabs/sys. Refer to the sar(1) and sadc(1m) man pages # for more information. if [ -z "$_INIT_RUN_LEVEL" ]; then set -- `/usr/bin/who -r` _INIT_RUN_LEVEL="$7" _INIT_RUN_NPREV="$8" _INIT_PREV_LEVEL="$9" fi if [ $_INIT_RUN_LEVEL -ge 2 -a $_INIT_RUN_LEVEL -le 4 -a \ $_INIT_RUN_NPREV -eq 0 -a \( $_INIT_PREV_LEVEL = 1 -o \ $_INIT_PREV_LEVEL = S \) ]; then /usr/bin/su sys -c "/usr/lib/sa/sadc /var/adm/sa/sa`date +%d`" fi # sar SunOS dev-app-s00 5.9 Generic_112233-11 sun4u 04/13/2004 16:04:55 %usr %sys %wio %idle 16:04:55 unix restarts 16:14:00 unix restarts 16:15:00 1 2 7 91 # date Tue Apr 13 16:50:19 EDT 2004 # cat /var/cron/log ! *** cron started *** pid = 164 Tue Apr 13 16:14:01 2004 ! unexpected pid returned 174 (ignored) Tue Apr 13 16:14:02 2004 > CMD: /usr/lib/sa/sa1 > sys 295 c Tue Apr 13 16:15:00 2004 < sys 295 c Tue Apr 13 16:15:00 2004 > CMD: /usr/lib/sa/sa1 > sys 328 c Tue Apr 13 16:20:00 2004 < sys 328 c Tue Apr 13 16:20:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 333 c Tue Apr 13 16:25:00 2004 < sys 333 c Tue Apr 13 16:25:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 335 c Tue Apr 13 16:30:00 2004 < sys 335 c Tue Apr 13 16:30:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 340 c Tue Apr 13 16:35:00 2004 < sys 340 c Tue Apr 13 16:35:01 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 346 c Tue Apr 13 16:40:00 2004 < sys 346 c Tue Apr 13 16:40:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 355 c Tue Apr 13 16:45:00 2004 < sys 355 c Tue Apr 13 16:45:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 363 c Tue Apr 13 16:50:00 2004 < sys 363 c Tue Apr 13 16:50:00 2004 ts=13 There are no error messages in syslog or in my mail. Any ideas? TIA! I will summarize. 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 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Daping.Xia at united.com Wed Apr 14 18:20:19 2004 From: Daping.Xia at united.com (Daping.Xia at united.com) Date: Wed, 14 Apr 2004 22:20:19 -0000 Subject: Summary: falrcreate problem Message-ID: Thanks for A.J.Caines at halplant.com and tkevans at tkevans.com for their quick responses. Here is the suggestion: /var/tmp itself is needed for the OS, so you can't exclude it. If you want to put the image in /var/tmp, simply create a subdirectory and put it in there, excluding that subdirectory, ie. # flarcreate -n "Solaris 9 Web Server Image" -R / -x /var/tmp/flash /var/tmp/flash/S9-web.image > root at ch0-s1 # flarcreate -n "Solaris 9 Web Server Image" \ > -R / \ > -x /var/tmp \ > /var/tmp/S9-web.image ... > ERROR: File /var/tmp registred in package database - can not be > excluded. I was following the instruction from the blueprint book JumpStart Technology, it seems this is a mistake in the original book. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Thu Apr 15 12:47:18 2004 From: levins at westnet.com (Adam Levin) Date: Thu, 15 Apr 2004 16:47:18 -0000 Subject: SUMMARY: /etc/notrouter and multi-homed hosts Message-ID: Kind listmembers, I'd like to once again thank the following: Charles Rawls Terry Gardner Richard Rodriguez Crist Clark Darren Dunham I would also like to apologize for wasting their time. I'm at our data center today, and the network guy is here. We poured over tcpdump output on the various interfaces and machines involved, and everything looks correct. The packets coming in to the FTP server look fine, and the server is responding on the other interface, but the datagram IP is correct. The problem is that our two subnets are not connected via router -- they're connected via a *firewall*. It's the firewall that's preventing the crosstalk, which is exactly what it's supposed to do. Since it's a stateful firewall (Cisco PIX), it sees the response go out on a different subnet from the request. It doesn't know that it's a valid response to a request, and therefore blocks the datagrams. Everything is working exactly as it should, and so You Can't Get There >From Here. Thank you very much for the help and replies. If nothing else, I've learned even more about diagnosing network issues. -Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mnashish at hotmail.com Thu Apr 15 13:29:35 2004 From: mnashish at hotmail.com (ashish madhavan) Date: Thu, 15 Apr 2004 17:29:35 -0000 Subject: summary:: root disk crashed!!!! Message-ID: hi all,i have solved the issue with the suggestions provided by Andres Martinez ..... i am also greatful to Ric Anderson who had a nice suggestion but the first method worked so i cldnt try out the one provided by him....also i was pleasently surprised by number the replies i got..thanx to all..the solution is as foll: recover from backup aditionally you must 1) boot from CDROM 2) mount root partition (for example under /a) 3) rm /a/etc/path_to_inst 4) rm /a/dev/dsk/c* 5) rm /a/dev/rdsk/c* 6) devfsadm -r /a -p /a/etc/path_to_inst then "reboot" thanx n regardsashish ------------------------------------------------------------------------ Easiest Money Transfer to India . Send Money To 6000 Indian Towns. Easiest Way To Send Money Home! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From Michael.Connolly at itt.com Fri Apr 16 09:39:23 2004 From: Michael.Connolly at itt.com (Connolly, Michael) Date: Fri, 16 Apr 2004 13:39:23 -0000 Subject: SUMMARY: problem with cron and sendmail Message-ID: <52EA22100573564994D8AA2EE6BB3C5C58E407@FWETGEXCH1.itt.net> Well, the cron job ran fine last night after modifying sendmail.cf: "...replace the DontBlameSendmail line by O DontBlameSendmail=GroupWritableDirPathSafe after uncommenting it. Restart sendmail and the error should stop." The suggestion above came from John Ballard. While this may represent a security issue I can at least breath easier knowing the cron job works and this gives me time to dig deeper. Other suggestion were to check/verify the permissions on / /etc and /etc/mail which I will do. Also several folks mentioned mentioned that this should not have aborted the cron job but I see no other reason as I have not changed the script and now it works again. Note: the same error appeared in the cron log, too. Thanks to: John Ballard Scott endley Andrew Hall Crist Clark Jonathan Voight Lars Hecking Chan Cao Mark Cain Andy Scutt Andrew Hay Regards, Michael J. Connolly Worldwide PLM/PDM Manager ITT Industries, Cannon 617-969-3700 x8302 -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of Connolly, Michael Sent: Thursday, April 15, 2004 2:50 PM To: sunmanagers at sunmanagers.org Subject: problem with cron and sendmail Recently added a cron job for user webusr: 15 3 * * * /home/users/webusr/restart_weblogic > /dev/null this kills 3 java processes and then restarts them. Well, it was working fine for a few weeks but now it kills the processes but does not restart them. (the script has not changed). dmesg output shown below Apr 15 03:15:00 cksun sendmail[963]: [ID 801593 mail.crit] NOQUEUE: SYSERR(webusr): /etc/mail/sendmail.cf: line 77: fileclass: cannot open '/etc/mail/local-host-names': Group writable directory Apr 15 03:15:00 cksun sendmail[963]: [ID 801593 mail.crit] NOQUEUE: SYSERR(webusr): /etc/mail/sendmail.cf: line 492: fileclass: cannot open '/etc/mail/trusted-users': Group writable directory permission on /etc/mail are drwxr-xr-x Could someone point me in the right direction to correct this? This IS urgent as it is affecting an application accessed worldwide! Regards, Michael J. Connolly Worldwide PLM/PDM Manager ITT Industries, Cannon 617-969-3700 x8302 ************************************ This email and any files transmitted with it are proprietary and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the sender. Please note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of ITT Industries, Inc. The recipient should check this email and any attachments for the presence of viruses. ITT Industries accepts no liability for any damage caused by any virus transmitted by this email. ************************************ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers ************************************ This email and any files transmitted with it are proprietary and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the sender. Please note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of ITT Industries, Inc. The recipient should check this email and any attachments for the presence of viruses. ITT Industries accepts no liability for any damage caused by any virus transmitted by this email. ************************************ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rmillisl at gdcanada.com Fri Apr 16 11:32:55 2004 From: rmillisl at gdcanada.com (Randy Millis) Date: Fri, 16 Apr 2004 15:32:55 -0000 Subject: SUMMARY: Dump prom settings on a SunFire 280R? References: <017301c4233c$1061a0a0$67fc1eac@gdcan.com> Message-ID: <00c401c423c7$0efa8660$67fc1eac@gdcan.com> Most who responded suggested using "eeprom" and redirecting to a file: e.g. eeprom > filename Joe Fletcher said: -cut- eeprom > output.txt will give you some but not all of it. -cut- Eugene Schmidt expanded on what Joe said with: -cut- Most, but not all can be seen by doing: eeprom Alternatively, an RS232 cross cable to the serial port to another system. (Sun/Laptop/..?) Set capture on on Hyperterm/ "script /tmp/mylog; tip hardwire" or whatever as required by tty session. Reboot the server you wish to capture. Break as system wishes to boot. printenv boot Exit from tty emulator and close script/log Make sure system boots. -cut- Thanks to: Vincent Bolay, Dirk Hamilton, Michael Schulte, Nicolas Dorfsman, Mark McManus, Stan Pietkiewicz, Brent Mcdaniel, krenzischek at Encompasserve.org, Alan Pae, Jim Seymour, Matt Gentry, Bob Cummings, Adrian Cole, and Tim Evans for their very speedy replies. ----- Original Message ----- From: "Randy Millis" To: Sent: Thursday, April 15, 2004 4:50 PM Subject: Dump prom settings on a SunFire 280R? > How can I dump prom settings on a SunFire 280R to a text file for > documentation purposes from a running OS? Googleing and searching the > archives did not reveal an answer. > > Thank you. > _______________________________________________ > 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 Omymama at sudatel.net Thu Apr 15 12:32:23 2004 From: Omymama at sudatel.net (Omyma Mahmoud Abelhafeez) Date: Thu, 15 Apr 2004 16:32:23 -0000 Subject: SUMMARY: sar not working in Solaris 9 on V880 or V240 Message-ID: <54C0965E3F89D811AAE2000BCD4FA7E0012137@MAIL> Hi I have afuny case that I have 2 sun blade 2000 machine, running Solaris 8 and when entered sar command in one machine it working, in anther one reported as (( <<<>>>)). By using find tools also the command not found!! Note that I do the same scenario in both machines!!! So please do you have any explanation? Thanks and Regard -----Original Message----- From: Eric Paul [mailto:epaul at profitlogic.com] Sent: Wednesday, April 14, 2004 5:53 PM To: Subject: SUMMARY: sar not working in Solaris 9 on V880 or V240 Many thanks to everyone who replied. On Darren Dunham's suggestion, I trussed the running cron pid and found that it was having pam troubles. After some googling, I was able to come across some sunforum posts saying that somewhere along the line, Sun changed how PAM behaves in Sol 9. I added these lines to pam.conf: cron account required pam_projects.so.1 cron account required /usr/lib/security/pam_unix.so.1 And restarted cron. NOTE: You MUST restart cron. I made the changes to pam.conf and did not restart cron, and was still seeing it not working. So after all this, it turned out to be not a SAR problem, not even really a CRON problem, but a PAM problem :P Many jeers to: Pohl, Stefan [Stefan.Pohl at Dresdner-Bank.com] Kulkarni Pawan-W19278 [Pawz at motorola.com] Obst, Thomas [Thomas.Obst at t-mobile.de] Ian Pease [I.Pease at esteem.co.uk] Broderick, Sean [Sean.Broderick at irl.xerox.com] for not having properly configured auto-responders! May a million spambots chew on their emails!! -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Eric Paul Sent: Wednesday, April 14, 2004 10:01 AM To: sunmanagers at sunmanagers.org Subject: UPDATE: sar not working in Solaris 9 on V880 or V240 Thanks to everyone who has responded so far. The suggestions have been good sanity checks, but so far no one has figured out the problem. Here is some additional information. - If I "su - sys" and manually run /usr/lib/sa/sa1, it works fine. So it appears to be a problem in cron. - If I try adding a truss to the cronjob, I get no results back. If I tell truss to output to a file "/bin/truss -f -o /tmp/trace /usr/lib/sa/sa1" the file never gets created. Ditto without the -f. Ditto if I don't use -o and just use >>. - I received many suggestions saying that cron wasn't running for a number of reasons. /var/cron/log shows my jobs being executed. The user can run cron jobs. I have tried changing the password in /etc/shadow to all manner of things, no difference. - Syslog shows nothing. No mails are being generated, not to root, not to sys. - I received one suggestion saying there was a problem with openssh on sol9, and that I should only change the sys crontab from the console. I dragged my laptop over and plugged into the serial port. No dice. - I received a suggestion telling me about a problem with the Virtual SCSI Device Driver on V240s. I can't see that I am loading that. # pkginfo |grep SCSI system SUNWfcp Sun FCP SCSI Device Driver system SUNWfcpx Sun FCP SCSI Device Driver (64-bit) system SUNWses SCSI Enclosure Services Device Driver system SUNWsesx SCSI Enclosure Services Device Driver (64-bit) If anyone has further suggestions, I am open to anything at this point. Thanks! Eric -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org]On Behalf Of Eric Paul Sent: Tuesday, April 13, 2004 4:56 PM To: sunmanagers at sunmanagers.org Subject: sar not working in Solaris 9 on V880 or V240 Ok, I have a problem that is stumping myself and the other admin here. We just got some new servers in, a couple of V880s and some V240s. We installed Solaris 9 on them (12/03 on the V240, 06/03 on the V880 when we saw problems on the 240) For some reason, sar is not working as it does everywhere else. I am using the same crontab entries that I have working on a V120 running Sol 9, and still no dice. It starts up fine, but the incremental updates don't seem to be happening. Here are the relevant files and info: # cat /var/spool/cron/crontabs/sys #ident "@(#)sys 1.5 92/07/14 SMI" /* SVr4.0 1.2 */ # # The sys crontab should be used to do performance collection. See cron # and performance manual pages for details on startup. # # 0 * * * 0-6 /usr/lib/sa/sa1 # 20,40 8-17 * * 1-5 /usr/lib/sa/sa1 # 5 18 * * 1-5 /usr/lib/sa/sa2 -s 8:00 -e 18:01 -i 1200 -A 00,05,10,15,20,25,30,35,40,45,50,55 * * * * /usr/lib/sa/sa1 01 00 * * * /usr/lib/sa/sa1 59 23 * * * /usr/lib/sa/sa1 01 00 * * * find /var/adm/sa -name "sa*" -type f -ctime +15 -exec rm -f {} \; # cat /etc/rc2.d/S21perf #!/sbin/sh # # Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T. # All rights reserved. # # THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF AT&T # The copyright notice above does not evidence any # actual or intended publication of such source code. # # Copyright (c) 1997 by Sun Microsystems, Inc. # All rights reserved. # #ident "@(#)perf.sh 1.7 97/12/08 SMI" # Uncomment the following lines to enable system activity data gathering. # You will also need to uncomment the sa entries in the system crontab # /var/spool/cron/crontabs/sys. Refer to the sar(1) and sadc(1m) man pages # for more information. if [ -z "$_INIT_RUN_LEVEL" ]; then set -- `/usr/bin/who -r` _INIT_RUN_LEVEL="$7" _INIT_RUN_NPREV="$8" _INIT_PREV_LEVEL="$9" fi if [ $_INIT_RUN_LEVEL -ge 2 -a $_INIT_RUN_LEVEL -le 4 -a \ $_INIT_RUN_NPREV -eq 0 -a \( $_INIT_PREV_LEVEL = 1 -o \ $_INIT_PREV_LEVEL = S \) ]; then /usr/bin/su sys -c "/usr/lib/sa/sadc /var/adm/sa/sa`date +%d`" fi # sar SunOS dev-app-s00 5.9 Generic_112233-11 sun4u 04/13/2004 16:04:55 %usr %sys %wio %idle 16:04:55 unix restarts 16:14:00 unix restarts 16:15:00 1 2 7 91 # date Tue Apr 13 16:50:19 EDT 2004 # cat /var/cron/log ! *** cron started *** pid = 164 Tue Apr 13 16:14:01 2004 ! unexpected pid returned 174 (ignored) Tue Apr 13 16:14:02 2004 > CMD: /usr/lib/sa/sa1 > sys 295 c Tue Apr 13 16:15:00 2004 < sys 295 c Tue Apr 13 16:15:00 2004 > CMD: /usr/lib/sa/sa1 > sys 328 c Tue Apr 13 16:20:00 2004 < sys 328 c Tue Apr 13 16:20:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 333 c Tue Apr 13 16:25:00 2004 < sys 333 c Tue Apr 13 16:25:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 335 c Tue Apr 13 16:30:00 2004 < sys 335 c Tue Apr 13 16:30:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 340 c Tue Apr 13 16:35:00 2004 < sys 340 c Tue Apr 13 16:35:01 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 346 c Tue Apr 13 16:40:00 2004 < sys 346 c Tue Apr 13 16:40:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 355 c Tue Apr 13 16:45:00 2004 < sys 355 c Tue Apr 13 16:45:00 2004 ts=13 > CMD: /usr/lib/sa/sa1 > sys 363 c Tue Apr 13 16:50:00 2004 < sys 363 c Tue Apr 13 16:50:00 2004 ts=13 There are no error messages in syslog or in my mail. Any ideas? TIA! I will summarize. 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 _______________________________________________ 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 joe_fletcher at btconnect.com Fri Apr 16 11:38:09 2004 From: joe_fletcher at btconnect.com (joe_fletcher at btconnect.com) Date: Fri, 16 Apr 2004 15:38:09 -0000 Subject: SUMMARY: hostids on F15K Message-ID: <407ffc4a.2abb.0@btconnect.com> Hi, Ok. I'm a muppet. The correct braincell woke up about 30 seconds after I sent the message. On this domain the hostid is a function of the MAC address on dman0. The serial number is just the hostid represented in decimal. I've checked the app and it's getting back the correct serial number. I've still no idea why it doesn't like the license however. Back to the app vendors. Thanks to messrs Riddick and McDaniel for their speedy replies. And is it me or has the holiday season started early? Cheers Joe ---------------------------- originally --------------------------------------------- Hi, Got one here for the experts on the big toys. We have an app running of an F15K domain. The app is licensed against hostname and hostid. These values are obtained via "hostid" and "uname -n". Easy enough so far. The problem is that we've just built a new domain and the licenses provided (we had several attempts) never work. The difference between the new domain and the old working instance is that the new one does all it's networking via a couple of Cassini cards (ce gigabits) whereas the old one had these plus a QFE card. I can't recall exactly how a 15K domain picks up it's hostid value. I'm also wondering if the app licensing software is picking up the eri card used for backplane comms. I've tried to truss the licensing steps and it looks like the software calls sysinfo and gets SI_HW_SERIAL. The number is comes back with doesn't look anything like the hostid so I'd like to know where this value comes from. If anyone can tell me how these components may interact I'd appreciate it. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From thesunlover2002 at yahoo.com Fri Apr 16 11:48:37 2004 From: thesunlover2002 at yahoo.com (Melissa Young) Date: Fri, 16 Apr 2004 15:48:37 -0000 Subject: SUMMARY: "crontab -e" returns "711" error In-Reply-To: <407FFB5D.3040204@pandora.be> Message-ID: <20040416154155.68113.qmail@web14910.mail.yahoo.com> This is it. Sorry I was confused momentarily. I knew this method, but thought if VI is working fine, it couldn't be the editor setting problem. Thanks a lot everyone!! Melissa --- Franky wrote: > Melissa Young wrote: > > Hi Gurus, > > > > I got a "small" problem here: > > > > After "crcrontabe" is inputed by either root or a > > user, "711" appears with process hanging. Vi is > > working fine. > > > > I gogoogledut found nothing. > > Please help. Thanks. > > > > Melissa Young > > System Admin > > try doing this: > > EDITOR=vi > export EDITOR > crcrontabe > > Franky __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From gunia at calumet.purdue.edu Fri Apr 16 12:17:19 2004 From: gunia at calumet.purdue.edu (Anthony Gunia) Date: Fri, 16 Apr 2004 16:17:19 -0000 Subject: SUMMARY: Starting CDE from SSH Shell Prompt Message-ID: Hi all, Thanks to all the following for responding to my question regarding starting CDE from SSH shell prompt: Jonathan Voigt Crist Clark Larye Parkins Paul Greidanus Blake Williford Kamal Ray Pengelly Here is my original question: "Is there a way to start a CDE session from a remote SSH session using something like Reflections for Windows or from a shell on Linux? I can bring up SMC, and other GUI tools this way, but would also like to bring up the entire CDE." In order to start a CDE session via an SSH shell, an Xwindows session is required for the PC (Linux should have this already). Three products that can accomplish this on the PC are Hummingbird Exceed, Reflections, and Xmanager. Reflections and Exceed have great Xsession wizards for setting up your connection over SSH. Xmanager does not have these, but works just as well and is lower in price. The command that seemed to work for me to access the CDE that was recommended the most was: /usr/dt/bin/Xsession Some other commands that folks used were as follows: /usr/dt/bin/dtwm Xnest -query localhost (man page at: http://www.xfree86.org/current/Xnest.1.html ) I had to make sure the following was set in the sshd_config on the server: X11Forwarding yes X11DisplayOffset 10 After the appropriate changes are made to the sshd_config, and have the Xsession software on the PC, accessing the CDE worked. Ray had a good "step by step" for using Exceed as follows: On PC: 1. Exceed->Xconfig->Security - Set to allow any host access. 2. Exceed->Xconfig->Display Set to Single Window mode. 2. Started Exceed in passive mode. 3. In SSH client made sure X11 tunneling was turned on. On the Solaris Machine 1. Set "X11Forwarding yes" in /etc/ssh/sshd_config 2. Restarted sshd 3. Made sure my login files (.login, .cshrc and .tcshrc) didn't set the DISPLAY environment variable. On PC 1. Used ssh client to login to Solaris machine. 2. Issued "echo $DISPLAY" to make sure X11forwarding is working on Solaris Machine. It should return ":10.0" 10 refers to the X11DisplayOffset variable from sshd_config. 3. ran /usr/dt/bin/Xsession 4. CDE came up and functioned. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mnashish at hotmail.com Fri Apr 16 13:49:49 2004 From: mnashish at hotmail.com (ashish madhavan) Date: Fri, 16 Apr 2004 17:49:49 -0000 Subject: summary: root disk crashed!!!! Message-ID: hi all, i have solved the issue with the suggestions provided by Andres Martinez & a late suggestion on similar lines by mr joe..... i am also greatful to Ric Anderson who had a nice suggestion but the first method worked so i cldnt try out the one provided by him....also i was pleasently surprised by number the replies i got..thanx to all..the solution is as foll: recover from backup aditionally you must 1) boot from CDROM 2) mount root partition (for example under /a) 3) rm /a/etc/path_to_inst 4) rm /a/dev/dsk/c* 5) rm /a/dev/rdsk/c* 6) devfsadm -r /a -p /a/etc/path_to_inst then "reboot" thanx n regards ashish BELOW IS THE MY ORIGINAL POST : hi all, i am having an E3500 sun server with solaris 8 (patch level 12) with a D1000 sun storedge attached to it....the root disk of this server having /,swap,/var & /export/home has crashed. i am trying to restore the same from my backups which was taken in multiuser mode with all the users logged off & no activity on the disk during the period. i am doing restoration by booting through the sun cdrom in single user mode & restoring it through "ufsrestore rvf". now my new disk is getting detected at " probe-fcal-all" & also while booted through cd but on loading through the root disk on which all data was recenly backed up,the following happens: 1.>it starts loading the os 2.>shows hostname 3.>then gives errors like c0t0d0s1 not found c0t0d0s0 not found , these are my swap & root partitions respectively. then it comes to single user mode and asks for passwd: here it accepts my old root passwd but on giving " format" command it shows me the six harddisks on my D1000 but the local hard disk is not shown....& the booting process stops & doesnt proceed further. also # cd /dev/dsk #ls -l shiows the entries made on todays date....as i had given boot -r...i have also tried restoring from a backup taken 6 months back the effect being the same. this server being an NIS client many users are being affected by it ..... any suggestions are welcome. i will summarise thanks n regards ashish ------------------------------------------------------------------------ Contact brides & grooms FREE! Only on www.shaadi.com. Register now! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jon.hudson at finisar.com Fri Apr 16 18:03:26 2004 From: jon.hudson at finisar.com (Jon Hudson) Date: Fri, 16 Apr 2004 22:03:26 -0000 Subject: SUMMARY: user not showing up.... Message-ID: <6044ECEB90CDBB4AAC19E512FDB6A94636BD79@EXCHANGE.ads.finisar.com> Ok, well that's embarrassing =) Thanks to.. Mark Cain Donald Mayers Darren Dunham ....for so rapidly pointing out exactly my malfunction. However, I must say I'm a bit surprised that utmpx does not get updated till the lookup times out. Thanks much, and sorry for the noise. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From gary at catapult.com Fri Apr 16 18:39:54 2004 From: gary at catapult.com (Gary Lopez) Date: Fri, 16 Apr 2004 22:39:54 -0000 Subject: Summary: rcp question Message-ID: <4080603F.3010005@catapult.com> Thanks to all who replied. Everyone gave me good clues. The answer was my .rhosts file was incorrect. on hostB is should be "hostname username" and then in the users home directory create a .rhosts file that has "remote-host username". Thanks again all.. I appreciate all your answers.. Gary Good afternoon, I know there is a simple answer, but I cannot remember. If I wanted to rcp a file from hostA to hostB as a user, what needs to be done to hostA to make it accept that file. I have added the .rhost file with hostname root username root Still getting permission denied on hostB. I even made the dir 777 on hostB. What am I forgetting ? Thanks .. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jon.hudson at finisar.com Fri Apr 16 19:08:59 2004 From: jon.hudson at finisar.com (Jon Hudson) Date: Fri, 16 Apr 2004 23:08:59 -0000 Subject: SUMMARY-REDUX: User not showing up..... Message-ID: <6044ECEB90CDBB4AAC19E512FDB6A94636BD7B@EXCHANGE.ads.finisar.com> So the original problem was that after logging into a server with openssh, the user would not show up using either "w" or "who". Only after waiting about 60sec would the user show up. Further investigation showed that it was /var/adm/utmpx that was not updating in a timely matter. What is somewhat interesting is that testing showed that telnet, would not actually let you log into until utmpx was updated, where openssh would just let you log in right away, allowing utmpx to take its time to update. The problem as it turned out what the ipaddr of the dns server in /etc/resolv.conf was incorrect. So any look-ups were failing after a timeout. After fixing this simple issue, the problem went away. Utmpx is now updating immediately and telnet logs in right away (or did before I shut it off again). While my initial issue is solved, I know why users were not showing up after log in, I am curious about what this has shown. I'm surprised that utmpx does not update till the lookup fails. I'm further surprised that telnet makes you wait till utmpx has been updated, but openssh does not. Thus leaving a window of time where a user is essentially invisible. So while the answer was simple, and I'm a bit embarrassed for not realizing this, the results are a bit odd. Once again thanks to Mark Cain Donald Mayers Darren Dunham _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From filipe_litaiff at optiglobe.com.br Mon Apr 19 10:49:23 2004 From: filipe_litaiff at optiglobe.com.br (Filipe Litaiff) Date: Mon, 19 Apr 2004 14:49:23 -0000 Subject: SUMMARY: Resuming tar restore Message-ID: Greetings Forumers, This is not quite an summary. We weren't able to test the solutions provided, but I think there is some information that deserves to be shared. 1 - One solution would be using exclude or include files for tar (-I and -X options). It doesn4t help us, mostly because the tape have to be read anyway (time consuming) to find the 'file headers' tar writes on the archive at the beggining of every file. 2 - If you are trying to extract some files from a tepe with a big archive, and you are time constrained, the option suggested on the tar man page to extract archives based no regular expressions is a no-brainer. It executes tar twice. "Filename substitution wildcards cannot be used for extracting files from the archive. Rather, use a command of the form: tar xvf ... /dev/rmt/0 `tar tf ... /dev/rmt/0 | \ grep 'pattern' ` " Well, it should work nicely with disk archives. 3 - The only way one could extract files from any given point on the archive would be: 3.1: Find the block size. Do the following: "dd if=TAPEDRIVE of=/tmp/tape.block count=1 bs=128k where the 'bs=128k' is some arbitrarily large value -- maybe 256k -- it is needed to pre-establish a buffer size for dd. Then 'ls -l /tmp/tape.block' and see how big the file is -- that's your tape block size." Thanks Bill R. Williams for the procedure. 3.2: Forward to any point right before the first file you want to download. - Considering the block size you found in the last procedure is the record the mt man page referes to, and you know how much in kbytes the tape should be forwarded to find the spot, do the following: mt -f fsr = / =non-rewindable tape device. 3.3: Use GNU tar to try to extract the files from that point. The POSIX tar provided with Solaris can't do this. You'll probably see something like this: ficus# /usr/local/bin/tar -xvf /dev/rmt/0cn Blocksize = 126 records Hmm, this doesn't look like a tar archive. Skipping to next file header... drwxr-xr-x root/other Jul 27 09:51 1999 / drwxr-xr-x root/sys Jul 27 07:36 1999 /etc/ drwxrwxr-x root/sys Mar 8 14:27 1999 /etc/default/ -r--r--r-- root/sys Oct 30 16:58 1996 /etc/default/sys-suspend -r-xr-xr-x bin/bin Mar 5 12:45 1999 /etc/default/cron -r--r--r-- bin/bin Mar 5 12:45 1999 /etc/default/fs (...) NOTE: I haven't tried this. I'm not sure even if the record thing mt uses is actually the size you get using the 3.1 procedure. If you are willing to use this, please double-check every information provided. Thanks to: Darren Dunham Bill R. Williams Terry L Moore Anatoliy Lisovskiy Cheers, Filipe Litaiff -----Original Message----- From: Filipe Litaiff Sent: Thursday, April 15, 2004 3:20 PM To: sunmanagers at sunmanagers.org Subject: Resuming tar restore Hi Forumers, We were restoring a very big tar archive from tape when something broke the command and the restore was interrupted. If we restart the tar we4ll take too long to restore files we already have extracted. We know where the tar was broken. What we want to figure out a way to resume the extraction, from given file to the end of the archive. Since restoring it from the beggining is not a good option, and the tar man page doesn4t help much, what could we do? We have a list of the remaining files that are still on the archive. Would it help? The tar man page states something that might help, but I think it4ll have lousy restore times since it runs tar twice: "Filename substitution wildcards cannot be used for extracting files from the archive. Rather, use a command of the form: tar xvf ... /dev/rmt/0 `tar tf ... /dev/rmt/0 | \ grep 'pattern' ` " Any help would be appreciated. Thanks in advance, Filipe. _______________________________________________ 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 mmata at multimatic.com Wed Apr 21 12:31:38 2004 From: mmata at multimatic.com (Marcelino Mata) Date: Wed, 21 Apr 2004 16:31:38 -0000 Subject: Summary : 4GB RAM on Blade1000 Message-ID: <07370339B1FAD31184C800508B670EE50343C1DA@mtcmail> Many thanks to those who gave suggestions. I can not give the exact reason or solution to the problem but I did upgrade the OBP to the latest and it worked. As the computer is needed, I could not waste any more time finding out the root cause. I did learn that setting diag-switch?=true and diag-level=max requires a lot of patience. With 1.5Gb RAM on OBP 4.13, it takes around 10 minutes before the graphics card kicks in. You basically stare at a blank screen thinking the workstation has locked up. I suspect it would take longer than 10 minutes with 5Gb RAM but I had someone standing over my shoulder waiting for his computer back. In the end I do not believe 100% that my problem was OBP related. With 1.5GB RAM it appeared to hang at the same location as 5GB a few times....by hanging I mean it takes longer than 4+ minutes instead of the normal 3-5 seconds... Marcelino >-----Original Message----- >From: Marcelino Mata >Sent: Tuesday, April 20, 2004 3:45 PM >To: Sun Managers >Subject: 4GB RAM on Blade1000 > > > >We just purchased an additional 4Gb RAM for our Blade1000 (Sun >licensed 3rd party). We have removed all existing RAM to keep >things simple. While the system starts to boot, it hangs >shortly after the "Print service starting" message. > >Has anyone else run across this type of problem? I can not >find any memory related issues resolved in OpenBoot level >upgrades but I am running version 4.5.4 while the latest is 4.13.0 > >Is there any useful diagnostics routines at the OpenBoot level? > >Marcelino > >System Information > ># /usr/sbin/prtconf -V >OBP 4.5.4 2001/12/19 14:37 > ># eeprom >test-args: data not available. >diag-passes=1 >enclosure-type=540-3256-15 >banner-name=SUNW,Sun-Blade-1000 >energystar-enabled?=true >pcia-probe-list=4,1 >pcib-probe-list=5,6,1,2,3,4 >local-mac-address?=false >fcode-debug?=false >silent-mode?=false >scsi-initiator-id=7 >oem-logo: data not available. >oem-logo?=false >oem-banner: data not available. >oem-banner?=false >ansi-terminal?=true >screen-#columns=80 >screen-#rows=34 >ttyb-rts-dtr-off=false >ttyb-ignore-cd=true >ttya-rts-dtr-off=false >ttya-ignore-cd=true >ttyb-mode=9600,8,n,1,- >ttya-mode=9600,8,n,1,- >output-device=screen >input-device=keyboard >load-base=16384 >auto-boot?=true >boot-command=boot >diag-file: data not available. >diag-device=net >boot-file: data not available. >boot-device=disk0:a >use-nvramrc?=true >nvramrc=devalias mydev /pci at 8,700000/SUNW,XVR-500 at 1 >devalias xvr500 /pci at 8,700000/SUNW,XVR-500 at 1 >security-mode=none >security-password: data not available. >security-#badlogins=0 >#power-cycles=65 >diag-script=none >diag-level=min >diag-switch?=false >error-reset-recovery=boot _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Wed Apr 21 14:54:26 2004 From: levins at westnet.com (Adam Levin) Date: Wed, 21 Apr 2004 18:54:26 -0000 Subject: SUMMARY: new to SAN -- need some direction Message-ID: Well, here I go again, replying to myself. Turns out everything is working correctly. It was a matter of getting the correct controller on my end (c11/c12/c13/c14) and matching that up with the correct WWN on the other end. When I re-read the install guide, I decided to just do a cfgadm -c configure on every WWN listed in the cfgadm -al output. Four of them matched, and I got my luns partitioned and newfs'd. -Adam ---------- Forwarded message ---------- Date: Wed, 21 Apr 2004 14:06:22 -0400 (EDT) From: Adam Levin To: Sun Managers Mailing List Subject: new to SAN -- need some direction So we've got an Enterprise E3500 running Solaris 8 04/01 with the latest recommended patch cluster (as of last week). I've got a brandy-nifty new NetApp FAS960c with SAN functionality. It's connected to two Brocade SilkWorm 8 port switches. I've got two Sun SBus dual-port HBAs in my E3500 (both are the QLOGIC cards). They are likewise connected to the Brocade switches. I've installed the SFS 4.4 software packages and patches, and I rebooted the machine. I'm trying to follow the SFS install and config guides, but I'm getting a bit lost. I've never dealt with a SAN before. I run cfgadm -al and get this (I removed everything but the FC stuff): c11 fc-fabric connected unconfigured unknown c11::210100e08b299781 unknown connected unconfigured unknown c11::500a098183511e35 disk connected unconfigured unknown c11::500a098183f11ddf disk connected unconfigured unknown c11::500a098a83511e35 disk connected unconfigured unknown c11::500a098a83f11ddf disk connected unconfigured unknown c12 fc-fabric connected unconfigured unknown c12::210000e08b099781 unknown connected unconfigured unknown c12::500a098183511e35 disk connected unconfigured unknown c12::500a098183f11ddf disk connected unconfigured unknown c12::500a098a83511e35 disk connected unconfigured unknown c12::500a098a83f11ddf disk connected unconfigured unknown c13 fc-fabric connected unconfigured unknown c13::210100e08b27ecbb unknown connected unconfigured unknown c13::500a098383511e35 disk connected unconfigured unknown c13::500a098383f11ddf disk connected unconfigured unknown c13::500a098c83511e35 disk connected unconfigured unknown c13::500a098c83f11ddf disk connected unconfigured unknown c14 fc-fabric connected unconfigured unknown c14::210000e08b07ecbb unknown connected unconfigured unknown c14::500a098383511e35 disk connected unconfigured unknown c14::500a098383f11ddf disk connected unconfigured unknown c14::500a098c83511e35 disk connected unconfigured unknown c14::500a098c83f11ddf disk connected unconfigured unknown Now, the 500xxx stuff is confirmed to be the WWNs on the NetApp. I'm not sure what the 210xxx stuff is -- are those the WWNs of my HBAs? Why are they unknown? More importantly, if they *aren't* the WWNs of my HBAs, then how do I get the WWNs so that I can configure the NetApp LUNs correctly? Secondly, I've tried to configure the LUNs using cfgadm -c configure, but I get this: [14:05:18]root at oracle1.stage:/root$ cfgadm -c configure c11::500a098183511e35 cfgadm: Library error: failed to create device node: 500a098183511e35: Invalid argument Could this be because the LUNs don't have the correct WWNs associated with them, or is there perhaps another issue? Any help with this would be greatly appreciated -- it's all new to me. Thanks very much, -Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From trk at cray.com Wed Apr 21 19:08:03 2004 From: trk at cray.com (Tim Kirby) Date: Wed, 21 Apr 2004 23:08:03 -0000 Subject: Summary: Strange jumpstart behavior In-Reply-To: <1082563910.2810.45.camel@wh0g3d3901> Message-ID: Originally posted as > Has anyone seen an issue where the symbolic link > > .../Solaris_9/Tools/Boot/etc/inet/hosts > > (which should point at .../Solaris_9/tmp/root/etc/inet/hosts) > > gets overwritten as a regular file at the beginning of the jumpstart? > I really didn't think I was doing anything unusual here but this has > got me going round in circles; apart from busting the jumpstart tree > it leaves the jumpstarted machine with a brain damaged hostfile. It is, as most of the world who isn't on vacation or out of the office... come on, people - if you're subscribing to this list you ought to be able to work out how to filter mail from a mailing list and not send back vacation responses to - well, who knows who) ... pointed out: That jumpstart share *shouldn't* have been exported rw. I'll swear it wasn't, but had brain been engaged I would have realized the implication of the question and not publicly demonstrated that I wasn't paying attention. Sorry for the waste of bandwidth, folks. Tim -- Tim Kirby 651-605-9074 trk at cray.com Cray Inc. Information Systems _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mgreene at aci.on.ca Wed Apr 21 21:33:23 2004 From: mgreene at aci.on.ca (Marco Greene (Home)) Date: Thu, 22 Apr 2004 01:33:23 -0000 Subject: SUMMARY: A1000 Weekly messages Message-ID: <001201c42809$b967c9a0$f6c8a8c0@pepstep> Sorry for the late summary. Basically, it does seem to be a weekly thing. A few people pointed out that it was a health check and the error is a result of some drivers that are not updated. A few more people indicated that they get the same thing and would like to see a summary. If anyone has anymore information on updated drivers and where the health check job is controlled from, that would be great. Regards, Marco _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From neil at quiogue.com Thu Apr 22 04:59:53 2004 From: neil at quiogue.com (neil quiogue) Date: Thu, 22 Apr 2004 08:59:53 -0000 Subject: SUMMARY: Help Regarding Veritas References: <016801c427ff$8bb18ff0$6501a8c0@frodoprecision> Message-ID: <0b2101c42847$f52813d0$5a1f5f40@shire> Hello, It seems that the internal mapping (on how the OS sees the device) was changed somehow. So what I did was the following: 1. Remove the Disk for Replacement option (vxdiskadm) 2. Placed the Disk in the control of VxVM (through vxdisksetup -i [device]) 3. Replaced the failed or removed disk option (vxdiskadm) 4. Placed the volume online again (vxvol -o bg -f start [volume name]) And voila, I'm able to access the disk again. Hope that helps someone. Regards, Neil ----- Original Message ----- From: "neil quiogue" To: Sent: Thursday, April 22, 2004 12:19 AM Subject: Help Regarding Veritas > I'm having a dilemma wherein the Veritas VM is not able to see the T3 > properly but when I execute the format program, the Solaris box sees it. > Whenever I powercycle the T3, the Solaris box sees the offline/online status > (through /var/adm/messages) also so what could possibly make this happen? > > On bootup, I see this strange information: > Starting VxVM restore daemon... > VxVM starting in boot mode... > NOTICE: vxvm:vxdmp: added disk array 60020F20000048DA0000000000000000, > datype = T3 > > vxvm:vxconfigd: WARNING: Disk t3disk01 in group rootdg: Disk device not > found > > Any tips would be appreciated. > > Regards, Neil > _______________________________________________ > 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 d.matinyarare at iorin.gov.pl Thu Apr 22 09:08:44 2004 From: d.matinyarare at iorin.gov.pl (David Matinyarare) Date: Thu, 22 Apr 2004 13:08:44 -0000 Subject: SUMMARY:Disk partition Message-ID: <6682A0A20A720942AC26337E5E5C03EA1435C6@www.iorin.gov.pl> Thanks a lot to all those who corrected my mistake. I.e. sending an empty mail to all sunmanagers. What I intended to do was: Thanks to all. All of you pointed somehow to manual layout the slices. My original problem was: At what stage In the installation of solaria 8 do I do the disk partition. After doing an auto-install I noticed that there is the root (/) and the "/export/home" directory partioned and mounted. When I wanted to disk partition, with different sizes of slices using "format" and label it I got something which says "it cannot be labeled when mounted or it is in use. In addition to the root (/) and /export/home, I wanted to disk partition /usr, /var, /opt , do a file system and then mount them. Does anyone know any documentation which goes step-by-step in solaris installation. i.e: Disk partition Creating files systems on slices Mounting files systems, and then Installation of the solaris to the disk partitions. David; I can't be specific, but after the time when you select the boot disk, you are given an opportunity to accept the disk, configured as the system suggests. But, one of the options (#3, I think) is to manually layout the disk. You need to select the manual option to re-partition the disk. If you don't do it then, you're screwed (as you found out). Dave H -----Original Message----- From: sunmanagers-bounces at sunmanagers.org [mailto:sunmanagers-bounces at sunmanagers.org] On Behalf Of David Matinyarare Sent: Thursday, April 22, 2004 05:01 To: sunmanagers at sunmanagers.org Subject: Disk partition _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From ghenry at cmi.univ-mrs.fr Thu Apr 22 09:15:20 2004 From: ghenry at cmi.univ-mrs.fr (Gerard Henry) Date: Thu, 22 Apr 2004 13:15:20 -0000 Subject: SUMMARY: strange disk capacity on blade100 In-Reply-To: <20040422110932.GA13898@scr.cmi.univ-mrs.fr> References: <20040422110932.GA13898@scr.cmi.univ-mrs.fr> Message-ID: <20040422131217.GA17690@scr.cmi.univ-mrs.fr> very thanks to all who responded, now it works! i found another old contribution here: http://sunportal.sunmanagers.org/pipermail/summaries/2003-November/004570.html for me, the correct thing is to do: hobbit-root% dd if=/setup.log of=/dev/dsk/c0t2d0s2 count=16 bs=65536 0+1 records in 0+1 records out if i try to use raw device, it didn't work hobbit-root% dd if=/dev/null of=/dev/rdsk/c0t2d0s2 count=16 bs=65536 0+0 records in 0+0 records out to p. bauer, i don't see any "autoconfigure" on format! here is repoonses: Ric Anderson: I'm running a Western Digital 120GB in a blade-100 with no problems. My guess is the disk you got was mislabled internally. If there is no data on it, do a dd if=/dev/zero of=/dev/rdsk/c0tXd0s2 count=16 bs=65536 to wipe any disk label on the device, then let format look at it. Another possibility is that there are some jumper settings on the drive to limit its capacity for older systems, and those may also need to be changed. Peter Stokes: Sounds like you need to delete the label on the disk. Try dd if=/dev/null of=/dev/rdsk/c0t0d0s2 count=20 Peter Bauer: The disk was probably used as a mirror for another 20G disk. You might look at the partition table you'll get in format and take a look what you find on the disk. :-) To fix your prob, launch format, select your disk, and then select "type" -> "autoconfigure". Afterwards, write the new (correct) disk label on the disk. On Thu, Apr 22, 2004 at 01:09:32PM +0200, Gerard Henry wrote: > hello all, > i buyed a 40gb from sun (P/N 370-4440-01) but os see only 20gb. > At the ok prompt, i see > ok probe-ide > Device 0 > ATA Model: ST340016A > > and when i boot from sol9, i see: > # iostat -En > c0t0d0 > Model ST340016A > > but with format: > # format > Searching for disks > 0. c0t0d0 > from seagate, the first reference ST340016A is 40gb, and ST320414A is 20gb. I don't understand why solaris displays a bad reference. > i just upgraded openboot to last update 4.16 > > i read on another forum that sunblade100 can support capacity more than 20gb > > Any clue? > > gerard > _______________________________________________ > 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 krecik at e-wro.net Mon Apr 26 03:52:55 2004 From: krecik at e-wro.net (Grzegorz Nowakowski) Date: Mon, 26 Apr 2004 07:52:55 -0000 Subject: SUMMARY: Generating status message from RM6. Message-ID: <1082966016.8660.14.camel@rohlik.e-wro.net> Hi, Thanks for fast reply, esp. to Alan Pae and Hutin Bertrand (sorry guys for not replying directly). Alan, as for my understanding, problem lies not in the script itself. When the script is invoked, event is outstanding for the whole time of scirpt execution (I found no way to clear it earlier) and healthck somewhow triggers this event again so we have endless loop (till system resources are exhausted). That's my hypothesis. Bertrand suggested using Big Brother plugin (http://www.deadcat.net/viewfile.php?fileid=616) and after all this seems to be the best solution (modulo monitoring system of course - I'm using Nagios (http://www.nagios.org)). Thanks again and best regards. Original message: > Hi Managers, > > I'm administering a few Sun servers with A1000 arrays under control of > the Raid Manager 6.22 software. The problem is that array status > logging/reporting is insufficient for my needs. Servers are remotely > operated but emails sent to root account (and forwarded to where they > can be read from) contain no details about events, so in each case I > have to login, run GUI application and check status log. > > Recently I tried to hack the 'rmscript' to supply output from 'healthck' > utility but there is strange effect: very invocation of 'healthck' > causes to generate 100 or so event messages. It seems that if > 'arraymon' (or whatever invokes 'rmscript') detects that execution of > 'rmscript' takes too long - it invokes another one. Or maybe 'healthck' > somehow causes 'rmscript' to be invoked. > > Any ideas how can I get around it? Or better, maybe there's another way > to get to see 'array xxx, disk yyy failed' in my inbox? Or, at least, > is there any way via command line to get event entry (like in rmlog.log) > nicely described like in GUI? -- Grzegorz Nowakowski e-Wro _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From leiv.jarle.larsen at ementor.no Thu Apr 29 02:16:48 2004 From: leiv.jarle.larsen at ementor.no (Leiv Jarle Larsen) Date: Thu, 29 Apr 2004 06:16:48 -0000 Subject: SUMMARY: fsck, corrupt lost+found Message-ID: <6D659526F82AB74792B2270F5F579F8DA0A37C@100NOOSLMSG002.common.alpharoot.net> I didn't find any workaround for this problem. Ended up making a new filesystem and restore from backup. Thanks to Andrew Rotrame and Lars Hecking regards -----Original Message----- From: Leiv Jarle Larsen Sent: 28. april 2004 11:11 To: sunmanagers at sunmanagers.org Subject: fsck, corrupt lost+found I have a corrupt filesystem (concatinated VxVM-volume) on a machine running Solaris 2.6 Running fsck on the filesystem, I get the message "SORRY. NO SPACE IN lost+found". Doing some more checking on the filesystem, I find that lost+found on the filesystem is corrupt, and is unable to delete it in order to make a new. Has anyone experienced this? Will summarize Leiv Jarle Larsen Systems Consultant ______________________________ Ementor Norge AS, Skvadronvn. 2, NO-4050 SOLA Tel +47 51 64 00 00, Dir +47 51 64 00 32, Mobile +47 93 03 30 62 leiv.jarle.larsen at ementor.no www.ementor.no _______________________________________________ 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 PhippsAC at telkom.co.za Thu Apr 29 09:40:37 2004 From: PhippsAC at telkom.co.za (Adrian Phipps (AC)) Date: Thu, 29 Apr 2004 13:40:37 -0000 Subject: SUMMARY: Sendmail novice Message-ID: Hi there, Thanks to Nicolas Figaro for his information. Apparently sendmail does not have this failover function but using a third party tool like bind can help. "sendmail doesn't know how to do this, but setting mx entries in your dns can help. Ask your dns administrator for this config, or grab bind at www.isc.org." Adrian Phipps > -----Original Message----- > From: Adrian Phipps (AC) > Sent: Thursday, April 29, 2004 9:25 AM > To: 'sunmanagers at sunmanagers.org' > Subject: Sendmail novice > > Hi all, > > I have received a request from one of my users and it has got me > stumped since my knowledge of sendmail is very poor. He wants me to > setup sendmail that if the main relay server does not respond then > sendmail should use a failover server. Is this possible and if so > how do I specify to sendmail to use "SERVER2" when "SERVER1" is not > responding? > > > > Regards > > > Adrian Phipps _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mmata at multimatic.com Fri Apr 30 10:47:38 2004 From: mmata at multimatic.com (Marcelino Mata) Date: Fri, 30 Apr 2004 14:47:38 -0000 Subject: SUMMARY : RE: RAM usage Message-ID: <07370339B1FAD31184C800508B670EE50343C212@mtcmail> Many thanks to : Darren Dunham Brent McDaniel Doug Bell Each supplied helpful information. Brent recommended the excellent tool RMCmem package which quickly identifed the amount of reserved swap vs used swap. I would assume that another tool called memtool from SE toolkit would identify the software which reserved the swap but I did not try it out. Here the responses I received... Get the RMCmem package from Sun. The latest version I have is 3.9.4. I don't remember exactly where, but you can do a search for it on sunsolve and find it and download it. Once installed, run the command "prtmem" This will break the memory down for you and is a very useful tool. There are commands in the package that will help you also. The problem with tools like top, vmstat and prstat and a bunch that cost a lot of money is they report just about all memory always used. This is because the way Sun uses it's file cache. When you open a file, it is kept in file cache for faster retrieval later. This file cache is actually free memory that can be allocated at any given time. But some tools see it as "used" memory. So they report the memory incorrectly. Brent Generally what happens is that less-commonly used system / user libraries get put into swap in case a burst of memory is needed. The best tool to use is going to be 'memtool' from the SE toolkit (available at www.setoolkit.com). This will give you a complete list of what binaries and libraries are in menory, and how many are resident / shared. Douglas Bell Most memory is "allocated". The program is loaded into allocated memory. It does a malloc for allocated memory. This memory is in RAM and might be paged to disk later. Some programs (usually oracle) instead "reserve" memory. It says "I don't need it right now, but later I need to get at least 200MB or things will be bad. Please reserve it for me". The OS marks it "reserved" so it cannot be allocated to another program. However, since it's not immediately used, it is not reserved in RAM, but only on disk. If it later gets used, it will move to RAM. Darren Dunham >-----Original Message----- >From: Marcelino Mata >Sent: Wednesday, April 21, 2004 3:57 PM >To: 'Sun Managers' >Subject: RAM usage > > > >This might be a very basic question but... > >Can someone tell me why various tools under Solaris show 2.3Gb >SWAP used and 3.3Gb RAM used when it has 5GB RAM? I know it >is not a tmpfs issue since /tmp only has 2Mb of files. > >I get those numbers from xosview, top, swap -s and SUN's CDE >tool, /usr/dt/bin/sdtwsinfo. > >Running vmstat 5 I noticed that scanrate "sr" is 0 so >everything is good... Sure beats 200+ scanrate when the >system had 1.5GB RAM. > >Searching SunManagers archive, I read this response to this >type of question : > >"swap is reserved when processes are placed in memory. This >is just in case the entire process needs to be swapped out." > >Does that mean that 2.3Gb is not actually in use? swap -s >reports 286Mb reserved so what's the other 2GB? > >Does anyone know the command line tool for better information >under Solaris 8? > >I tried using, > >mdb -k >> ::memstat > >but I think that option is only for Solaris 9. I tried using >::kmastat but the output was 3 pages long. > >As per our application vendor, /etc/systems is > >forceload: sys/msgsys >forceload: sys/pipe >forceload: sys/semsys >forceload: sys/shmsys >set shmsys:shminfo_shmseg=32 >set shmsys:shminfo_shmmax=67108864 >set msgsys:msginfo_msgmax=32768 >set msgsys:msginfo_msgmnb=65536 >set msgsys:msginfo_msgseg=8192 >set rlim_fd_cur=256 > >This is a Blade1000 used for running a CAD/CAM application. > >Thanks for any help, > >Marcelino _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From noriega at gwu.edu Fri Apr 30 16:39:02 2004 From: noriega at gwu.edu (Eric Noriega) Date: Fri, 30 Apr 2004 20:39:02 -0000 Subject: SUMMARY: Netra X1, DHCP jumpstart install In-Reply-To: <408F02F9.2060301@gwu.edu> References: <408F02F9.2060301@gwu.edu> Message-ID: <4092B953.3090007@gwu.edu> It seems that the DHCP server (Sun DHCP) does not pick up changes to macros w/o a restart (dhtadm). This is not the case with the client tables (pntadm), which is what through me. Eric Noriega wrote: > We've been having an issue with using jumpstart with our Netra > X1's. Recently we migrated from the std. RARP boot to a DHCP boot. > We have been getting the following errors. On one system it seemed to > clear up mysteriously (or at least I can figure out what I did :) ). > > When booting we get the following: > > ok boot net:dhcp - install > Boot device: /pci at 1f,0/ethernet at c:dhcp File and args: - install > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Timeout waiting for BOOTP/DHCP reply. Retrying ... > Failed to receive config params > Restarting DHCP process ... > TFTP server's IP address not known! > > We do have the correct definitions in the dhcp server (Sun dhcp), > and the machine is being assigned the entry. I wanted to see if > others had encountered this problem, before I dive too deep. > > Thanks, > Eric Noriega > _______________________________________________ > 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 Thu Apr 8 11:33:06 2004 From: serafino at cshl.edu (Sal Serafino) Date: Thu, 08 Apr 2004 15:33:06 -0000 Subject: SUMMARY: utmpd/wtmpx/last problem Message-ID: <200404081526.i38FQ9T17833@cshl.edu> Hi All- Thanks to all for their help. Sorry for the late response, but it does take awhile for the problem to happen and I've also done a little testing. I even verified /usr/bin/last with a copy that is known to work. The problem is unsolved as yet. If I solve this one I'll let you all know. This box is known to be hack free, but thanks to all who suggested I check for that. The two main suggestions were OpenSSH 64-bit problems and /var overflows. Since this is a user's workstation SSH should have had no effect in the first place because all logins occur from the console. I turned off SSH entirely and still had the problem. Casper suggested I zap wtmpx and start from scratch. Rather than zap the file, I have resigned myself to fixing it with bvi. Messy, but at least you get some history. Thanks, -Sal Original Posting: > >I have an Ultra-5 running Solaris-8, 108528-29. The problem is that after about >a week and a half, or about a dozen logons, 'last' reports a mystery session >that started Dec 31 and is still logged on. Subsequent sessions are logged in >/var/adm/wtmpx (I checked with a binary editor) but the database appears to be >corrupted somehow. The /var filesystem is about 56% full and has over 400MB >free. It's not a space issue because /var/adm/wtmpx is being updated. I looked >in the software registry, and the file /usr/bin/last checks out. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From egold at fsa.com Thu Apr 8 14:26:17 2004 From: egold at fsa.com (egold at fsa.com) Date: Thu, 08 Apr 2004 18:26:17 -0000 Subject: SUMMARY: how to log failed ftp login attempts? Message-ID: i have to run in.ftpd with the -dl flags AND run syslog with *.debug now i can see failed ftp logins, thanks to all who helped! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jesse-carroll at usa.net Mon Apr 12 09:14:28 2004 From: jesse-carroll at usa.net (JESSE CARROLL) Date: Mon, 12 Apr 2004 13:14:28 -0000 Subject: SUMMARY: E4500 mystery crash Message-ID: <412iDLNgn9488S03.1081775199@uwdvg003.cms.usa.net> Thanks to those who responded, especially those out of the office. All who sent real replies said its a CPU and pointe out http://www.itworld.com/nl/unix_sys_adm/04032002/ As we didn't have much information we couldn't pin down any one part. We ended up moving the applications to another system (I love fibre channel). Once the old system was freed Sun came in a checked every component. They found things like a missing SCSI terminator on the first I/O board, really old SAMBRA modules, DIMM's that where suspect, and really old CPU/memory boards. All boards had all cards reseated or retorqued, suspect and old components have been replaced, and the firmware is at current levels. We ran VTS (yeah I know, niced exerciser, so-so diagnostic tool) all week ened with no errors. At this point we'll probably re-deploy the system for something else. JC Orignal message: Several times in the past 6 or so weeks one of our E4500's had either hung, requiring a power off/on, or suddenly rebooted. The only indication we got is the following console messages: TL=0000.0000.0000.0005 TT=0000.0000.0000.0068 TPC=0000.0000.f000.3014 TL=0000.0000.0000.0004 TT=0000.0000.0000.0034 TPC=0000.0000.1000.8574 TL=0000.0000.0000.0003 TT=0000.0000.0000.0068 TPC=0000.0000.f000.3014 TL=0000.0000.0000.0002 TT=0000.0000.0000.0030 TPC=0000.0000.1000.8574 TL=0000.0000.0000.0001 TT=0000.0000.0000.0068 TPC=0000.0000.60fe.44d0 Software Power ON The OS is 2.6, Sun Cluster 2.2 and Resource Manager. The other node in the cluster is running just fine so I'm inclined to say it is a hardware issue rather than software, but I'm not completly ruling that out either. Our suspicion is a CPU but without any further information Sun wont venture a real guess. As this is a mission critical (actual dollars earned) 24x7 system we can not afford to have an extensive outage for hardware testing for another 1 = weeks. We are implementing a contingency in case the beast dies again, but Id prefer to fix the current system. As anyone seen similar symptoms? If so was there a viable solution? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From terryfranklin at sbcglobal.net Mon Apr 12 11:18:48 2004 From: terryfranklin at sbcglobal.net (Terry Franklin) Date: Mon, 12 Apr 2004 15:18:48 -0000 Subject: SUMMARY: sendmail and smart relays Message-ID: <20040412151132.57985.qmail@web80309.mail.yahoo.com> Figured this one out on my own. For some reason DNS was timing out, at least from the perspective of sendmail, on occasion. I have never had trouble with Dig or nslookup but nontheless I solved the problem by uncommenting O DirectSubmissionModifiers=C from submit.cf and bouncing sendmail. Thanks Terry _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From lance.tost at respironics.com Mon Apr 12 13:44:37 2004 From: lance.tost at respironics.com (Lance Tost) Date: Mon, 12 Apr 2004 17:44:37 -0000 Subject: SUMMARY: MemTool on a SunFire 4800 In-Reply-To: Message-ID: Original question follows summary. I got a couple of responses related to needing a higher kernel patch, but that's not it since the same version of memtool runs fine on a Sol8 box at patch level 17. William D. Hathawa hit the nail on the head by pointing me to a post related to the line "exclude: lofs" in /etc/system. Apparently, this has something to do with Sun Cluster (which I forgot to mention in my original post -- our 4800s have Cluster installed). I cannot test on the 4800 since I cannot reboot it... but I added the "exclude: lofs" line to /etc/system of a host where prtmem worked... rebooted with the exclude and now it does not work. Here's part of the reply with the solution: I also found the page you got the 404 on using the google cache, a URL that works is: http://216.239.41.104/search?q=cache:E8Z18sLbiAcJ:sunportal.sunmanagers.org/pip rmail/summaries/2002-September/003867.html+003867.html+memtool&hl=en&ie=UTF-8 Thanks Also, a special thanks to the following vacationers: STELLA KONG SANTA Giovanni Obst, Thomas Ian Pease Klas.Erlandsson Broderick, Sean Pohl, Stefan Jonathan Jackson Kulkarni Pawan-W19 On Mon, 12 Apr 2004, Tost, Lance wrote: > Date: Mon, 12 Apr 2004 12:36:45 -0400 > From: "Tost, Lance" > To: Sun Managers > Subject: MemTool on a SunFire 4800 > > > Anyone try using MemTool (ftp://playground.sun.com/pub/memtool/) on a > SunFire 4800? I have the latest version (3.9.4) and it works on all of > my > servers except my two 4800s. > > root at riuxerp1[/]# uname -a > SunOS riuxerp1 5.8 Generic_108528-22 sun4u sparc SUNW,Sun-Fire > root at riuxerp1[/]# prtmem > > The MemTool device driver is not loaded > > Run /opt/RMCmem/drv/bunyipload to load the driver > > root at riuxerp1[/]# > root at riuxerp1[/]# /opt/RMCmem/drv/bunyipload > can't load module: No such file or directory > root at riuxerp1[/]# > > > >From messages while doing the above: > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'nfs3_vnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'pcfs_dvnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'ufs_vnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'nfs_vnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'tmp_vnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'pcfs_fvnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 995402 kern.notice] > /opt/RMCmem/drv/5.8/sparcv9/bunyipmod: undefined symbol 'lo_vnodeops' > Apr 12 12:31:13 riuxerp1 krtld: [ID 472681 kern.notice] WARNING: > mod_load: > cannot load module 'bunyipmod' > > > Anyone have any ideas? Googling turned up this question but no answer > here: http://www.netsys.com/sunmgr/2002-08/msg00248.html and a SUMMARY > here: > http://sunportal.sunmanagers.org/pipermail/summaries/2002-September/003867.html > > which gives me a 404. Also turned up this, > http://groups.google.com/groups?hl=en&lr=lang_en&ie=UTF-8&oe=UTF-8&c2coff=1&sel > =K9ut5.678%24zx3.9811%40iad-read.news.verio.net, > > which has no answer/summary. > > Thanks > > -- > Lance Tost, Systems Engineer > lance.tost at respironics.com > > > > ======================================================================== > CONFIDENTIALITY NOTICE > ---------------------- > This message, together with any attachments, may be legally privileged > and is confidential information intended only for the use of the > individual or entity to which it is addressed. It is exempt from > disclosure under applicable law including court orders. If you are not > the intended recipient, you are hereby notified that any use, > dissemination, distribution or copy of this message, or any attachment, > is strictly prohibited. If you have received this message in error, > please notify the original sender and delete this message, along with > any attachments, from your computer. > _______________________________________________ > sunmanagers mailing list > sunmanagers at sunmanagers.org > http://www.sunmanagers.org/mailman/listinfo/sunmanagers > > > -- Lance Tost, Systems Engineer lance.tost at respironics.com ======================================================================== CONFIDENTIALITY NOTICE ---------------------- This message, together with any attachments, may be legally privileged and is confidential information intended only for the use of the individual or entity to which it is addressed. It is exempt from disclosure under applicable law including court orders. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution or copy of this message, or any attachment, is strictly prohibited. If you have received this message in error, please notify the original sender and delete this message, along with any attachments, from your computer. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From terryfranklin at sbcglobal.net Mon Apr 12 13:55:13 2004 From: terryfranklin at sbcglobal.net (Terry Franklin) Date: Mon, 12 Apr 2004 17:55:13 -0000 Subject: SUMMARY: sendmail and smart relays (resent) Message-ID: <20040412174817.76538.qmail@web80303.mail.yahoo.com> Sorry, neglected to include the original message. I had written: *********************** All, I have sporadic behavior with sendmail on Solaris 9 and I am hoping someone can point me in the right direction. Basically I can't figure out what determines when you will get a stat=Sent vs stat=queued. I am getting a dsn=4.4.3 which usually relates to an unresolvable domain, but it's sporadic. My sendmail.cd is pointing to a smart relay host. If I do a sendmail -C/etc/mail/sendmail.cf -t -v and enter everything manually, the email goes through correctly and isntantly. Using mailx though, works sporadically. Sometimes an email that gets the stat=queued won't be sent for as much as 9 hours. But it's not a consistant behavior. And the fact that I can either telnet to port 25 on the smtp server and send directly or use my sendmail.cf file as above leads me to think it's something to do with the sendmail process. How or what is determining what is queued here? Why would I be able to send an email to my pager (an internet address) and have it work fine one minute, but get queued for an extrememly long time the next? And why would sendmail behave differently between doing it as sendmail -C/etc/mail/sendmail.cf -t -v vs it's interaction with CLI mail tools like mailx? Any insight is appreciated. ********************** The solution: Figured this one out on my own. For some reason DNS was timing out, at least from the perspective of sendmail, on occasion. I have never had trouble with Dig or nslookup but nontheless I solved the problem by uncommenting O DirectSubmissionModifiers=C from submit.cf and bouncing sendmail _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From cmjohnson at uslec.com Mon Apr 12 14:11:00 2004 From: cmjohnson at uslec.com (Johnson, Chad) Date: Mon, 12 Apr 2004 18:11:00 -0000 Subject: SUMMARY: Automatic installation of packages / drivers Message-ID: The sum conclusion seems to be: 1. Install SUNCWall or/and SUNCWxall just after the flash image is restored 2. rm /etc/path_to_inst (as it probably won't contain the correct dev's anyway) 2. Touch /reconfigure (or a reconfigure reboot) These seemed to work quite well. I have received multiple persons responding with "touch /reconfigure". This is normally fine, but the situation is such that the drivers for the new system may not exist on the image which is installed onto this new system. Since these systems are restored with jumpstart / flash install the original CD images are present at the time of the image load. If I 'touch /reconfigure' then reboot, the cd images with all the relavent drivers are not present and the reconfigure fails. Normally I did not 'touch /reconfigure', I just did a boot -r, but it is the same net effect. ~~~~~~~~~~~~~~~ Original Post Below ~~~~~~~~~~~~~~~~ Hello all! Here is my situation, frequently I have to replicate systems, using jumpstart / flash install, from one machine to another, not necesssarily the same class box. The majority of the issues I have are quite simple driver issues, but definatly cost me a lot of time. My idea is this, when booting to a cd or network image into single user mode the system which comes up detects the devices and loads the appropriate drivers. I would like to do this same soft of thing on a newly restored system. Essentially, restore the flash archive onto the disk, then run 'whatever' to detect and load the appropriate drivers. Even if I cannot load the drives automatically, how could I tell what I need to load so I can do it later? TIA, Chad Johnson _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From peter at ashlyn.co.uk Tue Apr 13 06:23:28 2004 From: peter at ashlyn.co.uk (Peter Stokes) Date: Tue, 13 Apr 2004 10:23:28 -0000 Subject: SUMMARY: Scroll Wheel/Wireless mouse for Sun Ray? In-Reply-To: <1079628518.1064.91.camel@sunsys.localdomain> References: <1079628518.1064.91.camel@sunsys.localdomain> Message-ID: <1081851207.17372.21.camel@sunsys.localdomain> Hi All Bit of a late Summary, but have been waiting for delivery of a couple of mice to confirm. I received a few replies (thank you, as ever the best Sun admin list out there), which confirmed that some of the Logitech mice will work fine using the scroll wheel/wireless connection. Where it works, I was using Sol 9, latest SRS software and Gnome 2.0 with Mozilla and evolution 1.4.5. Acrobat does not work. Also the wheel does not appear to work with SunPCI II, which I guess is down to the Sun driver for Win2k (maybe there is an update?) I then decided to buy a few of them to test them. Results are Logitech Click! Optical (wired) USB mouse - Works 100% Logitech Cordless Mouseman Optical - Works 100% Logitech Cordless Mouse (roller ball) - does not work I also found some references to wheel support for systems (not Sun Ray) and that appears to be something of a problem on earlier versions of Solaris (Sun Ray should be ok). No idea if recent Sol 9/10 still has the problem. On Thu, 2004-03-18 at 16:48, Peter Stokes wrote: > Hi > > As I now use the Sun Ray's for my day to day work, I really would like > to use a scroll wheel/wireless mouse for browsing etc. Anyone have > experience of using one on Sun Rays and how to do it? > > Thanks > > Peter > _______________________________________________ > 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 PhippsAC at telkom.co.za Tue Apr 13 07:19:13 2004 From: PhippsAC at telkom.co.za (Adrian Phipps (AC)) Date: Tue, 13 Apr 2004 11:19:13 -0000 Subject: SUMMARY: Veritas Volume Manger 3.5 GUI problems Message-ID: Thanks to Philippe Ghazarian who suggested I first install the VRTSfspro package before opening the GUI. I am able to create and see all my file systems via the VEA GUI now. Thanks again Philippe Adrian Phipps 90 Durban Road IT Building Bellville Work 021-949 4642 Cell 082 787 7321 Fax 021-945 4740 <> > -----Original Message----- > From: Adrian Phipps (AC) > Sent: Tuesday, April 13, 2004 12:30 PM > To: 'sunmanagers at sunmanagers.org' > Subject: Veritas Volume Manger 3.5 GUI problems > > HI all, > > I have a very strange problem, on the VEA GUI I am unable to create > a file system. All the documentation points me to the action menu > tab but those options just do not appear. Veritas.com have no > information regarding this or I am searching for the wrong phrase. > > I can create a volume but it basically skips the stage when I have > to create the file system? There are also no options to enable, > basically Veritas File system is missing from the VEA gui. > > Your help will be greatly appreciated. > > Regards > > > Adrian Phipps > 90 Durban Road > IT Building > Bellville > Work 021-949 4642 > Cell 082 787 7321 > Fax 021-945 4740 > > << OLE Object: Picture (Metafile) >> [demime 1.01b removed an attachment of type image/bmp which had a name of ole0.bmp] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mikelist at sky.net Tue Apr 13 11:51:20 2004 From: mikelist at sky.net (Mike's List) Date: Tue, 13 Apr 2004 15:51:20 -0000 Subject: SUMMARY: HP and CDROM issues Message-ID: Thanks to Darren Dunham, Peter Ondruska, Annamria Heupke, krenzischek, and hmnguyen. The "HPdds options value invalid bits set: 0x4000" is normal in Sol9 and can be ignore. The tape works fine and I have not experience any problems with ufsdump over the weekend. As for the second issue, I grab another CDROM and replace the previous. Also, I was able to ftp over the necessary packages. I ran fstyp on /dev/dsk/c1t6d0s0 on CD 1/2 and keeps on returning ufs versus hsfs, hmnguyen recommended mounting as cdfs but I didn't have the opportunity to attempt this. Regards, - Mike [ In addition to www.sunfreeware.com more packages at ftp.patriots.net ] [ Furthermore, you can explore www.openpkg.org -- ease of management. ] ---------- original message ---------- I'm installing a new server, Solaris 9 4/04 with core installation, with the following hardware controllers/config. c0t0 - boot disk c1t5 - HP DAT-24i (id5, term power jumper set on) c1t6 - NEC CDROM (id6, term power jumper set on) issue #1 -------- During installation I received the following error message: (and after installation, during boot) HPdds3 optons value invalid bits set: 0x4000 I believed I have all the jumpers set correctly on the HP DAT-24i, according to HP user/installation guide. Is the above normal error message and can be ignore or do I need to investigate the rest of the SCSI devices within this controller? There are two devices on this specific SCSI controller, HP DDS3 on target 5, and NEC CDROM on target 6. I think there's a parity option on the CDROM, but it's not set, not sure if this matter? issue #2 -------- After installation (no patch installation or anything else added) I can't seem to mount my Solaris CD 1/2 (as HSFS) to add some packages. I can mount my Solaris CD 2/2 just fine (same command/syntax). # mount -F hsfs /dev/dsk/c1t6d0s0 /cdrom hsfs mount: /dev/dsk/c1t6d0s0 is not an hsfs file system. I can boot with the CD 1/2 and do installation, and even burn another CD 1/2 to make certain it's not the media, I just can't mount as HSFS. solutions tried --------------- I've searched the archive, and it's not a vold issue. I attempted to mount different slice s0/s1/s2/etc. and still the same error message. Looking forwards to some hints/solutions, thanks in advance. - Mike [ In addition to www.sunfreeware.com more packages at ftp.patriots.net ] [ Furthermore, you can explore www.openpkg.org -- ease of management. ] _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From tbw at geo.hunter.cuny.edu Tue Apr 13 12:55:32 2004 From: tbw at geo.hunter.cuny.edu (Thomas Walter) Date: Tue, 13 Apr 2004 16:55:32 -0000 Subject: SUMMARY: Sol8 - local disk won't mount In-Reply-To: Message-ID: Thank you all for the quick responses! fuser showed no open files. The disk was not mounted anywhere else. I could successfully mount the drive on /mnt. Chad Johnson suggested I remove/remake the mount point which worked! Thanks Chad! Tom ================================================================================= On Tue, 13 Apr 2004, Thomas Walter wrote: > > Good Morning, > > I just rebooted a Solaris 8 E250 and one of the hard drives didn't mount. > When I try to mount manually I get the following message: > > mount -a /data/moon/a > mount: /dev/dsk/c0t0d0s4 is already mounted, /data/moon/a is busy, > or the allowable number of mount points has been exceeded > > I need access to this drive asap so I'm hoping someone has the secret. > > TIA > > Tom > > > ================================================================================== > > Thomas Walter > Geography & Computer Science Departments > Hunter College of the City University of New York > 695 Park Avenue > New York, NY 10021 > > (212)772-5457 Office > (212)772-5268 Fax > tbwalter at geo.hunter.cuny.edu > http://geography.hunter.cuny.edu/~tbw > > > -- ================================================================================== Thomas Walter Geography & Computer Science Departments Hunter College of the City University of New York 695 Park Avenue New York, NY 10021 (212)772-5457 Office (212)772-5268 Fax tbwalter at geo.hunter.cuny.edu http://geography.hunter.cuny.edu/~tbw _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sloewenthal at gemini.edu Tue Apr 20 11:39:34 2004 From: sloewenthal at gemini.edu (LOEWENTHAL Simon) Date: Tue, 20 Apr 2004 15:39:34 -0000 Subject: SUMMARY: Broadcast address on Solaris mysteriously changes. In-Reply-To: Message-ID: Hi all, Thanks to everyone who replied to me. Original post is included at the end of this mail. * Decided that it was a 3rd party app calling ifconfig. Added an ifconfig wrapper to log ifconfig usage. * Removed extra entry from /etc/netmasks. * Rebooted machine. After reboot watched the logs and saw something change the NIC. I am in the process of hunting down the s/w thaat did it. Regards, Simon. --- Simon Loewenthal Gemini Observatory Information Systems Group ORIGINAL POST: -----Original Message----- From: LOEWENTHAL Simon Sent: Friday, April 16, 2004 11:24 AM To: Sun Mangers Subject: Broadcast address on Solaris mysteriously changes. Hi all, We have a Sun system whose broadcast address changes occasionally from a Class C broadcast address to a Class B broadcast address. I.E from 172.17.2.255 to 172.17.255.255. I have entered this (See below) into the /etc/netmasks to try and keep the broadcast address on track, but because the address mysteriously changes during normal day-to-day operation and the /etc/netmasks is only (I think?) consulted at during boot time, the broadcast address still changes. $ cat netmasks 172.17.0.0 255.255.255.0 172.17.2.0 255.255.255.0 172.17.0.0 255.255.255.0 172.17.3.0 255.255.255.0 $ ifconfig -a lo0: flags=1000849 mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 hme0: flags=1000843 mtu 1500 index 2 inet 172.17.2.64 netmask ffffff00 broadcast 172.17.255.255 ether 8:0:20:b2:f4:27 hme1: flags=1000843 mtu 1500 index 3 inet 172.17.3.64 netmask ffffff00 broadcast 172.17.255.255 ether 8:0:20:b2:f4:27 $ showrev Hostname: trecsk-ctl Hostid: 80b2f427 Release: 5.8 Kernel architecture: sun4u Application architecture: sparc Hardware provider: Sun_Microsystems Domain: cl.gemini.edu Kernel version: SunOS 5.8 Generic 108528-03 August 2000 $ uname -a SunOS trecsk-ctl 5.8 Generic_108528-03 sun4u sparc SUNW,UltraSPARC-IIi-cEngine Has anyone seen this before, or can anyone think of a process that would change this? Many thanks and regards for any time spent on this, Simon. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.661 / Virus Database: 424 - Release Date: 19/04/2004 _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From egold at fsa.com Tue Apr 20 11:40:19 2004 From: egold at fsa.com (egold at fsa.com) Date: Tue, 20 Apr 2004 15:40:19 -0000 Subject: SUMMARY: shell script help Message-ID: thanks to all who helped, i recieved over 20 responses in 10 minutes. Several people mentioned the list is not for this type of question, sorry about that. if anyone knows a good shell script list id like to know. the solution was to to use the IFS variable, which i think tells cat to use that value as its field seperator, just make to sure to set IFS back to its original value when done or you can break alot of other stuff. thanks again! Well, in the simplest case, you simply need to alter 'IFS' (Inter-field separator) It defaults to 'whitespace'. And I'd recommend 'saving' it because otherwise stuff can break eg. OLDIFS=$IFS IFS=" " IFS=$OLDIFS You may also wish to consider doing something like (after setting the IFS): for i in `cat file` do set $i KEY=$1 EDATE=$2 ... done (Although if you're doing that kind if thing, I might suggest having a look at perl) ORIGINAL MESSAGE: Subject: shell script help i need help parsing a text file in korn shell on solaris 7. I cant use perl. i want to grab each line of an input file then break out the columns (probably using awk). here is a sample input file: TEXT,04/04/04,04/20/04,05/01/03,a description with spaces in it. JUNK, 4324, 4323432, 324231, another description with spaces in it. . . . i was using the following code, but then the users started putting spaces in the last column and broke my script: for i in `cat inputfile` do KEY=`echo $i|awk -F, '{print $1}'` EDATE=`echo $i|awk -F, '{print $2}'` MR=`echo $i|awk -F, '{print $3}'` ENDDATE=`echo $i|awk -F, '{print $4}'` LASTCOL=`echo $i | awk '{print $5}'` done thanx in advance! E _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers :._______________ CONFIDENTIALITY : This e-mail and any attachments are confidential and may be privileged. If you are not a named recipient, please notify the sender immediately and do not disclose the contents to another person, use it for any purpose or store or copy the information in any medium. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From sudhakar.peram at corpusinc.com Tue Apr 20 16:06:53 2004 From: sudhakar.peram at corpusinc.com (Sudhakar) Date: Tue, 20 Apr 2004 20:06:53 -0000 Subject: SUMMARY:rpc.pcnfsd or alternative for it Message-ID: <032001c42712$f69ff830$be9c8071@its.verizon.com> Hi Managers, I do apologize for the delay that in the summary. I would like to thank the following Managers for their replies first: Tim Evans James Matt Rob Kevin Korb Howard Hurst Hendrik Visage Vihn Cao Mike Salehi Anthony Talltree David Foster The following are the suggestions that I got from the gurus. - Use samba http://www.samba.org - hclnfsd ftp://ftp.hummingbird.com/pub/maestro/hclnfsd/ - pcnfsd http://support.wrq.com//techdocs/1429.html I went with pcnfsd as it was easy to install and run. Regards, Sudhakar _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mrb at bmyster.com Tue Apr 20 16:40:08 2004 From: mrb at bmyster.com (Brent Bailey) Date: Tue, 20 Apr 2004 20:40:08 -0000 Subject: SUMMARY: how to do stop-a from the service console or SC Message-ID: <10217.207.5.142.198.1082493847.squirrel@new.host.name> First ..i want to than everyone who helped me :-) My question was : Im setting up a sunfire v440 w/ solaris 9 12/03 ...the machine didnt come with a keyboard so i have done the install from the SC (service console) and a laptop ...now that the system has been installed i want to apply the recommended updates however ..the way i have done it in the past is # init 6 then do "stop-a" and do boot -s then apply the patches...i dont have a sun keyboard so i cant do a stop-a.. is there a equivilent to the stop-a keystroke on a regular keyboard ? THE ANSWER TO THIS WAS: you need to send a break from the PC terminal. But rather than init 6 (reboots) use init 0 (shutsdown to the 'ok' prompt). Then you can boot -s -- Brent Bailey _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From versun2001 at yahoo.com Wed Apr 21 04:06:29 2004 From: versun2001 at yahoo.com (Shailendra Dawane) Date: Wed, 21 Apr 2004 08:06:29 -0000 Subject: SUMMARY: NIS+ Message-ID: <20040421080557.87315.qmail@web41410.mail.yahoo.com> Many Many Thanks to Tim Villa for a quicky. My questions were 1) to rename an existing NIS+ user to a different answer for eg rename ABC1 to ABC 2) To Modify the home directory. for eg change /home/ABC1 to /home/ABC Solution : # nistbladm -e name=ABC \[name=ABC1\],passwd.org_dir # nistbladm -e home=/home/ABC \[name=ABC\],passwd.org_dir Please Note :If you have any credentials kindly apply to cred.org_dir. Thanks again TIM . Thanks shailendra Yahoo! Photos: High-quality 4x6 digital prints for 25" _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From slautier at amadeus.net Wed Apr 21 09:45:28 2004 From: slautier at amadeus.net (Sabrina Lautier) Date: Wed, 21 Apr 2004 13:45:28 -0000 Subject: SUMMARY: How to check the update version of a solaris 8 server ? Message-ID: Dear admins, First and foremost thanks a lot for all your answers ! Here is the solution: $ cat /etc/release | Solaris 8 10/01 s28s_u6wos_08a SPARC | Copyright 2001 Sun Microsystems, Inc. All Rights Reserved. | Assembled 12 September 2001 The current update version of our Sol8 server is then 6 (u6). FYI, to update from 6 to 7 I have to install the Maintenance Updates MU7 S8MU7_sparc.zip. Regards, Sabrina ----- Forwarded by Sabrina Lautier/NCE/AMADEUS on 21/04/2004 15:24 ----- From: Sabrina Lautier @sunmanagers.org on 21/04/2004 11:17 ZE2 Sent by: sunmanagers-bounces at sunmanagers.org To: sunmanagers at sunmanagers.org cc: Subjec How to check the update version of t: a solaris 8 server ? Dear admins, Does anybody know how to check what the update version is on my solaris 8 server ? I'm not asking for the release version (which I can easily get with the uname command) but for the update one. To move to a new disk array, one of the prerequisites is to have a 7 solaris 8 OS update version. And I don't really know how to check this... We installed the OS from a Sol8 10/01 CD-ROM. I guess the update version is calculated according to the patches applied since the installation. Any idea of how to check this ? Thanks a lot for your help, Sabrina _______________________________________________ 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 d.matinyarare at iorin.gov.pl Mon Apr 26 06:31:19 2004 From: d.matinyarare at iorin.gov.pl (David Matinyarare) Date: Mon, 26 Apr 2004 10:31:19 -0000 Subject: SUMMARY: ufsrestore Message-ID: <6682A0A20A720942AC26337E5E5C03EA1435C9@www.iorin.gov.pl> By the way how do I do a summary, then send to all sunmanagers' list? Here I am editing a new mail, which is not what I want. Thank you very much to all. Especially those who spared their time to help me. I got mails within minutes which pointed either using the "mt" command with the "fsf" option to move the header to the correct position or "ufsrestore" with the "-s" option to get to the same position as with "mt" command. i.e. # mt rewind # mt -f /dev/rmt/0n fsf 3 # ufsrestore -if /dev/rmt/0 or # mt rewind # ufsrestore -is 4 My orginal question was: >I use ufsdump to backup solaria 8 file systems on one tape. Of late I > wanted to go the restore procedure preparedness using ufsrestore. When > I do "ufsstore itf /dev/rmt/0" I get root (/) file system listing > only. If I run "ufsrestore vt /dev/rmt/0", I get "Level 0 dump of / on > [debango]: /dev/dsk/c1t0d0s0. What about the rest, /dev/dsk/c1t0d0s3, > (/usr), /dev/dsk/c1t0d0s4, (/var), /dev/dsk/c1t0d0s6, (/opt) and > /dev/dsk/c1t0d0s7, (/export/home)? When I did the backup, i.e. ufsdump > 0ucf /dev/rmt/0n [file systems to be backed-up], I saw these file > systems being dumped to tape. In other words, how do I get to the > /usr, /var, /opt, and /export/home file systems on tape using > ufsrestore? _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From rob.delanghe at telindus.be Tue Apr 27 02:46:02 2004 From: rob.delanghe at telindus.be (Rob De Langhe) Date: Tue, 27 Apr 2004 06:46:02 -0000 Subject: SUMMARY: advise wanted for multi-terabytes filesystems Message-ID: Needless to say that there were again numerous valuable replies ... Thx to Grzegorz Bakalarski, Octave Orgeron, Steve Starer, Alex Madden, Przemol, and specially Dik Casper who mentioned the most important aspect : at creation-time of a filesystem (with Solaris' "newfs") that is intended to grow past the 1TB limit, use the "-T" option to allow it to grow as such. Opinions were generally ok with using Solaris-9's UFS type of filesystem, creating it with large blocks-per-inode ("newfs"-option "-i", set to multi-megabyte size), and using the logging feature (mount-option "logging") to accelerate/avoid any FSCK. Altough maybe technically possible, I guess we will follow anyhow some people's advice to keep not everything on a single filesystem but store the Oracle data-files over multiple approx-1TB filesystems. Several people suggested using Veritas' VxFS, but to us that's no added value compared to the features already present in Solaris-9, and above all it's quite expensive as well. If anyone can still point me to an objective performance-comparison between VxFS and (Solaris-)UFS, that would be very welcome. Someone also mentioned the possibility to bypass the filesystem-layer, and let Oracle use the raw disk access method available in Oracle-9. Again, if someone can present objective performance-comparisons, would be very nice. Performance-wise we have the LUNs set up as RAID-5 sets over 10 disks for each LUN (giving good striped performance while reading, sure we know it's slower during writes bcos of the RAID-5 operations), and the LUNs are balanced over the two controllers in the EMC array, each controller being connected to a separate HBA in the server using FCAL. At this moment, that renders service-times of 1-4 msec for 200-800 KB read/sec plus 300-2000 KB written/sec, the CPU (SF480R, 2x900MHz) meanwhile busy for 20% of its time blocked on I/O. Is this good, or could it be (much) better ? Rob -----Original Message----- From: Rob De Langhe Sent: Friday, April 23, 2004 9:17 AM To: 'sunmanagers at sunmanagers.org' Subject: advise wanted for multi-terabytes filesystems Hi, we have a pair of (Solaris-9) Oracle-servers connected via FCAL to EMC storage, with RAID-5 LUNs from that storage array managed by Veritas Volume Manager on the hosts. On these LUNs, we have create filesystems to hold the Oracle files. One of them, that contains the actual data-files, is mounted as "/oraAdb" with some subdirectories for each database instance : /oraAdb/instanceA /oraAdb/instanceB and so on This filesystem is created using plain Solaris-command ("newfs -i 65536 -f 8192 /dev/vx/rdsk/..."), so no Veritas-Filesystem is used. The option "logging" is used in "/etc/vfstab" to have it journalling its changes. Since our database will grow from (currently) 1 TB to approx 14 TB, I was wondering how I should keep this stable : is it recommended/safe practice to keep all the Oracle-DBF files on a single, therefor multi-terabyte filesystem (mount on "/oraAdb") ? What if the server crashes suddenly : is it sufficient/safe to have the "logging" option so that a FSCK will not take days to complete ? Are there any Solaris limits we will hit when growing this filesystem to is max capacity ? Or should we split the Oracle-DBF files over multiple -say 1TB capacity- filesystems, and just add more of these filesystems ? TIA for any suggestions and/or experiences. Rob Visit us at the Telecom cITy Fair - The largest IT Fair in Belgium! 25, 26, 27 May - Brussels Expo Get your free tickets here! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers Visit us at the Telecom cITy Fair - The largest IT Fair in Belgium! 25, 26, 27 May - Brussels Expo Get your free tickets here! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From lhecking at nmrc.ucc.ie Tue Apr 27 12:51:51 2004 From: lhecking at nmrc.ucc.ie (Lars Hecking) Date: Tue, 27 Apr 2004 16:51:51 -0000 Subject: SUMMARY: Ultra 60 and TFT Message-ID: <20040427165115.GA1181@nmrc.ie> Original question: http://www.sunmanagers.org/pipermail/sunmanagers/2004-April/030225.html Thanks to Peter Stokes for the only reply: | Use the OBP value for output-device and set to | | screen:r1280x1024x76 | | That should do it. I did some more digging on sunsolve, and the only reference I could find on sunsolve describes this method in the context of SPARCstation 4. It works fine on the U 60, though. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From jElser at ck8.uscourts.gov Tue Apr 27 16:03:11 2004 From: jElser at ck8.uscourts.gov (John Elser) Date: Tue, 27 Apr 2004 20:03:11 -0000 Subject: SUMMARY: Boot Problem with RAID subsystem Message-ID: <059701c42c92$9e16fad0$61617f9c@JELSER> Thanks Joe Fletcher, Chris Hoogendyk, Bruce Newcomer and Chuck for your replies and suggestions. A couple of people suggested that I simply put a startup script in rc3.d on the internal drive and have it check the status of the raid system. If the raid system isn't running, pause and check it again. If, after a 30 second pause, the raid is up and ready, then reboot the system. Otherwise send an e-mail to notify me that there is a problem with the raid system. Another suggestion was to us a startup script on the internal drive that puts up a note on the console saying "really start up from internal drive?" and then pauses for, say, 30 seconds. if no answer in 30 seconds, then issue a reboot off the raid device. that way, you aren't locked out of booting off the internal drive, but the switchover is automated to the raid if you aren't explicitly asking otherwise. Another suggestion was to use a power sequencer to make sure the raid system is up before the system. The last suggestion was to disable booting from internal drive; check eeprom settings. Probably have multiple boot drives specified. Thanks for all of the suggestions. I haven't had a chance to determine which path I'll take, but I'm sure one of these is the method we will use. Thanks again!!! _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From mario.b.agustin at lmco.com Tue Apr 27 19:58:13 2004 From: mario.b.agustin at lmco.com (Agustin, Mario B) Date: Tue, 27 Apr 2004 23:58:13 -0000 Subject: Sun's SRC/P card, Solaris upgrade SUMMARY Message-ID: <869557E0E4BD3C4FB993526E15C869597523A8@emss02m06.us.lmco.com> Ted John you wrote " I upgrade a similar box (e450, with non-os slices in raid through the HW raid card) from Sol 2.6->8 about a month ago. I simply did the upgrade ; confirmed the OS stuff was good ; then uninstalled the appropriate packages, applied the new ones (downloaded from sunsolve) which provided support to this hardware under solaris 8. After a reboot, everything was back to normal (ie, mounts on the HW raid controller were visible again). " Where can I download the packages? Mario Agustin Staff Unix Administrator l Enterprise Information Systems 1300 S. Litchfield Rd., Goodyear, AZ 85338 Phone - (623) 925-7890, Pager 623-580-2712 From joe.philip at verizon.net Wed Apr 28 06:41:47 2004 From: joe.philip at verizon.net (Joe Philip) Date: Wed, 28 Apr 2004 10:41:47 -0000 Subject: Summary: solaris password settings References: <00e001c4296f$f456e2a0$1c01a8c0@miniwork> Message-ID: <003301c42d0d$41f75710$13051eac@miniwork> Thanks to everyone who responded. For setting the strong password requirements, mostly everyone suggested to use "npasswd." For cracking the weaker passwords which are already used on the system, the general advice was to use "Crack," "John the ripper" etc. programs. BTW, I installed Crack 5.0 and it was kind of "cool" to find out how this program cracks passwords. To change the expiration date, warning date etc. currently I wrote a script which calls "passwd -x #days -w #warn_days " I will try to set the npasswd sometime. ----- Original Message ----- From: "Joe Philip" To: Sent: Friday, April 23, 2004 4:17 PM Subject: solaris password settings > I have sent an earlier question on strong passwords. Later I realized that my > question was kind of vague. > > I am re-wording my question: > > What should I set on my Solaris 7 and Solaris 8 servers so that when people > try to change their passwords, the system does not accept a weaker password. > It asks the user to type in a password which confirms to the strong password > policy. How do I achieve this? > _______________________________________________ > 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 harald.husemann at materna.de Wed Apr 28 06:51:55 2004 From: harald.husemann at materna.de (Harald Husemann) Date: Wed, 28 Apr 2004 10:51:55 -0000 Subject: SUMMARY: ssh doesn't care about password aging In-Reply-To: <1083147391.17803.16.camel@yellow> References: <6D659526F82AB74792B2270F5F579F8D0D3F37@100NOOSLMSG002.common.alpharoot.net> <1083147391.17803.16.camel@yellow> Message-ID: <1083149493.17803.25.camel@yellow> Hi agn, and thanks to Matt (mlh at zipworld.com.au) for the fast reply! It seems we need a newer openssh or a patch for the old one. It's a known behavior of the old-fashioned ssh servers, :-) Matt's reply: ==================/snip/====================== Get a recent openssh, i.e. 3.8 or later. Handling expired passwords required a patch up to 3.7.1 See this page for more info: http://www.zip.com.au/~dtucker/openssh/ Matt ==============/snap/=========================== Thanks again to the list and Matt for the fast and reliable help, keep on hackin', Harald On Wed, 2004-04-28 at 12:16, Harald Husemann wrote: > Hi gurus, > > we have a problem here with one of our systems concerning password aging > and ssh. System is a V240, running Solaris 8 and openssh. > Password aging is working when a user connects the machine with e. g. > telnet, he/ she gets a warning when the password is about to expire, and > the access is denied when the password has been expired. > But, ssh doesn't care about this, there is no warning abt. the expiring, > and, even more worse, it's still possible to connect the system with ssh > when the password is expired... > > Any ideas?? > > Thanks in advance, > > have a nice hackin', > > Harald _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From stuart at IBS-PUBLIC-SERVICES.CO.UK Wed Apr 28 08:12:07 2004 From: stuart at IBS-PUBLIC-SERVICES.CO.UK (Stuart Morris) Date: Wed, 28 Apr 2004 12:12:07 -0000 Subject: SUMMARY: exe request Message-ID: Many thanks to all who assisted. Stuart. -----Original Message----- From: Stuart Morris [mailto:stuart at IBS-PUBLIC-SERVICES.CO.UK] Sent: 28 April 2004 11:33 To: 'sunmanagers at sunmanagers.org' Subject: exe request Hi all, Is there any chance someone could email me a copy of the /usr/bin/more executable for an I386 Solaris 8 build. Ours inadvertantly got piped to and wiped out, Doh!, and the media is not to hand. Of course, in time honoured fashion no backups are made of this dev box :o). tia, Stuart. _______________________________________________ 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 Mitchell.Lewars at penske.com Wed Apr 28 14:26:50 2004 From: Mitchell.Lewars at penske.com (Lewars, Mitchell (EM, PTL)) Date: Wed, 28 Apr 2004 18:26:50 -0000 Subject: SUMMARY: Solaris 8 Patch Breaks Commercial SSH ? Message-ID: I replaced my pam.conf and restarted sshd, I will continue to monitor but for now the cpu load is holding. -------------------- Probably a change to the pam.conf file -- and you may well need to restart the whole server. Bummer. Can you log in locally? Also, if there were local changes to the stock pam.conf, you'll need to reapply those to the new one supplied in the patch. _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From martini at mrpeabody.llnl.gov Wed Apr 28 17:22:46 2004 From: martini at mrpeabody.llnl.gov (Dave Martini 1) Date: Wed, 28 Apr 2004 21:22:46 -0000 Subject: SUMMARY: What versions of Solaris will run as LDAP clients? Message-ID: <200404282116.i3SLGIU11419@raider.llnl.gov> Looks like you must be at Solaris 8 or higher to run the LDAP client on Solaris. There are two main LDAP inplementations, Sun One Directory Server and Open LDAP. If you have Solaris systems, Sun One Directory Server is SUN's version of LDAP. Also, there is an article in the May issue of SysAdmin magazine regarding using LDAP to manage unix accounts. www.samag.com. Sun also has a blueprint book out on the subject. http://www.bookpool.com/.x/dcpehhtnfr/sm/0131456938 Dave Martini LLNL _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From PhippsAC at telkom.co.za Thu Apr 8 04:04:17 2004 From: PhippsAC at telkom.co.za (Adrian Phipps (AC)) Date: Thu, 08 Apr 2004 08:04:17 -0000 Subject: SUMMARY: Port errors - solaris 6 Message-ID: Hi all, Annamaria provided me with the correct command to get my ports enabled. start sac: /usr/lib/saf/sac -t 300 Regards Adrian Phipps 90 Durban Road IT Building Bellville Work 021-949 4642 Cell 082 787 7321 Fax 021-945 4740 -----Original Message----- From: Annamaria Heupke/Denic [mailto:heupke at denic.de] Sent: Thursday, April 08, 2004 9:55 AM To: Adrian Phipps (AC) Subject: Re: Port errors - solaris 6 start sac: /usr/lib/saf/sac -t 300 ah sunmanagers-bounces at sunmanagers.org schrieb am 07.04.2004 11:44:49: > > Hi guys, > I am busy trying to enable a serial port on a E250 by using the > following command "sacadm -e -p zsmon" but get the following message > "Can not contact SAC". Has anybody seen this before? I cannot find any > documents on this anywhere. This is the output from "sacadm -l" > PMTAG PMTYPE FLGS RCNT STATUS COMMAND > zsmon ttymon - 0 NO_SAC /usr/lib/saf/ttymon > #TTY Ports a & b > I am having difficulty getting a tip session established because of > this. > regards > Adrian Phipps > _______________________________________________ > 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 fabrizio at golconda.homeunix.net Thu Apr 15 02:48:42 2004 From: fabrizio at golconda.homeunix.net (fabrizio at golconda.homeunix.net) Date: Thu, 15 Apr 2004 06:48:42 -0000 Subject: SUMMARY: Authenticating users against OpenLDAP on Solaris 8 with PAM In-Reply-To: <38811.192.168.73.9.1081936391.squirrel@golconda.homeunix.net> References: <38811.192.168.73.9.1081936391.squirrel@golconda.homeunix.net> Message-ID: <35336.192.168.73.9.1082011165.squirrel@golconda.homeunix.net> SUMMARY: - http://www4.rcf.bnl.gov/~rpetkus/LDAP/LDAP.setup.html - http://www.ypass.net/solaris8/openldap/ Everything works now. Thanks to Robert Petkus. Original Post: > Hi all, > I've a working mail server (postfix+openldap+courier) on my SunBlade 150 > w/ Sol8. Now I would like to authenticate users of various services like > ssh, login, ftp, login against my openldap through pam. I compiled > nss_ldap and pam_ldap from padl.com. My purpose is to get a single > username and pwd for all services and later implenting SSO with kerberos. > What should I do to get auth works against PAM/LDAP? > Thanks in advance for your help. > -fabrizio -- How much for your women? I want to buy your daughter... how much for the little girl? -- Jake Blues, "The Blues Brothers" _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From G.Bakalarski at icm.edu.pl Thu Apr 15 10:15:22 2004 From: G.Bakalarski at icm.edu.pl (Grzegorz Bakalarski) Date: Thu, 15 Apr 2004 14:15:22 -0000 Subject: SUMMARY: Solaris + USB 2.0 IDE disk + NTFS = problems ? Message-ID: <20040415140821.GP18002@burza.icm.edu.pl> Dear Friends! Here is summary on the topic in "subject". Unfortunately I was right. Most agreed "no way on solaris sparc". Some suggested linux box. (We tried this but our standard distribution of linux did not work - one needs SCSI support enabled for USB storage and also ntfs support (which was not default in suse 8.2)). Linux by Sun i.e. Sun Java Desktop should work with ntfs but I'm not sure if with USB 2.0 Storage. Finally I gave up and borrowed a laptop with USB 2.0 and MS Windows XP. Then I copied data (80GB) through the net. Original question & detailed summary follows. Great thanks to all who tried to help. All you are great! GB -------------- Q ---------------------- I have a following problem: our supplier of data sent us data on IDE harddisk in USB 2.0 enclosure. The disk has NTFS file system. We are almost MS Windows free site. But the vendor is almost MS Windows only site. I have new V440 with Solaris 9 12/03 which has 4 USB 2.0 ports. I found on SunSolve that Solaris can "talk" with USB 2.0 Storage devices. But the problem is NTFS. I found only support for FAT32 (pcfs). Nothing about NTFS and Solaris Sparc. Is there any way to read data (from NTFS file system) directly on my V440 ? Or should I look for laptop with USB 2.0 ? Any other solutions? -------------- A ----------------------- 1) From Michael.Horton The sun java desktop "live" cdrom disc will permit one to manually mount an ntfs partition and manipulate files. The sjd "live" disc is a freebie. Ask your sun rep for a copy. A full installation of sun java desktop should have the same capabilities as well as other linux distributions. (sjd is Suse 8.x) ***** 2) From: Casper Dik You could use it on a PC and use something like partition magic to convert to FAT32 format; or ask teh vendor to reship on in UDF format or ISO format. Linux, I think, has some NTFS support. Casper ***** 3) From: Ryan Krenzischek You need to look for a machine running Windows NT or better and get the information off of the drive. Ryan ***** 4) From joe_fletcher If you can get hold of a linux system then it's capabale of reading NTFS. Failing that find a PC, hook the drive up and copy the contents across via SAMBA or similar. ***** 5) From: Tim Chipman FYI, USB2.0 is backwards compatible to USB1 spec, so you can plug the device on any PC which has USB. IF the PC doesn't support USB2.0 the bandwidth/throughput will be slower, of course, but it should still work... in the future // if at all possible, the folks at the remote site should (ideally) use FAT32 filesystem on this drive, since it is a "non secure" filesystem and more accessible to other operating systems. Or, consider using some more "standard" intermediate format (such as burned CD or DVD discs, with data archived into .zip or .tar files as desired to preserve longnames / user owner info etc etc as desired). IIRC there are ways to mount NTFS on linux without too much trouble, and even to do so with read-write access ... however, I don't recall that this is so easily done on Solaris (alas). Hope this helps a bit, Tim ***** 6) From: Jason Grove Solaris can not read NTFS. Only options are a PC/laptop with windows or maybe a linux box that can read ntfs. ***** 7) From: David Foster Check out the mtools package, it's kinda old but lets you read DOS/Windows data under older versions of Solaris. Dave Foster ***** 8) From: Peter Stokes As far as I know NTFS is a problem with Solaris. Suggest you go down the PC system route, or maybe Linux. The later Linux systems can read NTFS disks. Peter --------------------------EoT------------------------------------------ _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers From levins at westnet.com Thu Apr 15 12:31:35 2004 From: levins at westnet.com (Adam Levin) Date: Thu, 15 Apr 2004 16:31:35 -0000 Subject: partial summary and more details re: notrouter and multi-home Message-ID: Thanks to: Darren Dunham Crist Clark Brent Mcdaniel Charles Rawls bob.cummings Terry Gardner krenzischek at Encompasserve.org cprice at its.to First, /etc/notrouter has nothing to do with this. That file causes other machines to be unable to route packets through my multi-homed machine, but that's not the problem. More info: My ifconfig looks like this: lo0: flags=1000849 mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 hme0: flags=1000843 mtu 1500 index 2 inet 10.20.21.72 netmask ffffff00 broadcast 10.20.21.255 ether 8:0:20:a6:f2:f8 hme0:1: flags=1000843 mtu 1500 index 2 inet 10.20.21.92 netmask ffffff00 broadcast 10.20.21.255 hme1: flags=1000843 mtu 1500 index 3 inet 10.20.20.72 netmask ffffff00 broadcast 10.20.20.255 ether 8:0:20:ea:9b:43 hme0 and hme1 are on two physically different networks, both handled by a core Cisco switch. 10.20.21 and 10.20.20 are two different VLANs. tcpdump reports that traffic is coming in from one server, 10.20.20.100, via the hme0 interface (which is correct), but the return traffic is going out the hme1 interface, which is what I want to stop. The hme1 interface actually changes depending on which VLAN I need to jumpstart machines on -- we have half a dozen VLANs up there, and the idea was to move that port to whichever VLAN needed jumpstarts (and also to change the IP address appropriately, of course). /etc/netmasks is set correctly, with each subnet listing 255.255.255.0. Also, local-mac-address is set on, as you can see from the ifconfig output. So, the question really is: is there any way to force the multi-homed machine to respond via the same interface it received the traffic on? -Adam _______________________________________________ sunmanagers mailing list sunmanagers at sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers