INFO-VAX Thu, 04 Oct 2007 Volume 2007 : Issue 541 Contents: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: backup problem Re: DFU on the freeware disks Re: DSPP PAK diffs I64/Alpha Re: DSPP PAK diffs I64/Alpha Re: Guidance with OpenVMS IA64 8.3 with Java 1.5 Problems Re: SmartArray errors on system boot Re: VMS Apache on older versions? We're VMS - Please do not discuss the outside world Re: We're VMS - Please do not discuss the outside world Re: We're VMS - Please do not discuss the outside world Re: We're VMS - Please do not discuss the outside world Re: Web-based VMS Que Monitor for Deathrow Re: Web-based VMS Que Monitor for Deathrow Re: Web-based VMS Que Monitor for Deathrow Re: Web-based VMS Que Monitor for Deathrow Re: Web-based VMS Que Monitor for Deathrow ---------------------------------------------------------------------- Date: Wed, 03 Oct 2007 13:14:02 -0700 From: jhjr4381 Subject: backup problem Message-ID: <1191442442.812461.92500@g4g2000hsf.googlegroups.com> I'm still getting an error while running BACKUP, as follows(ALPHA VMS 6.2-1h3 ): $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - /NOALIAS - /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - RDAXP$MKB500:DUA510.BCK %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 -RMS-E-EXT, ACP file extend failed -SYSTEM-F-EXQUOTA, process quota exceeded I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a listing of files stored in the save-set. However, I'm at a loss as to how to determine which "process" QUOTA is being exceeded here. It looks like backup was trying to make the file larger in order to write to it, but failed to do so (maybe). Is this an error because I need x blocks of contiguous space (to extend the file)? Did this error stop the backup, or just the writing to this .log file? I looked at the process as it was executing (batch job) and got the following: RDAXP::SYSTEM> show process /quotas/identification=20400230 3-OCT-2007 09:30:22.28 User: SYSTEM Process ID: 20400230 Node: RDAXP Process name: "BATCH_5" Process Quotas: Account name: SYSTEM CPU limit: Infinite Direct I/O limit: 4096 Buffered I/O byte count quota: 148976 Buffered I/O limit: 400 Timer queue entry quota: 39 Open file quota: 995 Paging file quota: 292800 Subprocess quota: 10 Default page fault cluster: 64 AST quota: 99 Enqueue quota: 2000 Shared file limit: 0 Max detached processes: 0 Max active jobs: 0 Does this show me what "system's" quotas are for this process, or is this the process' quotas (or are they one and the same)? Is possibly a system quota that the process itself is exceeding? I've been searching through the FAQ's, HP's site, etc and haven't found a thing. Any words of wisdom out there? Thanks! ------------------------------ Date: Wed, 03 Oct 2007 20:51:09 GMT From: Rob Brown Subject: Re: backup problem Message-ID: On Wed, 3 Oct 2007, jhjr4381 wrote: > I'm still getting an error while running BACKUP, as follows(ALPHA > VMS 6.2-1h3 ): > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > /NOALIAS - > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > RDAXP$MKB500:DUA510.BCK > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded > > > ... > However, I'm at a loss as to how to determine which "process" > QUOTA is being exceeded here. In addition to the SHOW PROCESS/QUOTA you did, you can also get the quota and the "amount left" using the F$GETJPI lexical. But there may be an easier first step. Check the _OpenVMS System Manager's Manual: Essentials_ chapter (10?) on using backup, in particular the section (10).7 Setting process Quotas for Efficient Backups. According to the DCL dictionary, the quotas shown by SHOW PROCESS /QUOTA are the initial quotas, not the amount used or left. Some of the values in the example you gave look quite low compared to the advice in section 10.7. hth - Rob -- Rob Brown b r o w n a t g m c l d o t c o m G. Michaels Consulting Ltd. (780)438-9343 (voice) Edmonton (780)437-3367 (FAX) http://gmcl.com/ ------------------------------ Date: Wed, 03 Oct 2007 13:55:18 -0700 From: IanMiller Subject: Re: backup problem Message-ID: <1191444918.425416.158250@50g2000hsm.googlegroups.com> The SHOW PROCESS/Q command showed you the quotas remaining of that process at the time you did the command. I expect the error stopped the backup as the error had fatal severity (-F-) %BACKUP-F-WRITEERR, There is a relationship between the various quotas that should be maintained for the backup to work. Check the docs. ------------------------------ Date: Wed, 03 Oct 2007 14:42:40 -0700 From: deano Subject: Re: backup problem Message-ID: <1191447760.403828.94710@w3g2000hsg.googlegroups.com> Heres a pointer to the doc. I usually do the penny in the fusebox approach and up quotas. Dean http://h71000.www7.hp.com/doc/82FINAL/aa-pv5mj-tk/aa-pv5mj-tk.HTMl ------------------------------ Date: Wed, 03 Oct 2007 15:57:45 -0700 From: AEF Subject: Re: backup problem Message-ID: <1191452265.360287.323780@o80g2000hse.googlegroups.com> On Oct 3, 4:14 pm, jhjr4381 wrote: > I'm still getting an error while running BACKUP, as follows(ALPHA VMS > 6.2-1h3 ): > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > /NOALIAS - > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > RDAXP$MKB500:DUA510.BCK > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded > > I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a > listing of files stored in the save-set. > However, I'm at a loss as to how to determine which "process" QUOTA is > being exceeded here. It looks like backup was trying to make the file > larger in order to write to it, but failed to do so (maybe). > > Is this an error because I need x blocks of contiguous space (to > extend the file)? I doubt it. You could run ANAL/DISK and see if there are any filesystem errors, but I don't think that's your problem. Hmmm, could it be disk quota?! Run the SHOW QUOTA command to see. I think that would do it! > > Did this error stop the backup, or just the writing to this .log file? Being a Fatal error, it probably stopped the backup, as another poster pointed out. > > I looked at the process as it was executing (batch job) and got the > following: > > RDAXP::SYSTEM> show process /quotas/identification=20400230 > > 3-OCT-2007 09:30:22.28 User: SYSTEM Process ID: > 20400230 > Node: RDAXP Process name: > "BATCH_5" > > Process Quotas: > Account name: SYSTEM > CPU limit: Infinite Direct I/O limit: 4096 > Buffered I/O byte count quota: 148976 Buffered I/O limit: 400 > Timer queue entry quota: 39 Open file quota: 995 > Paging file quota: 292800 Subprocess quota: 10 > Default page fault cluster: 64 AST quota: 99 > Enqueue quota: 2000 Shared file limit: 0 > Max detached processes: 0 Max active jobs: 0 > > Does this show me what "system's" quotas are for this process, or is > this the process' > quotas (or are they one and the same)? A quick test shows that it is what's left. > Is possibly a system quota that the process itself is exceeding? I don't know what this means. > > I've been searching through the FAQ's, HP's site, etc and haven't > found a thing. Any words of wisdom out there? Thanks! As I said last time, try increasing FILLM and/or DIOLM. Also, as another poster suggested, look up the process quotas recommended for efficient backups. (If it's not the disk quota causing the problem, that is.) AEF ------------------------------ Date: Wed, 03 Oct 2007 20:16:52 -0500 From: David J Dachtera Subject: Re: backup problem Message-ID: <47043F04.24CFFDDA@spam.comcast.net> jhjr4381 wrote: > > I'm still getting an error while running BACKUP, as follows(ALPHA VMS > 6.2-1h3 ): > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > /NOALIAS - > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > RDAXP$MKB500:DUA510.BCK > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded > > I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a > listing of files stored in the save-set. > However, I'm at a loss as to how to determine which "process" QUOTA is > being exceeded here. It looks like backup was trying to make the file > larger in order to write to it, but failed to do so (maybe). > > Is this an error because I need x blocks of contiguous space (to > extend the file)? > > Did this error stop the backup, or just the writing to this .log file? > > I looked at the process as it was executing (batch job) and got the > following: > > RDAXP::SYSTEM> show process /quotas/identification=20400230 > > 3-OCT-2007 09:30:22.28 User: SYSTEM Process ID: > 20400230 > Node: RDAXP Process name: > "BATCH_5" > > Process Quotas: > Account name: SYSTEM > CPU limit: Infinite Direct I/O limit: 4096 > Buffered I/O byte count quota: 148976 Buffered I/O limit: 400 > Timer queue entry quota: 39 Open file quota: 995 > Paging file quota: 292800 Subprocess quota: 10 > Default page fault cluster: 64 AST quota: 99 > Enqueue quota: 2000 Shared file limit: 0 > Max detached processes: 0 Max active jobs: 0 > > Does this show me what "system's" quotas are for this process, or is > this the process' > quotas (or are they one and the same)? > Is possibly a system quota that the process itself is exceeding? > > I've been searching through the FAQ's, HP's site, etc and haven't > found a thing. Any words of wisdom out there? Thanks! Well SHOW PROC/QUOTA will show the quotas, but not the "counts" (how much used or how much remaining, depending on the quota being considered). AMDS or a suitably crafted DCL proc. would be useful for that. Since the message states: -RMS-E-EXT, ACP file extend failed ..., first guess time would be something related to either memory (BYTLM exceeded) or I/O (DIOLM or BIOLM being exceeded), possibly even ENQLM being exceeded since BACKUP is so aggressive about resource usage. Without knowing how close to the max.'es the process had been running, I don't know if there's a way to tell without a process dump. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Wed, 03 Oct 2007 21:50:31 -0400 From: "Richard B. Gilbert" Subject: Re: backup problem Message-ID: <470446E7.2010701@comcast.net> jhjr4381 wrote: > I'm still getting an error while running BACKUP, as follows(ALPHA VMS > 6.2-1h3 ): > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > /NOALIAS - > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > RDAXP$MKB500:DUA510.BCK > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded > > > I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a > listing of files stored in the save-set. > However, I'm at a loss as to how to determine which "process" QUOTA is > being exceeded here. It looks like backup was trying to make the file > larger in order to write to it, but failed to do so (maybe). > > Is this an error because I need x blocks of contiguous space (to > extend the file)? > No! > Did this error stop the backup, Yes! Try this command procedure. I wrote it many years ago but I think it should still be usable. It creates a BACKUP account for you with quotas and limits appropriate for your system. NOTE THAT MY MAILER HAS WRAPPED SOME LINES! YOU WILL HAVE TO UNWRAP THEM BEFORE USING. $! MAKE_BACKUP_ACCT.COM $! $! Copyright 1995, 1996, 1997, 1999, 2001 $! Richard B. Gilbert $! All rights reserved. $! $! $! A license to use and distribute this software is granted under the $! following terms and conditions: $! 1. You may not sell it. This is not intended to preclude a reasonable $! charge for: media, copying, shipping, and handling. $! 2. You may not claim you wrote it. $! 3. You may not remove or alter my Copyright notice or these terms and $! conditions. $! 4. You agree that this software is offered without warranty of any $! kind whatever. If it's broken, you fix it. $! 5. You may not modify this software unless: $! a. You make your modifications in such a way that it is clear $! which parts of the modified software are your changes and $! which are part of the original. $! b. You make your changes freely available to all other $! licensees. $! Posting your modifications to the Info-VAX mailing list, the $! comp.os.vms newsgroup, or communicating them to me shall be $! deemed sufficient to meet this requirement. $! $! Writes a DCL command procedure to create an account for use with the $! BACKUP utility. It calculates appropriate values of the UAF Quotas $! and Limits using rules of thumb from the article "Setting Up Params $! For OpenVMS BACKUP Operation (V5.2 or above" which is available from $! DSNlink. Some of these values may have to be hand tuned for optimum $! results. $! $! 02-DEC-1996 - Added prompt for the number of backup jobs. $! 11-MAR-1997 - Fixed to calculate WSQUOTA <= $! - MIN ((VIRTUALPAGECNT, PAGEFILE_FREE) - 3000). $! - Translated Alpha PAGEFILE_FREE to pagelets. $! 17-JUN-1999 - Create logicals in executive mode. $! 14-SEP-1999 - Apply Andy Goldstein's recommendations for $! calculating DIOLM. $! 16-MAR-2001 - Added terms and conditions, corrected the $! maximum for DIOLM and corrected the $! calculation for ASTLM. Removed references to $! a missing label. Modified password generation $! to ensure that original privileges were $! restored. Used F$FAO to fancy up the $! output a little. $! $! $ READ /PROMPT="Maximum Simultaneous BACKUP Jobs? " - SYS$COMMAND MAX_BACKUP_JOBS $ MAX_BACKUP_JOBS = F$INTEGER(MAX_BACKUP_JOBS) $ WSMAX = F$GETSYI("WSMAX") $! $! Note that VIRTUALPAGECNT is no longer a tunable SYSGEN parameter $! under OpenVMS/ALpha V7.0 and above. Its value is set to 2147483647 $! and ignored. This should have no effect on the following calculation $! unless your pagefile is really BIG. $! $ VIRTUALPAGECNT = F$GETSYI("VIRTUALPAGECNT") $ PAGEFILE_FREE = F$GETSYI("PAGEFILE_FREE") $ IF F$GETSYI("ARCH_TYPE") .EQ. 2 $ THEN $ PAGEFILE_FREE = PAGEFILE_FREE *F$GETSYI("PAGE_SIZE")/512 $ ENDIF $ PAGEFILE_FREE = PAGEFILE_FREE / MAX_BACKUP_JOBS $ IF VIRTUALPAGECNT .GT. PAGEFILE_FREE $ THEN $ PGFLQUO = PAGEFILE_FREE $ ELSE $ PGFLQUO = VIRTUALPAGECNT $ ENDIF $ WSDEF = 1024 $ WSQUOTA = WSMAX $ IF WSQUOTA .GT. (PGFLQUO - 3000) THEN WSQUOTA=PGFLQUO-3000 $ WSEXTENT = WSMAX $ FILLM = F$GETSYI("CHANNELCNT")-15 $! $! Calculate DIOLM. The maximum usable value of DIOLM is 32747 $! (required to allow ASTLM=32767). The minimum value of DIOLM is $! MAX(4096, FILLM*3). The rule that DIOLM = WSQUOTA / BLOCKSIZE is due $! to Andy Goldstein . The rule given in $! the System Manager's Manual (and other places) yields the MINIMUM $! rather than the recommended value and fails for very large values $! of WSQUOTA. $! $! Assume BLOCKSIZE (in blocks) =32256/512 or 63. The minimum is 4 and $! the maximum is 127. A blocksize of 32256 is the recommended value. $! N.B. The setting of a suitable blocksize is one of the the MOST $! important tuning steps. $! $ BLOCKSIZE = 32256/512 ! In blocks $ MAX_DIOLM = 32747 $ DIOLM = WSQUOTA / BLOCKSIZE $ IF DIOLM .GT. MAX_DIOLM THEN DIOLM=MAX_DIOLM $! This calculates a minimum value for DIOLM. $ IF 3*FILLM .LE. 4096 $ THEN $ MIN_DIOLM = 4096 $ ELSE $ MIN_DIOLM = 3*FILLM $ ENDIF $ IF DIOLM .LT. MIN_DIOLM THEN DIOLM=MIN_DIOLM $ ASTLM = DIOLM + 20 ! Minimum value. $! $! BYTLM may need to be hand tuned. It should never be less than the $! minimum but may need to be increased. If BYTLM cannot be increased $! then FILLM must be decreased. $! $ BYTLM = (256*FILLM) + (6*DIOLM) ! Minimum value. $ BIOLM = FILLM $ ENQLM = FILLM + 2 ! ENQLM > FILLM $! $! Now write the ADD command for authorize. $! $! Note: we assume that the file being written will be executed from an $! account with adequate privileges to define executive mode logical $! names and to modify SYSUAF.DAT. We also assume that you will check $! the generated file and make any necessary changes before executing $! it! $! $ OPEN /WRITE FILE BACKUP_ACCT.COM $ WRITE FILE "$! Suggested UAF Parameters for the account running BACKUP on ''F$ GETSYI("NODENAME")'." $ WRITE FILE F$FAO("$!! We assume !UB simultaneous backup job!%S and the SYSGEN parameters in !/$!! effect at !%D", MAX_BACKUP_JOBS, 0) $! $! Make the User Authorization File (UAF) accessable. We use F$PARSE $! to allow for a logical name pointing to the file. $! $ WRITE FILE "$ ASSIGN /EXECUTIVE_MODE ''F$PARSE("SYSUAF", "SYS$SYSTEM", ".DAT") ' SYSUAF" $ WRITE FILE "$ ASSIGN /EXECUTIVE_MODE ''F$PARSE("RIGHTSLIST", "SYS$SYSTEM", ".D AT")' RIGHTSLIST" $ WRITE FILE "$ ASSIGN /EXECUTIVE_MODE ''F$PARSE("NETPROXY", "SYS$SYSTEM", ".DAT ")' NETPROXY" $ WRITE FILE "$ ASSIGN /EXECUTIVE_MODE ''F$PARSE("NET$PROXY", "SYS$SYSTEM", ".DA T")' NET$PROXY" $ WRITE FILE "$ RUN SYS$SYSTEM:AUTHORIZE" $ GOSUB GENERATE_PASSWORDS $ WRITE FILE "ADD BACKUP /UIC=[16,1] /PASSWORD=''PWD3' -" $ WRITE FILE "/DEVICE=SYS$SYSROOT: /DIRECTORY=[SYSMGR] -" $ WRITE FILE "/FILLM=''FILLM' /BIOLM=''BIOLM' /DIOLM=''DIOLM' -" $ WRITE FILE "/ASTLM=''ASTLM' /ENQLM=''ENQLM' /BYTLM=''BYTLM' -" $ WRITE FILE "/WSDEF=''WSDEF' /WSQUOTA=''WSQUOTA' -" $ WRITE FILE "/WSEXTENT=''WSEXTENT' /PGFLQUO=''PGFLQUO'" $ WRITE FILE "MODIFY BACKUP /PRIVILEGE=(OPER,READALL,SYSPRV,VOLPRO) -" $ WRITE FILE "/DEFPRIV=SYSPRV /FLAGS=NODISUSER /NOINTERACTIVE -" $ WRITE FILE "/NOREMOTE /NONETWORK" $ WRITE FILE "$ DEASSIGN /EXECUTIVE_MODE SYSUAF" $ WRITE FILE "$ DEASSIGN /EXECUTIVE_MODE RIGHTSLIST" $ WRITE FILE "$ DEASSIGN /EXECUTIVE_MODE NETPROXY" $ WRITE FILE "$ DEASSIGN /EXECUTIVE_MODE NET$PROXY" $ CLOSE FILE $ EXIT $! $! The method for generating passwords in a command procedure was taken $! from NETCONFIG.COM. $! $GENERATE_PASSWORDS: $ MYPID = F$GETJPI("", "PID") $ ON ERROR THEN GOTO GEN_PWD_DONE $ DEFINE /USER_MODE SYS$OUTPUT SYS$SCRATCH:A'MYPID'.TMP $ ON ERROR THEN GOTO ERROR_EXIT $ ON CONTROL_Y THEN GOTO ERROR_EXIT $! Privs are a temporary workaround for a bug in OpenVMS/Alpha $! V6.2 -- V7.0. $ PREV_PRIVS = F$SETPRV("LOG_IO,PHY_IO") $ SET PASSWORD /GENERATE=8 INVALID $GEN_PWD_DONE: $ IF F$SEARCH("SYS$SCRATCH:A''MYPID'.TMP") .EQS. "" THEN - GOTO ERROR_EXIT $ SET PROTECTION=(G,W) SYS$SCRATCH:A'MYPID'.TMP $ PWD1 = "" $ PWD2 = "" $ PWD3 = "" $ PWD4 = "" $ PWD5 = "" $ PWDN = 1 $ OPEN/READ PWD_FILE SYS$SCRATCH:A'MYPID'.TMP $PWD_LOOP: $ READ /END=END_OF_FILE PWD_FILE RECORD $ PWD = F$ELEMENT(0," ",RECORD) $ IF PWD .EQS. "" THEN GOTO PWD_LOOP $ PWD'PWDN' = PWD $ SHOW SYMBOL PWD'PWDN' $ PWDN = PWDN + 1 $IF PWDN .LE. 5 THEN GOTO PWD_LOOP $END_OF_FILE: $ CLOSE PWD_FILE $ DELETE SYS$SCRATCH:A'MYPID'.TMP;* $ JUNK = F$SETPRV(PREV_PRIVS) $ RETURN $ERROR_EXIT: $ EXIT_STATUS = $STATUS $ JUNK = F$SETPRV(PREV_PRIVS) $ WRITE SYS$OUTPUT "Error generating paswords!" $ EXIT EXIT_STATUS ------------------------------ Date: Wed, 03 Oct 2007 18:52:54 -0700 From: AEF Subject: Re: backup problem Message-ID: <1191462774.523707.326480@y42g2000hsy.googlegroups.com> On Oct 3, 6:57 pm, AEF wrote: > On Oct 3, 4:14 pm, jhjr4381 wrote: > > > I'm still getting an error while running BACKUP, as follows(ALPHA VMS > > 6.2-1h3 ): > > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > > /NOALIAS - > > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > > RDAXP$MKB500:DUA510.BCK > > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > > -RMS-E-EXT, ACP file extend failed > > -SYSTEM-F-EXQUOTA, process quota exceeded > > > I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a > > listing of files stored in the save-set. Correct. > > However, I'm at a loss as to how to determine which "process" QUOTA is > > being exceeded here. It looks like backup was trying to make the file > > larger in order to write to it, but failed to do so (maybe). I think it may be a disk quota. Use the SHOW QUOTA/DISK=disk/USER=user command to see. Unfortunately you've run into one of VMS's few weaknesses. Which quota is exceeded? Check the disk quota and please report back. > > Is this an error because I need x blocks of contiguous space (to > > extend the file)? > > I doubt it. You could run ANAL/DISK and see if there are any > filesystem errors, but I don't think that's your problem. Hmmm, could > it be disk quota?! Run the SHOW QUOTA command to see. I think that > would do it! > > > > > Did this error stop the backup, or just the writing to this .log file? > > Being a Fatal error, it probably stopped the backup, as another poster > pointed out. You can check by mounting the tape (write-protected!) and running $ BACKUP/LIST and see if it abruptly ends on error or gives you the expected full listing. [...] AEF ------------------------------ Date: Wed, 03 Oct 2007 23:39:34 -0400 From: JF Mezei Subject: Re: backup problem Message-ID: jhjr4381 wrote: >> %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded From a programming point of view: would it be fair to state that BACKUP was doing a simple "write" (SYS$PUT or whatever) and it got a "RMS-E-EXT" status code and then called LIB$SIGNAL with it own error message (the WRITEERR) and the RMS-E-EXT status code it got from LIB$PUT ? I am curious as the mechanism that resulted in the underlying error code (EXQUOTA) to be included in the LIB$SIGNAL output ? Or does the fact that multiple error messages was issued mean that BACKUP called LIB$PUT with a error procedure and the later would have had access to an error vector that would have contained both the EXQUOTA and the RMS-E-EXT ? ------------------------------ Date: Thu, 04 Oct 2007 07:25:30 +0200 From: Jur van der Burg <"vdburg at hotmail dot com"> Subject: Re: backup problem Message-ID: <470479b4$0$227$e4fe514c@news.xs4all.nl> > CPU limit: Infinite Direct I/O limit: 4096 > Buffered I/O byte count quota: 148976 Buffered I/O limit: 400 > Default page fault cluster: 64 AST quota: 99 Well, a diolim of 4096 and astlim of 99 (probably 100) is asking for trouble. Lower diolim to about 100 and make sure that astlim is at least as big as the max of biolim and diolim. But reading the docs about backup quotas should be the best thing to do. Jur. jhjr4381 wrote, On 3-10-2007 22:14: > I'm still getting an error while running BACKUP, as follows(ALPHA VMS > 6.2-1h3 ): > > $ BACKUP/NOREWIND/LIST=DKA100:[BACKUP]DUA510FUL.LOG/IMAGE - > /NOALIAS - > /RECORD/IGNORE=(INTERLOCK,LABEL)/NOASSIST $1$DUA510: - > RDAXP$MKB500:DUA510.BCK > %BACKUP-F-WRITEERR, error writing DKA100:[BACKUP]DUA510FUL.LOG;202 > -RMS-E-EXT, ACP file extend failed > -SYSTEM-F-EXQUOTA, process quota exceeded > > > I understand that this /LIST file DKA100:[BACKUP]DUA510FUL.LOG is a > listing of files stored in the save-set. > However, I'm at a loss as to how to determine which "process" QUOTA is > being exceeded here. It looks like backup was trying to make the file > larger in order to write to it, but failed to do so (maybe). > > Is this an error because I need x blocks of contiguous space (to > extend the file)? > > Did this error stop the backup, or just the writing to this .log file? > > I looked at the process as it was executing (batch job) and got the > following: > > RDAXP::SYSTEM> show process /quotas/identification=20400230 > > 3-OCT-2007 09:30:22.28 User: SYSTEM Process ID: > 20400230 > Node: RDAXP Process name: > "BATCH_5" > > Process Quotas: > Account name: SYSTEM > CPU limit: Infinite Direct I/O limit: 4096 > Buffered I/O byte count quota: 148976 Buffered I/O limit: 400 > Timer queue entry quota: 39 Open file quota: 995 > Paging file quota: 292800 Subprocess quota: 10 > Default page fault cluster: 64 AST quota: 99 > Enqueue quota: 2000 Shared file limit: 0 > Max detached processes: 0 Max active jobs: 0 > > Does this show me what "system's" quotas are for this process, or is > this the process' > quotas (or are they one and the same)? > Is possibly a system quota that the process itself is exceeding? > > I've been searching through the FAQ's, HP's site, etc and haven't > found a thing. Any words of wisdom out there? Thanks! > ------------------------------ Date: Thu, 04 Oct 2007 07:27:35 +0200 From: Jur van der Burg <"vdburg at hotmail dot com"> Subject: Re: DFU on the freeware disks Message-ID: <47047a30$0$227$e4fe514c@news.xs4all.nl> > No OpenVMS VAX. Is it really that hard to write portable code? Not at all, it's called progress. Support for 64-bit things on Vax is tricky and not worth the effort, to name one thing. Jur. Brian Tillman wrote, On 2-10-2007 14:40: > "Jur van der Burg" <"lddriver at digiater dot nl"> wrote in message > news:46e548bc$0$232$e4fe514c@news.xs4all.nl... > >> It was not removed, but never put there. Don't count on the freeware >> to be maintained anymore. >> >> You can get it from my site: >> >> http://www.digiater.nl/dfu.html > > No OpenVMS VAX. Is it really that hard to write portable code? ------------------------------ Date: Wed, 03 Oct 2007 20:43:59 GMT From: "John Vottero" Subject: Re: DSPP PAK diffs I64/Alpha Message-ID: "Tom Linden" wrote in message news:op.tzmpr7wuhv4qyg@murphus.linden... > FYI, made a list paks in Alpha but not IA64 > http://www.kednos.com/missing.txt Your list seems to be off somewhat. It looks like maybe you have the list of differences followed by the full list? ------------------------------ Date: Wed, 03 Oct 2007 18:05:46 -0700 From: "Tom Linden" Subject: Re: DSPP PAK diffs I64/Alpha Message-ID: On Wed, 03 Oct 2007 13:43:59 -0700, John Vottero wrote: > "Tom Linden" wrote in message > news:op.tzmpr7wuhv4qyg@murphus.linden... >> FYI, made a list paks in Alpha but not IA64 >> http://www.kednos.com/missing.txt > > Your list seems to be off somewhat. It looks like maybe you have the > list of differences followed by the full list? > That could be, I just did a diff and then a little emacs, but may have got it wrong. It should contain only those in Alpha but not in IA64. I didn't list what ios common to both -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Wed, 03 Oct 2007 13:54:09 -0700 From: thierry.uso@wanadoo.fr Subject: Re: Guidance with OpenVMS IA64 8.3 with Java 1.5 Problems Message-ID: <1191444849.731117.111200@n39g2000hsh.googlegroups.com> On 3 oct, 15:22, "Tom Linden" wrote: > On Mon, 01 Oct 2007 22:00:46 -0700, wrote: > > Note that the architecture of Itanium and Java are big memory > > consumers. 256 Mb ram on Intel x86 32bits is roughly equivalent to 4 > > Gb ram on Itanium. > > How would you compare it Alpha? > > -- > PL/I for OpenVMSwww.kednos.com ~ 1 Gb ram on Alpha ------------------------------ Date: Wed, 3 Oct 2007 16:43:53 -0400 From: "David Turner, Island Computers" Subject: Re: SmartArray errors on system boot Message-ID: <13g7vnq128qbg8d@news.supernews.com> Are your disks HP/Compaq firmwared disks? Are they all identical sizes? wrote in message news:1191425849.767217.202610@57g2000hsv.googlegroups.com... > Hiya, > > I've got about five logs open with support at the moment, but I'm > wondering if anyone else has experienced similar... > > I've got a mix of 5302A and 6402A SmartArray cards in Alphas and, more > recently, a couple of A9890A 6402 SmartArrays in a couple of Integrity > servers (one per server). > > When the system boots, the system logs errors against the PK port > device (SCSI bus resets maybe?) then, usually, calms down and behaves > itself. The disks never log any errors. > > At first we thought maybe a bad card. When a second one did it we > thought maybe the MSA30 shelves that they are connected to were the > culprits, maybe having their interface cards damaged in transit. This > weekend, we had problems on an Integrity which has caused the other > SmartArray and MSA30 installations that were planned to be pulled, > pending a response from HP. > > Anybody seen this kind of behaviour before? > > Full config is VMS (either Alpha or I64), 8.2 on Alpha and 8.3 on > I64. RAID software v3.0 on both platforms to enable us to partition > the disk that the SmartArray is configured with so that we can size > the DPA devices that VMS sees appropriately. > > Thanks in advance > > Steve > ------------------------------ Date: Wed, 03 Oct 2007 19:58:11 -0000 From: Rich Jordan Subject: Re: VMS Apache on older versions? Message-ID: <1191441491.552930.15130@g4g2000hsf.googlegroups.com> On Oct 3, 10:39 am, mor...@osmium.mv.net (Skipper W. Morris) wrote: > I have an Alpha Multia I've just installed VMS 7.2 on (AFAIK the latest > version of VMS it will run). > > I'd like to install Apache on it (aka HP Secure Web Server for OpenVMS), > but the HP web page claims it runs on VMS 7.3-1 and higher. > > Anyone know if it will run on an earlier version? Or what I have to do > to make it run on 7.2? > > I guess the only other option is to install WASD instead. > > thanks > /Skip You almost certainly need to get an older version of CSWS (Compaq SWS). CSWS V1.2 required OpenVMS V7.2-1 Alpha (and TCPIP Services V5.0A or newer) at minimum, while CSWS V1.1 and updates would run on (at least) V7.1-2 so would _probably_ support V7.2, though i don't see any mention of it (I don't have V1.1 release notes; the V1.2 release notes simply mention dropping support for V7.1-2). Honestly you would probably be better off running one of the other webservers (OSU or WASD) unless your purpose is to learn to use and configure CSWS. Rich ------------------------------ Date: Thu, 4 Oct 2007 05:53:04 +0800 From: "Richard Maher" Subject: We're VMS - Please do not discuss the outside world Message-ID: Hi Robert, > $ SHOW QUEUE works pretty well for me. I guess I'll leave the > Mickey-Clicky stuff to the kids. Oh yes, that point and click GUI stuff is just for kids and will never last. FIMS is coming back, you mark my words! We're from Digital and we still believe that the industry is waiting for our next prognostication about what the future holds. "DECforms and VT dumb terminals" forever! Regards Richard Maher PS. Why is VMS's customer base disappearing again? "Robert Deininger" wrote in message news:rdeininger-04F7FD.07550626092007@032-478-847.area7.spcsdns.net... > In article <7arKi.10$fT2.4@newsfe12.lga>, VAXman- @SendSpamHere.ORG > wrote: > > > In article , "Richard Maher" > > writes: > > > > > > > > >Hi, > > > > > >If any of you Inmates out there would like to monitor the state of your > > >batch jobs on the Deathrow cluster via a Web Browser interface, then please > > >click on the following web page: - > > > > > >http://manson.vistech.net/t3$examples/demo_client_web.html > > > > > >In order to be able to use this Queue Manager example successfully you will > > >need:- > > > > > >1) Javascript enabled > > >2) Java Applets enabled > > >3) Can't be behind a Firewall that forbids all unknown outgoing connections > > >(otherwise open-up 5255) > > >4) Must be running SUN's JRE 1.4.2_13 or later (1.6 is advised) > > >http://java.sun.com/javase/downloads/index.jsp > > >5) Must be running Internet Explorer (6 or later) or Firefox (Haven't tested > > >other browsers) > > > > > >If need be, turn on the Java Console (Settings/Controle pannel/Java > > >Plug-in/Console on) and, if using Firefox, the Error Console. (You just > > >gotta love FireBug!!!) > > > > > >All things being equal you should then be prompted (via Java modal dialogue > > >box) for:- > > > > > >Username: TIER3_DEMO > > >Password: QUEUE > > > > > >If things are still going well, enter an asterix "*" for the Queue Name and > > >you will see the List-of-values appear with all the available queues on the > > >Deathrow cluster. Now click on the "Get Job Info" button and we're away! > > > > FYI RIchard, > > > > Safari aborts after entering the Username and Password. > > Congrats! What incantation did you use? I couldn't even get the login > box using Safari. > > Am I brave enough to download yet another browser (Firefox) and try this > again? > > $ SHOW QUEUE works pretty well for me. I guess I'll leave the > Mickey-Clicky stuff to the kids. ------------------------------ Date: Wed, 03 Oct 2007 20:08:06 -0500 From: David J Dachtera Subject: Re: We're VMS - Please do not discuss the outside world Message-ID: <47043CF6.1FE502E9@spam.comcast.net> Richard Maher wrote: > [snip] > PS. Why is VMS's customer base disappearing again? Google this group for "stealth marketing" and "ISVs I64 UX" -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Wed, 03 Oct 2007 23:03:13 -0400 From: Robert Deininger Subject: Re: We're VMS - Please do not discuss the outside world Message-ID: In article , "Richard Maher" wrote: > Hi Robert, > > > $ SHOW QUEUE works pretty well for me. I guess I'll leave the > > Mickey-Clicky stuff to the kids. > > Oh yes, that point and click GUI stuff is just for kids and will never last. > FIMS is coming back, you mark my words! The distinguishing feature of SHOW QUEUE is that it works, and the interface you are demonstrating DOES NOT WORK out of the box. (And the demo GUI kind of stinks, but I understand you're showing off the underlying stuff, and the GUI isn't supposed to be polished.) I'm following the advice from various posters about fiddling with Java on my client side. Dunno when I'll find time to do that, but maybe someone will offer an easy explanation. I find it odd that one of the suggestions offered to make this "portable" Java run on a Macintosh was to hunt for a software DEVELOPERS kit. This is not a way to make a client-server application popular with the masses. Java is wonderful (so I read), but different versions seem to have compatibility problems. If I install the Java that makes your stuff happy, will it break something else? Could your Java application query the environment for version information, and give us a message if we don't have the required bits? That seems marginally more helpful than CRASHING the browser. -- Robert ------------------------------ Date: Thu, 04 Oct 2007 07:34:45 +0200 From: "P. Sture" Subject: Re: We're VMS - Please do not discuss the outside world Message-ID: In article <47043CF6.1FE502E9@spam.comcast.net>, David J Dachtera wrote: > Google this group for "stealth marketing" and "ISVs I64 UX" Google seems to have messed up big time. I'm getting such pathetic results from Group searches, it's hardly worth the effort. example 1: comp.os.vms ISVs I64 UX 6 results example 2: comp.os.vms "stealth marketing" 63 results -- Paul Sture Sue's OpenVMS bookmarks: http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: Thu, 4 Oct 2007 05:15:54 +0800 From: "Richard Maher" Subject: Re: Web-based VMS Que Monitor for Deathrow Message-ID: Hi Bob, As Graham has already pointed out, the protection on the queue(s) is such that you do in fact have privilege to see the jobs. Try deleting one of the jobs (via the "X" button after you've drilled-down into the