Client connection severed
I keep getting the "session is lost" errors with a certain box. I already changed the maxsessions, maxschedule, idletimeout, but no changes. The errors keep coming. Only specials about this box is the fact that it has large files (oracle database) and that it is behind a firewall. I assume the big files (2 and 4 gb) are making the problem here. Any hints where I can look next? Thanks, Geert Server side: 10/26/04 03:15:01 ANR0480W Session 4563 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:26:22 ANR0480W Session 4566 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:45:54 ANR0480W Session 4567 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 04:15:49 ANR0480W Session 4568 for node AILDB (WinNT) terminated - connection with client severed. Client side dsmerror.log 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ANS1810E TSM session has been reestablished. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ANS1810E TSM session has been reestablished. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ANS1810E TSM session has been reestablished. 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. Client side dsmsched.log 26-10-2004 03:07:58 --- SCHEDULEREC OBJECT BEGIN BACKUP_INC_AIL 26-10-2004 03:00:00 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ... successful 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ... successful 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ... successful 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:44:13 ... successful 26-10-2004 05:11:36 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:11:51 ... successful 26-10-2004 05:17:11 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:17:26 ... successful 26-10-2004 05:32:41 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:32:56 ... successful 26-10-2004 05:35:47 --- SCHEDULEREC STATUS BEGIN 26-10-2004 05:35:47 Total number of objects inspected: 78,805
Re: Client connection severed
Geert, Sounds like you might want to get your network guys to double check the network config settings between the client and the server - are your local machine and switch/router ports all hard coded to the same media settings (e.g. 100Mb/Full Duplex) and not reliant upon auto detect? Depending upon platform, can you check with netmon or a `netstat -rn` or similar for Unix. >From my experiences (and I'm not a network guy in any way, shape or form), when backing up large files using the BA client, I'd expect to see many more of these TCP drop errors than when backing up many thousands of smaller files. Also, I've found the TSM API client to be much less tolerant of packet drops too (perhaps because we're normally using it to back up large files again via a TDP). Hope that gives you somewhere to start looking. Rgds, David McClelland Reuters Ltd., London -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Geert De Pecker Sent: 26 October 2004 08:30 To: [EMAIL PROTECTED] Subject: Client connection severed I keep getting the "session is lost" errors with a certain box. I already changed the maxsessions, maxschedule, idletimeout, but no changes. The errors keep coming. Only specials about this box is the fact that it has large files (oracle database) and that it is behind a firewall. I assume the big files (2 and 4 gb) are making the problem here. Any hints where I can look next? Thanks, Geert Server side: 10/26/04 03:15:01 ANR0480W Session 4563 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:26:22 ANR0480W Session 4566 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:45:54 ANR0480W Session 4567 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 04:15:49 ANR0480W Session 4568 for node AILDB (WinNT) terminated - connection with client severed. Client side dsmerror.log 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ANS1810E TSM session has been reestablished. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ANS1810E TSM session has been reestablished. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ANS1810E TSM session has been reestablished. 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. Client side dsmsched.log 26-10-2004 03:07:58 --- SCHEDULEREC OBJECT BEGIN BACKUP_INC_AIL 26-10-2004 03:00:00 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ... successful 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ... successful 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ... successful 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:44:13 ... successful 26-10-2004 05:11:36 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:11:51 ... successful 26-10-2004 05:17:11 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:17:26 ... successful 26-10-2004 05:32:41 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:32:56 ... successful 26-10-2004 05:35:47 --- SCHEDULEREC STATUS BEGIN 26-10-2004 05:35:47 Total number of objects inspected: 78,805 --- - Visit our Internet site at http://www.reuters.com Get closer to the financial markets with Reuters Messaging - for more information and to register, visit http://www.reuters.com/messaging Any views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of Reuters Ltd.
dir include ?
hi all, how to backup directories with default management class ? (or should i say with mc into which are all the files below that dirs are going) it seems that default is that dirs are backuped with mc which has the longest retain only version .. am i right ? thanks in advance
extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked]
Hi, I want to extend my primary diskstorage pool on SSA-disks. Environment: AIX 5.2 TSM 5.1.8 primary storage on SSA-disks secondary storage on tape (3494 library) copy storage pool on tape (3583 library) Now I must extend the primary disk storage pool. In older TSM versions (TSM Version 4) I start the following procedure to this - create log. volumes (AIX-LVM) - create jfs (AIX-LVM) - dsmfmt -data (AIX-Command-Line) - then in TSM: DEFINE VOLUME Is this in TSM-Version 5.1.8 the same procedure and the same syntax? Best regards, Frank
ANR9999D
Hello All! I am getting a very bizarre message and I was just wondering if anyone has ever seen this before? The TSM server is on AIX 5.2 and is at level 5.2.2.5. We have ACSLS 7.1 and EDT 6.4.6. Just yesterday I had been able to mount tapes, but this morning I tried to do a DB backup and it failed with the following message. I do have temp license keys and I was wondering if this could be an expiration of the license or if this is a TSM issue or a server issue? Thanks in advance! 10/26/04 07:05:56 ANRD mmsext.c(2195): ThreadId<17> Invalid response fr om exit (DEMO_EXPIRED). Callchain of previous message: 0 x000100017a04 outDiagf <- 0x0001005602c8 ExtHand leTextResults <- 0x00010055f5e4 ExtSendMount <- 0x00 010055ffb4 ExtMount <- 0x000100560eb8 ExtMountVo lume <- 0x00010023a76c MmsMountVolume <- 0x00010 0667804 LtoOpen <- 0x0001000e750c AgentThread <- 0x0 0017fc8 StartThread <- 0x092f9244 _pthre ad_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANR8769E External media management function MOUNT returned result=**Undefined**. (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANRD icstream.c(1654): ThreadId<28> Error 87 opening output stream. Callchain of previous message: 0x0001 00017a04 outDiagf <- 0x000100674290 OpenOutputStream <- 0x000100675500 IcOpenOutputStream <- 0x00010 06729f8 IcStreamThread <- 0x00017fc8 StartThread <- 0x092f9244 _pthread_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:57 ANRD icback.c(386): ThreadId<27> Backup rc=6. Callcha in of previous message: 0x000100017a04 outDiagf <- 0 x000100678620 IcBackupThread <- 0x00017fc8 S tartThread <- 0x092f9244 _pthread_body <- (SESS ION: 721, PROCESS: 46) 10/26/04 07:05:57 ANR4581W Database backup/restore terminated - internal se rver error detected. (SESSION: 721, PROCESS: 46) Joni Moyer Highmark Storage Systems Work:(717)302-6603 Fax:(717)302-5974 [EMAIL PROTECTED]
Re: ANR9999D
Joni, I think DEMO_EXPIRED is prett Joni Moyer <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 26/10/2004 12:36 Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject ANRD Hello All! I am getting a very bizarre message and I was just wondering if anyone has ever seen this before? The TSM server is on AIX 5.2 and is at level 5.2.2.5. We have ACSLS 7.1 and EDT 6.4.6. Just yesterday I had been able to mount tapes, but this morning I tried to do a DB backup and it failed with the following message. I do have temp license keys and I was wondering if this could be an expiration of the license or if this is a TSM issue or a server issue? Thanks in advance! 10/26/04 07:05:56 ANRD mmsext.c(2195): ThreadId<17> Invalid response fr om exit (DEMO_EXPIRED). Callchain of previous message: 0 x000100017a04 outDiagf <- 0x0001005602c8 ExtHand leTextResults <- 0x00010055f5e4 ExtSendMount <- 0x00 010055ffb4 ExtMount <- 0x000100560eb8 ExtMountVo lume <- 0x00010023a76c MmsMountVolume <- 0x00010 0667804 LtoOpen <- 0x0001000e750c AgentThread <- 0x0 0017fc8 StartThread <- 0x092f9244 _pthre ad_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANR8769E External media management function MOUNT returned result=**Undefined**. (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANRD icstream.c(1654): ThreadId<28> Error 87 opening output stream. Callchain of previous message: 0x0001 00017a04 outDiagf <- 0x000100674290 OpenOutputStream <- 0x000100675500 IcOpenOutputStream <- 0x00010 06729f8 IcStreamThread <- 0x00017fc8 StartThread <- 0x092f9244 _pthread_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:57 ANRD icback.c(386): ThreadId<27> Backup rc=6. Callcha in of previous message: 0x000100017a04 outDiagf <- 0 x000100678620 IcBackupThread <- 0x00017fc8 S tartThread <- 0x092f9244 _pthread_body <- (SESS ION: 721, PROCESS: 46) 10/26/04 07:05:57 ANR4581W Database backup/restore terminated - internal se rver error detected. (SESSION: 721, PROCESS: 46) Joni Moyer Highmark Storage Systems Work:(717)302-6603 Fax:(717)302-5974 [EMAIL PROTECTED] ** The information in this E-Mail is confidential and may be legally privileged. It may not represent the views of Scottish and Southern Energy Group. It is intended solely for the addressees. Access to this E-Mail by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited and may be unlawful. Any unauthorised recipient should advise the sender immediately of the error in transmission. Scottish Hydro-Electric, Southern Electric, SWALEC and S+S are trading names of the Scottish and Southern Energy Group. **
Re: ANR9999D
Joni, I think "DEMO_EXPIRED" is pretty conclusive Joni Moyer <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 26/10/2004 12:36 Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject ANRD Hello All! I am getting a very bizarre message and I was just wondering if anyone has ever seen this before? The TSM server is on AIX 5.2 and is at level 5.2.2.5. We have ACSLS 7.1 and EDT 6.4.6. Just yesterday I had been able to mount tapes, but this morning I tried to do a DB backup and it failed with the following message. I do have temp license keys and I was wondering if this could be an expiration of the license or if this is a TSM issue or a server issue? Thanks in advance! 10/26/04 07:05:56 ANRD mmsext.c(2195): ThreadId<17> Invalid response fr om exit (DEMO_EXPIRED). Callchain of previous message: 0 x000100017a04 outDiagf <- 0x0001005602c8 ExtHand leTextResults <- 0x00010055f5e4 ExtSendMount <- 0x00 010055ffb4 ExtMount <- 0x000100560eb8 ExtMountVo lume <- 0x00010023a76c MmsMountVolume <- 0x00010 0667804 LtoOpen <- 0x0001000e750c AgentThread <- 0x0 0017fc8 StartThread <- 0x092f9244 _pthre ad_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANR8769E External media management function MOUNT returned result=**Undefined**. (SESSION: 721, PROCESS: 46) 10/26/04 07:05:56 ANRD icstream.c(1654): ThreadId<28> Error 87 opening output stream. Callchain of previous message: 0x0001 00017a04 outDiagf <- 0x000100674290 OpenOutputStream <- 0x000100675500 IcOpenOutputStream <- 0x00010 06729f8 IcStreamThread <- 0x00017fc8 StartThread <- 0x092f9244 _pthread_body <- (SESSION: 721, PROCESS: 46) 10/26/04 07:05:57 ANRD icback.c(386): ThreadId<27> Backup rc=6. Callcha in of previous message: 0x000100017a04 outDiagf <- 0 x000100678620 IcBackupThread <- 0x00017fc8 S tartThread <- 0x092f9244 _pthread_body <- (SESS ION: 721, PROCESS: 46) 10/26/04 07:05:57 ANR4581W Database backup/restore terminated - internal se rver error detected. (SESSION: 721, PROCESS: 46) Joni Moyer Highmark Storage Systems Work:(717)302-6603 Fax:(717)302-5974 [EMAIL PROTECTED] ** The information in this E-Mail is confidential and may be legally privileged. It may not represent the views of Scottish and Southern Energy Group. It is intended solely for the addressees. Access to this E-Mail by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited and may be unlawful. Any unauthorised recipient should advise the sender immediately of the error in transmission. Scottish Hydro-Electric, Southern Electric, SWALEC and S+S are trading names of the Scottish and Southern Energy Group. **
Re: ANR9999D
Joni Definitely due to your temporary EDT license expiring. Check the key in ELM.CNF and compare with details on http://license.greshamstorage.com/license/ to confirm. Regards Neil Take a tour of a virtual waste water treatment works at http://www.wastewatertour.com The information in this e-mail is confidential and may also be legally privileged. The contents are intended for recipient only and are subject to the legal notice available at http://www.keldagroup.com/email.htm Yorkshire Water Services Limited Registered Office Western House Halifax Road Bradford BD6 2SZ Registered in England and Wales No 2366682
dir include ?
goc wrote: > it seems that default is that dirs are backuped with mc which has the > longest retain only version .. Absolutely right. Use the 'dirmc' option to override the default behaviour. There's info on it in the docs. Regards Neil Schofield Yorkshire Water Services Ltd Take a tour of a virtual waste water treatment works at http://www.wastewatertour.com The information in this e-mail is confidential and may also be legally privileged. The contents are intended for recipient only and are subject to the legal notice available at http://www.keldagroup.com/email.htm Yorkshire Water Services Limited Registered Office Western House Halifax Road Bradford BD6 2SZ Registered in England and Wales No 2366682
Re: TDP for Notes Restore Problem
Gerald, When you say "Full".. do you mean backing up all databases using "DOMDSMC SELECTIVE"? And when you say "INCREMENTAL", do you mean using the "DOMDSMC INCREMENTAL" command? ... or when you say incremental are you referring to running DOMDSMC ARCHIVELOG to get the logs backed up? Remember... "DOMDSMC INCREMENTAL" does "complete database" backups, i.e. full .NSF file backups. It is similar to the TSM BA client concept of INCREMENTAL.. which means back up the entire file if something has changed. If TEST was a "new" TSM node starting on March 8, the very first DOMDSMC INCREMENTAL that you ran will back up everything... ... since it is comparing the files to an empty filespace on the TSM Server. It is the equivalent of running a SELECTIVE. All of these backups should be restorable as long as your management classes are set to keep multiple versions. Thanks, Del "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 10/25/2004 10:51:31 PM: > We have Lotus Notes backups done with the TDP 5.1.5 (Notes v5.x). Back > in January the server name was called "TEST". In March, the server was > renamed to "TEST-OLD" and we continued to do backups by creating a new > node "TEST". So now we have data from Jan through March under > "TEST-OLD" and from March on called "TEST"... > > We have not expired any data since since January.. > > Our normal rotation is to do a weekly full and daily incrementals. > There is a problem though that i'm finding. The first full backup for > TEST was done March 15th. The server rename was done on March 7th. Now > when I look back i see: > > March 1st - Full taken now under TEST-OLD > March 2-7 - Incrementals now under TEST-OLD > March 8-14 - Incrementals now under TEST > March 15 - Full taken now under TEST > > So the question is whether or not March 8-14 can be restored? It seems > not unless someone has any ideas. TSM thinks there is no full for TEST > prior to the 15th so how can it? Have we lost access to the data on > these dates?
Re: Restore a Lotus Domino via TSM
Nick, Was this a restore across different Domino servers? What was the version of the Domino server that you took the backup on and what was the version that you are performing the restore on? Was this a restore across different OS platforms? Was this a logged database? Did you apply logs to the database after it was restored? Have you tried other database restores? Does this occur with all databases that you restore? Do I understand correctly that the restore worked successfully but when you opened the database with Notes it failed? Thanks, Del "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 10/25/2004 10:37:14 AM: > Hi to all, > > I have one question regarding TSM and TDP for Lotus Domino. > > I am using Tivoli Data Protection for Lotus Domino to take online backups. > I have performed successfully a lot of restores but currently when i tried > to restore an inactive backup of a database and then open it, i got an > error message "Invalid NSF version". The only change i have done in dsm.opt > file is enabling compression. > > Additional information: > > TSM SERVER: 5.1.5.0 > TDP for Lotus Domino: 5.1.5.0 > Lotus Domino: 6.5.2 > > Thanks in advance > > Nick
Re: dir include ?
thanks, i just wasnt sure ... - Original Message - From: "Neil Schofield" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Tuesday, October 26, 2004 2:23 PM Subject: dir include ? goc wrote: it seems that default is that dirs are backuped with mc which has the longest retain only version .. Absolutely right. Use the 'dirmc' option to override the default behaviour. There's info on it in the docs. Regards Neil Schofield Yorkshire Water Services Ltd Take a tour of a virtual waste water treatment works at http://www.wastewatertour.com The information in this e-mail is confidential and may also be legally privileged. The contents are intended for recipient only and are subject to the legal notice available at http://www.keldagroup.com/email.htm Yorkshire Water Services Limited Registered Office Western House Halifax Road Bradford BD6 2SZ Registered in England and Wales No 2366682
extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked]
Hi, any information about the syntax? Please :-) Best regards, Frank - Weitergeleitet von Frank Mueller/AG/OERAG am 26.10.2004 15:41 - Frank Mueller An: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 26.10.2004 13:18 Kopie: Thema:extend primary diskstorage pool on SSA-disks in TSM 5.1.8 Hi, I want to extend my primary diskstorage pool on SSA-disks. Environment: AIX 5.2 TSM 5.1.8 primary storage on SSA-disks secondary storage on tape (3494 library) copy storage pool on tape (3583 library) Now I must extend the primary disk storage pool. In older TSM versions (TSM Version 4) I start the following procedure to this - create log. volumes (AIX-LVM) - create jfs (AIX-LVM) - dsmfmt -data (AIX-Command-Line) - then in TSM: DEFINE VOLUME Is this in TSM-Version 5.1.8 the same procedure and the same syntax? Best regards, Frank
Re: extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked]
Frank, The procedure to define disk based volumes has not changed, so you're right ! You could eventually get a rid of the "dsmfmt blah ." step while using : "define volume formatsize=xxx"xxx being the size in MB you want for your volume. Hope this helped ... Cheers. Arnaud ** Panalpina Management Ltd., Basle, Switzerland, CIT Department Viadukstrasse 42, P.O. Box 4002 Basel/CH Phone: +41 (61) 226 11 11, FAX: +41 (61) 226 17 01 Direct: +41 (61) 226 19 78 e-mail: [EMAIL PROTECTED] ** -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Frank Mueller Sent: Tuesday, 26 October, 2004 15:42 To: [EMAIL PROTECTED] Subject: extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked] Hi, any information about the syntax? Please :-) Best regards, Frank - Weitergeleitet von Frank Mueller/AG/OERAG am 26.10.2004 15:41 - Frank Mueller An: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 26.10.2004 13:18 Kopie: Thema:extend primary diskstorage pool on SSA-disks in TSM 5.1.8 Hi, I want to extend my primary diskstorage pool on SSA-disks. Environment: AIX 5.2 TSM 5.1.8 primary storage on SSA-disks secondary storage on tape (3494 library) copy storage pool on tape (3583 library) Now I must extend the primary disk storage pool. In older TSM versions (TSM Version 4) I start the following procedure to this - create log. volumes (AIX-LVM) - create jfs (AIX-LVM) - dsmfmt -data (AIX-Command-Line) - then in TSM: DEFINE VOLUME Is this in TSM-Version 5.1.8 the same procedure and the same syntax? Best regards, Frank
Re: Client connection severed
Talk to the person who manages the firewall. It is common for firewall software to have its own "idletimeout" value. If a session goes quiet for a time, the firewall terminates the session. And it is not uncommon for TSM backups to exceed the firewall timeout value - can happen easily if TSM spends a long time searching directories before sending data, or if waiting on a response from Oracle, etc.. In my experience, it results in the errors you saw - the firewall closes the session, TSM reopens it - the backup eventually completes OK, but takes forever. Ask them to check the firewall, turn off their timeout (or change it to something very very large), and test running a backup. If that clears up your problem, then you can play around with the timeout value until you find something that will work most of the time. Good luck! Wanda Prather "I/O, I/O, It's all about I/O" -(me) -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Geert De Pecker Sent: Tuesday, October 26, 2004 3:30 AM To: [EMAIL PROTECTED] Subject: Client connection severed I keep getting the "session is lost" errors with a certain box. I already changed the maxsessions, maxschedule, idletimeout, but no changes. The errors keep coming. Only specials about this box is the fact that it has large files (oracle database) and that it is behind a firewall. I assume the big files (2 and 4 gb) are making the problem here. Any hints where I can look next? Thanks, Geert Server side: 10/26/04 03:15:01 ANR0480W Session 4563 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:26:22 ANR0480W Session 4566 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 03:45:54 ANR0480W Session 4567 for node AILDB (WinNT) terminated - connection with client severed. 10/26/04 04:15:49 ANR0480W Session 4568 for node AILDB (WinNT) terminated - connection with client severed. Client side dsmerror.log 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ANS1810E TSM session has been reestablished. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ANS1810E TSM session has been reestablished. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ANS1810E TSM session has been reestablished. 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. Client side dsmsched.log 26-10-2004 03:07:58 --- SCHEDULEREC OBJECT BEGIN BACKUP_INC_AIL 26-10-2004 03:00:00 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:17:23 ... successful 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 03:41:23 ... successful 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:11:18 ... successful 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 04:44:13 ... successful 26-10-2004 05:11:36 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:11:51 ... successful 26-10-2004 05:17:11 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:17:26 ... successful 26-10-2004 05:32:41 ANS1809W Session is lost; initializing session reopen procedure. 26-10-2004 05:32:56 ... successful 26-10-2004 05:35:47 --- SCHEDULEREC STATUS BEGIN 26-10-2004 05:35:47 Total number of objects inspected: 78,805
select cmd question
We have a mixed tapes in our 3494 library we have 3590 20Gb tape (shorts) and we have 3590 extended tapes and I need to make a count of the 3590 20Gb tapes, my question is there a select cmd to an mtlib cmd that would return the count of the 3590 20Gb tape even if they are in use or scratch tapes ? Thanks Justin
Re: select cmd question
>We have a mixed tapes in our 3494 library we have 3590 20Gb tape (shorts) >and we have 3590 extended tapes >and I need to make a count of the 3590 20Gb tapes, my question is there a >select cmd to an mtlib cmd that would >return the count of the 3590 20Gb tape even if they are in use or scratch >tapes ? Justin - Issue the mtlib command with '-qI' for volume inventory (short form) and examine the value of the last column byte relative to your /usr/include/sys/mtlibio.h vol_class, or '-vqI' (long form) for descriptions. Richard Sims
Re: Client connection severed
>...It is common for firewall software to have its own "idletimeout" value. ... As always, excellent experience-based info from Wanda. You might also look into the possibility of tuning the TCP keepalive value for the client operating system. For example, AIX has a network tuning option called tcp_keepintvl, which defaults to 75 seconds. Note that the client programming might also set the SO_KEEPALIVE socket option. Richard Sims
Re: Client connection severed
Geert - Some further info on your firewall problem...
Your posting did not specify your software levels, but look
into IBM site Technote 1109798
("How to make use of "keepalive" network option with TSM clients")
which is new as of August.
Richard Sims
AW: extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked]
Hi Frank, simply: yes. There is the possibility to use raw-volumes (logical volumes without filesystems), see the docs. Some enhancements to your (my) procedure: - use jfs2 (gives you some additional space) - after defining the volume in TSM, the TSM server uses direct IO (dio). - dsmfmt -data ... works on AIX-Command-Line, so BEFORE dsmfmt change the filesystem to use dio (chfs -a options=dio[,other_options] /Filesystem ; umount ; mount) (without dio the lrud -last recent use daemon, the Filesystem-Cache-Administrator- runs 'amok', if you format more than ? GB/ ??? MB (depends on your Memory)) I found out the above AFTER setting up two new TSM servers, formatting ca. 300GB log -, database- and diskpool volumes ;-((( Best regards, Peter Sternecker R+V Allgemeine Versicherung AG * mailto:[EMAIL PROTECTED] > -Ursprüngliche Nachricht- > Von: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Im > Auftrag von Frank Mueller > Gesendet: Dienstag, 26. Oktober 2004 13:18 > An: [EMAIL PROTECTED] > Betreff: extend primary diskstorage pool on SSA-disks in TSM > 5.1.8 [Virus checked] > > > Hi, > > I want to extend my primary diskstorage pool on SSA-disks. > Environment: > AIX 5.2 > TSM 5.1.8 > primary storage on SSA-disks > secondary storage on tape (3494 library) > copy storage pool on tape (3583 library) > > Now I must extend the primary disk storage pool. In older TSM > versions (TSM Version 4) I start the following procedure to this > - create log. volumes (AIX-LVM) > - create jfs (AIX-LVM) > - dsmfmt -data (AIX-Command-Line) > - then in TSM: DEFINE VOLUME > > Is this in TSM-Version 5.1.8 the same procedure and the same syntax? > > Best regards, > Frank >
Tips on backing up over slow networks please
Has anybody has experience with backing up over slow N/W's (min. 256k)? Unfortunately the clients are NOT Windows boxes so I can not use adaptive backup or journaling features. Currently I am not aware of the daily change rate of data on each system. My initial thoughts are to perform a staggered initial backup and then once complete implement a daily incremental backup. Things that I am looking for would be advice on the best performance tuning parameters to use. Thanks in advance Regards Neil Sharp <[EMAIL PROTECTED]>, Technical Support & TSM Consultant
Re: Client connection severed
Wanda, Richard, David, Thanks for the help. I was in the opinion that the firewall was set to a session timeout of 1 hour, which would have excluded these error occuring 12 minutes apart, but I checked and it was set to 10 minutes. I've changed the session timeout to 1 hour for the tsm ports 1500 and 1501. Wait and see I'll let you know how it went tonight (the backup I mean). Regards and thanks, Geert "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 26/10/2004 16:05:11: > Talk to the person who manages the firewall. > > It is common for firewall software to have its own "idletimeout" value. > If a session goes quiet for a time, the firewall terminates the session. > > And it is not uncommon for TSM backups to exceed the firewall timeout > value - can happen easily if TSM spends a long time searching > directories before sending data, or if waiting on a response from > Oracle, etc.. > > In my experience, it results in the errors you saw - the firewall closes > the session, TSM reopens it - the backup eventually completes OK, but > takes forever. > > Ask them to check the firewall, turn off their timeout (or change it to > something very very large), and test running a backup. If that clears > up your problem, then you can play around with the timeout value until > you find something that will work most of the time. > > Good luck! > Wanda Prather > "I/O, I/O, It's all about I/O" -(me) > > > > > -Original Message- > From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of > Geert De Pecker > Sent: Tuesday, October 26, 2004 3:30 AM > To: [EMAIL PROTECTED] > Subject: Client connection severed > > > I keep getting the "session is lost" errors with a certain box. > > I already changed the maxsessions, maxschedule, idletimeout, but no > changes. The errors keep coming. > > Only specials about this box is the fact that it has large files (oracle > database) and that it is behind a firewall. I assume > the big files (2 and 4 gb) are making the problem here. > > Any hints where I can look next? > > Thanks, > > Geert > > Server side: > > 10/26/04 03:15:01 ANR0480W Session 4563 for node AILDB (WinNT) > terminated - > connection with client severed. > 10/26/04 03:26:22 ANR0480W Session 4566 for node AILDB (WinNT) > terminated - > connection with client severed. > 10/26/04 03:45:54 ANR0480W Session 4567 for node AILDB (WinNT) > terminated - > connection with client severed. > 10/26/04 04:15:49 ANR0480W Session 4568 for node AILDB (WinNT) > terminated - > connection with client severed. > > Client side dsmerror.log > > 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:17:23 ANS1810E TSM session has been reestablished. > 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:41:23 ANS1810E TSM session has been reestablished. > 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 04:11:18 ANS1810E TSM session has been reestablished. > 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session > reopen > procedure. > > Client side dsmsched.log > > 26-10-2004 03:07:58 --- SCHEDULEREC OBJECT BEGIN BACKUP_INC_AIL > 26-10-2004 > 03:00:00 > 26-10-2004 03:17:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:17:23 ... successful > 26-10-2004 03:41:08 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 03:41:23 ... successful > 26-10-2004 04:11:03 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 04:11:18 ... successful > 26-10-2004 04:43:58 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 04:44:13 ... successful > 26-10-2004 05:11:36 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 05:11:51 ... successful > 26-10-2004 05:17:11 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 05:17:26 ... successful > 26-10-2004 05:32:41 ANS1809W Session is lost; initializing session > reopen > procedure. > 26-10-2004 05:32:56 ... successful > 26-10-2004 05:35:47 --- SCHEDULEREC STATUS BEGIN > 26-10-2004 05:35:47 Total number of objects inspected: 78,805
Informix and Tivoli
Hi Everybody, I'm currently testing our my Informix TDP and Tivoli setup and I get the following issue. I initiate the onbar command and the backup begins but looking at the onbar log I see the following : 2004-10-26 11:23:20 1437942 3235868 Begin level 0 backup rootdbs. 2004-10-26 11:23:20 1437942 3235868 Successfully connected to Storage Manager. Problme is it stays at this spot forever ( greater than 1 hour and my rootdbs is only about 500Mb ) or else I get a BAR_TIMEOUT error. I'm not sure that my client is communicating with the server even though I was able to porperly register my clien ton the server. My question is what is the proper BAR_BSALIB_PATH variable users have setup in their onconfig for an Informix 9.3 setup ? The ever useful Informix TDP states what it should be for an Informix 7 config : BAR_BSALIB_PATH=/usr/tivoli/tsm/client/informix/bin/bsashr10.o but not for any other subsequent versions. Any help would be greatly appreciated. Rich
space utilization
Hi all I need to create reports for space utilization I need to know how many space use all my sql databases .. do I need to add all of them one by one.. node by node ??? I need to know how many space use all my windows nodes ,, Domino nodes ... Sql nodes ... is there a easier way than add them one by one thanks Luc
Versions Data Exists Question
Our management class is setup that the maximum number of versions permitted is 3, I have a customer requesting that wants to know if can keep let's say...6 active versions of only one particular file on a server. Is that even possible? Background: TSM Server Version 5.2.3.4 AIX Operating System: 5.1.5 Nancy Backhaus Enterprise Systems [EMAIL PROTECTED] Office: (716) 887-7979 Cell: (716) 609-2138 CONFIDENTIALITY NOTICE: This email message and any attachments are for the sole use of the intended recipient(s) and may contain proprietary, confidential, trade secret or privileged information. Any unauthorized review, use, disclosure or distribution is prohibited and may be a violation of law. If you are not the intended recipient or a person responsible for delivering this message to an intended recipient, please contact the sender by reply email and destroy all copies of the original message.
Re: Open file and image backups on Linux
Hi Zoltan, I agree with you in regards to making a backup of file systems that contain System files like /, /usr, /var, etc. However, I don't think it is the backing them up that is a problem. It is how do you restore them? If you need to restore an image of /, /usr or other System related file system then you almost certainly will have to restore the entire OS any way. So I am not quite sure how effective an image backup of these file systems would be. You will need an OS with a TSM B/A Client installed in order to restore the image. You might be able to get this to work in AIX or Linux (not familiar enough with HP-UX or Solaris) using a dual boot system and boot one OS while you restore the other, but it will be tricky making the new target disk or partition bootable. In Linux using GRUB you should be able to get it done without too much trouble. In AIX, I am not sure how you would get the bosboot command to set the boot sector of the disk where the restored file systems were. This is an interesting problem, but there is already an alternative why to solve this problem using BMR or Sysback, so why reinvent the wheel. Zoltan Forray/AC/VCU wrote: Since when You can use the IMAGETYPE=dynamic option ! From the 5.2.x book: dynamic Replaces the dependency on the copy serialization value in the management class to perform an image backup without unmounting and remounting the file system read-only. Use this option only if the volume cannot be unmounted and remounted read-only. Tivoli Storage Manager backs up the volume as is without remounting it read-only. Corruption of the backup may occur if applications write to the volume while the backup is in progress. In this case, run fsck after a restore. This option is valid for AIX, Solaris, HP-UX, Linux86, Linux IA64, Linux pSeries, and Linux iSeries. include.image /home MYMC imagetype=dynamic "Stapleton, Mark" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 10/25/2004 11:27 AM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject Re: Open file and image backups on Linux You can only take image backups of Linux filesystems if you can unmount the filesystem. This requirement will prevent you from image backups of filesystems containing system files, such as /, /var, /etc/, and /usr. -- Mark Stapleton ([EMAIL PROTECTED]) Berbee Information Networks Office 262.521.5627 -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Sandra Sent: Monday, October 25, 2004 10:33 AM To: [EMAIL PROTECTED] Subject: Open file and image backups on Linux Dear All, Please share your experience on taking open file backup and image backup on Linux (including system files). I have been going through documentation and manuals, but there are so many IFs and BUTs involved. My scenario is to take cold backup of Oracle 9i on Linux, which is fine, but with this I need to take entire system backup so that I can restore my server real soon in case of disaster. Please share practical implementations that you gurus are using !!! Thanking you! This list has been a source of real good knowledge to me ! Kind Regards Sandra -- Regards, Mark D. Rodriguez President MDR Consulting, Inc. === MDR Consulting The very best in Technical Training and Consulting. IBM Advanced Business Partner SAIR Linux and GNU Authorized Center for Education IBM Certified Advanced Technical Expert, CATE AIX Support and Performance Tuning, RS6000 SP, TSM/ADSM and Linux Red Hat Certified Engineer, RHCE ===
Re: Open file and image backups on Linux
Dear Mark, I am confused again. See, I need to take backup (image or incremental) of entire system. IMAGETYPE is one things, and running fsck later on after restore is another. Now question is, BMR is actually something, that will backup and then restore the entire system in one go. Image is the solution for this probably. Now, If i take Image backup and my machine crashes, all i need to do is to install my OS and BA client, and start the restore of the image of all the file systems (including system file system). This way when i am going to restart after this, i will be on the same level as the image was, onlly thing that i would do is to run fsck if there is any problem. IMPORTANT: when i will initiate the image backup, the only thing that would be running would be those OS files and no application other than that. If there is some system file that is open in for exampple in /usr, and there is no application that is using that file, then it will be backed up?? Am i right? I m using Linux... Kind Regards, Sandra "Mark D. Rodriguez" wrote: > Hi Zoltan, > > I agree with you in regards to making a backup of file systems that > contain System files like /, /usr, /var, etc. However, I don't think it > is the backing them up that is a problem. It is how do you restore > them? If you need to restore an image of /, /usr or other System > related file system then you almost certainly will have to restore the > entire OS any way. So I am not quite sure how effective an image backup > of these file systems would be. You will need an OS with a TSM B/A > Client installed in order to restore the image. You might be able to > get this to work in AIX or Linux (not familiar enough with HP-UX or > Solaris) using a dual boot system and boot one OS while you restore the > other, but it will be tricky making the new target disk or partition > bootable. In Linux using GRUB you should be able to get it done without > too much trouble. In AIX, I am not sure how you would get the bosboot > command to set the boot sector of the disk where the restored file > systems were. > > This is an interesting problem, but there is already an alternative why > to solve this problem using BMR or Sysback, so why reinvent the wheel. > > Zoltan Forray/AC/VCU wrote: > > >Since when > > > >You can use the IMAGETYPE=dynamic option ! > > > >From the 5.2.x book: > > > >dynamic > >Replaces the dependency on the copy serialization value in the management > >class to perform an image backup without unmounting and remounting the > >file system read-only. Use this option only if the volume cannot be > >unmounted and remounted read-only. Tivoli Storage Manager backs up the > >volume as is without remounting it read-only. Corruption of the backup may > >occur if applications write to the volume while the backup is in progress. > >In this case, run fsck after a restore. This option is valid for AIX, > >Solaris, HP-UX, Linux86, Linux IA64, Linux pSeries, and Linux iSeries. > > > >include.image /home MYMC imagetype=dynamic > > > > > > > > > > > >"Stapleton, Mark" <[EMAIL PROTECTED]> > >Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> > >10/25/2004 11:27 AM > >Please respond to > >"ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> > > > > > >To > >[EMAIL PROTECTED] > >cc > > > >Subject > >Re: Open file and image backups on Linux > > > > > > > > > > > > > >You can only take image backups of Linux filesystems if you can unmount > >the filesystem. This requirement will prevent you from image backups of > >filesystems containing system files, such as /, /var, /etc/, and /usr. > > > >-- > >Mark Stapleton ([EMAIL PROTECTED]) > >Berbee Information Networks > >Office 262.521.5627 > > > > > > > >>-Original Message- > >>From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On > >>Behalf Of Sandra > >>Sent: Monday, October 25, 2004 10:33 AM > >>To: [EMAIL PROTECTED] > >>Subject: Open file and image backups on Linux > >> > >>Dear All, > >>Please share your experience on taking open file backup and > >>image backup on Linux (including system files). > >> > >>I have been going through documentation and manuals, but there > >>are so many IFs and BUTs involved. > >> > >>My scenario is to take cold backup of Oracle 9i on Linux, > >>which is fine, but with this I need to take entire system > >>backup so that I can restore my server real soon in case of disaster. > >> > >>Please share practical implementations that you gurus are using !!! > >> > >>Thanking you! > >>This list has been a source of real good knowledge to me ! > >> > >>Kind Regards > >>Sandra > >> > >> > >> > > > > > > > > -- > Regards, > Mark D. Rodriguez > President MDR Consulting, Inc. > > === > MDR Consulting > The very best in Technical Training and Consulting. > IBM Advanced Business Partner > SAIR Linux and GNU Authorized Center for Education > IBM Certified Advanc
ANS1076E
I cannot get rid of the directory Path not found error. I am running a simple archive backup under windows 5.1.6.7 TSM client. The backup seems to run ,but we use other software to schedule the backup and it is seeing it as a 12 error. Has anyone else had this problem.
Re: Versions Data Exists Question
>Our management class is setup that the maximum number of versions >permitted is 3, I have a customer requesting that wants to know if can >keep let's say...6 active versions of only one particular file on a >server. Is that even possible? Yes - by backing up that file again, with a new Include which names that file and associates it with a different management class having such a policy. Richard Sims
Re: select cmd question
Hi Justin, Select mediatype,count(*) from libvolumes group by mediatype Then.. you will need to know which mediatype is which. Matt. _-'-_ -|- -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Justin Case Sent: Tuesday, October 26, 2004 3:10 PM To: [EMAIL PROTECTED] Subject: select cmd question We have a mixed tapes in our 3494 library we have 3590 20Gb tape (shorts) and we have 3590 extended tapes and I need to make a count of the 3590 20Gb tapes, my question is there a select cmd to an mtlib cmd that would return the count of the 3590 20Gb tape even if they are in use or scratch tapes ? Thanks Justin ___ Disclaimer Notice __ This message and any attachments are confidential and should only be read by those to whom they are addressed. If you are not the intended recipient, please contact us, delete the message from your computer and destroy any copies. Any distribution or copying without our prior permission is prohibited. Internet communications are not always secure and therefore Powergen Retail Limited does not accept legal responsibility for this message. The recipient is responsible for verifying its authenticity before acting on the contents. Any views or opinions presented are solely those of the author and do not necessarily represent those of Powergen Retail Limited. Registered addresses: Powergen Retail Limited, Westwood Way, Westwood Business Park, Coventry, CV4 8LG. Registered in England and Wales No: 3407430 Telephone +44 (0) 2476 42 4000 Fax +44 (0) 2476 42 5432
To know what are empty scratch volumes in a pool
Dear All, Please give me the knowledge of a query that can give me the names of tapes, that in my scratch pool. I know there are 3 tapes that are scratch, but i don't know the volume labels. What command would give me the volume labels for those tapes? Kind Regards, Sandra
Re: To know what are empty scratch volumes in a pool
select volume_name, status from libvolumes where status='Scratch' or, if you have multiple libraries attached: select library_name, volume_name, status from libvolumes where status='Scratch' Sandra wrote: Dear All, Please give me the knowledge of a query that can give me the names of tapes, that in my scratch pool. I know there are 3 tapes that are scratch, but i don't know the volume labels. What command would give me the volume labels for those tapes? Kind Regards, Sandra
File space size is 0
Dear List, I have taken backup of oracle database using TDP for oracle. The filespace for this when expanded displays PCT util as 0MB, whereas there are backedup files in the tape as seen from "query content of storage pool volume" from web administration. Also how do i se when files were put on tape (backedup) for specific node, because query contenets of storage pool volume shows only the name and no size or time? Kind Regards, Sandra
Re: To know what are empty scratch volumes in a pool
select volume_name from libvolumes where status='Scratch' Debra I. Wagy PHONE:(314)252-6443 Storage Management & RecoveryTIE-LINE: 513-6443 UNIX Design and ImplementationMC: S306-6270 IBM Global Services EMAIL: [EMAIL PROTECTED] Sandra <[EMAIL PROTECTED] K> To Sent by: "ADSM: [EMAIL PROTECTED] Dist Stor cc Manager" <[EMAIL PROTECTED] Subject .EDU> To know what are empty scratch volumes in a pool 10/26/2004 01:35 PM Please respond to "ADSM: Dist Stor Manager" Dear All, Please give me the knowledge of a query that can give me the names of tapes, that in my scratch pool. I know there are 3 tapes that are scratch, but i don't know the volume labels. What command would give me the volume labels for those tapes? Kind Regards, Sandra <><><>
Poor restore performance Windows-client
Hello, TSM-server 5.1.7.3 on AIX 5.2. W2K-client 5.1.5.x and 5.2.x Last week we did a restore of 15 MB and 4000 small html-files. The performance was bad, it took 57 minutes to restore the 4000 html-files and 15 MB. The back-up is very fast. FTP between client and server is also fast. A restore of a big(ger) file is no problem. Collocation is on and data is only on one tape. During restore we see no mediawaits, so that's not the problem. I checked the fragmentationperc. and the database has a fragemented for only 0.99%, so also no problem. We restored the same data to another client with the same poor performance. So that's why I think this is a tuning problem for Windows-clients when restoring lots of small files. I tried to change some settings like tcpwindowsize and uselargecombuffers, but the result is still the same. We are not using the option 'dirmc', but when we start the restore the directory-structure is still there or we recreate it in another way, so the restore can start immediately. Restores of UNIX-clients or TDP-clients are fast. Does anyone has some parameters we can change/tune to speed up the restore-time? Thank you very much. Mike _ Get ready for school! Find articles, homework help and more in the Back to School Guide! http://special.msn.com/network/04backtoschool.armx
Re: File space size is 0
Additionally, the rest of the filespaces are displaying the right amount of space utilization. sandra Sandra wrote: > Dear List, > I have taken backup of oracle database using TDP for oracle. The filespace for > this when expanded displays PCT util as 0MB, whereas there are backedup files in > the tape as seen from "query content of storage pool volume" from web > administration. > > Also how do i se when files were put on tape (backedup) for specific node, > because query contenets of storage pool volume shows only the name and no size > or time? > > Kind Regards, > Sandra
Re: Poor restore performance Windows-client
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Mike Hagery >Last week we did a restore of 15 MB and 4000 small html-files. >The performance was bad, it took 57 minutes to restore the >4000 html-files and 15 MB. > >The back-up is very fast. FTP between client and server is >also fast. A restore of a big(ger) file is no problem. >Collocation is on and data is only on one tape. During restore >we see no mediawaits, so that's not the problem. Try this select statement: select distinct volume_name from volumeusage where stgpool_name='PRIMARYSTORAGEPOOL' and node_name='YOURNODENAME' All node and storagepool entries must be in uppercase. This will tell you how many tapes you'd have to mount to restore an entire node's worth of data. Your restore problem probably stems from how scattered your data is across your primary storage pool. It has nothing to do with database fragmentation. -- Mark Stapleton ([EMAIL PROTECTED]) Berbee Information Networks Office 262.521.5627
Re: File space size is 0
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Sandra >Additionally, >the rest of the filespaces are displaying the right amount of >space utilization. > >sandra > >Sandra wrote: > >> Dear List, >> I have taken backup of oracle database using TDP for oracle. The >> filespace for this when expanded displays PCT util as 0MB, whereas >> there are backedup files in the tape as seen from "query content of >> storage pool volume" from web administration. >> >> Also how do i se when files were put on tape (backedup) for specific >> node, because query contenets of storage pool volume shows only the >> name and no size or time? Data backed up through the TSM API does not show up in QUERY FILESPACE commands. It does show up, however, when you run QUERY OCCUPANCY. -- Mark Stapleton ([EMAIL PROTECTED]) Berbee Information Networks Office 262.521.5627
Re: Open file and image backups on Linux
On Tuesday 26 October 2004 19:37, Sandra wrote: > Dear Mark, > I am confused again. Me too :) > See, I need to take backup (image or incremental) of entire system. > IMAGETYPE is one things, and running fsck later on after restore is > another. Why? What do you want to do? Do a disaster recovery? Then you never need images, only incremental backups. > Now question is, BMR is actually something, that will backup and then > restore the entire system in one go. Image is the solution for this > probably. > > Now, If i take Image backup and my machine crashes, all i need to do is to > install my OS and BA client, and start the restore of the image of all the > file systems (including system file system). This way when i am going to > restart after this, i will be on the same level as the image was, onlly > thing that i would do is to run fsck if there is any problem. My DR scenario: - AIX: mksysb rules :)). Mksysb is a bootable tape (or cd, dvd, or network) that restores a single AIX server in 1 go. I did a test restore of a cluster node with TSM and domino and after 45 min the minutes was up and running !!! I started HACMP and TSM and domino were up and running without any problems. Use mksysb on a local tape for the TSM server (+ TSM db backup to disk ona filesystem in rootvg) and nim (network mksysb) for the AIX clients. - Windows: CBMR is cool. I used CBMR to restore a TSM server and it woked flawless. This was on a local, single, dedicated lto tape. CBMR is also network aware so you can backup over the network directly to a TSM server. If you don't use CBMR, you have to do it manually as described in the docs. Or ASR, available in 2003 and XP. With ASR, you can boot from a CD (windows + TSM client) and recover everything directly from the TSM server. - Linux: I never tried it, but there are a lot of tricks you can use. I should install a basic linux in a partition, recreate the original partition, mounts them and restore everything. You can boot from cd or change the bootloader to load the fresh restored linux. I know Cristie was working on a linux version of the CBMR software, but I don't know if it's finished. > IMPORTANT: when i will initiate the image backup, the only thing that would > be running would be those OS files and no application other than that. If > there is some system file that is open in for exampple in /usr, and there > is no application that is using that file, then it will be backed up?? Am i > right? Yes, with *unix you can backup any file you want, even open files. But the backup may be inconsistent. It also depends if the file is open read-only of read-write and if it's changed or not. Also, TSM will try by default 3 times to backup a changed file. So, unless you have a database running, only log files are changed. And you don't care about log files, they will not prevent you from doing a disaster restore. > I m using Linux... Good choice :) Stef -- [EMAIL PROTECTED] "Using Linux as bandwidth manager" http://www.docum.org/
backup complete, server still has session in run state
... anyone seen this? I have a client that completes it's backup successfully, but the server still has the session in "run" state many hours later. TSM Server Environment Platform Z/os TSM Server 5.2.2.4 TSM Client Platform SunOS 5.8 B/A client 5.2 If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/
Re: Versions Data Exists Question
You bet. For the TSM domain where this server lives (you may only have 1 domain, STANDARD, which is the default), you create a new management class in the non-active policy set. You define new retention rules in the new management class/backup copy group. You activate the policy set you just edited. On the client machine, you relate the file (or files) you want to be governed by the non-default rules to the new management class with an INCLUDE in the dsm.opt file: include *:\pathname\filename NEW_MGMT_CLASS Stop and restart the scheduler to pick up the changes to dsm.opt, if you are not using dsmcad. You can read all about management classes and setting rules in the Admin Guide. -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Nancy L Backhaus Sent: Tuesday, October 26, 2004 12:03 PM To: [EMAIL PROTECTED] Subject: Versions Data Exists Question Our management class is setup that the maximum number of versions permitted is 3, I have a customer requesting that wants to know if can keep let's say...6 active versions of only one particular file on a server. Is that even possible? Background: TSM Server Version 5.2.3.4 AIX Operating System: 5.1.5 Nancy Backhaus Enterprise Systems [EMAIL PROTECTED] Office: (716) 887-7979 Cell: (716) 609-2138 CONFIDENTIALITY NOTICE: This email message and any attachments are for the sole use of the intended recipient(s) and may contain proprietary, confidential, trade secret or privileged information. Any unauthorized review, use, disclosure or distribution is prohibited and may be a violation of law. If you are not the intended recipient or a person responsible for delivering this message to an intended recipient, please contact the sender by reply email and destroy all copies of the original message.
Re: Open file and image backups on Linux
GREAT Stef, Yes, this backup will be used in disaster recovery. That means i should not go for image backup, but just incremental every would do fine. !!! Sandra Stef Coene wrote: > On Tuesday 26 October 2004 19:37, Sandra wrote: > > Dear Mark, > > I am confused again. > Me too :) > > > See, I need to take backup (image or incremental) of entire system. > > IMAGETYPE is one things, and running fsck later on after restore is > > another. > Why? What do you want to do? Do a disaster recovery? Then you never need > images, only incremental backups. > > > Now question is, BMR is actually something, that will backup and then > > restore the entire system in one go. Image is the solution for this > > probably. > > > > Now, If i take Image backup and my machine crashes, all i need to do is to > > install my OS and BA client, and start the restore of the image of all the > > file systems (including system file system). This way when i am going to > > restart after this, i will be on the same level as the image was, onlly > > thing that i would do is to run fsck if there is any problem. > My DR scenario: > - AIX: mksysb rules :)). Mksysb is a bootable tape (or cd, dvd, or network) > that restores a single AIX server in 1 go. I did a test restore of a cluster > node with TSM and domino and after 45 min the minutes was up and running !!! > I started HACMP and TSM and domino were up and running without any problems. > Use mksysb on a local tape for the TSM server (+ TSM db backup to disk ona > filesystem in rootvg) and nim (network mksysb) for the AIX clients. > > - Windows: CBMR is cool. I used CBMR to restore a TSM server and it woked > flawless. This was on a local, single, dedicated lto tape. CBMR is also > network aware so you can backup over the network directly to a TSM server. > If you don't use CBMR, you have to do it manually as described in the docs. > Or ASR, available in 2003 and XP. With ASR, you can boot from a CD (windows + > TSM client) and recover everything directly from the TSM server. > > - Linux: I never tried it, but there are a lot of tricks you can use. I > should install a basic linux in a partition, recreate the original partition, > mounts them and restore everything. You can boot from cd or change the > bootloader to load the fresh restored linux. > I know Cristie was working on a linux version of the CBMR software, but I > don't know if it's finished. > > > IMPORTANT: when i will initiate the image backup, the only thing that would > > be running would be those OS files and no application other than that. If > > there is some system file that is open in for exampple in /usr, and there > > is no application that is using that file, then it will be backed up?? Am i > > right? > Yes, with *unix you can backup any file you want, even open files. But the > backup may be inconsistent. It also depends if the file is open read-only of > read-write and if it's changed or not. Also, TSM will try by default 3 times > to backup a changed file. > So, unless you have a database running, only log files are changed. And you > don't care about log files, they will not prevent you from doing a disaster > restore. > > > I m using Linux... > Good choice :) > > Stef > > -- > [EMAIL PROTECTED] > "Using Linux as bandwidth manager" > http://www.docum.org/
Re: Poor restore performance Windows-client
>Last week we did a restore of 15 MB and 4000 small html-files. The >performance was bad, it took 57 minutes to restore the 4000 html-files and >15 MB. > >The back-up is very fast. ... During restore we see no mediawaits, so that's >not the problem. Arrgh: No one who posts a question about restoral performance ever includes the details of the restoral, number of tapes mounted, drive type, etc. It all makes a big difference. See "Restoral performance" in http://people.bu.edu/rbs/ADSM.QuickFacts for common factors. We don't know how you checked for media waits, but if via Query SEssion, it may be the case that they were not long enough for the server to mention - but cumulatively it could add up. Backups are fast if they are streaming, and/or if the drive technology is excellent at start-stop. Restorals are fast only if the files being restored are contiguous - which is rarely the case over time, as expirations cause spread and repositioning. Repopulating conventional client directories can also be very painful in terms of time. Many small files are the bane of backup-restoral tasks. Richard Sims
Re: Open file and image backups on Linux
I can't agree with the statement "Do a disaster recovery? Then you never need images, only incremental backups." That is EXACTLY why we need to do IMAGE backups. When our mail systems disk array died, it took >15 DAYS to restore >15MILLION individual files. The slowness was due to OS overhead trying to restore/register that many files. TSM was constantly idle/waiting. Total storage was only 400GB. If we would have had an IMAGE backup to restore the affected filesystem and then did incremental restore from that point, we feel the recovery would have been considerably shorter ! Stef Coene <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 10/26/2004 03:13 PM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject Re: Open file and image backups on Linux On Tuesday 26 October 2004 19:37, Sandra wrote: > Dear Mark, > I am confused again. Me too :) > See, I need to take backup (image or incremental) of entire system. > IMAGETYPE is one things, and running fsck later on after restore is > another. Why? What do you want to do? Do a disaster recovery? Then you never need images, only incremental backups. > Now question is, BMR is actually something, that will backup and then > restore the entire system in one go. Image is the solution for this > probably. > > Now, If i take Image backup and my machine crashes, all i need to do is to > install my OS and BA client, and start the restore of the image of all the > file systems (including system file system). This way when i am going to > restart after this, i will be on the same level as the image was, onlly > thing that i would do is to run fsck if there is any problem. My DR scenario: - AIX: mksysb rules :)). Mksysb is a bootable tape (or cd, dvd, or network) that restores a single AIX server in 1 go. I did a test restore of a cluster node with TSM and domino and after 45 min the minutes was up and running !!! I started HACMP and TSM and domino were up and running without any problems. Use mksysb on a local tape for the TSM server (+ TSM db backup to disk ona filesystem in rootvg) and nim (network mksysb) for the AIX clients. - Windows: CBMR is cool. I used CBMR to restore a TSM server and it woked flawless. This was on a local, single, dedicated lto tape. CBMR is also network aware so you can backup over the network directly to a TSM server. If you don't use CBMR, you have to do it manually as described in the docs. Or ASR, available in 2003 and XP. With ASR, you can boot from a CD (windows + TSM client) and recover everything directly from the TSM server. - Linux: I never tried it, but there are a lot of tricks you can use. I should install a basic linux in a partition, recreate the original partition, mounts them and restore everything. You can boot from cd or change the bootloader to load the fresh restored linux. I know Cristie was working on a linux version of the CBMR software, but I don't know if it's finished. > IMPORTANT: when i will initiate the image backup, the only thing that would > be running would be those OS files and no application other than that. If > there is some system file that is open in for exampple in /usr, and there > is no application that is using that file, then it will be backed up?? Am i > right? Yes, with *unix you can backup any file you want, even open files. But the backup may be inconsistent. It also depends if the file is open read-only of read-write and if it's changed or not. Also, TSM will try by default 3 times to backup a changed file. So, unless you have a database running, only log files are changed. And you don't care about log files, they will not prevent you from doing a disaster restore. > I m using Linux... Good choice :) Stef -- [EMAIL PROTECTED] "Using Linux as bandwidth manager" http://www.docum.org/
Re: extend primary diskstorage pool on SSA-disks in TSM 5.1.8 [Virus checked]
On Tuesday 26 October 2004 13:18, Frank Mueller wrote: > Hi, > > I want to extend my primary diskstorage pool on SSA-disks. > Environment: > AIX 5.2 > TSM 5.1.8 > primary storage on SSA-disks > secondary storage on tape (3494 library) > copy storage pool on tape (3583 library) > > Now I must extend the primary disk storage pool. In older TSM versions (TSM > Version 4) I start the following procedure to this > - create log. volumes (AIX-LVM) > - create jfs (AIX-LVM) > - dsmfmt -data (AIX-Command-Line) > - then in TSM: DEFINE VOLUME > > Is this in TSM-Version 5.1.8 the same procedure and the same syntax? skip the dsmfmt part. do help define volume for the right syntax. Stef -- [EMAIL PROTECTED] "Using Linux as bandwidth manager" http://www.docum.org/
Re: Open file and image backups on Linux
On Tuesday 26 October 2004 21:33, Zoltan Forray/AC/VCU wrote: > I can't agree with the statement "Do a disaster recovery? Then you never > need images, only incremental backups." > > That is EXACTLY why we need to do IMAGE backups. I don't agree. You need image backup for the disks with mails (I hope you don't store them on the os disks) and incremental backups for the os files. > When our mail systems disk array died, it took >15 DAYS to restore > >15MILLION individual files. The slowness was due to OS overhead trying to > restore/register that many files Was this for the os of the client or the os of the TSM server? > TSM was constantly idle/waiting. Total > storage was only 400GB. If we would have had an IMAGE backup to restore > the affected filesystem and then did incremental restore from that point, > we feel the recovery would have been considerably shorter ! Ok, you are right, there is no single solution for the problem. Stef -- [EMAIL PROTECTED] "Using Linux as bandwidth manager" http://www.docum.org/
Re: Open file and image backups on Linux
The mail system resides on an AIX server. The storage is in a SAN ! For the OS, it is irrelevant since you would have to reinstall AIX, anyway !. However, we have seen lots of applications that store their files and data as part of /etc, /usr, /home, etc. We even have one "client" that decided to link/share /usr/local among many systems. Stef Coene <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 10/26/2004 04:27 PM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject Re: Open file and image backups on Linux On Tuesday 26 October 2004 21:33, Zoltan Forray/AC/VCU wrote: > I can't agree with the statement "Do a disaster recovery? Then you never > need images, only incremental backups." > > That is EXACTLY why we need to do IMAGE backups. I don't agree. You need image backup for the disks with mails (I hope you don't store them on the os disks) and incremental backups for the os files. > When our mail systems disk array died, it took >15 DAYS to restore > >15MILLION individual files. The slowness was due to OS overhead trying to > restore/register that many files Was this for the os of the client or the os of the TSM server? > TSM was constantly idle/waiting. Total > storage was only 400GB. If we would have had an IMAGE backup to restore > the affected filesystem and then did incremental restore from that point, > we feel the recovery would have been considerably shorter ! Ok, you are right, there is no single solution for the problem. Stef -- [EMAIL PROTECTED] "Using Linux as bandwidth manager" http://www.docum.org/
Re: space utilization
I am very bad at SQL statements, but I do it this way: select sum(total_mb) from auditocc where node_name='node1' or node_name=' node2' or node_name... Just select which nodes you want to add to the total. Etienne Brodeur Coordonnateur Technique - Technical Coordinator Serti Informatique Inc. Tel.: (514) 493-1909 Fax: (514) 493-3575 http://www.serti.com Luc Beaudoin <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 26/10/2004 11:52 AM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To [EMAIL PROTECTED] cc Subject space utilization Hi all I need to create reports for space utilization I need to know how many space use all my sql databases .. do I need to add all of them one by one.. node by node ??? I need to know how many space use all my windows nodes ,, Domino nodes ... Sql nodes ... is there a easier way than add them one by one thanks Luc
Re: ANS1076E
Backup and archive operations can accept multiple file specifications for backup. For example: dsmc selective filespec1 filespec2 filespec3 If any of the filespec in the list do not exist, you will get ANS1076E. Therefore you should carefully examine the file specifications on your backup or archive operation and make sure that they are correct, and that the specifications exist. Regards, Andy Andy Raibeck IBM Software Group Tivoli Storage Manager Client Development Internal Notes e-mail: Andrew Raibeck/Tucson/[EMAIL PROTECTED] Internet e-mail: [EMAIL PROTECTED] The only dumb question is the one that goes unasked. The command line is your friend. "Good enough" is the enemy of excellence. "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 10/26/2004 10:49:45: > I cannot get rid of the directory Path not found error. I am running a > simple archive backup under windows 5.1.6.7 TSM client. The backup seems > to run ,but we use other software to schedule the backup and it is seeing it > as a 12 error. Has anyone else had this problem.
Windows Small Business Server 2003 is it supported?
Hi, I can't get anybody at IBM to give me a definite answer on this, so I will ask the list. I think that Windows Small Business server 2003 is just a variation on the Windows 2003 server code, i.e. it just has some additional features, but it is not listed as a supported OS. Therefore, my question is will an ITSM server run on it properly? Will my customer be able to call IBM support and still get help? Is anybody out there already using this OS and if so have they had any problems with it? Any feedback on this would be greatly appreciated. Thank You. -- Regards, Mark D. Rodriguez President MDR Consulting, Inc. === MDR Consulting The very best in Technical Training and Consulting. IBM Advanced Business Partner SAIR Linux and GNU Authorized Center for Education IBM Certified Advanced Technical Expert, CATE AIX Support and Performance Tuning, RS6000 SP, TSM/ADSM and Linux Red Hat Certified Engineer, RHCE ===
Archive fails with >4GB files
We're seeing archive attempts failing on files in excess of 4GB. The archive seems to be fine until 4GB are transferred and then stops as if things were finished in a normal fashion. This entry in the DSMERROR.LOG file corresponds with the stop point: 10/26/2004 15:47:54 Attempt to free unknown pointer. Addr a855af00, File procopts.cpp, Line 6901 10/26/2004 15:47:54 Attempt to free unknown pointer. Addr a855af00, File procopts.cpp, Line 6903 The TSM Server is at 5.2.3.2 and the client is on a NetWare 6.5 SP2 server going through a clustered NetWare volume. The TSM client code is 5.02.03 and the TSAFS version is 6.50.09. Has anyone else seen this or better yet solved it? Andrew Ferris Network Support Analyst iCAPTURE Research Centre University of British Columbia
AW: Archive fails with >4GB files
Its a documented bug. Ask Novell ? regards Joachim -Ursprüngliche Nachricht- Von: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Auftrag von Andrew Ferris Gesendet: Mittwoch, 27. Oktober 2004 02:03 An: [EMAIL PROTECTED] Betreff: Archive fails with >4GB files We're seeing archive attempts failing on files in excess of 4GB. The archive seems to be fine until 4GB are transferred and then stops as if things were finished in a normal fashion. This entry in the DSMERROR.LOG file corresponds with the stop point: 10/26/2004 15:47:54 Attempt to free unknown pointer. Addr a855af00, File procopts.cpp, Line 6901 10/26/2004 15:47:54 Attempt to free unknown pointer. Addr a855af00, File procopts.cpp, Line 6903 The TSM Server is at 5.2.3.2 and the client is on a NetWare 6.5 SP2 server going through a clustered NetWare volume. The TSM client code is 5.02.03 and the TSAFS version is 6.50.09. Has anyone else seen this or better yet solved it? Andrew Ferris Network Support Analyst iCAPTURE Research Centre University of British Columbia
