Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... 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 Steve Bennett Sent: Friday, 11 March, 2005 23:51 To: ADSM-L@VM.MARIST.EDU Subject: SUSPECT: (MSW) review of recent addition of sata array storage pool For what it is worth I'll provide my experience with installing a sata storage pool and some observations about potential issues. Dell 2550, 1gb ram, 2 x 1ghz pentium cpus TSM server v5.2.3.2 Windows 2000 sp4 160gb local scsi disk 6.4tb (16 x 400gb) local sata array fiber attached IBM 3494 with 2 dedicated scsi attached 3590-e1a drives 600 IBM 3590J tapes about 4.6tb of TSM data stored in primary storage pools about 100gb compressed client data stored daily Our 3494 was filling up and management did not want to spend the $ to upgrade to 3592 drives and media. We added a 6.4tb fiber attached sata array which has about 5.3tb usable when configured for raid5. Clients backup daily to the local scsi diskpool and once a day we migrate that storage pool to the sata diskpool. The sata diskpool is defined to TSM as a sequential with maxscr=260, file size of 20gb, reusedelay=8 and migrdelay=33. Once a day we migrate about 2% of the sata pool to the collocated tapepool and do sata file reclamation and tapepool reclamation. We see 85 to 90gb per hour throughput when migrating from the scsi disk to the sata. Running two migration processes doesn't seem to increase the throughput so I suspect the interface or pci bus is pretty well maxed with one migration process. Sata file reclamation runs about 100gb per hour. Sata migration to tape throughput is dependent on the number of tape mounts and how much tape seek there is. Process displays indicate 10-20gb per hour is the norm for us. Tapepool reclamation can see as high as 60gb per hour. Overall it was fairly easy to implement. As far as tape use relief we are able to keep about 4tb of data on the sata so we now have less than 100 tapes used in the 3494. Cost for the sata, interface, cable, etc. was about $15k. No comment yet on the reliability of this brand of sata. The only real issue I see right now is the limited throughput when migrating from the sata to tape. The migration is done one sata volume at a time which causes some collocated tapes to be mounted multiple times to receive client data from multiple sata volumes. Unless I missed something, multiple concurrent migration processes are not allowed (migpr=2 is invalid) for the sata diskpool so I'm not sure how I could increase this migration throughput. Perhaps I could define the sata volumes larger which reduces the number of volumes to be migrated and results in fewer potential multiple mounts of the same tape, a minimal improvement at best. Questions, comments, suggestions? -- Steve Bennett, (907) 465-5783 State of Alaska, Enterprise Technology Services, Technical Services Section
Re: Windows TSM Client: PAE/AWE enabled?
Maybe it is good idea to try/test the NT 4.0 ( I know it's NT 4.0 ) Res KIT tool imagecfg.exe To make the executable large memory aware. http://support.microsoft.com/kb/171793 Imagecfg -l adsm.exe Maybe it works Best Regards, Met Vriendelijke Groet, Mit freundlichen Gruessen, Arnold van Wijnbergen Technical Engineer IT Supply Development METRO Cash Carry Nederland B.V. Information Services Dalsteindreef 101 t/m 139 Kamer 4.05 1112 XC Diemen ' Tel.: +31 (20) 398 0987 7 Fax: +31 (020) 398 09 80 * Mailto:[EMAIL PROTECTED] -Oorspronkelijk bericht- Van: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Namens Andrew Raibeck Verzonden: zondag 13 maart 2005 0:29 Aan: ADSM-L@VM.MARIST.EDU Onderwerp: Re: Windows TSM Client: PAE/AWE enabled? For what it's worth, we are looking at this, but I cannot commit to if or when we will do it. If this (or any other change request) is important for you, you should contact your IBM sales/marketing/account rep and ask them to open a formal requirement for you. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-12 12:25:12: Laurent Bendavid [EMAIL PROTECTED] wrote: Stapleton, Mark wrote: From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Egon Blouder I am running a few huge file servers (~6-9 million files, 600GB-1.2TB data, 2GB MEM). In order to enhance backup speed I tried to set resourceutilization to a higher value (5-10). After setting that value I monitored dsm process during backup. The process uses up to 2GB memory and exits with a memory error message. TSM Client ran out of memory. Now I'm thinking about installing up to 8GB (max. for W2K Adv. Server) additional memory for these clients. I'll set /PAE boot option in boot.ini in order to enable more than 4GB of memory. Unfortunately I cannot figure out whether TSM client (5.2/5.3) is able to use that additional memory using Address Window Extension (AWE) API. Does anybody know whether it's supported? If using PAE doesn't get you what you want, try this workaround: 1. Create multiple dsm.opt option files, each using a different nodename and pointing at distinct disk drives. Example: nodename nodeA domain c: d: e: nodename nodeB domain f: g: systemobjects 2. Create a TSM scheduler service for each nodename. 3. Run backups (and restores) in parallel. -- Mark Stapleton ([EMAIL PROTECTED]) Office 262.521.5627 We have the same problem. If you use TSM Win32 bits, you could not use more than 2GB memory, after dsmc exit with ran out of memory. /PAE didn't work in our tests. You could use only one dsmc process for one drive, so you have to check if you don't exceed this limit for one drive (depends on the number of objects inspected, 4 millions in my situation). However, it's this requisite is ok, the workaround explain is ok. We are in the bad case, so we use other workaround : - split backup with // dsmc on different file tree - we realize that dsmc in command line have less memory consumption than dsmc sched (we have a PMR open ont this object) - you could use memoryefficientbackup = yes = impact on backup time Recently I read a Field Guide (or something like that) on how to run multiple schedulers at one time and how to feed them into one node. I know that I could use memoryefficientbackup=yes but my box has that much memory so why couldn't it be used? Additionally I disklike the implementation of memoryefficientbackup parameter. I would prefer setting the max memory limit for dsm processes. Why don't dsm processes use those 2GB memory and if more memory is needed to process backup dsm should move on memoryefficient behaviour. Btw.: Can anybody forward that to TSM Client RD? Regards -- __ Switch to Netscape Internet Service. As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register Netscape. Just the Net You Need. New! Netscape Toolbar for Internet Explorer Search from anywhere on the Web and block those annoying pop-ups. Download now at http://channels.netscape.com/ns/search/install.jsp
Re: Windows TSM Client: PAE/AWE enabled?
This will not work. A single Windows process can not grow to more than 2 Gb of memory. I have a similar issue on a box with 6Gb memory and multiple 1-2Tb file systems. Best option we could come up with was Journal based backups, which enhances the speed of backup and make things stable most of the time. Exception is if the system crashes and the journal needs to be recreated. Steve Young Information Services -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Andrew Raibeck Sent: Saturday, March 12, 2005 6:29 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: Windows TSM Client: PAE/AWE enabled? For what it's worth, we are looking at this, but I cannot commit to if or when we will do it. If this (or any other change request) is important for you, you should contact your IBM sales/marketing/account rep and ask them to open a formal requirement for you. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-12 12:25:12: Laurent Bendavid [EMAIL PROTECTED] wrote: Stapleton, Mark wrote: From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Egon Blouder I am running a few huge file servers (~6-9 million files, 600GB-1.2TB data, 2GB MEM). In order to enhance backup speed I tried to set resourceutilization to a higher value (5-10). After setting that value I monitored dsm process during backup. The process uses up to 2GB memory and exits with a memory error message. TSM Client ran out of memory. Now I'm thinking about installing up to 8GB (max. for W2K Adv. Server) additional memory for these clients. I'll set /PAE boot option in boot.ini in order to enable more than 4GB of memory. Unfortunately I cannot figure out whether TSM client (5.2/5.3) is able to use that additional memory using Address Window Extension (AWE) API. Does anybody know whether it's supported? If using PAE doesn't get you what you want, try this workaround: 1. Create multiple dsm.opt option files, each using a different nodename and pointing at distinct disk drives. Example: nodename nodeA domain c: d: e: nodename nodeB domain f: g: systemobjects 2. Create a TSM scheduler service for each nodename. 3. Run backups (and restores) in parallel. -- Mark Stapleton ([EMAIL PROTECTED]) Office 262.521.5627 We have the same problem. If you use TSM Win32 bits, you could not use more than 2GB memory, after dsmc exit with ran out of memory. /PAE didn't work in our tests. You could use only one dsmc process for one drive, so you have to check if you don't exceed this limit for one drive (depends on the number of objects inspected, 4 millions in my situation). However, it's this requisite is ok, the workaround explain is ok. We are in the bad case, so we use other workaround : - split backup with // dsmc on different file tree - we realize that dsmc in command line have less memory consumption than dsmc sched (we have a PMR open ont this object) - you could use memoryefficientbackup = yes = impact on backup time Recently I read a Field Guide (or something like that) on how to run multiple schedulers at one time and how to feed them into one node. I know that I could use memoryefficientbackup=yes but my box has that much memory so why couldn't it be used? Additionally I disklike the implementation of memoryefficientbackup parameter. I would prefer setting the max memory limit for dsm processes. Why don't dsm processes use those 2GB memory and if more memory is needed to process backup dsm should move on memoryefficient behaviour. Btw.: Can anybody forward that to TSM Client RD? Regards -- __ Switch to Netscape Internet Service. As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register Netscape. Just the Net You Need. New! Netscape Toolbar for Internet Explorer Search from anywhere on the Web and block those annoying pop-ups. Download now at http://channels.netscape.com/ns/search/install.jsp _ The information contained in this electronic e-mail transmission and any attachments are intended only for the use of the individual or entity to whom or to which it is addressed, and may contain information that is privileged, confidential and exempt from disclosure under applicable law. If the reader of this communication is not the intended recipient, or the employee or agent responsible for delivering this communication to the intended recipient, you are hereby notified that any dissemination, distribution, copying or
SkipNTsecurityCRC doesn't seem to work as I'd expect
First the problem: A Windows fileserver with about 560GB data in 13.500.000 files. (We also have one with 1,1TB data.) Always done (full/standard) incrementals. Consider a security update (local group addition; no discussion about that). A normal incremental backup treats me next run with a full backup of all. (Server v5.2.4 on W2K-sp4, Client v5.2.3.11 on W2K-sp4) Considerations: - I don't really need or want security on file level: via inheritance the files get the right rights from the directory - A dirsonly backup takes as long as a full incremental (so it seems, and because also file info is sent, doesn't save memory; And about an extra filesonly backup...) - A full incremental takes about a whole night - An incremental with option skipntpermissions treats me with a full backup (that I don't like) next run - An incremental with option skipntsecuritycrc treats me with a full backup (that I don't like) next run(is this a bug or something?) - I don't like full backups at all (except for databases; that why we chose TSM didn't we?) - Journal based backups also need a full incremental now and then Solutions: ? Ton Mol Corus IJmuiden Information Services
JR-Red Hat 7.2 ES cluster running ipvs
My experience with TSM is that I have set it up for a windows 2 node cluster, backing up the shared SAN disk. For this example I installed TSM on each node then registered the virtual node with TSM. Next I configure the cluster service to start TSM when a node fails over and setup the shared resource as well. Is the process similar for setting up TSM for a 2 Node Red Hat 7.2 ES cluster running ipvs? Thanks in advance
Re: If we all complain, do you think they will add the WEB gui back?
Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
I did not collocate the sata diskpool. I believe that collocation would restict the use of multi process restores, might cause more disk partition fragmentation and I really did not see any benefit. PAC Brion Arnaud wrote: Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... 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 Steve Bennett Sent: Friday, 11 March, 2005 23:51 To: ADSM-L@VM.MARIST.EDU Subject: SUSPECT: (MSW) review of recent addition of sata array storage pool For what it is worth I'll provide my experience with installing a sata storage pool and some observations about potential issues. Dell 2550, 1gb ram, 2 x 1ghz pentium cpus TSM server v5.2.3.2 Windows 2000 sp4 160gb local scsi disk 6.4tb (16 x 400gb) local sata array fiber attached IBM 3494 with 2 dedicated scsi attached 3590-e1a drives 600 IBM 3590J tapes about 4.6tb of TSM data stored in primary storage pools about 100gb compressed client data stored daily Our 3494 was filling up and management did not want to spend the $ to upgrade to 3592 drives and media. We added a 6.4tb fiber attached sata array which has about 5.3tb usable when configured for raid5. Clients backup daily to the local scsi diskpool and once a day we migrate that storage pool to the sata diskpool. The sata diskpool is defined to TSM as a sequential with maxscr=260, file size of 20gb, reusedelay=8 and migrdelay=33. Once a day we migrate about 2% of the sata pool to the collocated tapepool and do sata file reclamation and tapepool reclamation. We see 85 to 90gb per hour throughput when migrating from the scsi disk to the sata. Running two migration processes doesn't seem to increase the throughput so I suspect the interface or pci bus is pretty well maxed with one migration process. Sata file reclamation runs about 100gb per hour. Sata migration to tape throughput is dependent on the number of tape mounts and how much tape seek there is. Process displays indicate 10-20gb per hour is the norm for us. Tapepool reclamation can see as high as 60gb per hour. Overall it was fairly easy to implement. As far as tape use relief we are able to keep about 4tb of data on the sata so we now have less than 100 tapes used in the 3494. Cost for the sata, interface, cable, etc. was about $15k. No comment yet on the reliability of this brand of sata. The only real issue I see right now is the limited throughput when migrating from the sata to tape. The migration is done one sata volume at a time which causes some collocated tapes to be mounted multiple times to receive client data from multiple sata volumes. Unless I missed something, multiple concurrent migration processes are not allowed (migpr=2 is invalid) for the sata diskpool so I'm not sure how I could increase this migration throughput. Perhaps I could define the sata volumes larger which reduces the number of volumes to be migrated and results in fewer potential multiple mounts of the same tape, a minimal improvement at best. Questions, comments, suggestions? -- Steve Bennett, (907) 465-5783 State of Alaska, Enterprise Technology Services, Technical Services Section -- Steve Bennett, (907) 465-5783 State of Alaska, Enterprise Technology Services, Technical Services Section
Re: If we all complain, do you think they will add the WEB gui back?
really ? where ? i'm on it !!! - Original Message - From: Ben Bullock [EMAIL PROTECTED] To: ADSM-L@VM.MARIST.EDU Sent: Monday, March 14, 2005 4:20 PM Subject: Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] === This e-mail is intended only for its addressee and may contain information that is privileged, confidential, or otherwise protected from disclosure. If you have received this communication in error, please notify us immediately by e-mail: [EMAIL PROTECTED], and delete the original message. ===
Re: If we all complain, do you think they will add the WEB gui back?
Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
Please don't shoot the messenger, but as far as I know (and I have this on very good authority), this is not true. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 08:20:31: Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben
Re: If we all complain, do you think they will add the WEB gui back?
I see 5.3.0.1 and 5.3.0.2 but no 5.3.1 -Original Message- From: Iain Barnetson [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
Well...there is 5.3.0.2 but I don't see anything about going back to old GUI just fixes for ISC/Admin. Iain Barnetson wrote: Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
...and where did you download it from? There is no 5.3.1 directory in the maintenance directory on service.boulder.ibm.com. -- Mark Stapleton ([EMAIL PROTECTED]) Office 262.521.5627 -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Iain Barnetson Sent: Monday, March 14, 2005 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
ok, now i'm confused ... i looked over ptf's on support and something on PA , but i found nothing ... others ? - Original Message - From: Andrew Raibeck [EMAIL PROTECTED] To: ADSM-L@VM.MARIST.EDU Sent: Monday, March 14, 2005 4:54 PM Subject: Re: If we all complain, do you think they will add the WEB gui back? Please don't shoot the messenger, but as far as I know (and I have this on very good authority), this is not true. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 08:20:31: Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben
Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
Steve, I agree with you : collocation on this kind of storage pool doesn't seem to make much sense. However I've read something in IBM's Administrator guide for 5.2 (ref GC32-0768-01), page 206~207, that made me doubt. It states : If you decide to migrate data from one sequential access storage pool to another, ensure that: - Collocation is set the same in both storage pools. For example, if collocation is set to yes in the first storage pool, then collocation should be set to yes in the next storage pool. As I saw you where migrating data to a collocated tape-based pool, I was curious ! 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 Steve Bennett Sent: Monday, 14 March, 2005 16:34 To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool I did not collocate the sata diskpool. I believe that collocation would restict the use of multi process restores, might cause more disk partition fragmentation and I really did not see any benefit. PAC Brion Arnaud wrote: Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... 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 Steve Bennett Sent: Friday, 11 March, 2005 23:51 To: ADSM-L@VM.MARIST.EDU Subject: SUSPECT: (MSW) review of recent addition of sata array storage pool For what it is worth I'll provide my experience with installing a sata storage pool and some observations about potential issues. Dell 2550, 1gb ram, 2 x 1ghz pentium cpus TSM server v5.2.3.2 Windows 2000 sp4 160gb local scsi disk 6.4tb (16 x 400gb) local sata array fiber attached IBM 3494 with 2 dedicated scsi attached 3590-e1a drives 600 IBM 3590J tapes about 4.6tb of TSM data stored in primary storage pools about 100gb compressed client data stored daily Our 3494 was filling up and management did not want to spend the $ to upgrade to 3592 drives and media. We added a 6.4tb fiber attached sata array which has about 5.3tb usable when configured for raid5. Clients backup daily to the local scsi diskpool and once a day we migrate that storage pool to the sata diskpool. The sata diskpool is defined to TSM as a sequential with maxscr=260, file size of 20gb, reusedelay=8 and migrdelay=33. Once a day we migrate about 2% of the sata pool to the collocated tapepool and do sata file reclamation and tapepool reclamation. We see 85 to 90gb per hour throughput when migrating from the scsi disk to the sata. Running two migration processes doesn't seem to increase the throughput so I suspect the interface or pci bus is pretty well maxed with one migration process. Sata file reclamation runs about 100gb per hour. Sata migration to tape throughput is dependent on the number of tape mounts and how much tape seek there is. Process displays indicate 10-20gb per hour is the norm for us. Tapepool reclamation can see as high as 60gb per hour. Overall it was fairly easy to implement. As far as tape use relief we are able to keep about 4tb of data on the sata so we now have less than 100 tapes used in the 3494. Cost for the sata, interface, cable, etc. was about $15k. No comment yet on the reliability of this brand of sata. The only real issue I see right now is the limited throughput when migrating from the sata to tape. The migration is done one sata volume at a time which causes some collocated tapes to be mounted multiple times to receive client data from multiple sata volumes. Unless I missed something, multiple concurrent migration processes are not allowed (migpr=2 is invalid) for the sata diskpool so I'm not sure how I could increase this migration throughput. Perhaps I could define the sata volumes larger which reduces the number of volumes to be migrated and results in fewer potential multiple mounts of the same tape, a minimal improvement at best. Questions, comments, suggestions? -- Steve Bennett, (907) 465-5783 State of Alaska, Enterprise Technology
Re: If we all complain, do you think they will add the WEB gui back?
Hmmm, perhaps he meant '5.3.0.2' and not '5.3.2'? I can see a 5.3.0.2 version out there, for AIX at least: ftp://service.boulder.ibm.com/storage/tivoli-storage-management/patches/ server/AIX/5.3.0.2/ Hmm... Lemme touch base with the IBM rep again to see if he can give me the exact PTF... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Iain Barnetson Sent: Monday, March 14, 2005 8:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
5.3.1 is not yet available. Patch levels 5.3.0.1 and 5.3.0.2 are available. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 08:57:10: ...and where did you download it from? There is no 5.3.1 directory in the maintenance directory on service.boulder.ibm.com. -- Mark Stapleton ([EMAIL PROTECTED]) Office 262.521.5627 -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Iain Barnetson Sent: Monday, March 14, 2005 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
Hmm, Perhaps my IBM person is misinformed. Bad me for not double checking it before shot-gunning it out to the list Lemme check with him again. Sorry for the confusion. Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of goc Sent: Monday, March 14, 2005 8:58 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? ok, now i'm confused ... i looked over ptf's on support and something on PA , but i found nothing ... others ? - Original Message - From: Andrew Raibeck [EMAIL PROTECTED] To: ADSM-L@VM.MARIST.EDU Sent: Monday, March 14, 2005 4:54 PM Subject: Re: If we all complain, do you think they will add the WEB gui back? Please don't shoot the messenger, but as far as I know (and I have this on very good authority), this is not true. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 08:20:31: Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben
Re: If we all complain, do you think they will add the WEB gui back?
Yeah right enough it's 5.3.01 Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Andrew Raibeck Sent: 14 March 2005 16:00 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? 5.3.1 is not yet available. Patch levels 5.3.0.1 and 5.3.0.2 are available. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 08:57:10: ...and where did you download it from? There is no 5.3.1 directory in the maintenance directory on service.boulder.ibm.com. -- Mark Stapleton ([EMAIL PROTECTED]) Office 262.521.5627 -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Iain Barnetson Sent: Monday, March 14, 2005 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? Sorry but 5.3.1 is out I donwloaded it on Sunday :) Regards, Iain Barnetson IT Systems Administrator UKN Infrastructure Operations -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: 14 March 2005 15:46 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: If we all complain, do you think they will add the WEB gui back?
What is the PTF number??? -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of goc Sent: Monday, March 14, 2005 10:46 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] If we all complain, do you think they will add the WEB gui back? really ? where ? i'm on it !!! - Original Message - From: Ben Bullock [EMAIL PROTECTED] To: ADSM-L@VM.MARIST.EDU Sent: Monday, March 14, 2005 4:20 PM Subject: Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
AW: If we all complain, do you think they will add the WEB gui back?
This is the list of fixed APARS in 5.3.0.2 - the WEB GUI isn't mentioned: *** *APARs fixed in service level 5.3.0.2 * *** |IC44159 |QUERY STGPOOL REPORTS INVALID ESTIMATED CAPACITY AND |PK6 5698ISMVS 520 | |PK7 5698ISMVS 530 | |PK8 5698ISMVS 520 | |IC44159 5698ISMSV 52A | |IC44367 |TSM 5.3 UPGRADE FAILS WITH ANRD ADMINIT.C(3628): |IC44367 5698ISMSV 53A | |IC44818 |TSM SERVER MAY CRASH OR INVALIDLY ISSUE DATA INTEGRITY |IC44818 5698ISMSV 52A | *** *APARs fixed in service level 5.3.0.1 * *** |IC43118 |ANR8972E UPDATE PATH: Unable to find the element number for |IC43118 5698ISMSV 52A | |IC43738 |DEFINING A DRIVE PATH FAILS WITH: ANR8972E UNABLE TO FIND THE |IC43738 5698ISMSV 52W | |IC43796 |TSM SERVER MANAGES LAST INACTIVE VERSION BY RETEXTRA AFTER |PQ98755 5698ISMVS 520 | |IC43796 5698ISMSV 52A | |IC43954 |ANR8447E CAUSED BY INCORRECT LTO BARCODE LABEL HANDLING |IC43954 5698ISMSV 52A | |IC43019 |ANR9771E: Volume volName has incorrect barcode label... seen |IC43019 5698ISMSV 52A | |IC43118 |ANR8972E UPDATE PATH: Unable to find the element number for |IC43118 5698ISMSV 52A | |IC43729 |The TSM Linux Pass-through Driver failed leot case on 64bit |IC43729 5698ISMSV 53L | |IC43738 |DEFINING A DRIVE PATH FAILS WITH: ANR8972E UNABLE TO FIND THE |IC43738 5698ISMSV 52W | |IC43796 |TSM SERVER MANAGES LAST INACTIVE VERSION BY RETEXTRA AFTER |PQ98755 5698ISMVS 520 | |IC43796 5698ISMSV 52A | |IC43954 |ANR8447E CAUSED BY INCORRECT LTO BARCODE LABEL HANDLING |IC43954 5698ISMSV 52A | |IC44148 |DB/LOG Copy vols not added by spacetrigger |IC44148 5698ISMSV 53W | |IC44164 |Crash is seen after an INTERNAL SERVER ERROR. |IC44164 5698ISMSV 53S | Thomas Rupp
Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
We have a setup like: 2 x Windows 2003, 5.2.2.4 Server Clients backup to local DISK stgpool, this migrates to a Sequential File Disk Pool with a maxsize and this migrates to tape (3584 LTO1 and LTO2). We don't collocate the sequential file pool. But we really never migrate from sequential file to tape. The idea here is we configure maxsize so that all small files will be on disk - only large will be on tape. This takes a bit to configure the size based on the amount of storage you have but to me it is the best situation. If you don't have enough disk to keep all files, only store large ones on tape. On one of our servers the maxsize is set at 2GB and we currently have 186 files occupying 705 GB on tape. From Occupancy table: STGPOOL_NAMENUM_FILES PHYSICAL_MB LOGICAL_MB BACKUP-DISK Total 17,156,200 4,161,023 4,147,439 BACKUP-LTO Total186 705,469 705,469 With this setup everything flies, mutli-session client restores, migration to seq file disk, onsite plus offsite reclamation. Tim Rushforth City of Winnipeg -Original Message- From: PAC Brion Arnaud [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 9:59 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool Steve, I agree with you : collocation on this kind of storage pool doesn't seem to make much sense. However I've read something in IBM's Administrator guide for 5.2 (ref GC32-0768-01), page 206~207, that made me doubt. It states : If you decide to migrate data from one sequential access storage pool to another, ensure that: - Collocation is set the same in both storage pools. For example, if collocation is set to yes in the first storage pool, then collocation should be set to yes in the next storage pool. As I saw you where migrating data to a collocated tape-based pool, I was curious ! 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 Steve Bennett Sent: Monday, 14 March, 2005 16:34 To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool I did not collocate the sata diskpool. I believe that collocation would restict the use of multi process restores, might cause more disk partition fragmentation and I really did not see any benefit. PAC Brion Arnaud wrote: Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... 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 Steve Bennett Sent: Friday, 11 March, 2005 23:51 To: ADSM-L@VM.MARIST.EDU Subject: SUSPECT: (MSW) review of recent addition of sata array storage pool For what it is worth I'll provide my experience with installing a sata storage pool and some observations about potential issues. Dell 2550, 1gb ram, 2 x 1ghz pentium cpus TSM server v5.2.3.2 Windows 2000 sp4 160gb local scsi disk 6.4tb (16 x 400gb) local sata array fiber attached IBM 3494 with 2 dedicated scsi attached 3590-e1a drives 600 IBM 3590J tapes about 4.6tb of TSM data stored in primary storage pools about 100gb compressed client data stored daily Our 3494 was filling up and management did not want to spend the $ to upgrade to 3592 drives and media. We added a 6.4tb fiber attached sata array which has about 5.3tb usable when configured for raid5. Clients backup daily to the local scsi diskpool and once a day we migrate that storage pool to the sata diskpool. The sata diskpool is defined to TSM as a sequential with maxscr=260, file size of 20gb, reusedelay=8 and migrdelay=33. Once a day we migrate about 2% of the sata pool to the collocated tapepool and do sata file reclamation and tapepool reclamation. We see 85 to 90gb per hour throughput when migrating from the scsi disk to the sata. Running two migration processes doesn't seem to increase the throughput so I suspect the interface or pci bus is pretty well maxed with one migration process. Sata file reclamation runs about 100gb per hour.
Re: If we all complain, do you think they will add the WEB gui back?
Eek, I've stirred up a hornets nest I thought since it was coming from my official IBM channel, it was legit, but I'm not so sure now... I've got a message into him to see if I can get some clarification... Thanks, Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Monday, March 14, 2005 8:46 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? What PTF are they talking about? 5.3.1 isn't even out yet, much less 5.3.2. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED] Ben Bullock [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 03/14/2005 09:20 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: If we all complain, do you think they will add the WEB gui back? Hmm, I just got an e-mail from my IBM rep that seems to contradict the messages you are getting. Here is the text from him: ... Since there were several customers that complained about taking the oldTSM web GUI interface away, with the PTF that bring TSM to the 5.3.2 level, you have the option to continue to use the old interface that you like better. Please download the PTF and look at the read me file because there may be some considerations such as what port to use, etc. If you have problems with the PTF, I can have a technical TSM person get in contact with you. I have yet to download it (too busy with other things), but thought I'd share... Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Joe Howell Sent: Friday, March 11, 2005 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: If we all complain, do you think they will add the WEB gui back? OK - I just got back from a user group meeting where there was a briefing by IBM on ITSM 5.3. They stated categorically that under NO circumstances would the web interface be coming back. ITSM 5.3.1 is supposed to be out in April and will be largely made up of fixes to the Admin Console and the ISC. ### Joe Howell Shelter Insurance Companies (573)-214-6534 [EMAIL PROTECTED]
Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
Can I ask a silly question: Why do you back up to a disk pool, then migrate to a disk pool. It seems like it might be more efficient to back up to the SATA pool directly. Thanks! Rushforth, Tim wrote: We have a setup like: 2 x Windows 2003, 5.2.2.4 Server Clients backup to local DISK stgpool, this migrates to a Sequential File Disk Pool with a maxsize and this migrates to tape (3584 LTO1 and LTO2). We don't collocate the sequential file pool. But we really never migrate from sequential file to tape. The idea here is we configure maxsize so that all small files will be on disk - only large will be on tape. This takes a bit to configure the size based on the amount of storage you have but to me it is the best situation. If you don't have enough disk to keep all files, only store large ones on tape. On one of our servers the maxsize is set at 2GB and we currently have 186 files occupying 705 GB on tape. From Occupancy table: STGPOOL_NAMENUM_FILES PHYSICAL_MB LOGICAL_MB BACKUP-DISK Total 17,156,200 4,161,023 4,147,439 BACKUP-LTO Total186 705,469 705,469 With this setup everything flies, mutli-session client restores, migration to seq file disk, onsite plus offsite reclamation. Tim Rushforth City of Winnipeg -Original Message- From: PAC Brion Arnaud [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 9:59 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool Steve, I agree with you : collocation on this kind of storage pool doesn't seem to make much sense. However I've read something in IBM's Administrator guide for 5.2 (ref GC32-0768-01), page 206~207, that made me doubt. It states : If you decide to migrate data from one sequential access storage pool to another, ensure that: - Collocation is set the same in both storage pools. For example, if collocation is set to yes in the first storage pool, then collocation should be set to yes in the next storage pool. As I saw you where migrating data to a collocated tape-based pool, I was curious ! 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 Steve Bennett Sent: Monday, 14 March, 2005 16:34 To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool I did not collocate the sata diskpool. I believe that collocation would restict the use of multi process restores, might cause more disk partition fragmentation and I really did not see any benefit. PAC Brion Arnaud wrote: Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... 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 Steve Bennett Sent: Friday, 11 March, 2005 23:51 To: ADSM-L@VM.MARIST.EDU Subject: SUSPECT: (MSW) review of recent addition of sata array storage pool For what it is worth I'll provide my experience with installing a sata storage pool and some observations about potential issues. Dell 2550, 1gb ram, 2 x 1ghz pentium cpus TSM server v5.2.3.2 Windows 2000 sp4 160gb local scsi disk 6.4tb (16 x 400gb) local sata array fiber attached IBM 3494 with 2 dedicated scsi attached 3590-e1a drives 600 IBM 3590J tapes about 4.6tb of TSM data stored in primary storage pools about 100gb compressed client data stored daily Our 3494 was filling up and management did not want to spend the $ to upgrade to 3592 drives and media. We added a 6.4tb fiber attached sata array which has about 5.3tb usable when configured for raid5. Clients backup daily to the local scsi diskpool and once a day we migrate that storage pool to the sata diskpool. The sata diskpool is defined to TSM as a sequential with maxscr=260, file size of 20gb, reusedelay=8 and migrdelay=33. Once a day we migrate about 2% of the sata pool to the collocated tapepool and do sata file reclamation and tapepool reclamation. We see 85 to 90gb per hour throughput when migrating from the scsi disk to the sata. Running two migration processes doesn't seem to increase the throughput so I suspect the
test
test
unsucccessful DR test
We were unsuccessful over the weekend in the attempt to test a DR senario where an alternate TSM assumes responsibility for restoring data. The problem seems to be related to TSM losing information when some devconfig info for the 3494 tape libraries are deleted and recreated in order to adjust the information to correct it for how it appears on the DR server. Our production environment has 1 TSM server on AIX, and 2 3494 tape libraries in alternate sites. TSM accesses the drives via fibre and switches. Production: Library Name Drive Name Device Type On-Line --- --- IBM3494A 3590RMT0 3590Yes local IBM3494A 3590RMT1 3590Yes local IBM3494B 3590RMT2 3590Yes alternate site IBM3494B 3590RMT3 3590Yes alternate site IBM3494B 3590RMT4 3590Yes alternate site IBM3494B 3590RMT5 3590Yes alternate site lmcp0 Available LAN/TTY Library Management Control Point lmcp1 Available LAN/TTY Library Management Control Point rmt0 Available 1H-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt1 Available 1H-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt2 Available 1Z-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt3 Available 1Z-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt4 Available 1Z-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt5 Available 1Z-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) rmt6 Defined 1Z-08-01 IBM 3590 Tape Drive and Medium Changer (FCP) /* Device Configuration */ DEFINE DEVCLASS 3494ATAPE DEVTYPE=3590 FORMAT=DRIVE ESTCAPACITY=41943040K MOUNTL IMIT=2 MOUNTWAIT=60 MOUNTRETENTION=5 PREFIX=ADSMVOL LIBRARY=IBM3494A DEFINE DEVCLASS 3494BTAPE DEVTYPE=3590 FORMAT=DRIVE ESTCAPACITY=41943040K MOUNTL IMIT=DRIVES MOUNTWAIT=60 MOUNTRETENTION=5 PREFIX=ADSMVOL LIBRARY=IBM3494B DEFINE DEVCLASS ADSMDBBK DEVTYPE=FILE FORMAT=DRIVE MAXCAPACITY=9437184K MOUNTLIM IT=1 DIRECTORY=/var/tsmdbbk SHARED=NO SET SERVERNAME BACKUP DEFINE LIBRARY IBM3494A LIBTYPE=349X PRIVATECATEGORY=300 SCRATCHCATEGORY=302 SHA RED=NO DEFINE LIBRARY IBM3494B LIBTYPE=349X PRIVATECATEGORY=300 SCRATCHCATEGORY=302 SHA RED=NO DEFINE DRIVE IBM3494A 3590RMT0 ONLINE=Yes WWN=500507634DAF SERIAL=000 23294 DEFINE DRIVE IBM3494A 3590RMT1 ONLINE=Yes WWN=500507634D4A SERIAL=000 80797 DEFINE DRIVE IBM3494B 3590RMT2 ONLINE=Yes WWN=500507632D9B SERIAL=000 C0629 DEFINE DRIVE IBM3494B 3590RMT3 ONLINE=Yes WWN=500507633904 SERIAL=000 C0614 DEFINE DRIVE IBM3494B 3590RMT4 ONLINE=Yes WWN=5005076355D5 SERIAL=000 F4915 DEFINE DRIVE IBM3494B 3590RMT5 ONLINE=Yes WWN=500507632839 SERIAL=000 F4908 DEFINE PATH BACKUP IBM3494A SRCTYPE=SERVER DESTTYPE=LIBRARY DEVICE=/dev/lmcp0 ON LINE=YES DEFINE PATH BACKUP 3590RMT0 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494A DEVIC E=/dev/rmt0 ONLINE=YES DEFINE PATH BACKUP 3590RMT1 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494A DEVIC E=/dev/rmt1 ONLINE=YES DEFINE PATH BACKUP IBM3494B SRCTYPE=SERVER DESTTYPE=LIBRARY DEVICE=/dev/lmcp1 ON LINE=YES DEFINE PATH BACKUP 3590RMT2 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt2 ONLINE=YES DEFINE PATH BACKUP 3590RMT3 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt3 ONLINE=YES DEFINE PATH BACKUP 3590RMT4 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt4 ONLINE=YES DEFINE PATH BACKUP 3590RMT5 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt5 ONLINE=YES * The emergency DR TSM was expected to restore data from the tape library housing the primary storage pools. DR site and the device have different rmt0 from the production server. So dev info for the drives and libraries had to be deleted and redefined. lmcp1 Available LAN/TTY Library Management Control Point rmt0 Available 1Z-08-02 IBM 3590 Tape Drive and Medium Changer (FCP) rmt1 Available 1Z-08-02 IBM 3590 Tape Drive and Medium Changer (FCP) rmt2 Available 1Z-08-02 IBM 3590 Tape Drive and Medium Changer (FCP) rmt3 Available 1Z-08-02 IBM 3590 Tape Drive and Medium Changer (FCP) The consequence is that although the library showed on TSM, the info about where the vols resides was lost. a - q vol returned info about a volume - a q content vol returned info about a volume but: - q libvol showed empty rows for the library. Anyone have insight, experience, or suggestions related to this problem? I am going to try changing the logical names on the devices so they match the production server and should not require and deletions and redefginitions of devices. Larry Clark
Re: SUSPECT: (MSW) review of recent addition of sata array storage pool
We want clients to backup directly to disk (no tape). All of our backups do not end up on the sequential disk pool (it's actually not SATA, but FC SCSI). Some end up on tape as mentioned. If we went directly to the sequential disk pool, files bigger than 2GB would go to tape from the client - we don't want that. This would also mean that the stgpool backups would require tape mounts. I've seen an APAR (which I've mentioned on the list before) about some issues backing up directly to sequential files - another reason. I agree it seems like extra work but the migration from internal disk to our sequential file pool is currently taking about 1.5 hours to do 282 GB of compressed client data - not really an issue in our environment. One other thing this buys us is added redundancy - if something went wrong with the sequential file pool we could still run a day of backups (and we have some backups running hourly throughout the day). One thing I don't like about this setup is clients backup to disk overnight and we currently don't migrate this pool until 16:30 the next day (so all files will be on disk during the day) - but this means we don't get multisession restore from those backups on the DISK pool. This is a work in progress so I'm open to suggestions ... Thanks, Tim Rushforth City of Winnipeg -Original Message- From: Andy Carlson [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 10:28 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool Can I ask a silly question: Why do you back up to a disk pool, then migrate to a disk pool. It seems like it might be more efficient to back up to the SATA pool directly. Thanks! Rushforth, Tim wrote: We have a setup like: 2 x Windows 2003, 5.2.2.4 Server Clients backup to local DISK stgpool, this migrates to a Sequential File Disk Pool with a maxsize and this migrates to tape (3584 LTO1 and LTO2). We don't collocate the sequential file pool. But we really never migrate from sequential file to tape. The idea here is we configure maxsize so that all small files will be on disk - only large will be on tape. This takes a bit to configure the size based on the amount of storage you have but to me it is the best situation. If you don't have enough disk to keep all files, only store large ones on tape. On one of our servers the maxsize is set at 2GB and we currently have 186 files occupying 705 GB on tape. From Occupancy table: STGPOOL_NAME NUM_FILES PHYSICAL_MB LOGICAL_MB BACKUP-DISK Total 17,156,200 4,161,023 4,147,439 BACKUP-LTO Total 186 705,469 705,469 With this setup everything flies, mutli-session client restores, migration to seq file disk, onsite plus offsite reclamation. Tim Rushforth City of Winnipeg -Original Message- From: PAC Brion Arnaud [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 9:59 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool Steve, I agree with you : collocation on this kind of storage pool doesn't seem to make much sense. However I've read something in IBM's Administrator guide for 5.2 (ref GC32-0768-01), page 206~207, that made me doubt. It states : If you decide to migrate data from one sequential access storage pool to another, ensure that: - Collocation is set the same in both storage pools. For example, if collocation is set to yes in the first storage pool, then collocation should be set to yes in the next storage pool. As I saw you where migrating data to a collocated tape-based pool, I was curious ! 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 Steve Bennett Sent: Monday, 14 March, 2005 16:34 To: ADSM-L@VM.MARIST.EDU Subject: Re: SUSPECT: (MSW) review of recent addition of sata array storage pool I did not collocate the sata diskpool. I believe that collocation would restict the use of multi process restores, might cause more disk partition fragmentation and I really did not see any benefit. PAC Brion Arnaud wrote: Hi Steve, Just wanted to ask if your sata disk pool was collocated. We will soon implement the same kind of setup, and I'm still wondering if collocation using sequential type disks makes sense ... Cheers. Arnaud ** ** ** Panalpina Management Ltd., Basle, Switzerland, CIT Department Viadukstrasse 42, P.O. Box 4002 Basel/CH Phone: +41 (61) 226
Re: If we all complain, do you think they will add the WEB gui back?
ARGH... My apologies to all for getting everyone's hopes up and then dashing them on the rocks of reality. I just got a reply from my misinformed IBM rep: * Ben, Alex , one of our technical TSM folks, had previously been told that the Web GUI would be an option in 5.3.2, but has just sent me an email indicating that this is not the case any longer. His response is below: Joe is right Rich I thought they were going to put it back in patch 5.3.02 and the next APAR 5.3.1 but I researched and they have changed directions. Sorry for the bad info. Alex: Can you please comment / reply to Ben's issues below or make any recommendations to him? Thanks, * Sorry for believing my IBM rep and not checking before I sent it out to the group. My bad, no biscuit for me Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Rupp Sent: Monday, March 14, 2005 9:09 AM To: ADSM-L@VM.MARIST.EDU Subject: AW: If we all complain, do you think they will add the WEB gui back? This is the list of fixed APARS in 5.3.0.2 - the WEB GUI isn't mentioned: *** *APARs fixed in service level 5.3.0.2 * *** |IC44159 |QUERY STGPOOL REPORTS INVALID ESTIMATED CAPACITY AND |PK6 5698ISMVS 520 | |PK7 5698ISMVS 530 | |PK8 5698ISMVS 520 | |IC44159 5698ISMSV 52A | |IC44367 |TSM 5.3 UPGRADE FAILS WITH ANRD ADMINIT.C(3628): |IC44367 5698ISMSV 53A | |IC44818 |TSM SERVER MAY CRASH OR INVALIDLY ISSUE DATA INTEGRITY |IC44818 5698ISMSV 52A | *** *APARs fixed in service level 5.3.0.1 * *** |IC43118 |ANR8972E UPDATE PATH: Unable to find the element number for |IC43118 5698ISMSV 52A | |IC43738 |DEFINING A DRIVE PATH FAILS WITH: ANR8972E UNABLE TO FIND THE |IC43738 5698ISMSV 52W | |IC43796 |TSM SERVER MANAGES LAST INACTIVE VERSION BY RETEXTRA AFTER |PQ98755 5698ISMVS 520 | |IC43796 5698ISMSV 52A | |IC43954 |ANR8447E CAUSED BY INCORRECT LTO BARCODE LABEL HANDLING |IC43954 5698ISMSV 52A | |IC43019 |ANR9771E: Volume volName has incorrect barcode label... seen |IC43019 5698ISMSV 52A | |IC43118 |ANR8972E UPDATE PATH: Unable to find the element number for |IC43118 5698ISMSV 52A | |IC43729 |The TSM Linux Pass-through Driver failed leot case on 64bit |IC43729 5698ISMSV 53L | |IC43738 |DEFINING A DRIVE PATH FAILS WITH: ANR8972E UNABLE TO FIND THE |IC43738 5698ISMSV 52W | |IC43796 |TSM SERVER MANAGES LAST INACTIVE VERSION BY RETEXTRA AFTER |PQ98755 5698ISMVS 520 | |IC43796 5698ISMSV 52A | |IC43954 |ANR8447E CAUSED BY INCORRECT LTO BARCODE LABEL HANDLING |IC43954 5698ISMSV 52A | |IC44148 |DB/LOG Copy vols not added by spacetrigger |IC44148 5698ISMSV 53W | |IC44164 |Crash is seen after an INTERNAL SERVER ERROR. |IC44164 5698ISMSV 53S | Thomas Rupp
Re: SkipNTsecurityCRC doesn't seem to work as I'd expect
If you want to omit security changes from the criteria used to determine whether files have changed, then you need to use SKIPNTPERMISSIONS YES If this is not working for you, then ensure that you've updated the correct options file; verify that you've restarted the scheduler or client after you made the change; and check that the option is not being overridden from a client option set on the server. For journal-based backup, a good place to start is the JBB FAQ at URL: http://www-1.ibm.com/support/docview.wss?uid=swg21155524 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 05:31:12: First the problem: A Windows fileserver with about 560GB data in 13.500.000 files. (We also have one with 1,1TB data.) Always done (full/standard) incrementals. Consider a security update (local group addition; no discussion about that). A normal incremental backup treats me next run with a full backup of all. (Server v5.2.4 on W2K-sp4, Client v5.2.3.11 on W2K-sp4) Considerations: - I don't really need or want security on file level: via inheritance the files get the right rights from the directory - A dirsonly backup takes as long as a full incremental (so it seems, and because also file info is sent, doesn't save memory; And about an extra filesonly backup...) - A full incremental takes about a whole night - An incremental with option skipntpermissions treats me with a full backup (that I don't like) next run - An incremental with option skipntsecuritycrc treats me with a full backup (that I don't like) next run(is this a bug or something?) - I don't like full backups at all (except for databases; that why we chose TSM didn't we?) - Journal based backups also need a full incremental now and then Solutions: ? Ton Mol Corus IJmuiden Information Services
Re: unsucccessful DR test
== On Mon, 14 Mar 2005 11:47:50 -0500, Lawrence Clark [EMAIL PROTECTED] said: The consequence is that although the library showed on TSM, the info about where the vols resides was lost. a - q vol returned info about a volume - a q content vol returned info about a volume but: - q libvol showed empty rows for the library. Anyone have insight, experience, or suggestions related to this problem? What happened when you did the CHECKIN LIBVOLUMES? What category were the volumes in from the 3494? If you set the categories manually, it would still be the case that the TSM server might not know they are there. I'm not sure wether an 'AUDIT LIBRARY' or a 'CHECKIN LIBVOLUMES' would do this job faster. It's really an all-at-once alert the TSM server that it's got a bunch of new volumes thing... - Allen S. Rout
Solaris server won't backup
Hello All! I am having issues with a solaris server: Node Name: FJSU000 Platform: SUN SOLARIS Client OS Level: 5.9 Client Version: Version 5, Release 2, Level 2.0 Policy Domain Name: SOLARIS Last Access Date/Time: 03/14/05 12:50:13 that is trying to backup to my TSM 5.2.2.5 server on AIX 5.2. I do not see any errors within the activity logs other than that it has missed it's scheduled startup window. On the client side, there are no error messages either. We have recycled the scheduler and then tried a dsmc i incremental backup from the client command line, but it just sits there. I query sessions on the TSM server and it has a status of idlew which means it's waiting for the client to communicate more info. to the TSM server. Any suggestions on what I can check to see why the server isn't backup up? Thank you in advance! Joni Moyer Highmark Storage Systems Work:(717)302-6603 Fax:(717)302-5974 [EMAIL PROTECTED]
Re: unsucccessful DR test
On Mar 14, 2005, at 11:47 AM, Lawrence Clark wrote: The consequence is that although the library showed on TSM, the info about where the vols resides was lost. a - q vol returned info about a volume - a q content vol returned info about a volume but: - q libvol showed empty rows for the library. In a library containing a Library Manager, such as the 3494, TSM does not need to know (and does not know) where tapes reside within the library. As Allen suggests, the information needed in the problem definition is how the volumes were introduced to the library. Don't forget to physically verify library and content access with the mtlib command before presuming that TSM can get at it. Richard Sims
Solicitation of Tomatoes. (database unload procedure)
Greetings, all. I'm working on a procedure for database unload and reload, including sketching a bunch of caveats and look-for-this type advice, and reference to the fragmentation estimate query discussed here some time ago. I'd appreciate it if those of you who are interested take a glance at my instructions, and throw tomatoes if you care to, public or private. I'd also appreciate some experience numbers: How long it took, how much space before and after. I've got a test version of our website, with the postulated procedure, at: http://nersp.nerdc.ufl.edu/~asr/osg-test/services/NSAM/maint_docs/db_un_reload.html - Allen S. Rout
Re: Solaris server won't backup
Joni - A classic cause of this situation is that the previously scheduled backup was still running (stuck) when the next one tried to start. In your Query Session, the key here is in the numbers: if the Bytes Sent is large, it indicates that the server has sent the client its inventory of Active files and is waiting for the client to sort and start processing it as the client then starts traversing the file system. If no data is coming back from the client, then it suggests a problem either with general (memory) resources on the client, or issues in the file system. In such a case, I would, instead of just a 'dsmc i', do a dsmc i on a small file system, to start isolating the problem. If that works, escalate to larger file systems - and suspect the one that's first in the Domain list. You indicate no errors in the dsmerror.log. See if anything in the Solaris /var/log/messages. Also exercise other client functions such as 'dsmc q fi' and 'dsmc query backup ...' and 'dsmc archive SomeFile' to further isolate the problem. Richard Sims On Mar 14, 2005, at 1:23 PM, Joni Moyer wrote: Hello All! I am having issues with a solaris server: Node Name: FJSU000 Platform: SUN SOLARIS Client OS Level: 5.9 Client Version: Version 5, Release 2, Level 2.0 Policy Domain Name: SOLARIS Last Access Date/Time: 03/14/05 12:50:13 that is trying to backup to my TSM 5.2.2.5 server on AIX 5.2. I do not see any errors within the activity logs other than that it has missed it's scheduled startup window. On the client side, there are no error messages either. We have recycled the scheduler and then tried a dsmc i incremental backup from the client command line, but it just sits there. I query sessions on the TSM server and it has a status of idlew which means it's waiting for the client to communicate more info. to the TSM server. Any suggestions on what I can check to see why the server isn't backup up? Thank you in advance!
Image Backups and SANs
TSM server 5.2.2.5 on AIX 5.2 BAClient 5.2.2.0 on Windows 2003 I noticed the following in the help regarding BACKUP IMAGE: Image backup of IBM TotalStorage SAN File Systems is not supported. ... and in the Backup/Archive Installation manual, 5.2.2, Technical changes for version 5.2.2 - Dec 2003 section Backup and restore support for IBM TotalStorage SAN File systems Backup and restore for IBM TotalStorage SAN File Systems is supported on the Windows 2000 client. See Supported files systems ... and that section also indicates IBM TotalStorage SAN File System is supported on Windows 2000 only. Is data on an IBM Shark considered on an IBM TotalStorage SAN File System? Does the support also extend to Windows 2003? Thanks in advance
Re: unsucccessful DR test
The consequence is that although the library showed on TSM, the info about where the vols resides was lost. a - q vol returned info about a volume - a q content vol returned info about a volume but: - q libvol showed empty rows for the library. Anyone have insight, experience, or suggestions related to this problem? What happened when you did the CHECKIN LIBVOLUMES? What category were the volumes in from the 3494? The same definition was retained for the DR server I was trying to use. (see below). Initially a 'q libvol' would show volumes for the library, then they would disappear. I'm not certain but it may be at the point the lib defs were deleted, then redefined. If you set the categories manually, it would still be the case that the TSM server might not know they are there. I'm not certain what you mean by this I'm not sure wether an 'AUDIT LIBRARY' or a 'CHECKIN LIBVOLUMES' would do this job faster. It's really an all-at-once alert the TSM server that it's got a bunch of new volumes thing... I tried an audit library and a checkin libv but it did not add volumes I've decided to logically rename the devices for the library so they match the devices as they are on the production server, e.g., rmt2. rmt3,rmt4,rmt5 for libB. Then there should not be a need to delete and redefine the drives since they would match the existing ones in the DB when the server comes up. - Allen S. Rout [tsmserv2] /home/root/tsmfiles # more devconfig.info /* Device Configuration */ DEFINE DEVCLASS 3494ATAPE DEVTYPE=3590 FORMAT=DRIVE ESTCAPACITY=41943040K MOUNTL IMIT=2 MOUNTWAIT=60 MOUNTRETENTION=5 PREFIX=ADSMVOL LIBRARY=IBM3494A DEFINE DEVCLASS 3494BTAPE DEVTYPE=3590 FORMAT=DRIVE ESTCAPACITY=41943040K MOUNTL IMIT=DRIVES MOUNTWAIT=60 MOUNTRETENTION=5 PREFIX=ADSMVOL LIBRARY=IBM3494B DEFINE DEVCLASS ADSMDBBK DEVTYPE=FILE FORMAT=DRIVE MAXCAPACITY=9437184K MOUNTLIM IT=1 DIRECTORY=/var/tsmdbbk SHARED=NO SET SERVERNAME BACKUP DEFINE LIBRARY IBM3494A LIBTYPE=349X PRIVATECATEGORY=300 SCRATCHCATEGORY=302 SHA RED=NO DEFINE LIBRARY IBM3494B LIBTYPE=349X PRIVATECATEGORY=300 SCRATCHCATEGORY=302 SHA RED=NO DEFINE DRIVE IBM3494B 3590RMT0 ONLINE=Yes WWN=500507632D9B SERIAL=000 C0629 DEFINE DRIVE IBM3494B 3590RMT1 ONLINE=Yes WWN=500507633904 SERIAL=000 C0614 DEFINE DRIVE IBM3494B 3590RMT2 ONLINE=Yes DEFINE DRIVE IBM3494B 3590RMT3 ONLINE=Yes DEFINE PATH BACKUP IBM3494A SRCTYPE=SERVER DESTTYPE=LIBRARY DEVICE=/dev/lmcp0 ON LINE=YES DEFINE PATH BACKUP IBM3494B SRCTYPE=SERVER DESTTYPE=LIBRARY DEVICE=/dev/lmcp1 ON LINE=YES DEFINE PATH BACKUP 3590RMT0 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt0 ONLINE=YES DEFINE PATH BACKUP 3590RMT1 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=IBM3494B DEVIC E=/dev/rmt1 ONLINE=YES
Re: unsucccessful DR test
[EMAIL PROTECTED] 03/14/2005 1:31:49 PM On Mar 14, 2005, at 11:47 AM, Lawrence Clark wrote: The consequence is that although the library showed on TSM, the info about where the vols resides was lost. a - q vol returned info about a volume - a q content vol returned info about a volume but: - q libvol showed empty rows for the library. In a library containing a Library Manager, such as the 3494, TSM does not need to know (and does not know) where tapes reside within the library. Is this true? It does seem that when the contents of a Q LIBVOL came up emnpty was the point that TSM could not mount a volume. I mean, if you have 2 libraries defined, how does TSM know which library to send the command to for a specific volume? As Allen suggests, the information needed in the problem definition is how the volumes were introduced to the library. Don't forget to physically verify library and content access with the mtlib command before presuming that TSM can get at it. The volumes were not 'introduced' into the library. They were there and that library was disconnected from the production TSM server at the switch, and connected to the DR TSM server at the switch. Richard Sims
Error Allocating device !!!!!
Hi,
I have a problem ,after migrate to TSM Client 5.3 for Linux. When i try to
backup data base for Oracle.
First i Created a Link with this command:
$ ln -s /usr/lib/libobk.so $ORACLE_HOME/lib/libobk.so
i send the backup and then come out this error:
RMAN run
2 {
3 allocate channel t1 type 'sbt_tape' parms
4 'ENV=(TDPO_OPTFILE=/opt/tivoli/tsm/client/oracle/bin/tdpo.opt)';
5 backup
6 filesperset 15
7 format 'df_%t_%s_%p'
8 (database);
9 release channel t1;
10 }
11
RMAN-00571: ===
RMAN-00569: === ERROR MESSAGE STACK FOLLOWS ===
RMAN-00571: ===
RMAN-03009: failure of allocate command on t1 channel at 03/12/2005 11:38:35
ORA-19554: error allocating device, device type: SBT_TAPE, device name:
ORA-27000: skgfqsbi: failed to initialize storage subsystem (SBT) layer
Linux Error: 106: El otro extremo ya estaonectado
Additional information: 7011
ORA-19511: Error received from media manager layer, error text:
SBT error = 7011, errno = 106, sbtopen: system error
Recovery Manager complete.
I don4t know if exist TDP for Linux 5.3 , i have looked for an install but i
didn4t find it.
Would you help me pls
Eduardo Rivero
_
MSN Amor: busca tu = naranja http://latam.msn.com/amor/
Re: unsucccessful DR test
On Mar 14, 2005, at 3:02 PM, Lawrence Clark wrote: In a library containing a Library Manager, such as the 3494, TSM does not need to know (and does not know) where tapes reside within the library. Is this true? It does seem that when the contents of a Q LIBVOL came up emnpty was the point that TSM could not mount a volume. I mean, if you have 2 libraries defined, how does TSM know which library to send the command to for a specific volume? Yes, it's true. In a library with a Library Manager, such as the 3494, TSM does not mount the tape - the library does, per TSM simply asking the library to mount volume 001234. TSM knows which of your library definitions to send its commands to by virtue of your TSM definitions: you defined a Library instance, and then the physical Path to the library, where the Path specified a unique /dev/lmcp_, which goes through the lmcpd and the /etc/ibmatl.conf to reach the library. It is vital in initial phases of library use, such as your DR test, that you employ the mtlib command to verify these basics. A missing or faulty atl conf file, or lmcpd functionality issue, will keep the arrangement from working. This would be reflected in your AIX Error Log. It is also the case that the 3494 needs its authorization list updated to allow access by each host system in which TSM runs. And in some cases, the 3494 may be on a non-routed subnet, which means that the TSM system must also have an ethernet card on that same subnet. (Another path is via RS-232 rather than TCP/IP, but few people bother with that any more.) I'd recommend running tape inventory lists via the mtlib command to verify that the inventory can be seen from the DR system, and that the Category Codes reported match your TSM library definitions. If all looks good, try mounting a volume on a tape drive via the mtlib command and see if that works. In AIX, be sure that the tape drives are in an Available rather than Defined state. References: IBM TotalStorage Tape Device Drivers: Installation and User's Guide, (GC35-0154) (a renaming of the earlier manual IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers: Installation and User's Guide, of the same publication number) IBM TotalStorage Tape Device Drivers: Programming Reference (GC35-0346) (a renaming of the earlier manual IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers: Programming Reference (WB2107)) Available at ftp://ftp.storsys.ibm.com/devdrvr/Doc/ (refer to the .message or README file in that directory) or ftp://ftp.software.ibm.com/storage/devdrvr/Doc/ as files: IBM_TotalStorage_tape_IUG.ps or IBM_TotalStorage_tape_IUG.pdf IBM_TotalStorage_tape_PROGREF.ps or IBM_TotalStorage_tape_PROGREF.pdf Richard Sims
Re: unsucccessful DR test
Yes, we have two libraries We only defined one library to the DR server for the switch, library it could access. The problem was the incorrect I.D. from the ibmatl.conf was selected for the lmcp definition to AIX. It because available because it was acccessible from the lan. The drives became available because they were accessible from the switch. So the command was going to one library where the volume was not, the drives were not, and TSM did not know anything of it contents. It was confusing to see the volumes in the other library still showing with a 'q libvol', but the ones I was trying to access not. Now I have an explanation. [EMAIL PROTECTED] 03/14/2005 3:21:42 PM On Mar 14, 2005, at 3:02 PM, Lawrence Clark wrote: In a library containing a Library Manager, such as the 3494, TSM does not need to know (and does not know) where tapes reside within the library. Is this true? It does seem that when the contents of a Q LIBVOL came up emnpty was the point that TSM could not mount a volume. I mean, if you have 2 libraries defined, how does TSM know which library to send the command to for a specific volume? Yes, it's true. In a library with a Library Manager, such as the 3494, TSM does not mount the tape - the library does, per TSM simply asking the library to mount volume 001234. TSM knows which of your library definitions to send its commands to by virtue of your TSM definitions: you defined a Library instance, and then the physical Path to the library, where the Path specified a unique /dev/lmcp_, which goes through the lmcpd and the /etc/ibmatl.conf to reach the library. It is vital in initial phases of library use, such as your DR test, that you employ the mtlib command to verify these basics. A missing or faulty atl conf file, or lmcpd functionality issue, will keep the arrangement from working. This would be reflected in your AIX Error Log. It is also the case that the 3494 needs its authorization list updated to allow access by each host system in which TSM runs. And in some cases, the 3494 may be on a non-routed subnet, which means that the TSM system must also have an ethernet card on that same subnet. (Another path is via RS-232 rather than TCP/IP, but few people bother with that any more.) I'd recommend running tape inventory lists via the mtlib command to verify that the inventory can be seen from the DR system, and that the Category Codes reported match your TSM library definitions. If all looks good, try mounting a volume on a tape drive via the mtlib command and see if that works. In AIX, be sure that the tape drives are in an Available rather than Defined state. References: IBM TotalStorage Tape Device Drivers: Installation and User's Guide, (GC35-0154) (a renaming of the earlier manual IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers: Installation and User's Guide, of the same publication number) IBM TotalStorage Tape Device Drivers: Programming Reference (GC35-0346) (a renaming of the earlier manual IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers: Programming Reference (WB2107)) Available at ftp://ftp.storsys.ibm.com/devdrvr/Doc/ (refer to the .message or README file in that directory) or ftp://ftp.software.ibm.com/storage/devdrvr/Doc/ as files: IBM_TotalStorage_tape_IUG.ps or IBM_TotalStorage_tape_IUG.pdf IBM_TotalStorage_tape_PROGREF.ps or IBM_TotalStorage_tape_PROGREF.pdf Richard Sims
Re: unsucccessful DR test
== On Mon, 14 Mar 2005 15:21:42 -0500, Richard Sims [EMAIL PROTECTED] said: In a library with a Library Manager, such as the 3494, TSM does not mount the tape - the library does, per TSM simply asking the library to mount volume 001234. I'm going to have to dispute this. I just tested on my 3494, and if my library manager doesn't see the volume in its' Q LIBVOL, it doesn't acknowledge the existence of the tape. I did the following: 1) Restored a file from a client (and therefore determined which tape it was on) 2) CHECKOUT LIBVOLUME remove=no 3) Confirmed volume in scratch category 4) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. 5) Canceled restore, checked in, checked out again. 6) used mtlib to assign the volume to the correct 'private' category. (this would be analogous to the failed DR scenario) 7) Attempted restore. 8) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. ... Cleaned up. You may be confusing the library manager-ness of a 3494 with the library manager-ness of a TSM server. My configuration has a 3494 attached to a library-manager TSM instance (which I call 'CTRL', for controller), and then other TSM instances request mounts from the CTRL instance. When I Q libvol from CTRL, the volumes show up. When I Q libvol from a library client, it's empty. - Allen S. Rout
dsm scheduler on windows
Trying to set up an easier way to manage the configuration (dsm.opt) on windows files. One comment mentioned today is that anytime the dsm.opt changes the dsm sched service must be cycled. Is that true? I thought only the windows equivilant program to dsmc would read dsm.opt and as such it is the only think that might need cycling or tweaking of dsm.opt changes. What's the real answer? Mike
Re: unsucccessful DR test
Allen - There's really nothing to dispute... TSM asks the Library Manager to mount the tape - but, naturally, TSM has to believe that the tape is in the library in order to want to request the mount. When TSM restarts it performs the equivalent of an Audit Library, to assure that its last knowledge of the library's tape inventory is the same. If the library says that a tape isn't there, then TSM has to believe that and so deems the tape(s) unavailable. The original question involved a pondering, as I read it, as to whether TSM knows where the tapes are in the 3494. TSM doesn't get involved in that: it's up to the 3494 and its database to keep track of that. This relieves TSM of a lot of overhead, as compared with what it has to deal with in a SCSI type library. Richard Sims On Mar 14, 2005, at 4:07 PM, [EMAIL PROTECTED] wrote: == On Mon, 14 Mar 2005 15:21:42 -0500, Richard Sims [EMAIL PROTECTED] said: In a library with a Library Manager, such as the 3494, TSM does not mount the tape - the library does, per TSM simply asking the library to mount volume 001234. I'm going to have to dispute this. I just tested on my 3494, and if my library manager doesn't see the volume in its' Q LIBVOL, it doesn't acknowledge the existence of the tape. I did the following: 1) Restored a file from a client (and therefore determined which tape it was on) 2) CHECKOUT LIBVOLUME remove=no 3) Confirmed volume in scratch category 4) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. 5) Canceled restore, checked in, checked out again. 6) used mtlib to assign the volume to the correct 'private' category. (this would be analogous to the failed DR scenario) 7) Attempted restore. 8) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. ... Cleaned up. You may be confusing the library manager-ness of a 3494 with the library manager-ness of a TSM server. My configuration has a 3494 attached to a library-manager TSM instance (which I call 'CTRL', for controller), and then other TSM instances request mounts from the CTRL instance. When I Q libvol from CTRL, the volumes show up. When I Q libvol from a library client, it's empty. - Allen S. Rout
Re: dsm scheduler on windows
If you make a change in dsm.opt you have to restart the schedule service for scheduled operations to pick up the change. The schedule service reads the option file at startup. -Original Message- From: Mike [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 4:03 PM To: ADSM-L@VM.MARIST.EDU Subject: dsm scheduler on windows Trying to set up an easier way to manage the configuration (dsm.opt) on windows files. One comment mentioned today is that anytime the dsm.opt changes the dsm sched service must be cycled. Is that true? I thought only the windows equivilant program to dsmc would read dsm.opt and as such it is the only think that might need cycling or tweaking of dsm.opt changes. What's the real answer? Mike
DSMSERV FORMAT
Hi I have a TSM server 5.3.0 running on a Windows 2003 (not sure what patch level the OS is at). The sever use a router as its NTP time source. Last week the router was rebooted and its time and date defaulted to 1993 (another story). The TSM server would not run with the time set to '93 I assumed this was because TMS server's transactions are date and time stamped, and as a safety mechanism the server would not run... I wasnt too worried about it at the time. as the server is in development. No the time issue is resolved TSM still wont load' When I start the data base through dos prompt dsmserv I get the following out put. ANR4726I The ICC support module has been loaded. ANR0990I Server restart-recovery in progress. ANR0200I Recovery log assigned capacity is 100 megabytes. ANR0201I Database assigned capacity is 100 megabytes. ANR0306I Recovery log volume mount in progress. ANR0292W Database page shadow file dbpgshdw.bdt does not exist. ANR0353I Recovery log analysis pass in progress. ANR0354I Recovery log redo pass in progress. ANR0355I Recovery log undo pass in progress. ANR0352I Transaction recovery complete. ANR1635I The server machine GUID, 3a.ce.a4.71.68.e9.11.d9.a7.39.00.0e.0c.5c.43- .56, has initialized. ANRD adminit.c(1530): ThreadId0 Insufficient log space to update table Administrative.Attributes. ANRD ThreadId0 issued message from: -105642D8 outDiagf()+168 -1005EA24 admInitPart2()+13c4 I suspect the log space is full. From what I can tell I need to format some space for the log file then extend the log... simple right... This is what I get when attempting to format a log file.. :\Program Files\Tivoli\tsm\serverdsmserv format 1 c:\new_log.dsm 25mb NR0900I Processing options file C:\Program Files\Tivoli\tsm\server\dsmserv.o . NR7800I DSMSERV generated at 17:33:50 on Dec 7 2004. ivoli Storage Manager for Windows ersion 5, Release 3, Level 0.0 icensed Materials - Property of IBM C) Copyright IBM Corporation 1990, 2004. ll rights reserved. .S. Government Users Restricted Rights - Use, duplication or disclosure estricted by GSA ADP Schedule Contract with IBM Corporation. NR4726I The ICC support module has been loaded. issing format arguments. Any Ideas?? Thanks in advance
Re: DSMSERV FORMAT
Carl - You should NOT be using DSMSERV FORMAT to create additional Recovery Log space: use the DSMFMT command. You're lucky that the command failed! Refer to the Admin Guide manual topic Recovering When the Recovery Log Runs Out of Space for the steps to follow. And by all means review server configuration and maintenance topics in that manual to avoid major problems like this. Richard Sims On Mar 14, 2005, at 5:42 PM, Carl Lea wrote: ANRD adminit.c(1530): ThreadId0 Insufficient log space to update table Administrative.Attributes. ANRD ThreadId0 issued message from: -105642D8 outDiagf()+168 -1005EA24 admInitPart2()+13c4 I suspect the log space is full. From what I can tell I need to format some space for the log file then extend the log... simple right... This is what I get when attempting to format a log file.. :\Program Files\Tivoli\tsm\serverdsmserv format 1 c:\new_log.dsm 25mb NR0900I Processing options file C:\Program Files\Tivoli\tsm\server\dsmserv.o . NR7800I DSMSERV generated at 17:33:50 on Dec 7 2004. ivoli Storage Manager for Windows ersion 5, Release 3, Level 0.0 icensed Materials - Property of IBM C) Copyright IBM Corporation 1990, 2004. ll rights reserved. .S. Government Users Restricted Rights - Use, duplication or disclosure estricted by GSA ADP Schedule Contract with IBM Corporation. NR4726I The ICC support module has been loaded. issing format arguments.
Re: unsucccessful DR test
== On Mon, 14 Mar 2005 17:42:24 -0500, Richard Sims [EMAIL PROTECTED] said: There's really nothing to dispute... TSM asks the Library Manager to mount the tape - but, naturally, TSM has to believe that the tape is in the library in order to want to request the mount. When TSM restarts it performs the equivalent of an Audit Library, to assure that its last knowledge of the library's tape inventory is the same. If the library says that a tape isn't there, then TSM has to believe that and so deems the tape(s) unavailable. I'm sorry, I understood you to mean that the TSM server would ask for the volume regardless of wether it thought the volume to be checked in. Clearly, I misunderstood. - Allen S. Rout
Re: DSMSERV FORMAT
Thanks Richard for your answer howevere we are running Windows port of TSM 5.3 and the dsmfmt is not supported, we are having difficulies in getting the correct syntax of the dsmserv format and is not behaving as expected. Thanks Carl Richard Sims [EMAIL PROTECTED] wrote: Carl - You should NOT be using DSMSERV FORMAT to create additional Recovery Log space: use the DSMFMT command. You're lucky that the command failed! Refer to the Admin Guide manual topic Recovering When the Recovery Log Runs Out of Space for the steps to follow. And by all means review server configuration and maintenance topics in that manual to avoid major problems like this. Richard Sims On Mar 14, 2005, at 5:42 PM, Carl Lea wrote: ANRD adminit.c(1530): ThreadId0 Insufficient log space to update table Administrative.Attributes. ANRD ThreadId0 issued message from: -105642D8 outDiagf()+168 -1005EA24 admInitPart2()+13c4 I suspect the log space is full. From what I can tell I need to format some space for the log file then extend the log... simple right... This is what I get when attempting to format a log file.. :\Program Files\Tivoli\tsm\serverdsmserv format 1 c:\new_log.dsm 25mb NR0900I Processing options file C:\Program Files\Tivoli\tsm\server\dsmserv.o . NR7800I DSMSERV generated at 17:33:50 on Dec 7 2004. ivoli Storage Manager for Windows ersion 5, Release 3, Level 0.0 icensed Materials - Property of IBM C) Copyright IBM Corporation 1990, 2004. ll rights reserved. .S. Government Users Restricted Rights - Use, duplication or disclosure estricted by GSA ADP Schedule Contract with IBM Corporation. NR4726I The ICC support module has been loaded. issing format arguments.
Re: Anybody backing up MIRAPOINT Mail
== On Mon, 14 Mar 2005 15:14:31 -0600, fred johanson [EMAIL PROTECTED] said: Some weeks after those in charge installed a mail system from MIRAPOINT, I was asked if I (TSM) could back it up. It's a proprietary OS with proprietary filesystems, but so far we've come up with a number of off the wall possible schemes. Is there anybody out there who has accomplished such a thing? They claimed to do NDMP when they came by here. We weren't thrilled with them. :) - Allen S. Rout
Re: DSMSERV FORMAT
Thanks Richard for your answer howevere we are running Windows port of TSM 5.3 and the dsmfmt is not supported, It is most certainly supported. You will find it in the ..\TSM\console directory. You can also use the TSM Management Console to format a recovery log volume. Check your documentation, in particular the TSM Administrator's Reference, Appendix A. we are having difficulies in getting the correct syntax of the dsmserv format and is not behaving as expected. That is good, since Richard's response was absolutely correct: Assuming you still want to recover from the out of log space condition, running dsmserv format will put you in a world of hurt. Read the description for that command (also in the Admin Reference, Appendix A) and you will see that it formats not only the recovery log, but the database, too. What you want is to format a new recovery log *volume*, not the whole recovery log and database. In short, you don't want dsmserv format. The TSM Administrator's Guide has a section that discusses recovering from the out of log space condition. See the chapter on Managing the Database and Recovery Log, section Increasing the Size of the Database or Recovery Log, sub-section Recovering When the Recovery Log Runs Out of Space. 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 ADSM-L@VM.MARIST.EDU wrote on 2005-03-14 19:21:02: Thanks Richard for your answer howevere we are running Windows port of TSM 5.3 and the dsmfmt is not supported, we are having difficulies in getting the correct syntax of the dsmserv format and is not behaving as expected. Thanks Carl Richard Sims [EMAIL PROTECTED] wrote: Carl - You should NOT be using DSMSERV FORMAT to create additional Recovery Log space: use the DSMFMT command. You're lucky that the command failed! Refer to the Admin Guide manual topic Recovering When the Recovery Log Runs Out of Space for the steps to follow. And by all means review server configuration and maintenance topics in that manual to avoid major problems like this. Richard Sims On Mar 14, 2005, at 5:42 PM, Carl Lea wrote: ANRD adminit.c(1530): ThreadId0 Insufficient log space to update table Administrative.Attributes. ANRD ThreadId0 issued message from: -105642D8 outDiagf()+168 -1005EA24 admInitPart2()+13c4 I suspect the log space is full. From what I can tell I need to format some space for the log file then extend the log... simple right... This is what I get when attempting to format a log file?.. :\Program Files\Tivoli\tsm\serverdsmserv format 1 c:\new_log.dsm 25mb NR0900I Processing options file C:\Program Files\Tivoli\tsm\server\dsmserv.o . NR7800I DSMSERV generated at 17:33:50 on Dec 7 2004. ivoli Storage Manager for Windows ersion 5, Release 3, Level 0.0 icensed Materials - Property of IBM C) Copyright IBM Corporation 1990, 2004. ll rights reserved. .S. Government Users Restricted Rights - Use, duplication or disclosure estricted by GSA ADP Schedule Contract with IBM Corporation. NR4726I The ICC support module has been loaded. issing format arguments.
Anybody backing up MIRAPOINT Mail
Some weeks after those in charge installed a mail system from MIRAPOINT, I was asked if I (TSM) could back it up. It's a proprietary OS with proprietary filesystems, but so far we've come up with a number of off the wall possible schemes. Is there anybody out there who has accomplished such a thing? Fred Johanson ITSM Administrator University of Chicago 773-702-8464
Re: unsucccessful DR test
As I mentioned in the previous email, I determined the problem as being the wrong ID was used for the lmcp def. It was the ID of the other library. Doing a 'q libvol' showed the volumes of the library with the ident, but the ones from the library I was trying to use were missing. This was a bit confusing until I determined where my error was. It also suggest that table is populated from information TSM derives from the library, since it 'flushed' that info from the library that was not accessible. q vol, and q con vol still showed info on the 'flushed' volumes because that info is maintained solo by TSM and not the library manager. [EMAIL PROTECTED] 03/14/2005 4:07:06 PM == On Mon, 14 Mar 2005 15:21:42 -0500, Richard Sims [EMAIL PROTECTED] said: In a library with a Library Manager, such as the 3494, TSM does not mount the tape - the library does, per TSM simply asking the library to mount volume 001234. I'm going to have to dispute this. I just tested on my 3494, and if my library manager doesn't see the volume in its' Q LIBVOL, it doesn't acknowledge the existence of the tape. I did the following: 1) Restored a file from a client (and therefore determined which tape it was on) 2) CHECKOUT LIBVOLUME remove=no 3) Confirmed volume in scratch category 4) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. 5) Canceled restore, checked in, checked out again. 6) used mtlib to assign the volume to the correct 'private' category. (this would be analogous to the failed DR scenario) 7) Attempted restore. 8) Attempted restore. Library manager says CHECKIN LIBVOLUME required within 60 minutes. ... Cleaned up. You may be confusing the library manager-ness of a 3494 with the library manager-ness of a TSM server. My configuration has a 3494 attached to a library-manager TSM instance (which I call 'CTRL', for controller), and then other TSM instances request mounts from the CTRL instance. When I Q libvol from CTRL, the volumes show up. When I Q libvol from a library client, it's empty. - Allen S. Rout
