Re: [Veritas-bu] Disk backup - Storage Hardware configurations
we see 40-50MB/sec when doing disk to tape jobs. That is at the minimum required for LTO3. Is anyone using LTO3 for D2D2T? I'd like to understand what does and what does not work. There is some pressure to start using LTO3 as the price gap narrows, but I doubt many of our systems can deliver the data rate even off local disk. I think we will need to get some disk for staging so I too would like to know what works. William D L Brown ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Disk backup - Storage Hardware configurations
Hi Monte, I did one install with EMC DMX1000 disk with hardware mirroring used for a DSSU and I have to say I found performance wasn't great. Worse than LTO2 performance but this was writing to a fairly small number of physical drives, though performance wasn't an issue so I didn't spend a lot of time tuning it, cheers Andy. Ambrose, Monte [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 01:54 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] Disk backup - Storage Hardware configurations I wanted to get a feel for what type of disk storage is being used out there for disk based backups. I am specifically looking for SAN attached storage ATA, SATA and what type of RAID is used and how it is performing overall. The NBU 6.0 tuning guide shows RAID-5 but I am wondering if anyone is using RAID-10 or others. Thanks, Monte
[Veritas-bu] Virtual Tape Library
This email and any attachments are confidential and may contain privileged information. If you are not the person for whom they are intended please return the email and then delete all material from any computer. You must not use the email or attachments for any purpose, nor disclose its contents to anyone other than the intended recipient. Any statements made by an individual in this email do not necessarily reflect the views of the Yorkshire Building Society Group. Hi, We are currently looking at purchasing a virtual tape library ( HP vls6000 ), and i am finding conflicting information on how the virtual library is licensed, some documentation says per TB, some say per virtual drive. Has anyone any experience with Netbackup and a vtl and what is the actual situation regarding licensing. Regards Gary Yorkshire Building Society, which is authorised and regulated by the Financial Services Authority, chooses to introduce its customers to: -Legal General for the purposes of advising on and arranging life assurance and investment products bearing Legal General's name; and -Homeowners Friendly Society for the purpose of arranging stakeholder Child Trust Funds. We are entered in the FSA Register and our FSA registration number is 106085 http://www.fsa.gov.uk/register Head Office: Yorkshire Building Society, Yorkshire House, Yorkshire Drive, Bradford, BD5 8LJ Tel: 0845 1 200 100 Visit Our Website http://www.ybs.co.uk All communications with us may be monitored/recorded to improve the quality of our service and for your protection and security. This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Virtual Tape Library
The NetBackup Virtual tape option was originally priced in tiers with so many TB per tier. That pricing was simplified to a price per TB. NBU does not charge by the number of virtual drives since you can have hundreds of virtual drives for relatively small amounts of disk storage. Contact you local account representative or Systems Engineer if you have questions. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gary Williams Sent: Thursday, March 23, 2006 4:54 AM To: veritas-bu@mailman.eng.auburn.edu Subject: [Veritas-bu] Virtual Tape Library This email and any attachments are confidential and may contain privileged information. If you are not the person for whom they are intended please return the email and then delete all material from any computer. You must not use the email or attachments for any purpose, nor disclose its contents to anyone other than the intended recipient. Any statements made by an individual in this email do not necessarily reflect the views of the Yorkshire Building Society Group. Hi, We are currently looking at purchasing a virtual tape library ( HP vls6000 ), and i am finding conflicting information on how the virtual library is licensed, some documentation says per TB, some say per virtual drive. Has anyone any experience with Netbackup and a vtl and what is the actual situation regarding licensing. Regards Gary Yorkshire Building Society, which is authorised and regulated by the Financial Services Authority, chooses to introduce its customers to: -Legal General for the purposes of advising on and arranging life assurance and investment products bearing Legal General's name; and -Homeowners Friendly Society for the purpose of arranging stakeholder Child Trust Funds. We are entered in the FSA Register and our FSA registration number is 106085 http://www.fsa.gov.uk/register Head Office: Yorkshire Building Society, Yorkshire House, Yorkshire Drive, Bradford, BD5 8LJ Tel: 0845 1 200 100 Visit Our Website http://www.ybs.co.uk All communications with us may be monitored/recorded to improve the quality of our service and for your protection and security. This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Tape re-use and replacement
Title: Message http://www.fujifilm.com/JSP/fuji/epartners/bin/LTO_DATA_SHEET_6-1-05.pdf Avg 1,000,000 head passes. I read another spec somewhere that indicated 25,000 mounts. (it was at http://www.fujifilm.com/JSP/fuji/epartners/tds/pdf/LTO_Ultrium_Drive_Data_QA.pdfbut the link is dead now.) so 25,000 mount / 1,000,000 head passes. I'd say you can get more than 12 writes out of em. Paul -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tristan BallSent: March 22, 2006 11:58 PMTo: veritas-bu@mailman.eng.auburn.eduSubject: [Veritas-bu] Tape re-use and replacement We're reviewing our backup policy documentation, in particular we're trying to define how often we replace tapes. The tapes in question are LTO-3. Management wants to replace them after 12 write passes, but that seems a little low to me? What are other shops doing? Thanks, Tristan.
RE: [Veritas-bu] Virtual Tape Library
Does this apply in the case of a 3rd party VTL that presents itself to NBU as tape drives? Paul -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dave Carpe Sent: March 23, 2006 6:25 AM To: Gary Williams; veritas-bu@mailman.eng.auburn.edu Subject: RE: [Veritas-bu] Virtual Tape Library The NetBackup Virtual tape option was originally priced in tiers with so many TB per tier. That pricing was simplified to a price per TB. NBU does not charge by the number of virtual drives since you can have hundreds of virtual drives for relatively small amounts of disk storage. Contact you local account representative or Systems Engineer if you have questions. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
Title: Message Ok interesting on the licensing. I thought the VTL vendors licensed on a per TB basis, but in order to use it with NBU, you had to license with Veritas for the number of frices you are presenting to NBU. Yeah, you can still write multiple jobs to the VTL (mpx to a virtual tape) otherwise, you (or I) would want to create a LOT of virtual drives...and with a lot of drive simultaneously active, an "inline" VTL (need to go back and refresh on terminology) would write one physical tape for every virtual tape, so you could end up with a lot more tapes "in play" on any given day. the key reason I want to go to VTL at this point is that we have many different business lines here with different retention policies, so even though we mpx up to 8 for our LTO3 drives and 100Mb/s clients, we end up with a bunch of jobs queued up waiting for that onelast client in a policyto finishthe client is pushing over 100Mb/s, the drive is shoeshining itself to pieces, throughput sucks and it could take an hour before this job finshes to release a free drive,while there are 20 jobs in a different policy queued up waiting for a free drive to put in another tape (different pool or retention or whatever.) seems DSSU would be easy in this case, to just set the max number of jobs, and let er rip.different policies, retentions, etc, all write at once and get sorted, hopefully in some efficient manner, after the fact. with a VTL, things still need to be treated as tape drives, so there's a different way of engineering/designing things. Paul -Original Message-From: Steven L. Sesar [mailto:[EMAIL PROTECTED] Sent: March 23, 2006 8:14 AMTo: Paul KeatingCc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] on the subject of disk based backupsNot true about the licensing piece. VTL's are licensed on a per TB basis. I believe that you can also write multiple jobs to VTL, as well - I don't see why you couldn't.Paul Keating wrote: I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one optionbetter than the other. Paul-- === Steven L. Sesar Senior Operating Systems Programmer/Analyst UNIX Application Services R101 The MITRE Corporation 202 Burlington Road - KS101 Bedford, MA 01730 tel: (781) 271-7702 fax: (781) 271-2600 mobile: (617) 893-9635 email: [EMAIL PROTECTED] ===
Re: [Veritas-bu] on the subject of disk based backups....
Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul
RE: [Veritas-bu] Re Veritas-bu] LTO 1 recovery using LTO 2 drive.
Not here - In fact I stand by my previous response - in that you can configure a LTO2 Drive to accept an LTO1 drive. Must admit, not seen the problem you mention with regards to data_buffer_size, although maybe it can cause problems with recovering data on old tapes Simon Weaver Technical Support Windows Domain Administrator EADS Astrium Tel: 02392-708598 Email: [EMAIL PROTECTED] -Original Message- From: Viv [mailto:[EMAIL PROTECTED] Sent: 22 March 2006 21:06 To: veritas-bu@mailman.eng.auburn.edu Subject: [Veritas-bu] Re Veritas-bu] LTO 1 recovery using LTO 2 drive. Thanks for the replies on this. We have since found that the DATA_BUFFERS_SIZE on some of the media servers could be causing this issue. For some reason (may be to do with performance) they have been set to 4MB. Has anyone else experienced problems with such high block sizes? thanks __ Hmmm... Likely that the underlying device is issuing some sort of message that NetBackup doesn't like. Which FP/MP of 4.5? What is the OS of the media server? Check the underlying OS message file on the media server for clues. Speculation: There may be an issue with the LTO driver you're using not automatically reading the LTO1 tapes as it should. Whenever possible, the latest stable driver for the tape and the OS of the media server should be used. This sort of sounds like issues with an LTO driver or a generic scsi tape driver. Mickey Baker Fort Lauderdale, FL -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Tuesday, March 14, 2006 1:56 PM To: veritas-bu@mailman.eng.auburn.edu Subject: [Veritas-bu] LTO 1 recovery using LTO 2 drive. We are currently backing up a system with LTO gen 1 drives/media and trying to recover using LTO gen 2 drives (LTO 1 drives are IBM and LTO 2 drives are HP). We are using NBU 4.5 and when the LTO 1 tapes are loaded into the gen 2 drive, the drives are downed. Any ideas what can be causing this is appreciated. -- ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu Previous message: [Veritas-bu] LTO 1 recovery using LTO 2 drive. Next message: [Veritas-bu] determining drive that a backup used Messages sorted ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu This email is for the intended addressee only. If you have received it in error then you must not use, retain, disseminate or otherwise deal with it. Please notify the sender by return email. The views of the author may not necessarily constitute the views of EADS Astrium Limited. Nothing in this email shall bind EADS Astrium Limited in any contract or obligation. EADS Astrium Limited, Registered in England and Wales No. 2449259 Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Virtual Tape Library
NetBackup 5.n was licenced by tape drives. NetBackup 6.n is licenced by TB in the VTL. I don't know the detail, e.g. is that raw TB etc. I also don't know if they changed the licencing for 5.n, or if it is still needing drive licences. William D L Brown Gary Williams [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23-Mar-2006 09:53 To veritas-bu@mailman.eng.auburn.edu cc Subject [Veritas-bu] Virtual Tape Library This email and any attachments are confidential and may contain privileged information. If you are not the person for whom they are intended please return the email and then delete all material from any computer. You must not use the email or attachments for any purpose, nor disclose its contents to anyone other than the intended recipient. Any statements made by an individual in this email do not necessarily reflect the views of the Yorkshire Building Society Group. Hi, We are currently looking at purchasing a virtual tape library ( HP vls6000 ), and i am finding conflicting information on how the virtual library is licensed, some documentation says per TB, some say per virtual drive. Has anyone any experience with Netbackup and a vtl and what is the actual situation regarding licensing. Regards Gary Yorkshire Building Society, which is authorised and regulated by the Financial Services Authority, chooses to introduce its customers to: -Legal General for the purposes of advising on and arranging life assurance and investment products bearing Legal General's name; and -Homeowners Friendly Society for the purpose of arranging stakeholder Child Trust Funds. We are entered in the FSA Register and our FSA registration number is 106085 http://www.fsa.gov.uk/register Head Office: Yorkshire Building Society, Yorkshire House, Yorkshire Drive, Bradford, BD5 8LJ Tel: 0845 1 200 100 Visit Our Website http://www.ybs.co.uk All communications with us may be monitored/recorded to improve the quality of our service and for your protection and security. This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiod
Hi, Thanks, but I have done the rman "change ... delete ..." bit. What concerns me is that NB is not removing the backup images from tape. Regards Pete -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]Sent: 23 March 2006 14:39To: Hitchman, Peter (TS UK)Cc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiodIs this the command ur after? The change ... delete command removes references to a backup piece, datafile copy, or archivelog from the control file and recovery catalog. It physically deletes the file. This command works with or without a recovery catalog. [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 14:10 To: veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] Netbackup expiry of rman backups with infinite retention periodHi,I have recently set-up some Oracle rman backups with Netbackup 4.5 on Solaris.The retention period in the policy is set to infinite for both the automatic and application backup schedules. I wanted to check that Netbackup would clean up the backups when I obsolete them in rman. But having done the obsolete in rman I have found that Netbackup has not removed the backup images from the tapes, even though they are gone from the rman catalog. I tried running "bpimage -cleanup -client name" but the bpdbm log showed that nothing was removed.There was an oddity with the Netbackup configuration on the client, when the original backup was made, the bp.conf contained the wrong client name. Could this be causing a problem, now that the client name has been changed?Am I on the wrong track here? According to the manual when rman deletes a backup-piece Netbackup should get rid of it regardless of the retention period. Does anyone know of a way I can sort this out, so that when rman deletes the backup, Netbackup does as well?ThanksPete___Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.eduhttp://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] New (test) install of NB 6.0 on Red Hat AS3
Just to put this to bed, let me share a finding. When I re-installed, I put the FQDN in when it asked for a host name. I rebooted, and things seem to be working! I guess the 'secret' for my install was to: 1) Make sure the server is in DNS and /etc/hosts. 2) When it asks for 'servername.' enter the FQDN. 3) Use the FQDN when logging into the console. Cheers, -dd -Dan Dobbs, System Administrator -Mutual Materials -425.452.2387 Peter DrakeUnderkoffler [EMAIL PROTECTED] 3/22/2006 2:30:55 PM -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I'm pretty sure that the install also doesn't set up the vm.conf file, I think if you put a DEVICE_HOST entry. I ran into the same exact thing and of course I lost my notes, but I think that was the solution, don't quote me though. Thanks Peter Peter DrakeUnderkoffler Xinupro, LLC 617-834-2352 Dan Dobbs wrote: Greetings, list. I'm doing a test install of 6.0 on a spare server to get a look at the 6.0 install process, and to get familiarized with the new features. The install seemed to go ok, got the PBX piece installed, took all the defaults. I noticed at the end of the install, that it seemed to want a FQDN name for your server. At the time, I just took the defaults. I fired up jnbSA, and I could not log into the console. When I noticed that it actually put 'servername.(none)' in the bp.conf, I changed the 'servername.(none)' entries to 'servername'. Once I did that, it let me log into the console. However, I tried to add a storage unit (there are no devices attached, so I was going to add a simple disk-based storage area), and it said that it couldn't attach to vmd (error 70). I checked bpps -a, and sure enough, vmd wasn't there. I restarted the server, did a bpps -a, and vmd was present. But, once I fired up the console and tried to add a storage unit again, it gave me the same (70) error. Should I wipe, reinstall, and FQDN the server name when it asks? As it's a test box, it's no big deal. Or, is there something else I should check/fix to make VMD not die? Cheers, -dd -Dan Dobbs, System Administrator -Mutual Materials -425.452.2387 ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (FreeBSD) iD4DBQFEIdAfl+lekZRM55oRAuDqAJiDAtlmp0kZUYAWY4XCchjfCeiEAJ9JYR8o no56oqnYvVzYyXCzk+KlQw== =PPfD -END PGP SIGNATURE- ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] drives showing UNKNOWN
Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiod
I don't have anywhere to test this, but my understanding is that this should do what ur after. Can u test on test box (as u have some issues with ur client name or something)? [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 14:50 To:[EMAIL PROTECTED] cc:veritas-bu@mailman.eng.auburn.edu Subject:RE: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiod Hi, Thanks, but I have done the rman change ... delete ... bit. What concerns me is that NB is not removing the backup images from tape. Regards Pete -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: 23 March 2006 14:39 To: Hitchman, Peter (TS UK) Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiod Is this the command ur after? The change ... delete command removes references to a backup piece, datafile copy, or archivelog from the control file and recovery catalog. It physically deletes the file. This command works with or without a recovery catalog. [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 14:10 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] Netbackup expiry of rman backups with infinite retention period Hi, I have recently set-up some Oracle rman backups with Netbackup 4.5 on Solaris. The retention period in the policy is set to infinite for both the automatic and application backup schedules. I wanted to check that Netbackup would clean up the backups when I obsolete them in rman. But having done the obsolete in rman I have found that Netbackup has not removed the backup images from the tapes, even though they are gone from the rman catalog. I tried running bpimage -cleanup -client name but the bpdbm log showed that nothing was removed. There was an oddity with the Netbackup configuration on the client, when the original backup was made, the bp.conf contained the wrong client name. Could this be causing a problem, now that the client name has been changed? Am I on the wrong track here? According to the manual when rman deletes a backup-piece Netbackup should get rid of it regardless of the retention period. Does anyone know of a way I can sort this out, so that when rman deletes the backup, Netbackup does as well? Thanks Pete ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] bpduplicate question
The action i got from original post was that he duplicated to the wrong pool and wanted to correct it by removing the duplication and then duplicating to the correct pool. I took from this that he just wanted 1 duplication and so the copy number would be 2. Increasing the number of copies to 3 just means bpduplicate will create a new copy and it doesnt mean the 2nd copy he did do will be gotten rid of. Plus if he has a restore procedure to use the duplicated copies and needs to either change to primary or restore from copy 2 his documentation will be out for this copy which is copy number 3. It depends indeed how the restore process adopted works. I may have gotten the wrong end of the stick from the original email. Cheers Mickey Baker wrote: That’s the correct solution, IMO. Mickey Baker Storage Consultant Fort Lauderdale, FL 954.729.6464 *From:* [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] *On Behalf Of *BeDour, Wayne *Sent:* Wednesday, March 22, 2006 10:26 AM *To:* veritas-bu@mailman.eng.auburn.edu *Subject:* RE: [Veritas-bu] bpduplicate question Just to answer my own question. I increased the “Maximum backup copies” in the “Global Attributes” from the default of 2 to 3 and was able to rerun my bpduplcate command. I’m not sure this is the correct or only solution but it appears to take care of my problem. Any thoughts or recommendations would be appreciated. Thanks Wayne BeDour IT Unix System Administrator PH: 313-240-3374 FAX: 313-240-3065 Internet: [EMAIL PROTECTED] *From:* [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] *On Behalf Of *BeDour, Wayne *Sent:* Wednesday, March 22, 2006 8:43 AM *To:* veritas-bu@mailman.eng.auburn.edu *Subject:* [Veritas-bu] bpduplicate question Our environment is HP-UX 11.11, Netbackup 5.1 MP3 backing up mostly HP unix with some Sun and Windows backups. I set up duplication and duplicated one of my tapes to the wrong volume pool and now want to correct it. I have created the correct volume pool for my duplicate, changed the policy to use that volume pool and expired the dup copy I created. When I try to duplicate the tape again it is giving me the following messages: # bpduplicate -id 000773 -dstunit SureStore20_700_ys -dp NB_duplicates -s 01/01/2004 Duplicate started Wed Mar 22 2006 08:22:04 Activity monitor job id = 173659 INF - Skipping backup id xx_1142944372, it already has 2 copies. INF - Skipping backup id xx_1142335945, it already has 2 copies. INF - Skipping backup id xx_1141736210, it already has 2 copies. …….. …….. …….. INF - found no images or media matching the selection criteria # What needs to be done so NBU doesn’t think it still has 2 copies? Thanks in advance… Wayne BeDour IT Unix System Administrator PH: 313-240-3374 FAX: 313-240-3065 Internet: [EMAIL PROTECTED] ** ** LEGAL DISCLAIMER ** ** This E-mail message and any attachments may contain legally privileged, confidential or proprietary information. If you are not the intended recipient(s), or the employee or agent responsible for delivery of this message to the intended recipient(s), you are hereby notified that any dissemination, distribution or copying of this E-mail message is strictly prohibited. If you have received this message in error, please immediately notify the sender and delete this E-mail message from your computer. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] drives showing UNKNOWN
The mt status showing no tape can mean the tape is ejected, but still in the drive - have u tried moving the UNKNOWN tape from the drive using robtest? m d3 s?? ? I suspect that will work (and the library just wasnt able to read the barcode for these tapes). If it doesnt then you probably have a hardware fault. Dave Markham [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 15:14 Please respond to dave.markham To:veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] drives showing UNKNOWN Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Netbackup expiry of rman backups with infiniteretentionperiod
Hi, Yes I am setting up to try this out on another database. Regards Pete -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]Sent: 23 March 2006 15:22To: Hitchman, Peter (TS UK)Cc: veritas-bu@mailman.eng.auburn.eduSubject: RE: [Veritas-bu] Netbackup expiry of rman backups with infiniteretentionperiodI don't have anywhere to test this, but my understanding is that this should do what ur after. Can u test on test box (as u have some issues with ur client name or something)? [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 14:50 To: [EMAIL PROTECTED] cc: veritas-bu@mailman.eng.auburn.edu Subject: RE: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiodHi, Thanks, but I have done the rman "change ... delete ..." bit. What concerns me is that NB is not removing the backup images from tape. Regards Pete -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]Sent: 23 March 2006 14:39To: Hitchman, Peter (TS UK)Cc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] Netbackup expiry of rman backups with infinite retentionperiodIs this the command ur after? The change ... delete command removes references to a backup piece, datafile copy, or archivelog from the control file and recovery catalog. It physically deletes the file. This command works with or without a recovery catalog. [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 14:10 To: veritas-bu@mailman.eng.auburn.educc: Subject: [Veritas-bu] Netbackup expiry of rman backups with infinite retention periodHi,I have recently set-up some Oracle rman backups with Netbackup 4.5 on Solaris.The retention period in the policy is set to infinite for both the automatic and application backup schedules. I wanted to check that Netbackup would clean up the backups when I obsolete them in rman. But having done the obsolete in rman I have found that Netbackup has not removed the backup images from the tapes, even though they are gone from the rman catalog. I tried running "bpimage -cleanup -client name" but the bpdbm log showed that nothing was removed.There was an oddity with the Netbackup configuration on the client, when the original backup was made, the bp.conf contained the wrong client name. Could this be causing a problem, now that the client name has been changed?Am I on the wrong track here? According to the manual when rman deletes a backup-piece Netbackup should get rid of it regardless of the retention period. Does anyone know of a way I can sort this out, so that when rman deletes the backup, Netbackup does as well?ThanksPete___Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.eduhttp://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
Title: Message excellent info...Now that you mention itmy one small DSSU I have configured to offload to a specific tape pool, so I should have known that. That definately sways me in the direction of VTL then. -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: March 23, 2006 8:30 AMTo: Paul KeatingCc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] on the subject of disk based backupsHi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. "Paul Keating" [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To: veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backupsI'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul
Re: [Veritas-bu] drives showing UNKNOWN
Ops have been to the unit and seen there are physically no tapes in the drive. I just need to rule out netbackup thinking there is to prove there is a h/w problem. Dave [EMAIL PROTECTED] wrote: The mt status showing no tape can mean the tape is ejected, but still in the drive - have u tried moving the UNKNOWN tape from the drive using robtest? m d3 s?? ? I suspect that will work (and the library just wasnt able to read the barcode for these tapes). If it doesnt then you probably have a hardware fault. *Dave Markham [EMAIL PROTECTED]* Sent by: [EMAIL PROTECTED] 23/03/2006 15:14 Please respond to dave.markham To:veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] drives showing UNKNOWN Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] drives showing UNKNOWN
Well the information u have from robtest is what the drives are reporting to the library not direct to NBU, so IMHO this rules out NetBackup. I guess from ur question it does not, but normally u could check the library status on a front panel or web GUI to verify this, cheers Andy. Dave Markham [EMAIL PROTECTED] 23/03/2006 15:51 Please respond to dave.markham To:[EMAIL PROTECTED] cc:veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu Subject:Re: [Veritas-bu] drives showing UNKNOWN Ops have been to the unit and seen there are physically no tapes in the drive. I just need to rule out netbackup thinking there is to prove there is a h/w problem. Dave [EMAIL PROTECTED] wrote: The mt status showing no tape can mean the tape is ejected, but still in the drive - have u tried moving the UNKNOWN tape from the drive using robtest? m d3 s?? ? I suspect that will work (and the library just wasnt able to read the barcode for these tapes). If it doesnt then you probably have a hardware fault. *Dave Markham [EMAIL PROTECTED]* Sent by: [EMAIL PROTECTED] 23/03/2006 15:14 Please respond to dave.markham To:veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] drives showing UNKNOWN Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
Yeah, for me the biggest weakness of the NBU DSSUs. Paul Keating [EMAIL PROTECTED] 23/03/2006 15:48 To:[EMAIL PROTECTED] cc:veritas-bu@mailman.eng.auburn.edu Subject:RE: [Veritas-bu] on the subject of disk based backups excellent info...Now that you mention itmy one small DSSU I have configured to offload to a specific tape pool, so I should have known that. That definately sways me in the direction of VTL then. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: March 23, 2006 8:30 AM To: Paul Keating Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] on the subject of disk based backups Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul
RE: [Veritas-bu] drives showing UNKNOWN
Title: RE: [Veritas-bu] drives showing UNKNOWN Leave the drive down, and reset the tape library when you get chance. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Dave Markham Sent: Thursday, March 23, 2006 9:51 AM To: [EMAIL PROTECTED] Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] drives showing UNKNOWN Ops have been to the unit and seen there are physically no tapes in the drive. I just need to rule out netbackup thinking there is to prove there is a h/w problem. Dave [EMAIL PROTECTED] wrote: The mt status showing no tape can mean the tape is ejected, but still in the drive - have u tried moving the UNKNOWN tape from the drive using robtest? m d3 s?? ? I suspect that will work (and the library just wasnt able to read the barcode for these tapes). If it doesnt then you probably have a hardware fault. *Dave Markham [EMAIL PROTECTED]* Sent by: [EMAIL PROTECTED] 23/03/2006 15:14 Please respond to dave.markham To: veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu cc: Subject: [Veritas-bu] drives showing UNKNOWN Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu The information in this message may be proprietary and/or confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify First Data immediately by replying to this message and deleting it from your computer.
[Veritas-bu] RE: on the subject of disk based backups....
Good Morning -- I'm curious about the CPU hit and overall throughput of VTL vs DSU or DSSU backups. How many concurrent backups can run from a single server? The most I can run on my servers with DSUs is 10 per server. --kathy To: Paul Keating [EMAIL PROTECTED] Cc: veritas-bu@mailman.eng.auburn.edu Subject: RE: [Veritas-bu] on the subject of disk based backups From: [EMAIL PROTECTED] Date: Thu, 23 Mar 2006 15:59:33 + This is a multipart message in MIME format. --=_alternative 0057D9418025713A_= Content-Type: text/plain; charset=us-ascii Yeah, for me the biggest weakness of the NBU DSSUs. Paul Keating [EMAIL PROTECTED] 23/03/2006 15:48 To: [EMAIL PROTECTED] cc: veritas-bu@mailman.eng.auburn.edu Subject:RE: [Veritas-bu] on the subject of disk based backups excellent info...Now that you mention itmy one small DSSU I have configured to offload to a specific tape pool, so I should have known that. That definately sways me in the direction of VTL then. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: March 23, 2006 8:30 AM To: Paul Keating Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] on the subject of disk based backups Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Netbackup expiry of rman backups with infinite retention period
There's a special set of commands to sync the RMAN Netbackup catalogs. Check this technote: http://seer.support.veritas.com/docs/251686.htm -M -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of [EMAIL PROTECTED] Sent: Thursday, March 23, 2006 7:11 AM To: veritas-bu@mailman.eng.auburn.edu Subject: [Veritas-bu] Netbackup expiry of rman backups with infinite retention period Hi, I have recently set-up some Oracle rman backups with Netbackup 4.5 on Solaris. The retention period in the policy is set to infinite for both the automatic and application backup schedules. I wanted to check that Netbackup would clean up the backups when I obsolete them in rman. But having done the obsolete in rman I have found that Netbackup has not removed the backup images from the tapes, even though they are gone from the rman catalog. I tried running bpimage -cleanup -client name but the bpdbm log showed that nothing was removed. There was an oddity with the Netbackup configuration on the client, when the original backup was made, the bp.conf contained the wrong client name. Could this be causing a problem, now that the client name has been changed? Am I on the wrong track here? According to the manual when rman deletes a backup-piece Netbackup should get rid of it regardless of the retention period. Does anyone know of a way I can sort this out, so that when rman deletes the backup, Netbackup does as well? Thanks Pete ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] v6.0 NOM on Solaris 10 x86
It's not on the supported list but Sun says Solaris 10 x86 is binary = compatible with Solaris 10 Sparc. Where does it say that they're binary compatible? No version of Solaris x86 offers SPARC binary compatibility. Both are binary compatible with older releases on the same architecture, but not with each other. -- Darren Dunham [EMAIL PROTECTED] Senior Technical Consultant TAOShttp://www.taos.com/ Got some Dr Pepper? San Francisco, CA bay area This line left intentionally blank to confuse you. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Remote Admin Console Login
Hi, Did you restart the Master server services? Each time you add a server , you need to restart the services on the Master server. Regards, Jim From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Yang XiaoSent: Saturday, March 18, 2006 3:32 AMTo: veritas-bu@mailman.eng.auburn.eduSubject: [Veritas-bu] Remote Admin Console Login Hi all,I'm using NB6.0on win2003, I have configured user access(username/domain/host) and added the workstation to the server list. but I still can't login to the Master server and is getting "Unable to connect to the selected netbackup host" error. What am I missing? Thanks!- Yang
RE: [Veritas-bu] drives showing UNKNOWN
What's s s (status of the slotted tapes) show via robtest? Only one bad tape number or all UNKNOWN tape numbers. Might have a bad barcode reader. -M -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Dave Markham Sent: Thursday, March 23, 2006 9:43 AM To: Xu, Ying Cc: [EMAIL PROTECTED]; veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] drives showing UNKNOWN Already done :( Same result Xu, Ying wrote: Leave the drive down, and reset the tape library when you get chance. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Dave Markham Sent: Thursday, March 23, 2006 9:51 AM To: [EMAIL PROTECTED] Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] drives showing UNKNOWN Ops have been to the unit and seen there are physically no tapes in the drive. I just need to rule out netbackup thinking there is to prove there is a h/w problem. Dave [EMAIL PROTECTED] wrote: The mt status showing no tape can mean the tape is ejected, but still in the drive - have u tried moving the UNKNOWN tape from the drive using robtest? m d3 s?? ? I suspect that will work (and the library just wasnt able to read the barcode for these tapes). If it doesnt then you probably have a hardware fault. *Dave Markham [EMAIL PROTECTED]* Sent by: [EMAIL PROTECTED] 23/03/2006 15:14 Please respond to dave.markham To:veritas-bu@mailman.eng.auburn.edu veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] drives showing UNKNOWN Netbackup 5.0 mp5s2 on solaris 8 connected to L1000 with 4 DLT7000 drives in. Has anyone seen it where robtest shows drive with barcode UNKNOWN, yet there are no tapes in the drive? Invoking robotic test utility: /usr/openv/volmgr/bin/tldtest -r /dev/sg/c3t0l0 -d1 /dev/rmt/0cbn -d2 /dev/rmt/1cbn -d3 /dev/rmt/2cbn -d4 /dev/rmt/3cbn Opening /dev/sg/c3t0l0 MODE_SENSE complete Enter tld commands (? returns help information) s d drive 1 (addr 128) access = 1 Contains Cartridge = no SCSI ID from drive 1 is 2 drive 2 (addr 129) access = 1 Contains Cartridge = no SCSI ID from drive 2 is 3 drive 3 (addr 130) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 3 is 4 drive 4 (addr 131) access = 1 Contains Cartridge = yes Barcode = UNKNOWN SCSI ID from drive 4 is 5 READ_ELEMENT_STATUS complete The drives keep going down if i up them and a job tries to use them. I currently have them in a down state. The drives are flashing, but the akibia support after not responding for ages are trying to say its a netbackup problem as mt -f /dev/rmt/2cbn status shows there is no tape loaded. ( which it wouldnt because there isnt one ) # for d in 0 1 2 3; do mt -f /dev/rmt/$d'cbn' status; done /dev/rmt/0cbn: no tape loaded or drive offline /dev/rmt/1cbn: no tape loaded or drive offline /dev/rmt/2cbn: no tape loaded or drive offline /dev/rmt/3cbn: no tape loaded or drive offline I am thinking the drives are goosed and robtest cannot talk to them to get a clear picture. Is this on the right track? Anyone any ideas? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu * The information in this message may be proprietary and/or confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify First Data immediately by replying to this message and deleting it from your computer. * ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Cold Catalog Backup
Title: Message Hi All, From what i understand, what Veritas refers to as the Cold DB Backup in 6.0 is actually the Catalog backup as id was done in previous versions on NBU. The new Online option allows for catalog backup to run even when regular backups are running, and even better to run incrementals on the catalog, speeding up the process even more. Regarding COLD Catalog Backups. = If you are using a standalone drive for your catalog backup, you just inster a blank tape in the drive and you are ready. NBU will always assume that this tape is A0 (unless it is already labeled by NBU as something else) If you are using a robot, you need to write a script to change the tape the catalog uses. Regards, Jim From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of WEAVER, SimonSent: Monday, March 20, 2006 9:26 AMTo: 'Yang Xiao'; veritas-bu@mailman.eng.auburn.eduSubject: RE: [Veritas-bu] Cold Catalog Backup Yang If I understand correctly, then as soon as the catalogue has written to the specific tape (ie: A), simply eject the volume from your robot / library / tape drive and insert a new tape (blank) and reconfigure the Catalogue Backup Wizard again to use the newly inserted tape! HTH Simon WeaverTechnical SupportWindows Domain Administrator EADS AstriumTel: 02392-708598 Email: [EMAIL PROTECTED] -Original Message-From: Yang Xiao [mailto:[EMAIL PROTECTED] Sent: 18 March 2006 01:29To: veritas-bu@mailman.eng.auburn.eduSubject: [Veritas-bu] Cold Catalog BackupHi all,Can someone please explain how Cold backup of the catalog should work witn NB 6.0 Enterprise(win 2003)I have configured cold backup of the catalog using the wizard, picked a Tape A, and alternate DISK storage. My question is, what do I do when I need to unload the tape and ship the tape off site? how do I asign more tapes to the catalog backup? Many thanks,- Yang This email is for the intended addressee only.If you have received it in error then you must not use, retain, disseminate or otherwise deal with it.Please notify the sender by return email.The views of the author may not necessarily constitute the views of EADS Astrium Limited.Nothing in this email shall bind EADS Astrium Limited in any contract or obligation.EADS Astrium Limited, Registered in England and Wales No. 2449259Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England
[Veritas-bu] NBU versus Legato
I am being mandated to look at other options outside of Netbackup since it is budget time and the support maintenance dollars just hit my boss' desk. I was curious if anyone here switched from Legato to Netbackup and how you would compare each product? Thanks for your insights. Reneé Carlisle Sr. Systems Administrator675 Basket RoadWebster, NY 14580 Jer 29:11-13 585-216-0497 (w)585-472-2360 (c) The information contained in this message may be privileged, confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or any employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Paychex, Inc.
RE: [Veritas-bu] Disk backup - Storage Hardware configurations
We are just starting to set up DSU storage and have completed some extensive performance tests. So far, we have found RAID5 4+1 works best with our file system (small files) backups and RAID3 4+1 works best with our database and raw backups. There is still some testing going on as far as buffer settings, we are between 256k and 512K buffer settings and are still not sure which is going to work out best for us. These tests have all just been done in a test environment, we won't know if we have to tweak anything for production for another month or so. Reneé Carlisle Sr. Systems Administrator675 Basket RoadWebster, NY 14580 Jer 29:11-13 585-216-0497 (w)585-472-2360 (c) From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED]Sent: Thursday, March 23, 2006 4:57 AMTo: Ambrose, MonteCc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] Disk backup - Storage Hardware configurations Hi Monte, I did one install with EMC DMX1000 disk with hardware mirroring used for a DSSU and I have to say I found performance wasn't great. Worse than LTO2 performance but this was writing to a fairly small number of physical drives, though performance wasn't an issue so I didn't spend a lot of time tuning it, cheers Andy. "Ambrose, Monte" [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 01:54 To: veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] Disk backup - Storage Hardware configurationsI wanted to get a feel for what type of disk storage is being used out there for disk based backups. I am specifically looking for SAN attached storage ATA, SATA and what type of RAID is used and how it is performing overall. The NBU 6.0 tuning guide shows RAID-5 but I am wondering if anyone is using RAID-10 or others. Thanks, Monte The information contained in this message may be privileged, confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or any employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Paychex, Inc.
RE: [Veritas-bu] Disk backup - Storage Hardware configurations
Title: Message So you're building your LUNs RAID5 4+1,then presenting multiple LUNs to the server and striping across them with your VM, or what? or is each DSSU only the size of 4 disks. Paul -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Carlisle, D ReneeSent: March 23, 2006 1:55 PMTo: [EMAIL PROTECTED]; Ambrose, MonteCc: veritas-bu@mailman.eng.auburn.eduSubject: RE: [Veritas-bu] Disk backup - Storage Hardware configurations We are just starting to set up DSU storage and have completed some extensive performance tests. So far, we have found RAID5 4+1 works best with our file system (small files) backups and RAID3 4+1 works best with our database and raw backups. There is still some testing going on as far as buffer settings, we are between 256k and 512K buffer settings and are still not sure which is going to work out best for us. These tests have all just been done in a test environment, we won't know if we have to tweak anything for production for another month or so. Reneé Carlisle Sr. Systems Administrator675 Basket RoadWebster, NY 14580 Jer 29:11-13 585-216-0497 (w)585-472-2360 (c) From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED]Sent: Thursday, March 23, 2006 4:57 AMTo: Ambrose, MonteCc: veritas-bu@mailman.eng.auburn.eduSubject: Re: [Veritas-bu] Disk backup - Storage Hardware configurations Hi Monte, I did one install with EMC DMX1000 disk with hardware mirroring used for a DSSU and I have to say I found performance wasn't great. Worse than LTO2 performance but this was writing to a fairly small number of physical drives, though performance wasn't an issue so I didn't spend a lot of time tuning it, cheers Andy. "Ambrose, Monte" [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 01:54 To: veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] Disk backup - Storage Hardware configurationsI wanted to get a feel for what type of disk storage is being used out there for disk based backups. I am specifically looking for SAN attached storage ATA, SATA and what type of RAID is used and how it is performing overall. The NBU 6.0 tuning guide shows RAID-5 but I am wondering if anyone is using RAID-10 or others. Thanks, Monte The information contained in this message may be privileged,confidential, and protected from disclosure. If the reader of thismessage is not the intended recipient, or any employee or agentresponsible for delivering this message to the intended recipient,you are hereby notified that any dissemination, distribution, orcopying of this communication is strictly prohibited. If you havereceived this communication in error, please notify us immediatelyby replying to the message and deleting it from your computer.Thank you. Paychex, Inc.
RE: [Veritas-bu] v6.0 NOM on Solaris 10 x86
Compatibility is the word from the bossman - no independant verification on my part. Maybe I better... -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Darren Dunham Sent: Thursday, March 23, 2006 11:39 AM To: Veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] v6.0 NOM on Solaris 10 x86 It's not on the supported list but Sun says Solaris 10 x86 is binary = compatible with Solaris 10 Sparc. Where does it say that they're binary compatible? No version of Solaris x86 offers SPARC binary compatibility. Both are binary compatible with older releases on the same architecture, but not with each other. -- Darren Dunham [EMAIL PROTECTED] Senior Technical Consultant TAOShttp://www.taos.com/ Got some Dr Pepper? San Francisco, CA bay area This line left intentionally blank to confuse you. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] on the subject of disk based backups....
On Thu, Mar 23, 2006 at 08:05:53AM -0500, Paul Keating wrote: Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. We're using DSSU. It was free and that was a contributing factor. At one point, I was a proponent of VTL technology here but since then, I've changed my mind and now believe that VTL is a niche market that will be killed off by the free DSSU functionality. Veritas originally had a fairly week implementation of its DSSUs and even in 5.1MP4, it's not great. It's supposed to be better in NBU 6, and I expect that it will continue to evolve and eventually kill off the VTL requirement for the vast majority of customers. I think that VTLs will work for us at least as good as DSSUs do. It's just that we don't see a need to throw extra $ at the problem today. .../Ed -- Ed Wilts, Mounds View, MN, USA mailto:[EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] v6.0 NOM on Solaris 10 x86
On Thu, Mar 23, 2006 at 10:59:42AM -0700, [EMAIL PROTECTED] wrote: It's not on the supported list but Sun says Solaris 10 x86 is binary compatible with Solaris 10 Sparc. For v6.0's NOM, we'd rather run x86 (really, we'd rather Linux but that's right off the list). NOM on Linux is on the road map. You should contact Veritas and see where they're at in the development process. .../Ed -- Ed Wilts, Mounds View, MN, USA mailto:[EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] on the subject of disk based backups....
I'm curious: Why do people use different volume pools? We have one pool for Exchange backups, and three others to go with our Iron Mountain weekly rotations, but I'm not sure what that's buying us. For tracking Iron Mountain boxes, we use volume groups, so the three pools for that are redundant. It's nice to be able to easily break out the Exchange tapes, but there are other ways of getting that information. So I don't think the case for separate volume pools is all that compelling for us. How do other folks use them? other than being able to track easily how much On 3/23/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul -- Phil Rand[EMAIL PROTECTED]
Re: [Veritas-bu] v6.0 NOM on Solaris 10 x86
Compatibility is the word from the bossman - no independant verification on my part. Maybe I better... Start simple. Try copying over /bin/touch or something from a SPARC machine and see how well it works. # uname -i i86pc # file ./touch ./touch:ELF 32-bit MSB executable SPARC Version 1, dynamically linked, stripped # truss -texec -f sh -c './touch' 11073: execve(/sbin/sh, 0x08047DA0, 0x08047DB0) argc = 3 11074: execve(./touch, 0x08078268, 0x08078270) Err#22 EINVAL sh: ./touch: cannot execute # man execve [...] EINVAL The new process image file has the appropri- ate permission and has a recognized execut- able binary format, but the system does not support execution of a file with this for- mat. -- Darren Dunham [EMAIL PROTECTED] Senior Technical Consultant TAOShttp://www.taos.com/ Got some Dr Pepper? San Francisco, CA bay area This line left intentionally blank to confuse you. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] Weird tape drive behavior
I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] v6.0 NOM on Solaris 10 x86
OK - here's the scoop - and the boss was wrong. Solaris x86 Sparc are Source compatible not binary compatible. Here's an InfoWorld verification: === When it comes to app compatibility, Solaris isn't Java -- the binaries aren't portable between architectures (though they remain portable from earlier versions of Solaris on the same chip family, so you don't have to recompile a Solaris 2.6 app). However, on the small apps I tested, the source code was portable between 64-bit Sparc, 32-bit x86, and 64-bit x86. (InfoWorld, 4/2006) === Sorry for the confusion. -M -Original Message- From: Donaldson, Mark - Broomfield, CO Sent: Thursday, March 23, 2006 12:16 PM Compatibility is the word from the bossman - no independant verification on my part. Maybe I better... -Original Message- It's not on the supported list but Sun says Solaris 10 x86 is binary = compatible with Solaris 10 Sparc. Where does it say that they're binary compatible? No version of Solaris x86 offers SPARC binary compatibility. Both are binary compatible with older releases on the same architecture, but not with each other. -- Darren Dunham [EMAIL PROTECTED] Senior Technical Consultant TAOShttp://www.taos.com/ Got some Dr Pepper? San Francisco, CA bay area This line left intentionally blank to confuse you. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
I think this is due to be changed in an upcoming MP for NBU 6to allow multiple streams off one DSSU. I'm sure I heard they were doing something about that. William D L Brown [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23-Mar-2006 15:59 To Paul Keating [EMAIL PROTECTED] cc veritas-bu@mailman.eng.auburn.edu Subject RE: [Veritas-bu] on the subject of disk based backups Yeah, for me the biggest weakness of the NBU DSSUs. Paul Keating [EMAIL PROTECTED] 23/03/2006 15:48 To:[EMAIL PROTECTED] cc:veritas-bu@mailman.eng.auburn.edu Subject:RE: [Veritas-bu] on the subject of disk based backups excellent info...Now that you mention itmy one small DSSU I have configured to offload to a specific tape pool, so I should have known that. That definately sways me in the direction of VTL then. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: March 23, 2006 8:30 AM To: Paul Keating Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] on the subject of disk based backups Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Weird tape drive behavior
Which version of NetBackup? What OS? NetBackup 5.1 MP2 on Solaris had a problem with positioning I think. MP3 or MP3A fixed it. At least thats what our consultant told us. Austin On 3/23/06, Song, Young [EMAIL PROTECTED] wrote: I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
I think its more efficient and cost effective to consolidate as much as possible. Reasons for mulitiple volume pools: different retentions 1 group of servers has a separate pool for Full and incremental reason is, they require sending Full Backup tapes, occasionally, to our DR site to restore to servers there. They use those servers to test maintenance and so require them to be restored from the production server backups. We use bpduplicate to make a copy before sending them to DR site. When they were in the same pool I had trouble getting everything on the tape copied in a timely manner. Better option is to have enough tape drives to create 2 copies but we dontL. We use bpduplicate to copy almost everything. We keep a copy on site for a short period of time while sending a copy offsite everyday. Separate pools allows more than one dup job to run, from cron, after the backup window. Multiple media servers you might want to easily see which media belongs to which media server. From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Phil Rand Sent: Thursday, March 23, 2006 12:20 PM To: [EMAIL PROTECTED] Cc: Paul Keating; veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] on the subject of disk based backups I'm curious: Why do people use different volume pools? We have one pool for Exchange backups, and three others to go with our Iron Mountain weekly rotations, but I'm not sure what that's buying us. For tracking Iron Mountain boxes, we use volume groups, so the three pools for that are redundant. It's nice to be able to easily break out the Exchange tapes, but there are other ways of getting that information. So I don't think the case for separate volume pools is all that compelling for us. How do other folks use them? other than being able to track easily how much On 3/23/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul -- Phil Rand [EMAIL PROTECTED]
Re: [Veritas-bu] v6.0 NOM on Solaris 10 x86
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 The opposite would substantially cut into hardware sales at Sunall business all the time. Peter DrakeUnderkoffler Xinupro, LLC 617-834-2352 [EMAIL PROTECTED] wrote: OK - here's the scoop - and the boss was wrong. Solaris x86 Sparc are Source compatible not binary compatible. Here's an InfoWorld verification: === When it comes to app compatibility, Solaris isn't Java -- the binaries aren't portable between architectures (though they remain portable from earlier versions of Solaris on the same chip family, so you don't have to recompile a Solaris 2.6 app). However, on the small apps I tested, the source code was portable between 64-bit Sparc, 32-bit x86, and 64-bit x86. (InfoWorld, 4/2006) === Sorry for the confusion. -M -Original Message- From: Donaldson, Mark - Broomfield, CO Sent: Thursday, March 23, 2006 12:16 PM Compatibility is the word from the bossman - no independant verification on my part. Maybe I better... -Original Message- It's not on the supported list but Sun says Solaris 10 x86 is binary = compatible with Solaris 10 Sparc. Where does it say that they're binary compatible? No version of Solaris x86 offers SPARC binary compatibility. Both are binary compatible with older releases on the same architecture, but not with each other. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (FreeBSD) iD8DBQFEIv/sl+lekZRM55oRAj0PAJ9Qw6LAUsx8TYqAY9lUJqU+LgQfJwCeIW/F C+cTOE5/3JjjSm+tP9hCegw= =/dxE -END PGP SIGNATURE- ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Weird tape drive behavior
Just checking...you do have this directive in your /usr/openv/netbackup/bp.conf, right? ALLOW_MEDIA_OVERWRITE = TAR hth --Chris I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] on the subject of disk based backups....
On Thu, Mar 23, 2006 at 07:33:38PM +, [EMAIL PROTECTED] wrote: I think this is due to be changed in an upcoming MP for NBU 6to allow multiple streams off one DSSU. I'm sure I heard they were doing something about that. What you might have heard is simultaneous destages from a single DSSU. That's also in 5.1MP4 - by default, NetBackup will schedule 4 simultaneous jobs of 25GB each. Frankly, the defaults suck for people upgrading from an earlier release - it's a good way to fill your DSSUs and watch your jobs fail with error 84s, especially if you don't have a lot of idle tape drives. All of the streams from a single DSSU still go to the same volume pool. .../Ed -- Ed Wilts, Mounds View, MN, USA mailto:[EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] on the subject of disk based backups....
Another serious issue which may cause the proliferation of Volume Pools is business litigation. If your business is the victim of law school graduates, Volume Pools which segregate tainted from untainted data/applications/servers can be an effective way of containing the potential damage. It can also serve as a mechanism of measurement to charge back to the legal department for the cost of tapes required to satisfy external and internal data retention obligations. This is closely related to different retentions, but goes much further. Michael F Lavelle CIT Storage Engineering 100 Abbott Park Rd AP14B-1 GB16 Abbott Park, IL 60064-6042 Office (847) 937-1195 Cell (847) 878-0914 [EMAIL PROTECTED] King, Cheryl [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 03/23/2006 01:51 PM To veritas-bu@mailman.eng.auburn.edu cc Subject RE: [Veritas-bu] on the subject of disk based backups I think it’s more efficient and cost effective to consolidate as much as possible. Reasons for mulitiple volume pools: different retentions 1 group of servers has a separate pool for Full and incremental – reason is, they require sending Full Backup tapes, occasionally, to our DR site to restore to servers there. They use those servers to test maintenance and so require them to be restored from the production server backups. We use bpduplicate to make a copy before sending them to DR site. When they were in the same pool I had trouble getting everything on the tape copied in a timely manner. Better option is to have enough tape drives to create 2 copies but we don’tL. We use bpduplicate to copy almost everything. We keep a copy on site for a short period of time while sending a copy offsite everyday. Separate pools allows more than one dup job to run, from cron, after the backup window. Multiple media servers – you might want to easily see which media belongs to which media server. From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Phil Rand Sent: Thursday, March 23, 2006 12:20 PM To: [EMAIL PROTECTED] Cc: Paul Keating; veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] on the subject of disk based backups I'm curious: Why do people use different volume pools? We have one pool for Exchange backups, and three others to go with our Iron Mountain weekly rotations, but I'm not sure what that's buying us. For tracking Iron Mountain boxes, we use volume groups, so the three pools for that are redundant. It's nice to be able to easily break out the Exchange tapes, but there are other ways of getting that information. So I don't think the case for separate volume pools is all that compelling for us. How do other folks use them? other than being able to track easily how much On 3/23/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hi Paul, from my experience what you say about DSSU is incorrect, a single DSSU cannot stage the data off to multiple tape tape pools - if you have a requirement to stage to different tape pools you need to have a at least one DSSU per tape pool, cheers Andy. Paul Keating [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 23/03/2006 13:05 To:veritas-bu@mailman.eng.auburn.edu cc: Subject:[Veritas-bu] on the subject of disk based backups I'm also interested in thoughts regarding VTL vs DSSU. DSSU interests me mostly because I can write multiple jobs at once without any regard for tape pool, mpx or not, retention, etc. ...then it all gets sorted out to appropriate tapes during the destaging VTL still presents itself to NBU as a tape, so each virtual tape drive must be licensed with Veritas, in additional to the physical tape drives (unless the VTL is run inline, though there could be issues there regarding estimated compression, etc.) Just a couple things off the top of my head. Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. Paul -- Phil Rand [EMAIL PROTECTED]
[Veritas-bu] Scheduling Restores
Greetings, I have just attempted a restore on our Mirapoint appliance. What we inadvertantly discovered is that attempting a restore on a Mirapoint with a 100Mbit card will DOS the appliance. So, we are now looking to do a scheduled restore, which I don't find as of now. Otherwise, it's a midnight worknight. Thanks, Jason
Re: [Veritas-bu] Scheduling Restores
On Thu, Mar 23, 2006 at 03:41:31PM -0500, Brooks, Jason wrote: I have just attempted a restore on our Mirapoint appliance. What we inadvertantly discovered is that attempting a restore on a Mirapoint with a 100Mbit card will DOS the appliance. So, we are now looking to do a scheduled restore, which I don't find as of now. Otherwise, it's a midnight worknight. Do a command line bprestore and schedule it via cron (or whatever scheduler you have handy). -- Ed Wilts, Mounds View, MN, USA mailto:[EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Weird tape drive behavior
Hi Chris, Yes, I do have that directive in my bp.conf: ALLOW_MEDIA_OVERWRITE = tar Thanks. - Young -Original Message- From: Christopher Jay Manders [mailto:[EMAIL PROTECTED] Sent: Thursday, March 23, 2006 12:11 PM To: Song, Young Cc: 'veritas-bu@mailman.eng.auburn.edu' Subject: Re: [Veritas-bu] Weird tape drive behavior Just checking...you do have this directive in your /usr/openv/netbackup/bp.conf, right? ALLOW_MEDIA_OVERWRITE = TAR hth --Chris I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Weird tape drive behavior
Hi Austin, I'm using Data Center 4.5 FP8 on Sun Solaris 8. Thanks. - Young -Original Message- From: Austin Murphy [mailto:[EMAIL PROTECTED] Sent: Thursday, March 23, 2006 11:43 AM To: Song, Young Cc: veritas-bu@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] Weird tape drive behavior Which version of NetBackup? What OS? NetBackup 5.1 MP2 on Solaris had a problem with positioning I think. MP3 or MP3A fixed it. At least thats what our consultant told us. Austin On 3/23/06, Song, Young [EMAIL PROTECTED] wrote: I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] Solaris10 client on NBU4.5 FP8
Hi, Does the 4.5 FP8 client work on Solaris 10 ? /Steve ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] on the subject of disk based backups....
* Phil Rand [EMAIL PROTECTED] [2006-03-23 11:19]: I'm curious: Why do people use different volume pools? We have one pool for Exchange backups, and three others to go with our Iron Mountain weekly rotations, but I'm not sure what that's buying us. For tracking Iron Mountain boxes, we use volume groups, so the three pools for that are redundant. It's nice to be able to easily break out the Exchange tapes, but there are other ways of getting that information. So I don't think the case for separate volume pools is all that compelling for us. How do other folks use them? We back up 18 different customers on a shared 5.1 environment and use pools to separate the client data. -- David Rock [EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] on the subject of disk based backups....
* Ed Wilts [EMAIL PROTECTED] [2006-03-23 13:12]: On Thu, Mar 23, 2006 at 08:05:53AM -0500, Paul Keating wrote: Just wondering who is using DSSU, adn who is using VTL, adn why you made that choiceie.what were the key requirements for your environment that made one option better than the other. We're using DSSU. It was free and that was a contributing factor. At one point, I was a proponent of VTL technology here but since then, I've changed my mind and now believe that VTL is a niche market that will be killed off by the free DSSU functionality. Veritas originally had a fairly week implementation of its DSSUs and even in 5.1MP4, it's not great. It's supposed to be better in NBU 6, and I expect that it will continue to evolve and eventually kill off the VTL requirement for the vast majority of customers. I think that VTLs will work for us at least as good as DSSUs do. It's just that we don't see a need to throw extra $ at the problem today. The only thing that will kill VTLs is when DSSUs become shareable via SSO to multiple media servers. Until that happens, the only way to share the resource is for the virtual tape drives to be set up in SSO. DSSUs can't be shared (yet). -- David Rock [EMAIL PROTECTED] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] Seeking advise to convert Windows to Linux Master server
Anyone converted a Windows master server to a Linux master server? I am running NBU 5.1 MP3. What is the best way to accomplish this task as bprecover does not support dissimilar platform recoveries. Thanks Steve ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
RE: [Veritas-bu] Weird tape drive behavior
Hi, Young, Is there anything in /var/adm/messages? I'll bet there's a bit more detail about the nature of the error, if it's something returned from the drive or driver. You might want to use robtest to make sure the library robot can move tapes around and that it puts tapes in the correct drives. After you've eliminated a bad tape (media) and you're absolutely sure that everything is setup correctly, it is likely a hardware issue. Try swapping the drives around to see if the problem moves. (Assuming scsi direct attached drives). Have fun solving the problem and let us know what you find. Mickey -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Song, Young Sent: Thursday, March 23, 2006 2:29 PM To: 'veritas-bu@mailman.eng.auburn.edu' Subject: [Veritas-bu] Weird tape drive behavior I just installed a new tape library Sun L100 with 3 x LTO-2 drives and one of the drives keeps showing weird behaviors. I believe I have all the proper SCSI terminations, drive mapping, etc. It tries to mount a tape, position it, but sits there for a long time and fails with the following error: Positioning L01101 to file Error bptm (pid=9063) ioctl (MTWEOF) failed on media id L01101, drive index 0, I/O error (bptm.c.17692) Error bptm (pid=9063) FROZE media id L01101, could not write tape mark to begin new image When I tried using the mt command and run a tar backup on the drive, it went successful and was OK when I tried positioning a tape back and forth to see if I could back up or restore, so I tend to believe that the drive itself may not be bad. Any suggestion? Thanks! - Young ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Virtual Tape Library
[EMAIL PROTECTED] wrote: NetBackup 5.n was licenced by tape drives. NetBackup 6.n is licenced by TB in the VTL. I don't know the detail, e.g. is that raw TB etc. I also don't know if they changed the licencing for 5.n, or if it is still needing drive licences. The per-TB pricing for VTLs is available in 5.x. I know this because I purchased it. I based it on raw capacity since most VTLs are fairly efficient in their consumption. Obviously you do cheat yourself a little due to RAID overhead and such, but I don't think it's enough to lose sleep over... -- Michael L. Barrow michael at michaelbarrow dot name (Free security available at CAcert.org!) ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
