Re: Raid 1 vs Raid 5
I'll amplify what Skylar said: if your goal for this disk pool is short term storage then I probably wouldn't use any RAID protection as the data will be backed up to tape and then migrated to tape again. And as Skylar said, worst case, the client will send it again if it somehow escapes. Conserve space: don't RAID... Kelly J. Lipp O: 719-531-5574 C: 719-238-5239 kellyjl...@yahoo.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Skylar Thompson Sent: Monday, August 09, 2010 9:33 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Raid 1 vs Raid 5 Do you have tape in your primary storage hierarchy? If so, remember that even if part of your disk pool fails, you only lose access to the data that are on the failed volumes. You can then regenerate that data by either running another backup from the nodes that had backed up to that volume (if the backup to the copy pool hasn't happened yet) or from the copy pool. New backups can continue against the disk pool volumes that are still available, or can be cut through directly to tape if the entire pool is unavailable. On 08/09/10 08:23, Dana Holland wrote: Does anyone have opinions about setting up storage pools as Raid 1 as opposed to Raid 5? We have a very limited amount of disk space at the moment and don't know when we'll get approval to buy more. At the time we first started planning to implement TSM, we purchased what we thought would be plenty of storage. But, that was 4 years ago - and our usage has grown. Now, if I choose Raid 1, I barely have enough to create a primary and copy storage pool for one of our servers. And that isn't allowing for any growth at all. And I'm not sure how much additional space incremental backups would take. I know Raid 5 would give me more storage space, but I've also read that it's harder to recover from if there's a disk failure (read this on a TSM site somewhere). So, I'm wondering what some of you are using? __ Information from ESET NOD32 Antivirus, version of virus signature database 5352 (20100809) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com -- -- Skylar Thompson (skyl...@u.washington.edu) -- Genome Sciences Department, System Administrator -- Foege Building S048, (206)-685-7354 -- University of Washington School of Medicine
Re: PA-RISC system shelf life
To amplify what Wanda has already said about archiving products, I've been researching one from FileTek that has a ton of functionality. One bit relevant to this discussion is the ability to actually do archiving directly from Oracle. I don't know all the details, but this could start a pursuit for you. One other thing about long term database archival is to export the required data periodically to flat CSV (or CSV like) files and save those away. Worst case one can open these in a program like Excel and find the information. Brute force, very close to microfiche but possibly effective. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Prather, Wanda Sent: Tuesday, July 27, 2010 7:21 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] PA-RISC system shelf life The TDP for Oracle is just a connector/driver; it's actually Oracle RMAN that is selecting the data out of the data base on a backup, and writing it back to the DB on restore. So if you have issues restoring from PA-RISC to Itanium, that is actually a question for Oracle rather than TSM. But the answer to your question is NO. Don't plan on any server working for 20 years. TSM will happily store the data, because you can keep migrating it from one type of storage media to the next. But besides the question of keeping the server physically working, you will be reliant on the version of Oracle you have installed on it. If your restore fails (and again, it's RMAN handling the DB, not TSM), you aren't going to be able to get assistance from Oracle (or anybody else). I would also point out, if you have a requirement to keep the data for 20 years, it's probably not just 1 copy of the data- it's probably many copies over the years. If you do need to recover something 10 years from now, how are you going to FIND it? Restore every copy sequentially until you find what you're looking for? The way to handle legal retention of that sort of stuff is with one of the archiving products. (Tivoli has one, and there are others that use TSM as the backstore). They not only put the data out in a machine-independent form, they index it so you have a clue of finding it in 10 years. Barring implementation of an archiving system (which is not without cost), you need to have your DBA's set up some sort of periodic dump to a non-DB dependent format (e.g. an ASCII text flat file) and archive that. W -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Thomas Denier Sent: Tuesday, July 27, 2010 4:41 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] PA-RISC system shelf life -Duane Ochs wrote: - Thomas, Did you perform a TDP backup of the Oracle database to TSM? Or use a native oracle dump and backup the dump ? The backups were done with the software Tivoli used to market as TDP for Oracle, and now markets as part of TSM for Databases.
Re: why create a 12TB LUN
DEC Rainbow: 5MB. Upgraded to 10MB for about $500. Resides in Pueblo Reservoir as a boat anchor. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Steven Langdale Sent: Friday, May 28, 2010 4:46 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] why create a 12TB LUN Well If we're taking a trip down memory lane, I had an original IBM AT built like a tank! I used it up until a few years ago in the garage as a big step to get in the loft space. Steven Langdale Global Information Services EAME Storage Planning and Implementation CITA Backup Recovery Architect ( Phone : +44 (0)1733 584175 ( Mob: +44 (0)7876 216782 ü Conference: +44 (0)208 609 7400 Code: 331817 + Email: steven.langd...@cat.com Jacques Van Den Berg jvandenb...@pnp.co.za Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 28/05/2010 09:40 Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: [ADSM-L] why create a 12TB LUN Caterpillar: Confidential Green Retain Until: 27/06/2010 Had an original IBM 4.77MHz in 1991. 640KB Main memory. 360K Floppy drive a 10MB Hard drive. Kind Regards, Jacques van den Berg TSM / Storage / SAP Basis Administrator Pick 'n Pay IT Email : jvandenb...@pnp.co.za Tel : +2721 - 658 1711 Fax : +2721 - 658 1676 Mobile : +2782 - 653 8164 Dis altyd lente in die hart van die mens wat God en sy medemens liefhet (John Vianney). -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Skylar Thompson Sent: 27 May 2010 10:16 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] why create a 12TB LUN I'm around there too. 20MB Seagate MFM drive in an Epson QX-16. This was actually a dual-processor system (8088 for DOS and Z80 for Epson's CPM clone TPM). I had fired it up just for the heck of it a few years ago and it came up without problems. They don't make 'em like they used to. On 05/27/10 13:04, David McClelland wrote: I can beat than - I have a 20MB 'Winchester' HDD inside a working original Compaq Deskpro 8086 from c 1985. Fired her up last week for some photos, still works a treat. (No TSM client for it though...) /DMc Sent from my BlackBerry® wireless device -Original Message- From: Strand, Neil B.nbstr...@leggmason.com Date: Thu, 27 May 2010 14:45:51 To:ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] why create a 12TB LUN Gill, This sounds like an interesting environment. Could you share some of the particulars such as what storage device is providing the LUN, what server OS is using the LUN and what the general reason was for choosing the LUN? Historical note - My first hard disk in my home PC was 20GB Thank you, Neil Strand Storage Engineer - Legg Mason Baltimore, MD. (410) 580-7491 Whatever you can do or believe you can, begin it. Boldness has genius, power and magic. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Wednesday, May 26, 2010 7:04 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] why create a 12TB LUN I'm guessing many of you will find this quite odd, I know I did, but I had someone come to me and say they were going to ask for a 12TB LUN and wanted to back it up. Without even mentioning the product they want to use, obviously not TSM though, and I'm not even sure it would make difference, how would you manage to get a 12TB LUN backed up daily. I would expect it to be at least 75% full if not more, and even without knowing what percentage of data changes on it, it would seem to me the request seems strange. They're thinking of getting a VTL and backing up through fiber direct, not across the network, but no idea which one or what sort of throughput to expect. Have any of you been approached with this sort of request and if so what was your response? I'm sort of dumbfounded at this point since I've not heard or seen this anywhere. Thanks, Geoff Gill TSM/PeopleSoft Administrator SAIC M/S-B1P 4224 Campus Pt. Ct. San Diego, CA 92121 (858)826-4062 (office) (858)412-9883 (blackberry) IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy
Re: T950 Library experiences
I have about 10 years of experience with Spectra libraries and we are a reseller of them so take my comments with the appropriate grains. They are highly engineered libraries with tons of features. In general they are very reliable and Spectra support is quite good. I think Nick's comments reflect that as well. We have a couple of T950s in the field and they have been good to us and our customers. I have seen their brand new library (in fact I'm going to training tomorrow) and it is very nice. I think that's the T-Finity line. Very similar to the T950 but updated technology. These guys are very committed to tape. Steve, if you need more we can take this offline. I'm actually pretty objective when it comes to these guys. Over Sun? Take the Spectra every time. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nick Laflamme Sent: Wednesday, May 19, 2010 5:58 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] T950 Library experiences On May 19, 2010, at 1:21 AM, Steven Harris wrote: Hi All I'm looking at a Spectralogic T950 tape library instead of another Sun one to replace our aging Sun L700s Has anyone good or bad stories to tell about this? How does the user-replaceable spares offering work in practice? Are the tetrapaks robust or do they break easily? Is configuration and partitioning simple and straightforward? It's been a couple of years since I worked with one, but we used one as a remote tape library locked away in a wiring closet across campus. We didn't do any of our own servicing, and I don't think we partitioned the library at all, but the tetrapaks were just fine. We had a period when, as I recall, the library would start to experience intermittent hangs. Shortly before I left that employer, SpectraLogic flew someone out who determined that a good realignment would fix the problem; she was right. (That may have been related to us physically moving the library on our own from one wiring closet to another a couple of hundred yards away. I'm not saying it was, but I have to wonder.) SpectraLogic was moving customers to SuperDAT drives to IBM LTO drives as I moved away from that shop. That's been more than two years; I presume they're stable on their drive choice now. Six months ago, when my current employer was getting ready to get off a particular virtual tape library technology, I would have been thrilled if we'd gone to T950s as our library type. I have good memories of the one I used. (Instead, we simply changed virtual tape library families, but that's for another day.) It's hard to beat the density of a T950. Good luck, Nick
Re: two libraries in a storage pool?
Your answer is correct. You can't have a single pool in two libraries. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Mehdi Salehi Sent: Wednesday, April 07, 2010 3:22 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] two libraries in a storage pool? Hi all, Is there any way to assign tape drives in two separate LTO libraries in a single storage pool? I see in TSM reference that each storage pool has one device class and to each device class only one library can be assigned. My answer to above question is no. Am I mistaken? Thanks.
Re: Virtual TSM server - using disk only
Duane, Works, but isn't supported. So if/when it doesn't work... Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Ochs, Duane Sent: Thursday, March 11, 2010 12:19 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Virtual TSM server - using disk only Good day everyone, Has anyone explored using TSM server (windows) on a VM using Iscsi storage ? No library requirement at this time. I have multiple European sites within close proximity of each other and they have outgrown the WAN coming back to the states. Only storage available there is Iscsi and they have a substantial VMware implementation which would allow us to ride on a VM if feasible/functional. Thoughts ? Thanks, Duane
Re: Virtual TSM server - using disk only
My comment concerned the VM portion of the question, not the iSCSI portion. I concur with Gary on that. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gary Bowers Sent: Thursday, March 11, 2010 12:37 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Virtual TSM server - using disk only My experience with direct connected iSCSI storage on a TSM server is that it gets abysmal performance unless you turn off Direct IO in TSM. See other posts for that. It is technically possible, but with the iSCSI limitation you might not want to use RMD Raw Device Mapping in VMware. I am not sure on this, but it makes sense given what I have seen and read about here. By the way, NFS and CIFS were equally bad performers for disk pools with DirectIO turned on. They seem to really need the filesystem caching. I'm guessing that putting the disks in a VMFS would help buffer the writes, and give you decent performance. It is something that would need to be tested first. I'm confident that it would be much faster than WAN connection back to the States. Yuck. Good luck, Gary Bowers Itrus Technologies On Mar 11, 2010, at 1:18 PM, Ochs, Duane wrote: Good day everyone, Has anyone explored using TSM server (windows) on a VM using Iscsi storage ? No library requirement at this time. I have multiple European sites within close proximity of each other and they have outgrown the WAN coming back to the states. Only storage available there is Iscsi and they have a substantial VMware implementation which would allow us to ride on a VM if feasible/functional. Thoughts ? Thanks, Duane
Re: Data retention period 60days
Figure out which management class is governing your Oracle backups. Q node nodename where nodename is the name the client uses to identify itself. Check to see which policy domain the node is in. Q copy domainname f=d You'll probably see verexists=2, verdel=1 retextra=30 retonly=60. The most accurate settings for a 60 day retention (which is a long time for a database IMHO) should be verexists=60, verdel=60 retextra=60 retonly=60 So, Update copy domainname policysetname magagementclassname STANDARD verexists=60 verdel=60 retextra=60 retonly=60 Then activate the policyset: Activate policyset domainname policysetname And you're done. It is likely that the policyset and managementclassname is STANDARD as well. If I had set up your system the commands would look like Update copy ORACLE STANDARD STANDARD STANDARD verexists=60 verdel=60 retextra=60 retonly=60 Activate ORACLE STANDARD Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Sujatha chk Sent: Thursday, January 21, 2010 12:43 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Data retention period 60days Hi Gurus, we backup our oracle database using rman to TAPE by TSM media management. Currently Backup Retention policy has been set to 30 days. Please guide me on how to configure Back up Retention period to 60 days? -Sujatha +-- |This was sent by sujatha@gmail.com via Backup Central. |Forward SPAM to ab...@backupcentral.com. +--
Re: DataDomain VTL
We have a customer that insisted on buying one of these for his TSM environment. Promised 20:1 dedup. He saw about five to one. He was in our Level 2 class telling the story. At the end he said he wouldn't buy it again. I made him repeat that part of the story... Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Howard Coles Sent: Monday, January 11, 2010 2:58 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] DataDomain VTL Is anyone out there using a DataDomain VTL? I'm getting some pressure to look at this, and I'd like to find some honest opinions of them. I know that some time back there were some conversations around this, but some tech has been updated and DD has been bought by EMC, etc. So, if you have one, and would like to share your opinion I'd appreciate it. See Ya' Howard Coles Jr. Sr. Systems Engineer (615) 296-3416 John 3:16!
Re: More than 1 EXPORT NODE per tape
Stacked. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Thursday, January 07, 2010 12:34 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] More than 1 EXPORT NODE per tape Are they stacked on the same tape? I thought they were put on separate tapes? Regards, Shawn Shawn Drew Internet boatr...@memorialhealth.com Sent by: ADSM-L@VM.MARIST.EDU 01/07/2010 02:25 PM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] More than 1 EXPORT NODE per tape Export node node1,node2,node3 _ From: Mario Behring [mailto:mariobehr...@yahoo.com] Sent: Thursday, January 07, 2010 2:13 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] More than 1 EXPORT NODE per tape Hi list, Is it possible to use put more than 1 export on the same tape? I executed one export but there is room for much more data on the tape.I think it is not possible.didnt see anything that made me believe otherwise.so Ím asking... Mario This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.
Re: alternatives to TSM due to license costs
At least let IBM know you are thinking about jumping ship and that they need to help with the pricing on the new licenses or you are gone! You have the leverage at this point so you might as well use it to your advantage. Based on our many competitive situations along the way with our appliance we do this all the time. For the low end, TSM is not a good fit generally. However, if you have more than 50 clients then TSM is a good fit for the reasons Remco states. Taking a giant step back to the 20th century with your backups (which is what you'll be doing with any other product) is generally a bad idea. Oh, IBM! Are you listening? The current licensing scheme is just killing us! Stop the madness before we lose more of our installed base. Stop the madness before I lose another Appliance deal! Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Remco Post Sent: Wednesday, December 30, 2009 3:15 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] alternatives to TSM due to license costs On 30 dec 2009, at 20:11, woodbm wrote: Hello all, I have been tasked with looking at alternatives to TSM due to recent Audit from IBM and the amount of money we just shelled out for the TSM License. I am sure most of you have pertaken to this wonderful experience. I don't really want to move away from TSM, but have to provide alternatives to management. Could anyone direct me to where I can begin my search. I have only used TSM my entire time here. What is the industry leader? Any info or documents or links doing a comparison would be helpful. Also, is the license structure the same for other environments as well or is IBM way out of line? I am going to start with EMC's Avamar/Networker. there are a few products that people try to compare TSM to; Networker, Netbackup and CommVault come to mind. Keep in mind that all of these require full backups at some interval, occupying much more tape that TSM does. So when presenting alternatives, do not only consider licensing cost but also the cost of media, servers, network etc. Also keep in mind that some products are disk to tape only, requiring you to have at least one tape drive for each concurrent backup/restore session, or implement expensive VTL solutions while TSM doesn't need one. (The IBM VTS for the mainframe was running ADSM for a reason!). Licensing schemes for these products are completely different. While IBM charges for cpu's, competitors also charge for functionality, such as archiving? drm? The number of volumes you can manage in a library? The number of drives? etc. Look into what you need and make sure you have the total picture. As a word of advise, make sure that the audit is correct. I've seen IBM try to charge customers for TDP for databases for DB2, while everybody knows that there is no such product! If in doubt, challenge the audit results, this might be worth the effort. Thanks much, Bryan +-- |This was sent by woo...@nu.com via Backup Central. |Forward SPAM to ab...@backupcentral.com. +-- -- Met vriendelijke groeten/Kind Regards, Remco Post r.p...@plcs.nl +31 6 248 21 622
Re: IBM bureaucracy strikes again
I particularly enjoy the ten product limit. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Tuesday, December 22, 2009 7:57 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM bureaucracy strikes again Yet another change. Every time they improve things, they make it harder for us to find the same information we used to be able to get at. I'll give this a try before passing judgement. It looks like they are going to a portal model where you can customize your own experience. At 08:30 AM 12/22/2009, Richard Sims wrote: If you haven't been to the TSM Support Page (http://www.ibm.com/software/sysmgmt/products/support/IBMTivoliStorageManager.html) very recently, you'll be dismayed to learn that this helpfully product-specific page is going away - being replaced by a generalized facility which you get to from the current TSM Support Page by scrolling down a very long list of products to find TSM, then select one category of things you want to see, then wait for their website to grind and produce a page with just that information. You can navigate to other sub-areas from the left pane, but it can entail considerable delays. And because it's generalized, there's no page title to quickly let you know you're looking at TSM stuff. Someone at IBM will probably get a bonus this year for thinking up this method of improving IBM's internal web pages organization ... by obliterating the tailored product page that we all found so useful over the years, where we could quickly get at information we needed. IBM management's focus seems to be on satisfying organizational directions more than best meeting customer needs. Richard Sims -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: IBM bureaucracy strikes again
I'm fed up and I won't take it anymore! Oh, shoot. Where will I go? Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Wanda Prather Sent: Tuesday, December 22, 2009 9:36 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM bureaucracy strikes again OK. Can someone on the list, or someone from Tivoli who monitors this list, give us a contact point within IBM where we can take these complaints? We should be bombarding them directly... On Tue, Dec 22, 2009 at 11:26 AM, Kelly Lipp l...@storserver.com wrote: I particularly enjoy the ten product limit. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Tuesday, December 22, 2009 7:57 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM bureaucracy strikes again Yet another change. Every time they improve things, they make it harder for us to find the same information we used to be able to get at. I'll give this a try before passing judgement. It looks like they are going to a portal model where you can customize your own experience. At 08:30 AM 12/22/2009, Richard Sims wrote: If you haven't been to the TSM Support Page ( http://www.ibm.com/software/sysmgmt/products/support/IBMTivoliStorageManager.html) very recently, you'll be dismayed to learn that this helpfully product-specific page is going away - being replaced by a generalized facility which you get to from the current TSM Support Page by scrolling down a very long list of products to find TSM, then select one category of things you want to see, then wait for their website to grind and produce a page with just that information. You can navigate to other sub-areas from the left pane, but it can entail considerable delays. And because it's generalized, there's no page title to quickly let you know you're looking at TSM stuff. Someone at IBM will probably get a bonus this year for thinking up this method of improving IBM's internal web pages organization ... by obliterating the tailored product page that we all found so useful over the years, where we could quickly get at information we needed. IBM management's focus seems to be on satisfying organizational directions more than best meeting customer needs. Richard Sims -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Move Windows filespaces
I think it will certainly be easier to re-backup the data. Faster? Not sure, but probably as fast as an export would go. I guess I wouldn't over think this: set up new client after the data is moved and do the backup. Probably be done before you know it. You will obviously lose the previous versions of files, but you could keep the old filespaces around until the data expires. If this is mostly on tape I'd opt for this approach. Kelly Lipp Chief Technology Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of David Longo Sent: Monday, December 14, 2009 1:40 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Move Windows filespaces I have TSM sever 5.5.2.0. Also have a W2K3 client with version 5.5.0.0 BA client, doing backups only. This has about 2 dozen drives and therefore filespaces. There is many TB of data involved in total and these are all SAN disks. (Also many TB of data that is left, so not easy to say redo that either.) For several technical reasons, they need to move about half the SAN disks to another server and leave half on existing server. They will just rezone those disks to the new server, no copying involved. I certainly don't want to backup all this moved data again. If was ALL being moved to new server, I could just RENAME NODE and FILESPACES and be done with it, have done that before. But is not the case. As I see it, my only option is to EXPORT NODE for the filespaces that are being moved and then import again into new node. I have never used Export Node, but have almost a few times over the years. Am I correct that this is my only option? Other than backing up in new location and deleting in old - or would that be faster and easier? NO idea on the speed this works. I have 3584 library with LTO2 and 3 drives/tapes. Thanks, David Longo Health First, Inc. # This message is for the named person's use only. It may contain private, proprietary, or legally privileged information. No privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it, and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Health First reserves the right to monitor all e-mail communications through its networks. Any views or opinions expressed in this message are solely those of the individual sender, except (1) where the message states such views or opinions are on behalf of a particular entity; and (2) the sender is authorized by the entity to give such views or opinions. #
Backing up Oracle Hyperion Essbase
Folks, I've researched and mostly determined that backing this guy up is a matter of closing it down (or putting it into readonly mode) and backing up the files. I saw a couple of entries from folks doing this several years ago. Our competition for this is EMC Avamar/Data Domain. Getting hurt by source side deduplication which we'll have down the road. If any of you are still out there listening and can provide me with a bit of information, I would appreciate it. Private email or the phone number below. My indebtedness will be immeasurable! Kelly Lipp Chief Technology Officer www.storserver.comhttp://www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all.
Re: TSM 6.1 De-dupe + NDMP
If it's true that the Celera compresses then it probably won't look like the previous version of the file so it won't deduplicate. Then the question becomes: maybe don't compress but rather send all the bits and let it get deduplicated - will this result in better storage utilization? From: ADSM: Dist Stor Manager [ads...@vm.marist.edu] On Behalf Of Wanda Prather [wanda.prat...@jasi.com] Sent: Thursday, December 03, 2009 10:00 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM 6.1 De-dupe + NDMP Is the NDMP backup going via TCP/IP and into the same storage pool as your other backup data? How can you tell it is not getting deduped? On Thu, Dec 3, 2009 at 11:54 AM, Brian P. Boyd boydb...@duke.edu wrote: Hello, Currently we are using TSM 6.1 to do NDMP backups of our EMC Celerra. We are getting no de-duplication results from these backups. I'm wondering if it is because the celerra is already sending a compressed image snapshot file to TSM and TSM just doesn't de-dupe a file like that. We're thinking about going to a more traditional backup method for NDMP backups (going with weekly incrementally, and monthly fulls...etc) however it would be good to understand why this is happening. I find it a bit strange because we still do file-level restores just fine... If someone could educate me a bit more that would be great for piece of mind! Thanks! Brian P. Boyd Sr. SAN Admin DUKE - OIT boydbria (at) duke (dot) edu
Re: Copy Tape pool for a selected set of tapes
I'll echo what Richard said and amplify it: TSM requires planning. As it is about 1000 more powerful and flexible than anything else you might work with, you must think about what your business requirements are before implementing it. Then do the right thing. The next coolest thing about the product is the fact that if you don't do the right thing, with a little thought you can change it! Usually no harm no foul. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Sims Sent: Wednesday, November 25, 2009 7:38 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Copy Tape pool for a selected set of tapes This is a plan-ahead issue. Nodes which need special handling like that need to be assigned their own storage pool, where your site should have distinct management classes for data having differing requirements. You can define such a storage pool, create a management class copy group pointing to it, and define a management class for the node to use henceforth. The MOVe NODEdata command can be employed to shift existing data to that separate storage pool, whereafter the desired Backup Stgpool operation can be performed. If the data is static (no new incoming files), another approach could be to define a new stgpool at the bottom of the current storage pool hierarchy, with no migration into it, then MOVe NODEdata into that pool, and do a Backup Stgpool on it. This would avoid the need for a new management class (though you should still pursue that). Richard Sims
Re: tape library for TSM
Brands? IBM 3584 for the big boy! Great library, but somewhat expensive. You can work your dealer on price. I like the very small IBM libraries for folks that can get by with very little library. Qualstar XLS for a more cost effective big boy. Very reliable and very good value. TLS family been around a very long time and very reliable as well. If footprint isn't an issue, the TSL88132 or 264 is probably the least expensive, highest capacity library available. Spectralogic T950. Great technology, not a cost leader but perhaps a bit less than the IBM. The other T380 family of libraries is more cost effective and flexible enough for most of us. T120 is still a workhorse and if all you'll need is 120 slots and six drives, perhaps the best rack mount library in the class. And as Wanda suggests: it has a 10 slot I/O port. Overland? Not wild about that brand. And the rest are just variants of the same OEM in many cases. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Wanda Prather Sent: Sunday, November 15, 2009 3:12 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] tape library for TSM And in most cases, look for a library with more than 1 I/O slot. On Sat, Nov 14, 2009 at 2:59 PM, madunix madu...@gmail.com wrote: On Sat, Nov 14, 2009 at 8:02 AM, Mehdi Salehi iranian.aix.supp...@gmail.com wrote: What are the essentials of a tape library to fully work with TSM? - enough tape drives - barcode reader - cartridge barcode labels - FC connection for LAN-free ...? All above!
Re: tape library for TSM
On the supported device list... Perhaps that's obvious, but I would start there. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of madunix Sent: Saturday, November 14, 2009 12:59 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] tape library for TSM On Sat, Nov 14, 2009 at 8:02 AM, Mehdi Salehi iranian.aix.supp...@gmail.com wrote: What are the essentials of a tape library to fully work with TSM? - enough tape drives - barcode reader - cartridge barcode labels - FC connection for LAN-free ...? All above!
Re: TSM DB Size
The architectural limit is 500GB. Practically, one should be a good bit smaller. It really boils down to how long do you want a restore of the backed up DB to take? Figure about 150% of the backup time for a restore. Can you live with that? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebschman, George J. Sent: Tuesday, November 10, 2009 9:42 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM DB Size We have DB's over 190 Gb in 5.5, on AIX servers: TSM_Server CAP_GB MAX_EXT_GB PCT_UTIL MAX_UTIL --- --- -- AIXPRODXYZ 191.95 0.00 89.4 89.4 AIXPRODDMZ 219.76 4.17 90.3 91.1 -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Mochnaczewski Sent: Tuesday, November 10, 2009 11:35 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] TSM DB Size Hi *, Does anyone know what the maximum size database is for TSM 5.4 and TSM 5.5 ? We were told by IBM when we were at TSM 5.3 that 120Gb was the limit and was wondering what the limit is for 5.4 and 5.5. Currently running TSM 5.4.3.0 on AIX 5.3 TL 9. Rich IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error, please notify the author by replying to this message and then kindly delete the message. Thank you.
Re: TSM DB Size
I stand corrected, as usual, by Richard! 500, 530, big diff. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Sims Sent: Tuesday, November 10, 2009 9:47 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM DB Size The Admin Guide manual says that the maximum size for the db is 530 GB.
Re: TSM DB Size
And most importantly, if you get a room full of TSM gurus, IBM types and folks like us, nobody will really agree on a hard number. What I have seen during 10 years and hundreds of TSM environments is that the number has gradually increased. When I first got into this business, 50GB was huge. Now that's nothing. The more normal is 100-150GB and things seem to be working just fine. I recall a period of time, soon after Dave Cannon arrived, that IBM's engineering focus was on quality. They did a ton of great work fixing what ailed the product. We could see the results afterwards. This work set the base for what we're seeing today (up to 5.5.x anyway). Couple that with huge improvements in hardware performance and our favorite product has grown up nicely. Our problems are so much different than everybody else's. They're still worried about getting the weekly full backup done they can't think about anything else. That or adding the next freaking band-aid to their already half assed (and declining) solution. You can sure tell when I'm working on a customer presentation can't you? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Tuesday, November 10, 2009 10:10 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM DB Size 100-120GB was the IBM recommended limit based on their tests on a baseline AIX system (Specs are lost to history). I talked to the Watson Research guys at a Symposium quite a while ago on this. (The guy was from one of their east coast research places, and I'm assuming it was Watson. He was with IBM Global Services and described his job as basically playing with all the hardware that came out and wrote reports and guidelines for them.) The idea was that if you have a faster system than their baseline, you can go higher. It was a general rule-of-thumb type of thing. You can decide for yourself on your own hardware with your own data based on how long your expirations/db backups take and if that's acceptable to you. Regards, Shawn Shawn Drew Internet richard.mochnaczew...@standardlife.ca Sent by: ADSM-L@VM.MARIST.EDU 11/10/2009 11:51 AM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] TSM DB Size We had a health check done by IBM when we were at TSM 5.3 and were told they don't recommend a DB size higher than 120Gb for performance and restore purposes. Rich -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Tuesday, November 10, 2009 11:48 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM DB Size The architectural limit is 500GB. Practically, one should be a good bit smaller. It really boils down to how long do you want a restore of the backed up DB to take? Figure about 150% of the backup time for a restore. Can you live with that? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebschman, George J. Sent: Tuesday, November 10, 2009 9:42 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM DB Size We have DB's over 190 Gb in 5.5, on AIX servers: TSM_Server CAP_GB MAX_EXT_GB PCT_UTIL MAX_UTIL --- --- -- AIXPRODXYZ 191.95 0.00 89.4 89.4 AIXPRODDMZ 219.76 4.17 90.3 91.1 -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Mochnaczewski Sent: Tuesday, November 10, 2009 11:35 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] TSM DB Size Hi *, Does anyone know what the maximum size database is for TSM 5.4 and TSM 5.5 ? We were told by IBM when we were at TSM 5.3 that 120Gb was the limit and was wondering what the limit is for 5.4 and 5.5. Currently running TSM 5.4.3.0 on AIX 5.3 TL 9. Rich IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error
Re: de-duplicating compressed data
That assumes that the compression occurs file by file. Is that true or is on the transaction. I suppose it is on the files themselves and all clients would compress the file into the same set of bits. If it doesn't do that though, then your high dedup rates won't be realized. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Grigori Solonovitch Sent: Saturday, November 07, 2009 9:16 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] de-duplicating compressed data What is the effect of compression on de-duplication? Does it help to reach a more de-duplication level? This is my opinion (please correct, if something is wrong): 1) note we are talking about client compression (compression=yes for node or in dsm.opt). Hardware compression on drive level is tottally independent from dedup process; 2) client compression can be used for any primary storage pool (device type DISK, FILE or any tapes). In this case, compressed data is comming to copy pools as well and you need less number of tapes in copy pools; 3) client compression takes time during backups (backups are much longer), but amount of data sent to TSM server via network is much less (average compression rate is 2-4 times); 4) deduplication is working only with primary sequential disk storage pool (device class FILE) and can give compression rate 10-20 and more. Deduplication process is working with data from all nodes (not only from one) and compares ALL to ALL. So just imagine which comression rate you can reach for some cases, when there are a lot of similar Windows servers (like server in each bank branch) with the same level of Windows and the same applications. For 50 branches you can have compression rate 40; 5) I see only one reason why deduplication is only working with FILE and is not working with DISK - after software deduplication you need to run reclamation to release space. Reclamation is not applicaple for DISK with random access. By the way, this question is still open and only IBM can anwer, what is the real reason; 6) there is special protection for data in TSM server. Deduplication is not working with data, if there is less than 2 copies on tapes. So sequence of actions is: backup data to DISK, make at least 2 copies of data to tapes (without deduplication!!), start deuplication and start reclamation. Deduplication will never reduce data on copy pools; 7) deduplication and compression are working together, but overal compression rate will be more than with only compression, but much less than with only deduplication. For example, you will have compression rate N for compression only (backups and all copies), M for deduplication only (only backups, copies have full size) and K for compression/deduplication (K for backups and N for copies). In general, N is much less than M, K is more than N and less than K. Real values for N, M and K depend on type of data; Regards, Grigori Please consider the environment before printing this Email. This email message and any attachments transmitted with it may contain confidential and proprietary information, intended only for the named recipient(s). If you have received this message in error, or if you are not the named recipient(s), please delete this email after notifying the sender immediately. BKME cannot guarantee the integrity of this communication and accepts no liability for any damage caused by this email or its attachments due to viruses, any other defects, interception or unauthorized modification. The information, views, opinions and comments of this message are those of the individual and not necessarily endorsed by BKME.
Re: Performance and migration: AIX vs Linux
I studied this in some detail as it obviously comes up often in my practice. If one reads the IBM documentation on the latest x3850/x3950 M2 one will observe that the data paths within that architecture are actually faster than in the latest pSeries hardware. That said, Windows or Linux inefficiencies probably just balance out. Shawn's comment about multiple $25K x3850 vs. one $100K p is valid. I think that clearly multiples of those will outperform on p. I think this argument should morph to what OS is your shop using predominantly as you will have to support the OS while fooling with TSM. And perhaps even more with V6. The difference in relative performance is probably in the single digits and thus undetectable by most of us. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Friday, October 23, 2009 8:58 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Performance and migration: AIX vs Linux # IMNSHO, IBM pSeries hardware is the best there is for large I/O # workloads. I've seen AIX do things that Linux wouldn't survive. I've always wondered about this. We have p570s and we can throw anything at them, and they won't even breath hard. But if you spent $100K on p-series, and $100K on multiple Intel machines, which solution will get the better, cumulative I/O-Dollar ratio? This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.
Re: Performance and migration: AIX vs Linux
Paul, Yes, the bus structures are similar and in the x M2 variants faster. I just attempted to lay my hands on a the technical paper I read about this to no avail and I'm struggling to recreate the search I used to find it originally. When I went after that is was my intent to confirm that the overall bus structures were indeed similar and that is what I found. I'll try to find that document again and post it when I do. Thanks, Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Friday, October 23, 2009 12:43 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Performance and migration: AIX vs Linux Kelly, At 11:27 AM 10/23/2009, Kelly Lipp wrote: If one reads the IBM documentation on the latest x3850/x3950 M2 one will observe that the data paths within that architecture are actually faster than in the latest pSeries hardware. I'm curious - does your analysis include the number of busses available? I know that the pSeries has quite a lot of busses to drive all of its adapters. I'm not as familiar with the xSeries, so I don't know if there is a model available that has a comparable number of busses. It's not just the speed, it's the overall throughput that most concerns me. Sure, I could have more smaller xSeries servers, but then I would have more resource baskets to manage. Thanks for sharing. ..Paul -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Performance and migration: AIX vs Linux
I found it! http://www.redbooks.ibm.com/redpapers/pdfs/redp4362.pdf Actually a fairly fun read. Way cool technology. I'm sure it will show up in the pSeries soon anyway so the hardware issues become further blurred. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Friday, October 23, 2009 12:43 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Performance and migration: AIX vs Linux Kelly, At 11:27 AM 10/23/2009, Kelly Lipp wrote: If one reads the IBM documentation on the latest x3850/x3950 M2 one will observe that the data paths within that architecture are actually faster than in the latest pSeries hardware. I'm curious - does your analysis include the number of busses available? I know that the pSeries has quite a lot of busses to drive all of its adapters. I'm not as familiar with the xSeries, so I don't know if there is a model available that has a comparable number of busses. It's not just the speed, it's the overall throughput that most concerns me. Sure, I could have more smaller xSeries servers, but then I would have more resource baskets to manage. Thanks for sharing. ..Paul -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Performance and migration: AIX vs Linux
Seven slots, but you can stack up to four servers to get to 28 slots. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Friday, October 23, 2009 2:48 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Performance and migration: AIX vs Linux I didn't read the whole thing, but it looks like there are only 7 I/O slots? Does it have RIO drawers similar to the higher end pSeries? At 03:46 PM 10/23/2009, Kelly Lipp wrote: I found it! http://www.redbooks.ibm.com/redpapers/pdfs/redp4362.pdf Actually a fairly fun read. Way cool technology. I'm sure it will show up in the pSeries soon anyway so the hardware issues become further blurred. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Friday, October 23, 2009 12:43 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Performance and migration: AIX vs Linux Kelly, At 11:27 AM 10/23/2009, Kelly Lipp wrote: If one reads the IBM documentation on the latest x3850/x3950 M2 one will observe that the data paths within that architecture are actually faster than in the latest pSeries hardware. I'm curious - does your analysis include the number of busses available? I know that the pSeries has quite a lot of busses to drive all of its adapters. I'm not as familiar with the xSeries, so I don't know if there is a model available that has a comparable number of busses. It's not just the speed, it's the overall throughput that most concerns me. Sure, I could have more smaller xSeries servers, but then I would have more resource baskets to manage. Thanks for sharing. ..Paul -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Create holes in data
That would be the basic idea. Explore the command syntax to see if you can do for more than one client at a time. 500TB is a bunch of data to wade through, but you knew that! Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Strand, Neil B. Sent: Thursday, October 15, 2009 1:14 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Create holes in data I have a question - please do not ask for an explination - it is what the customer wants. Environment: -TSM Server 5.4 on AIX -TSM BA client on Linux, Solaris and Windows (various flavors) -TS3500 tape library with TS1120 drives (+1TB on each tape) The customer desires me to convert the last couple of years of backup data - all of which has 3 year retention - to look like periodic full backups. - Data between 1 and 6 months old - only the end of month backup - Data older than 6 months - only one backup at the end of the year Similar to a traditional backup where fulls are periodically taken The customer's data from their 160 clients occupies nearly 500TB of tape space (primary only) and is looking to cut down on the number of tapes that they would be required to purchase. This will be a one time effort since the customer is now performing their own backups and I would like the customer to take custody of this legacy data. One thought is to create backup sets of each client at the specified points in time. Your ideas are appreciated. Thank you Neil Strand Storage Engineer - Legg Mason Baltimore, MD. (410) 580-7491 Whatever you can do or believe you can, begin it. Boldness has genius, power and magic. IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error, please notify the author by replying to this message and then kindly delete the message. Thank you.
Re: Migration for Windows-based installation
Start from scratch. Install TSM at the level you would like, move and configure the library. Point the clients at the new server and go. That first backup will necessarily be a full. Will take longer than usual, but easy. This method allows you go clean up your database. Keep the old server around until the data expires. Clearly, there are more details, especially since you are going to re-use the library on the new STORServer (oops, I mean TSM Server). The overall concept is sound. For sites of your size I like this method as it is simple and gets me a brand new, pristine database. This will become important next year when you migrate to TSM 6.2. Besides, a clean start is always a good thing. This topic has been covered earlier and in much more detail. You might peruse the archives to see what you can find. My name will show up along with others like Wanda who have been through this many times. Thanks, Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kurt Buff Sent: Monday, October 12, 2009 1:02 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Migration for Windows-based installation I'd prefer to be able to migrate server data as much as possible, including the diskpool, but if it it would be an incredibly difficult maneuver (or take inordinate amounts of time, say on the order of more than a 4-day weekend) and/or the downside of losing the diskpool is relatively minor, I could potentially live without it. I would also contemplate keeping the current TSM server version on the new machine and upgrading immediately after implementation if that would be of benefit. We back up fewer than 10 servers, but one is our Exchange 2003 server at roughly 200gb and is a full backup every night, and the other is our file server at over 2tb, though on a nightly basis it normally does around 25-50gb. Does that answer your question? Kurt On Mon, Oct 12, 2009 at 11:18, David McClelland t...@networkc.co.uk wrote: Are you looking purely at a lift and shift hardware change here, or at a clean installation of TSM Server (at 5.5.3 for example) and migrating clients into the new instance? Cheers, /David Mc London, UK -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kurt Buff Sent: 12 October 2009 18:54 To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Migration for Windows-based installation All, We have a TSM server that's running out of steam, and nearing the end of its expected reliable life. We have Version 5, Release 3, Level 4.6 installed, with various levels of clients installed on our servers. The server has 1gb RAM, roughly 2tb of disk storage, but it's old/slow PATA, and the OS is Win2k Pro. Definitely not ideal. The tape robot is a Spectralogic T50, with two LTO3 drives and 25 slots, out of which we expect to get much more life. We expect to replace the server with a new Dell server with 3tb of SATA disk, 3gb RAM, Win2k3 (32bit), and use the current tape robot. I'd like to get to the newest in the 5.x series on the new server, since the talk on this list about moving to 6.x indicates to me that we'd be better off staying at 5.x for now. I've been casting about, and can't seem to find documentation on how to migrate the setup to the new machine. Does anyone have a pointer to good documentation on doing this? Frankly, we've been given a quote by a VAR, and though the number of hours they are quoting seem reasonable, the price they're asking to work with us on this is beyond our budget. Thanks Kurt No virus found in this incoming message. Checked by AVG - www.avg.com Version: 8.5.421 / Virus Database: 270.14.3/2415 - Release Date: 10/12/09 04:01:00
Re: AW: TSM 6.1 and the ever expanding DB
That last paragraph made my head hurt! I had the opportunity to take a database class in college. Didn't want to know it then, don't want to know it now. I recall one of the design centers for the DB2 thing was to ensure that a TSM admin didn't need to become a DB2 admin. I don't even know the lingo! I'll echo Rick's comments: you pioneers, you go! Those arrows don't hurt that much. That which doesn't kill you makes you and all of us stronger. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Rhodes Sent: Friday, October 02, 2009 9:26 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] AW: TSM 6.1 and the ever expanding DB I've been watching this discussion with great interest, and more than a little fear. We are going to implement the v6 ISC/SC shortly on a standalone Win server, but we aren't planning to upgrade the TSM servers until next year. A BIG thanks to all you bleeding edge types out there IBM has a interesting/hard problem - TSM is used to backup TSM. I assume the requirement for multiple backups before a archive log is deleted is to ensure that multiple backups occur for each archive log.They are effectively throwing disk space at the archive logs to ensure they have good over lapping backups of them. I wonder if IBM isn't eventually going to have to implement some process that will periodically backup archive logs, make a second copy of them on different media, generate a Vol_Hist, a prepare(?), then delete the archive logs. In other words, on some trigger make a good backup of all archive logs (multiple copies on separate media) and everything needed for a recovery from the last full - roll forward through archive logs - then the active log. Rick Zoltan Forray/AC/VCU zfor...@vcu.edu To Sent by: ADSM: ADSM-L@VM.MARIST.EDU Dist Stor cc Manager ads...@vm.marist Subject .EDU Re: AW: TSM 6.1 and the ever expanding DB 10/02/2009 10:52 AM Please respond to ADSM: Dist Stor Manager ads...@vm.marist .EDU Good luck with the PMR. Let us know how it works out. As I understand it, currently this is WAD (requiring 2+ full backups and corresponding volhist backups before clearing the archivelog). It is also my understanding (from my last conversation with an L2 tech at IBM) that they know this is a problem and hope to reduce the requirements, in future (V6.2?) releases. They still have numerous DB2/TSM interaction bugs to squash, first! From: Stefan Holzwarth stefan.holzwa...@adac.de To: ADSM-L@VM.MARIST.EDU Date: 10/02/2009 10:16 AM Subject: [ADSM-L] AW: TSM 6.1 and the ever expanding DB Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU We want also to go into production with 6.1.2. All setup is finished. But with about 20 nodes (all export/import) we continously have trouble with full active log and full archivelog. Our active log size is 16Gbyte and it seems to be enough for this small setup. But sometimes the log usage explodes and goes rapidly up to the limit. I opened my second PMR Regards Stefan Holzwarth -Ursprüngliche Nachricht- Von: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] Im Auftrag von Zoltan Forray/AC/VCU Gesendet: Freitag, 2. Oktober 2009 15:15 An: ADSM-L@VM.MARIST.EDU Betreff: Re: TSM 6.1 and the ever expanding DB Join the club. I am beginning to wonder if anyone is successfully using V6.1, trouble-free. Monday I
Open Letter to TSM Product Mangement. Was Per terabyte licensing
This has been a good discussion. I would like to change the tone a bit in order to help IBM product management as they ponder this issue. STORServer is an OEM of IBM TSM code and TSM is an integral part of our appliance. We compete in the marketplace against just about everyone else in the backup space. The most difficulty we encounter is with respect to our licensing which is necessarily identical to IBMs. I have thought long and hard about how to decouple client licensing from our product and stay in compliance with our OEM agreement. I have not come up with an idea. I postulate the following: a TSM client derives value from the TSM environment in two ways: 1. simply by having the ability to store and restore data on the TSM server and 2. from the intrinsic features the server uses to store maintain that data. Some clients use server features relatively less while others use them relatively more. The features used in the server are relevant to the overall business requirements rather than for a single client. At STORServer, we asses this value by determining how much it costs us to support an environment. We can expect to field a certain number of support calls per customer with client side issues and certain number with server side issues. The more clients a customer has, the more calls we’ll get and the more sophisticated the server side is (larger library, more disk, server to server, etc.) the more server side calls we'll get. To account for the client side calls is fairly simple since we have to pay IBM an annual support fee for the clients we've licensed from them. We uplift this slightly to cover our costs of support. On the server, we've taken the approach of basing the initial cost of our solution and ongoing support costs on the overall size (in Terabytes) of the server storage. We have four tiers: micro, up to 40TB of storage, small 40-80TB, medium 80-120TB and large over 120TB. The levels are somewhat arbitrary but reasonably reflect the STORServers in the field and correlated nicely with what our support numbers are telling us. I go into this as I think it would behoove IBM to consider a similar model. A client doesn't necessarily benefit more or less based on the number of cores it has. It does benefit, generally, from having the ability to backup and restore data. The overall environment benefits from the presence of the TSM server as it is that environment that allows for the secure maintenance of critical corporate data. It also provides services to recover after a disaster and finally, it provides a support organization to help a customer when it all goes wrong. The value of the solution is thus spread. A licensing scheme that spreads this value is appropriate. A client has a license no matter how big or small it is. Essentially a connection fee. The more clients you have the more you pay. The server is sized according to how much data is processed and stored. The more data that arrives each day and the more data that is stored necessarily results in a larger server environment and thus more value. It is very easy to count how much or how many of each. It is also easy to sell increments of licensing to accommodate growth. I would not be inclined to sell a per GB/month type scheme as this is too difficult for customers to budget. There must be a fixed component to licensing with a periodic true up period to make the scheme fair to IBM. Today, the licensing scheme is not fair to either party. Value as perceived by the customer is not tied to the number of cores in the processor and IBM cannot accurately determine if a customer is in compliance. This is not acceptable by either party. As I write this, I recall an earlier version of the licensing model: clients were free and we paid for the server stuff. It was priced by function. For instance, we paid for DRM and its support. That model wasn't correct as it rewarded the sites with large numbers of clients. One of you said it correctly: it's time to get this right once and for all. We need a fair licensing model that ensures TSM continues to be a viable product in the marketplace. That means one that rewards IBM for the hard work it does to provide the code and its support and one that provides real value to its customers. Subtract out the IBM bureaucracy and this is simple, right? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of John D. Schneider Sent: Tuesday, September 29, 2009 8:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Per terabyte licensing Kelly, You are right. IBM's pricing model also has in mind IBM customers that have dozens of Tivoli titles, Websphere, etc., which all use the PVU model. I think that IBM should build the license
Re: Per terabyte licensing
Really. How much does a TB of storage cost? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Steven Langdale Sent: Monday, September 28, 2009 11:02 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Per terabyte licensing My Tivoli S/W rep here in the UK is happy to sell by PVU or per TB. It sounds like it's not quite made it over the water yet. Steven Langdale Global Information Services EAME SAN/Storage Planning and Implementation ( Phone : +44 (0)1733 584175 ( Mob: +44 (0)7876 216782 ü Conference: +44 (0)208 609 7400 Code: 331817 + Email: steven.langd...@cat.com John D. Schneider john.schnei...@computercoachingcommunity.com Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 28/09/2009 15:38 Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: [ADSM-L] Per terabyte licensing Caterpillar: Confidential Green Retain Until: 28/10/2009 Duane, I asked our TSM rep this question, and he asked Ron Broucek, the North America Tivoli Storage Software Sales Leader. His response was: just a rumor at this time as we occasionally evaluate pricing strategies to make sure we're delivering the right value in the marketplace. Ron Broucek North America Tivoli Storage Software Sales Leader So if he says it is just a rumor, then how do you know IBM is offering both? Do you have this from a reliable source within IBM? Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: Re: [ADSM-L] Per terabyte licensing From: Ochs, Duane duane.o...@qg.com Date: Mon, September 28, 2009 9:07 am To: ADSM-L@VM.MARIST.EDU We are actually looking into the cost difference. From what I understand, IBM is offering both. However, per terabyte licensing eliminates sub-capacity licensing. And it is your entire site. Not just where it works out best. We are in the midst of passport renewals and found an increase due to core type upgrades. Previously we had older xeons using 50 PVUs per core. And the new machines replacing the older ones are either same cores but at xeon 5540 cores which are now 70 PVUs or double the cores. They brought up per TB licensing. Since then sales has sent me two E-mails inquiring total number of hosts, total TSM sites and total library capacity at each. I was hesitant to say the least. It's been about a week and I haven't heard back yet. When I hear more I'll drop a line. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Skylar Thompson Sent: Saturday, September 26, 2009 11:02 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: Per terabyte licensing We're in that boat too. We have a GPFS cluster we expect to grow into the petabyte range, so unless IBM sets the per-byte cost *really* low we'll get hammered with that licensing scheme. Zoltan Forray/AC/VCU wrote: Or more costly. We have test VM servers with quad-core processors running 15-VM guests. If I started counting by T-Bytes backed-up, it would cost a lot more than 4-CPU's! From: David Longo david.lo...@health-first.org To: ADSM-L@VM.MARIST.EDU Date: 09/25/2009 03:22 PM Subject: Re: [ADSM-L] Per terabyte licensing Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU Haven't heard that. My first thought is that it would make licensing a LOT easier to figure out! David Longo Thomas Denier thomas.den...@jeffersonhospital.org 9/25/2009 3:09 PM Within the last few months there was a series of messages on counting processor cores. A couple of the messages stated that TSM is moving to licensing based on terabytes of stored data rather than processor cores. Where can I find more information on this? # This message is for the named person's use only. It may contain private, proprietary, or legally privileged information. No privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it, and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Health First reserves the right to monitor all e-mail communications through its networks. Any views or opinions expressed in this message are solely those of the individual sender, except (1) where the message states such views or opinions are on behalf of a particular entity; and (2) the sender is authorized by the entity to give such views or opinions. # -- -- Skylar Thompson (skyl...@u.washington.edu) -- Systems
Re: Per terabyte licensing
And the key to that would be to add the phrase in some cases... No matter what IBM does there will be happy people and unhappy people. While a core based model doesn't make sense to many of us, a per TB model may turn out to make even less sense. To argue on their side, they must find a model that is compatible with the industry and that does not diminish their own cash flow. We need for IBM to continue to enhance the product. They do that by keeping us as customers and by attracting new customers. That balance is a lot harder than one may think. I was fairly vocal about this at a previous Oxford. While we're the loudest of the constituent parties, we also matter the least from a cash flow perspective: new customers actually spend more money (they've already gotten ours). The dance is tricky and sometimes comes down to a they won't really leave (where would they go?) so let's worry about them but not too much. As I own my own business I can understand the complexity they face. It's really hard, though, not to simply say it's their problem. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Steven Langdale Sent: Monday, September 28, 2009 12:38 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Per terabyte licensing He was a bit cagey about the actual cost, but said we should expect approx 20% reduction in overall cost. Not pursued it as yet. Steven Langdale Global Information Services EAME SAN/Storage Planning and Implementation ( Phone : +44 (0)1733 584175 ( Mob: +44 (0)7876 216782 ü Conference: +44 (0)208 609 7400 Code: 331817 + Email: steven.langd...@cat.com Kelly Lipp l...@storserver.com Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 28/09/2009 19:00 Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: [ADSM-L] Per terabyte licensing Caterpillar: Confidential Green Retain Until: 28/10/2009 Really. How much does a TB of storage cost? Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Steven Langdale Sent: Monday, September 28, 2009 11:02 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Per terabyte licensing My Tivoli S/W rep here in the UK is happy to sell by PVU or per TB. It sounds like it's not quite made it over the water yet. Steven Langdale Global Information Services EAME SAN/Storage Planning and Implementation ( Phone : +44 (0)1733 584175 ( Mob: +44 (0)7876 216782 ü Conference: +44 (0)208 609 7400 Code: 331817 + Email: steven.langd...@cat.com John D. Schneider john.schnei...@computercoachingcommunity.com Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 28/09/2009 15:38 Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: [ADSM-L] Per terabyte licensing Caterpillar: Confidential Green Retain Until: 28/10/2009 Duane, I asked our TSM rep this question, and he asked Ron Broucek, the North America Tivoli Storage Software Sales Leader. His response was: just a rumor at this time as we occasionally evaluate pricing strategies to make sure we're delivering the right value in the marketplace. Ron Broucek North America Tivoli Storage Software Sales Leader So if he says it is just a rumor, then how do you know IBM is offering both? Do you have this from a reliable source within IBM? Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: Re: [ADSM-L] Per terabyte licensing From: Ochs, Duane duane.o...@qg.com Date: Mon, September 28, 2009 9:07 am To: ADSM-L@VM.MARIST.EDU We are actually looking into the cost difference. From what I understand, IBM is offering both. However, per terabyte licensing eliminates sub-capacity licensing. And it is your entire site. Not just where it works out best. We are in the midst of passport renewals and found an increase due to core type upgrades. Previously we had older xeons using 50 PVUs per core. And the new machines replacing the older ones are either same cores but at xeon 5540 cores which are now 70 PVUs or double the cores. They brought up per TB licensing. Since then sales has sent me two E-mails inquiring total number of hosts, total TSM sites and total library capacity at each. I was hesitant to say the least. It's been about a week and I haven't heard back yet. When I hear more I'll drop a line. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Skylar Thompson Sent: Saturday, September 26, 2009 11:02 AM
Re: Per terabyte licensing
And remember, too, that the PVU thing contemplated something like a DB2 license. Perhaps you had two or three systems that would run DB2. It did not contemplate something like TSM where EVERY system in the environment would have the software running. Keeping track of a couple of systems and their various processor/core/PVU stuff is relatively simple. Keeping track of that same thing across several hundred (never mind your case!) is very difficult. The one size fits all mentality of Tivoli software clearly missed the mark with TSM. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of John D. Schneider Sent: Monday, September 28, 2009 4:47 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Per terabyte licensing Kelly, You are right, IBM must build their license model to ensure the profit they expect. We can't blame them for doing this as a business. They can't give their product away for free. But the PVU based licensing model is a huge problem for an environment like ours that has over 2000 clients of all different shapes and kinds. Lots of separate servers, but also VMWare partitions, and AIX LPARs, and NDMP clients, etc. Keeping up with the PVU rules is a huge effort, especially the way IBM did it. In Windows, the OS might tell you that you have 2 processors. But is that a single-core dual processor, or two separate processors. The OS can't tell, but IBM insists there is a difference, because it counts PVUs differently in this case. That is too nit-picky if you ask me, and places too difficult a burden on the customer. There are freeware utilities that will correctly count processors IBM's way, but to run them on 2000 servers is a pain, too. We ended up writing our own scripts to call a freeware tool IBM recommended, then parse the resulting answer to get the details into a summarized format. As if that wasn't enough, the freeware tool crashed about 20 of our servers before we realized it. Boy, was that hard to explain to management! It is also very objectionable to us that they don't have sub-processor licensing for large servers like pSeries 595s. We have a 128 processor p595, with a 2-processor LPAR carved out of it running Oracle. Even if we aren't running Oracle on any of the other LPARs, we have to pay for a 128 processor Oracle license. That is insane, and bad for everybody, including IBM. We also have to pay for 128 processors of regular TSM client licenses, even if we have only allocated half the processors in the p595. These are unfair licensing practices, and just make IBM look greedy. To simplify the license counting problem, we are looking at IBM License Metric Tool, but it is a big software product to install and deploy on 2000 servers, too, just to count TSM licenses. ILMT 7.1 was deeply flawed, and 7.2 just came out, so we are going to take a look at that. From my perspective, a total-TB-under-management model would be very easy on the customer, as long as it was reasonably fair. It would be easy to run 'q occ' on all our TSM servers and pull together the result. You could find out your whole TSM license footprint in 10 minutes. The first time we had to it counting PVUs, it took us two months. Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: Re: [ADSM-L] Per terabyte licensing From: Kelly Lipp l...@storserver.com Date: Mon, September 28, 2009 3:05 pm To: ADSM-L@VM.MARIST.EDU And the key to that would be to add the phrase in some cases... No matter what IBM does there will be happy people and unhappy people. While a core based model doesn't make sense to many of us, a per TB model may turn out to make even less sense. To argue on their side, they must find a model that is compatible with the industry and that does not diminish their own cash flow. We need for IBM to continue to enhance the product. They do that by keeping us as customers and by attracting new customers. That balance is a lot harder than one may think. I was fairly vocal about this at a previous Oxford. While we're the loudest of the constituent parties, we also matter the least from a cash flow perspective: new customers actually spend more money (they've already gotten ours). The dance is tricky and sometimes comes down to a they won't really leave (where would they go?) so let's worry about them but not too much. As I own my own business I can understand the complexity they face. It's really hard, though, not to simply say it's their problem. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor
Re: DR restore of Client Data to new hardware w/ new IP
Eric, Should have been update stg access=destroyed rather than unavailable. That's one thing. Why isn't the client seeing the TSM server at the new IP? H. The client is pointing at the new source server, not the target, correct? The new source/target combo may require updating as well since the target new the source at a different address. So the whole HLLL address thing needs to be fixed up on the new source and target. Are sessions enabled on the new source? Can the TSM client on the source talk to the source TSM server? These are all troubleshooting things I would try. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Eric Vaughn Sent: Wednesday, September 23, 2009 12:01 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] DR restore of Client Data to new hardware w/ new IP Hello, Our environment consist of source to target. I have recreated and restored the db onto a new piece of hardware that would be used as the replacement source server. My issue is that the IP is new and not the address that was assigned to the original source server. My assumption lead me to change the opt file on a client to the new server address. I restarted the services and tried to open up the gui and received a TCP/IP communication failure. Can anyone tell me what I am missing? I would like to restore from the copy pool. I issued the command update stgpool secondarypool access=unav so the client would retrieve data from the archivepool but no luck. Eric Vaughn Technical Administrator Stevenson University Office (443)334-2301 evau...@stevenson.edu
Re: the purpose of file device class
I don't think it's an either/or decision. I believe that tape will always have a place but that using some file pool will offer very nice RTO/RPO combos for some data structures. The allure of very inexpensive tape storage should always be there (and perhaps increase again with LTO5) while the allure of instant restore will intrigue us, especially as 2TB SATA becomes mainstream. Power and cooling disk is expensive and not green (I'm not green either, but thought I should at least try to be politically correct), while tape is still green (except for the oil needed to make the darn things). How about some hardware compression on the shelf based controllers? Perhaps target these products specifically to the backup space. To me, that's the biggest problem. Kelly Lipp Chief Technical Officer www.storserver.com 719-266-8777 x7105 STORServer solves your data backup challenges. Once and for all. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Rhodes Sent: Tuesday, September 22, 2009 5:41 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] the purpose of file device class ADSM: Dist Stor Hi TSM-ers! At this moment we are using a diskpool with a VTS-like (DL4106 by EMC) storage pool as nextpool. I too am looking at a FILE pool to replace this in the future, just to prevent a vendor lock-in for our TSM environment and of course the possibility to use de-dup. The only problem I see for using large FILE (100 Tb +) pools is the size of the filesystems on the host running the TSM server. Now it's AIX, but in the future we are likely to migrate to Linux. It would be interesting to know if the VTL vendors (which are basically implementing FILE type volumes inside the VTL) use a filesystem, raw logical vlumes, raw disk . . . or whatever. Does anyone have experience with 100+ Tb FILE storagepools? Thanks for any reply in advance! We keep looking at big pool of FILE devices vs VTL for our next big purchase down the road - whether for the initial pool where backups go directly, or are later migrated to. We keep hitting our heads against a wall in trying to come up with a way to make widespread use of a BIG FILE device pool: - dedup: IBM has solved this one with v6.1!!! - compression: To replace tape drive compression we need TSM server side compression (or should I say FILE device pool compression). We can't just push the job onto the clients. VTL's support hdwr compression via compression cards. I wonder if IBM will ever support server side hdwr compression via add-on cards for FILE devices (http://www.aha.com). - lanfree: Provide good lanfree with FILE devices that is not Sanergy. We've got a couple years yet on our tape systms, but as of today, we really don't see much way to effectively implement a large scale FILE pool to replace our current TAPE drives. FILE pools may be IBM strategic direction for TSM, but I think they have a lot more work to make it a true reality. Rick - The information contained in this message is intended only for the personal and confidential use of the recipient(s) named above. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please notify us immediately, and delete the original message.
Re: the purpose of file device class
During Oxford 2005, IBM stated that the File Device Class is the strategic data structure. One can see this in action with the dedup functionality in V6. Our products use DISK for cachepool usually deployed on JBOD arranged FAST disk (SAS or FC at 15K) as the initial target for client backups, particularly for large numbers of clients with small data movement. File Device class is used for large amounts of storage (since we can have reclamation there). Our pools are targeted by clients moving a lot of data daily, or are the target of migration from the cachepool. I think both are still valid for the reasons already described but in the long run DISK may go the way of the dinosaur. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of David McClelland Sent: Monday, September 21, 2009 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] the purpose of file device class Just to be uber-picky, FILE volumes now do allow multiple sessions/processes to read/write concurrently to a single FILE volume from TSM 5.5 onwards (http://publib.boulder.ibm.com/infocenter/tivihelp/v1r1/topic/com.ibm.itsmms munn.doc/anrsgd5515.htm#wq28). The big picture as I've read it is that IBM are perhaps angling the user-base towards using FILE volumes, and that new developments would be implemented against FILE technology rather than DISK. That being said, it's fair to say that many people are simply more comfortable with the ease, simplicity and habit of implementing and managing DISK volumes. /DMc -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: 21 September 2009 15:58 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] the purpose of file device class Disk Based (Random access) lets you have several sessions all sending data to it. This removes the need to queue backups like all the competitors. But it will fragment as data expires and there is no defrag for random access. File devices (Sequential access) do also fragment over time, just like tape, but you are able to run reclamation on them. But sequential access would cause queuing. They each have different pro's and con's which make them suited for different tasks. (long term vs short term storage) Regards, Shawn Shawn Drew This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc. No virus found in this incoming message. Checked by AVG - www.avg.com Version: 8.5.409 / Virus Database: 270.13.109/2384 - Release Date: 09/20/09 06:22:00
Re: backup encryption
Trust, my friend, trust... You'll know it's encrypted when you try to restore it without the key. Sorry folks, couldn't resist. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Geil Sent: Wednesday, August 19, 2009 9:30 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backup encryption I just turned on backup encryption on one of my clients using the include.encrypt option. Is there a way to tell if it is actually encrypting the data?
Re: backup encryption
On a more serious note, perhaps something in the schedule log will indicate the include statement. I've not done this so I don't know. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Geil Sent: Wednesday, August 19, 2009 9:30 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backup encryption I just turned on backup encryption on one of my clients using the include.encrypt option. Is there a way to tell if it is actually encrypting the data?
Re: I'm getting new disk storage.
It seems the perfect environment for the JBOD I suggested as you do not want all those sessions banging a RAID5 array. Of course work flow is an issue, but the moving a TB disk to disk or disk to tape is a couple of hour process if you work it correctly into your daily processing and probably isn't an issue for you. Ideally you wouldn't need to move it again, but you really can't have that many sessions banging RAID5 so what else do you do? I've arrived at this approach pragmatically and through trial and error. I know that I can easily run fifty or sixty (and perhaps more) sessions to 12 SAS 15K drives (use two disk pool volumes per drive to saturate each drive). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Michael Green Sent: Thursday, August 06, 2009 4:04 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. On Tue, Aug 4, 2009 at 7:41 PM, Kelly Lippl...@storserver.com wrote: You are exactly correct: modeling what will be rather than what is can be tricky. The problem really boils down to not having enough data to really play with it adequately. I can tell you from experience on probably 200 TSM servers (Windows 2003 based, which is just fine all of you AIX heads!) that your overall scheme is good. One additional, I like to run initial backups to SAS drives using a storage pool of device class disk. Don't protect those drives simply use JBOD as you will migrate the data out of them fairly soon either to tape or file device class disk. Use RAID5/6 for your SATA drives (We generally use 6 drive RAID5 sets in 12 bay shelves and 8 drive in 16 bay shelves) and keep a smallish number of simultaneous backups to pools there as large numbers of backups thrash the RAID set something awful. Your suggestion of using JBOD instead of RAID for DISKCLASS data is intriguing... I accept the reasoning behind not protecting these drives by RAID. But how does it affect the workflow performance-wise? What is the rational behind this? Economy (no wasted drives for parity), performance gains? How much data do you backup today? How many clients simultaneously? If you want to take this private, give me a call. I have a pretty good idea how to size this if I have some more information. Perhaps can save you a testing step (testing costs money that you could spend on additional storage...). That particular server backs up 700-1000GB (~400K+ affected objects) coming from just under 100 nodes nightly. Thanks for offering me a phone consultation :) I won't bother you for the time being, but maybe I will at a later time :) Thanks, Kelly Lipp CTO STORServer, Inc.
Re: I'm getting new disk storage.
Andy, In your case, you probably have a good bit of cache in the controller and your RAID sets are relatively small and you're using FC disks which are more reliable, tolerant and faster than SATA so all is good. My data comes from the SATA 7.2K world with six or eight drives per RAID5. Remember, our goal was to maximize capacity with these drives rather than performance. In your case, you chose performance as most of your data is being stored long term somewhere else (cheaper disk or tape). Failure in the disk pool is the question. Since the data is copied relatively soon after arrival (both to the copy pool and to the migration pool) it is not at risk for very long. If you do have a failure of a drive you probably have the data in the other pools, or still on the client. My rationale is as follows: if you lose a cachepool disk and the client loses data then you are in trouble. That's two bad things happening to one good person and what the chances? If you lose the cachepool then the next time the client backs up the data will move again. With the more reliable SAS/SCSI/FC drives, failures happen very infrequently. Are there holes in my rationale? Of course. Have I been bitten by them? No. Long term storage is different. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebner,Andy,FORT WORTH,IT Sent: Thursday, August 06, 2009 9:40 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. I must be missing something? I cannot saturate my 15 FC 10k disks (3x 4+1 RAID5) with 2GB Ethernet and 110 concurrent clients (330+ sessions). The Ethernet on the other hand is saturated. Out of curiosity, what happens with a disk failure? Andy Huebner -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Thursday, August 06, 2009 10:10 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. It seems the perfect environment for the JBOD I suggested as you do not want all those sessions banging a RAID5 array. Of course work flow is an issue, but the moving a TB disk to disk or disk to tape is a couple of hour process if you work it correctly into your daily processing and probably isn't an issue for you. Ideally you wouldn't need to move it again, but you really can't have that many sessions banging RAID5 so what else do you do? I've arrived at this approach pragmatically and through trial and error. I know that I can easily run fifty or sixty (and perhaps more) sessions to 12 SAS 15K drives (use two disk pool volumes per drive to saturate each drive). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Michael Green Sent: Thursday, August 06, 2009 4:04 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. On Tue, Aug 4, 2009 at 7:41 PM, Kelly Lippl...@storserver.com wrote: You are exactly correct: modeling what will be rather than what is can be tricky. The problem really boils down to not having enough data to really play with it adequately. I can tell you from experience on probably 200 TSM servers (Windows 2003 based, which is just fine all of you AIX heads!) that your overall scheme is good. One additional, I like to run initial backups to SAS drives using a storage pool of device class disk. Don't protect those drives simply use JBOD as you will migrate the data out of them fairly soon either to tape or file device class disk. Use RAID5/6 for your SATA drives (We generally use 6 drive RAID5 sets in 12 bay shelves and 8 drive in 16 bay shelves) and keep a smallish number of simultaneous backups to pools there as large numbers of backups thrash the RAID set something awful. Your suggestion of using JBOD instead of RAID for DISKCLASS data is intriguing... I accept the reasoning behind not protecting these drives by RAID. But how does it affect the workflow performance-wise? What is the rational behind this? Economy (no wasted drives for parity), performance gains? How much data do you backup today? How many clients simultaneously? If you want to take this private, give me a call. I have a pretty good idea how to size this if I have some more information. Perhaps can save you a testing step (testing costs money that you could spend on additional storage...). That particular server backs up 700-1000GB (~400K+ affected objects) coming from just under 100 nodes nightly. Thanks for offering me a phone consultation :) I won't bother you for the time being, but maybe I will at a later time :) Thanks, Kelly Lipp CTO STORServer, Inc. This e-mail (including any
Re: I'm getting new disk storage.
That is correct: you can lose a couple of storage pool volumes (on a failed disk) and still go on. The operation in progress writing to those volumes will stall/fail (I don't know which, but I'm guessing retries probably save your butt). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebner,Andy,FORT WORTH,IT Sent: Thursday, August 06, 2009 3:43 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. Thank you for the reply; I should have been more precise with the disk failure question. Doesn't a disk failure affect the backup run, or are your pools on many physicals so the loss of one is not fatal to the process? I understand the data loss part; you have a short window of time for a low probability set of disk failures. Thanks again. Andy Huebner -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Thursday, August 06, 2009 2:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. Andy, In your case, you probably have a good bit of cache in the controller and your RAID sets are relatively small and you're using FC disks which are more reliable, tolerant and faster than SATA so all is good. My data comes from the SATA 7.2K world with six or eight drives per RAID5. Remember, our goal was to maximize capacity with these drives rather than performance. In your case, you chose performance as most of your data is being stored long term somewhere else (cheaper disk or tape). Failure in the disk pool is the question. Since the data is copied relatively soon after arrival (both to the copy pool and to the migration pool) it is not at risk for very long. If you do have a failure of a drive you probably have the data in the other pools, or still on the client. My rationale is as follows: if you lose a cachepool disk and the client loses data then you are in trouble. That's two bad things happening to one good person and what the chances? If you lose the cachepool then the next time the client backs up the data will move again. With the more reliable SAS/SCSI/FC drives, failures happen very infrequently. Are there holes in my rationale? Of course. Have I been bitten by them? No. Long term storage is different. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebner,Andy,FORT WORTH,IT Sent: Thursday, August 06, 2009 9:40 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. I must be missing something? I cannot saturate my 15 FC 10k disks (3x 4+1 RAID5) with 2GB Ethernet and 110 concurrent clients (330+ sessions). The Ethernet on the other hand is saturated. Out of curiosity, what happens with a disk failure? Andy Huebner -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Thursday, August 06, 2009 10:10 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. It seems the perfect environment for the JBOD I suggested as you do not want all those sessions banging a RAID5 array. Of course work flow is an issue, but the moving a TB disk to disk or disk to tape is a couple of hour process if you work it correctly into your daily processing and probably isn't an issue for you. Ideally you wouldn't need to move it again, but you really can't have that many sessions banging RAID5 so what else do you do? I've arrived at this approach pragmatically and through trial and error. I know that I can easily run fifty or sixty (and perhaps more) sessions to 12 SAS 15K drives (use two disk pool volumes per drive to saturate each drive). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Michael Green Sent: Thursday, August 06, 2009 4:04 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm getting new disk storage. On Tue, Aug 4, 2009 at 7:41 PM, Kelly Lippl...@storserver.com wrote: You are exactly correct: modeling what will be rather than what is can be tricky. The problem really boils down to not having enough data to really play with it adequately. I can tell you from experience on probably 200 TSM servers (Windows 2003 based, which is just fine all of you AIX heads!) that your overall scheme is good. One additional, I like to run initial backups to SAS drives using a storage pool of device class disk. Don't protect those drives simply use JBOD as you will migrate the data out of them fairly soon either to tape
Re: I'm getting new disk storage.
You are exactly correct: modeling what will be rather than what is can be tricky. The problem really boils down to not having enough data to really play with it adequately. I can tell you from experience on probably 200 TSM servers (Windows 2003 based, which is just fine all of you AIX heads!) that your overall scheme is good. One additional, I like to run initial backups to SAS drives using a storage pool of device class disk. Don't protect those drives simply use JBOD as you will migrate the data out of them fairly soon either to tape or file device class disk. Use RAID5/6 for your SATA drives (We generally use 6 drive RAID5 sets in 12 bay shelves and 8 drive in 16 bay shelves) and keep a smallish number of simultaneous backups to pools there as large numbers of backups thrash the RAID set something awful. Our STORServers almost always have this mix of storage pool options for our customers to use: SAS 15K JBOD, SATA 7.2K RAID and some tape. DB/LOG on SAS using best practices. Next thing to consider is your V6.2 upgrade next spring. You gotta count on needing more space than you have allocated currently to the DB/LOG volumes. Probably 3-6x is a good guess but we're all still learning. And probably more spindles than you might consider for V5. The shelves we offer have the ability to mix and match SAS and SATA but I'm thinking for you a whole shelf of SAS at least and then SATA. I don't think you really need to much intelligence or features in the shelves themselves as you have plenty of options with TSM to do what you need to do. How much data do you backup today? How many clients simultaneously? If you want to take this private, give me a call. I have a pretty good idea how to size this if I have some more information. Perhaps can save you a testing step (testing costs money that you could spend on additional storage...). Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Michael Green Sent: Tuesday, August 04, 2009 10:06 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] I'm getting new disk storage. I know performance benchmarking is an art. And asking questions like how can I test this new storage may bring a smile on face of some of you. But still... The time has come and I need to replace the disk storage underlying the DB/DISKCLASS/FILECLASS files. I'm going to test two storages from Infortrend and DDN. Both storages allow mix and match of SAS (for DB/LOG) and SATA (for DISK/FILECLASS) drives. Both will be FC 4Gb attached. I've been thinking about replicating my biggest TSM server (Linux based) on a server that has been put aside for this purpose. I'll load DB/LOG onto SAS drives and some FILECLASS volumes onto SATA drives and time the following workloads: - expiration - db defrag - Reclamation To be honest I am not sure I know how to properly set up such testing environment. I'm concerned that my tests will not be representative enough and will not indicate the actual performance that I'll observe after the storage system will become production. Your thoughts on this topic are much appreciated... -- Warm regards, Michael Green
Re: Change TSM Platform
What is the concern with keeping a Windows TSM platform (I sat in the weeds on this as long as I could)? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Tuesday, August 04, 2009 8:46 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Change TSM Platform You will have to move the NAS clients over to the new TSM server, and wait for the old backups to expire before you retire the old backup server. That's what I figured, but I'm not keeping that thing around for 7 years. I guess we'll have to stick with Windows :( Regards, Shawn Shawn Drew Internet john.schnei...@computercoachingcommunity.com Sent by: ADSM-L@VM.MARIST.EDU 08/03/2009 06:38 PM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] Change TSM Platform Shawn, From my understanding, you are going to have to set up a new TSM server, and migrate the clients over to it. You can use the export commands to export policies and client data from one TSM server to another directly across the LAN. That will make it somewhat less painful, but depending on how many clients you have, this could take a few weeks. You will have to have enough storage capacity on the new system to absorb all this data. If you only have one tape library, you will have to set up library sharing, and have enough tapes to have two copies of some clients' data as you migrate clients over. If you want more detailed instructions, please ask. Many of us have been through such migrations before. According to the help on EXPORT NODE, you can't use it on nodes of type NAS. You will have to move the NAS clients over to the new TSM server, and wait for the old backups to expire before you retire the old backup server. Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: Re: [ADSM-L] Change TSM Platform From: Michael Green mishagr...@gmail.com Date: Mon, August 03, 2009 1:32 pm To: ADSM-L@VM.MARIST.EDU You will be moving from x86/64 architecture to Power. They are binary incompatible. You cannot upload DB from x86 to Power (this is what backup/restore essentially doees). Your only option is to export DB. Don't know about the NDMP. -- Warm regards, Michael Green On Mon, Aug 3, 2009 at 9:06 PM, Shawn Drewshawn.d...@americas.bnpparibas.com wrote: We are looking at the possibility of changing a branch's TSM 5.4 server from Windows to AIX. As far as I know, you can NOT backup a DB on Windows and restore it to an AIX platform.Is this still the case? If not, can anyone think of a way to move NDMP toc/backups from one TSM server to another? This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.
Re: Change TSM Platform
Perfectly rational! That one I can get behind... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Tuesday, August 04, 2009 11:21 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Change TSM Platform Just a standard. All of our TSM servers are based on AIX in our main data centers. We recently took control of a branch location who has been running on Windows. We are spending a disproportionate amount of time to get our automation, and everything else, to work with Windows just for this little branch. little things like no grep or mail causes headaches. I know there are ways to get this to work, just requires time to work on it, where I'd rather just spend the extra money for an AIX box. Regards, Shawn Shawn Drew Internet l...@storserver.com Sent by: ADSM-L@VM.MARIST.EDU 08/04/2009 12:42 PM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] Change TSM Platform What is the concern with keeping a Windows TSM platform (I sat in the weeds on this as long as I could)? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Tuesday, August 04, 2009 8:46 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Change TSM Platform You will have to move the NAS clients over to the new TSM server, and wait for the old backups to expire before you retire the old backup server. That's what I figured, but I'm not keeping that thing around for 7 years. I guess we'll have to stick with Windows :( Regards, Shawn Shawn Drew Internet john.schnei...@computercoachingcommunity.com Sent by: ADSM-L@VM.MARIST.EDU 08/03/2009 06:38 PM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] Change TSM Platform Shawn, From my understanding, you are going to have to set up a new TSM server, and migrate the clients over to it. You can use the export commands to export policies and client data from one TSM server to another directly across the LAN. That will make it somewhat less painful, but depending on how many clients you have, this could take a few weeks. You will have to have enough storage capacity on the new system to absorb all this data. If you only have one tape library, you will have to set up library sharing, and have enough tapes to have two copies of some clients' data as you migrate clients over. If you want more detailed instructions, please ask. Many of us have been through such migrations before. According to the help on EXPORT NODE, you can't use it on nodes of type NAS. You will have to move the NAS clients over to the new TSM server, and wait for the old backups to expire before you retire the old backup server. Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: Re: [ADSM-L] Change TSM Platform From: Michael Green mishagr...@gmail.com Date: Mon, August 03, 2009 1:32 pm To: ADSM-L@VM.MARIST.EDU You will be moving from x86/64 architecture to Power. They are binary incompatible. You cannot upload DB from x86 to Power (this is what backup/restore essentially doees). Your only option is to export DB. Don't know about the NDMP. -- Warm regards, Michael Green On Mon, Aug 3, 2009 at 9:06 PM, Shawn Drewshawn.d...@americas.bnpparibas.com wrote: We are looking at the possibility of changing a branch's TSM 5.4 server from Windows to AIX. As far as I know, you can NOT backup a DB on Windows and restore it to an AIX platform.Is this still the case? If not, can anyone think of a way to move NDMP toc/backups from one TSM server to another? This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.
Re: Another perspective on ridiculous retention
There is a huge difference between backup and archive, especially with TSM. One can actually make them different in TSM. The keys to effective archiving are three fold: 1. Only archive that which you are legally (or for business purposes) required to archive and then for only the time required. Rarely is anyone required to keep everything forever! The requirements are much more specific than that. Having data around that you are not required to have can be as bad or worse than not having something! 2. Archive in some commonly readable format. Remember, if you are going to need to retrieve something it is likely going to be sometime down the road. How you read that something is critical. Does the application that created the data still exist? Perhaps not. Thus a binary representation of the data is probably not going to be easily, if at all, readable. This is especially true for databases. Consider archiving an export of the data rather than the database files themselves. This seems a duh, but often is not considered. 3. IT does not own archiving. The business units that rely on IT own archiving and its requirements. To expect the IT staff to somehow magically know/understand the business and legal requirements is absurd. We must be provided with concise requirements for archiving for us to do a good job. We must understand exactly what data, how long and what the retrieve time requirements are. Everything forever is STEWPID! I wrote a paper on this some time ago. If anyone is interested, email me privately and I'll be happy to send it to you. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of David McClelland Sent: Tuesday, August 04, 2009 11:57 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Another perspective on ridiculous retention Questions flood into my head along the lines of 'what's the difference between a backup and an archive' (obviously not in a TSM sense) and if/ how should they be treated differently practically with TSM (e.g. a seperate TSM Server instance for archival purposes, as some places do etc). /David Mc Sent from my iPhone On 4 Aug 2009, at 18:45, Troy Barnhart tbarnh...@rcrh.org wrote: Only 7 years? We're in Healthcare. Seven is usually the minimum. If you're talking Minors, then it is 18+7= 25 years. Digital Mammography Research-related electronic medical records are FOREVER. There are lots of numbers on time-retention floating around out there - it just depends on the governing authority. We haven't completed our Retention Policy, so we have tapes from various Operating Systems and Applications from the 1990's. Regards, Troy Barnhart, Sr. Systems Programmer tbarnh...@rcrh.org Regional Health, Inc. 353 Fairmont Boulevard Rapid City, South Dakota 57701 PH: 605-716-8352 / FAX: 605-716-8302 -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Tuesday, August 04, 2009 11:15 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Another perspective on ridiculous retention Do I understand you to say you have to keep your NDMP backups around for 7 years? The tape media isn't even meant to last for 7 years. Do you have customers that actually think they will need 7 year old copies of you NAS data? That's a tough requirement. I thought I'd change this to a new topic. I hear this type of comment alot on backup forums. From an engineering perspective, it completely makes sense. It also makes sense that people in backup forums think like engineers! Just another perspective.When I started with TSM, I was working for a software development company named Tivoli who obviously cared about their backup data. The mantra of the backup guys was Restores are more important than backups! I.E. do periodic test restores, and if a restore request comes in and conflicts with a backup. cancel the backup in favor of the restore. Several years later, I start working for a bank. After working here for a few years, I realize the mantra is now the reverse: Backups are more important than restores. Meaning. the main reason we perform backups and retain them for 7 years, is so we can show an auditor our settings and say we've done it. We very rarely have to restore anything that old, but we very often have to show records of these backups. One last note, I have been involved in legal discovery projects where we actually did have to restore 7 year old data off of old DLT IV tapes. We found tapes with dried up BBQ sauce on them and all sorts of damage. Luckily, between the multiple storage pools we were able to rebuild all the data. The DLTs never actually failed due to age (only by a tomato-based attack!) Regards, Shawn
Re: Backup Stgpool using 2 drives taking too long
And on Windows you can use the task manager to look at bytes read or written per process. Find the TSM Server process and verify that it is indeed moving data. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of John D. Schneider Sent: Monday, August 03, 2009 2:32 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Backup Stgpool using 2 drives taking too long Mario, So, if I understand this correctly, the primary storage pool is LTO4, and the copy storage pool is LTO4, and so the whole backup stgpool is straight tape to tape. Is that correct? When you say the process becomes idle for 20 minutes, how can you tell it is idle? Do you mean the number of bytes copied doesn't get any bigger? The total byte count only gets updated at the end of each aggregated file. That is, if the copy stream hits a 300GB single file, then the copy will proceed to copy that file as fast as tape performance permits, but the total byte count won't increase until that huge file completes. And a multi-hundred GB file could easily take 20 minutes or more. Next time you catch it in this state, look at the size of the current file, and do the arithmetic. Is it possible that is what you are seeing? If you can look at the tape drives themselves, you can tell if the tape drives themselves are reading and writing, or sitting idle, can't you? I suspect that the tape drives will be buzzing away, even though it seems like you aren't making progress. If I have misunderstood, and you can tell that the drives are really idle, then I would look at drive firmware next, then at IBMtape drivers. Make sure you are up-to-date We have 6 TSM Windows servers with IBM libraries and drives, and we are not seeing the symptom you describe, that is, periods of actual stalled backup stgpools. Feel free to contact me offline if you have not done the firmware and driver updates before, and need any assistance. Best Regards, John D. Schneider The Computer Coaching Community, LLC Office: (314) 635-5424 / Toll Free: (866) 796-9226 Cell: (314) 750-8721 Original Message Subject: [ADSM-L] Backup Stgpool using 2 drives taking too long From: Mario Behring mariobehr...@yahoo.com Date: Mon, August 03, 2009 1:34 pm To: ADSM-L@VM.MARIST.EDU Hi list, The environment is TSM Server 5.5 running on a Windows 2003 box with a IBM ULT3580-TD4 SCSI (2 drives) connected. There is a backup stgpool process running on background for more than 12 hours (several schedules were missed during this time). The process is using both drives, one reading the data and the other writing, as expected. The problem is that, at some point, both drives stay IDLE for more than 20 minutes...then start reading/writing again (for a short time I might add). The process would be long over by now if this was not happening...what could be wrong and how can I put the drives to work non-stop, if possible...?? Any process that uses both drives in this fashion, like an EXPORT for example, is behaving the same way... Thanks Mario
Re: Tivoli disk only backup with data duplicated at two locations?
So are you thinking FCOE for the connection to the storage at the remote site? That way the storage is essentially locally mounted. Remco brings up an interesting point about recovery: how do you rationalize the copy pool volume locations from the original to the DR server but I think that can be worked out in advance by ensuring that the system at the DR site that takes over the mountpoints has the same drive letter (or if Unix device names) as the old system. Restore the database and voila. I liked, too, the notion of keeping the config files or perhaps even using OS duplication/mirror of the DB and log files. One could then fire the DR TSM server and mount the correct storage devices and go. Planning is everything! Remember, too, that TSM has server-to-server functionality. It isn't quite slick enough for me, but it does work and for many sites is completely adequate. Since it is an intrinsic part of the product, well documented and requires a bit less tom foolery than anything else. To me this always boils down to is it really cheap enough? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Davis, Adrian Sent: Wednesday, July 29, 2009 9:02 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Tivoli disk only backup with data duplicated at two locations? We are looking at a number of possibilities for a new backup system? Until now we have a used a traditional tape based backup strategy - but with falling disk and network bandwidth prices - we are looking at the possibility of using a disk only backup solution. One option is to use Tivoli. We need to be able to have two copies of all data each copy at a different physical location. One idea that comes to mind is to backup to a disk storage pool, then BACKUP STGPOOL this storage pool to another (copy) disk storage pool - The second disk pool being a volume on a separate disk array at a remote location. We would also BACKUP DB the TSM database to the remote location. In the event of the loss of the primary site - I'm thinking we could build a TSM server, using the database backup up to the remote site, mount the remote copy disk storage pool volume on the new server and (as at that time the primary storage pool is unavailable) access the data on the disk copy storage pool directly in order to perform our DR restores. This seems like a (relatively) simple solution. However, as I have not seen this documented in any of the manuals, I'm guessing that there is a problem with this approach? Does anybody have any comments -or- better ideas for a Tivoli disk only backup with data duplicated at two sites? Many Thanks, =Adrian= DISCLAIMER This message is confidential and intended solely for the use of the individual or entity it is addressed to. If you have received it in error, please contact the sender and delete the e-mail. Please note that we may monitor and check emails to safeguard the Council network from viruses, hoax messages or other abuse of the Council’s systems. To see the full version of this disclaimer please visit the following address: http://www.lewisham.gov.uk/AboutThisSite/EmailDisclaimer.htm For advice and assistance about online security and protection from internet threats visit the Get Safe Online website at http://www.getsafeonline.org
Re: Used 3494 / 3590 equip
I'll take it! Just kidding: I'm thinking this has some rather significant value. Try eBay... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Druckenmiller, David Sent: Friday, July 24, 2009 12:51 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Used 3494 / 3590 equip Just wondering if anyone knows what the market might be for a 3-frame 3494 with 8 3590-H1A drives. Also, is there market for about 1800 used 3590E cartridges? We just finished migrating off and my boss just wants to trash the whole thing. I'd be curious to know just how much money they're throwing away. Thanks. - CONFIDENTIALITY NOTICE: This email and any attachments may contain confidential information that is protected by law and is for the sole use of the individuals or entities to which it is addressed. If you are not the intended recipient, please notify the sender by replying to this email and destroying all copies of the communication and attachments. Further use, disclosure, copying, distribution of, or reliance upon the contents of this email and attachments is strictly prohibited. To contact Albany Medical Center, or for a copy of our privacy practices, please visit us on the Internet at www.amc.edu.
Re: Data Domain - questions from a meeting with DD
Ben, How much data do you have in DD and what sort of ratios are you seeing? What about performance during restores? Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Ben Bullock Sent: Monday, June 15, 2009 9:52 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Data Domain - questions from a meeting with DD We use a DD580 on our TSM servers. We use a device type of FILE and set it to a 50GB file size. Our TSM server is on AIX, so we use an NFS mount for the storage pools on the DD. It works well, we can get very good throughput over the 1GB NIC to the DD. If your environment is larger and you need more throughput, you could get a DD with the VTL option and then I think you can attach it to your TSM server though fibre. We try not to compress anything we send to TSM. Yes, it takes longer to send over the network, but it then dedupes well. In some cases we do have to receive pre-compressed data that will not depupe/compress. In those cases, we don't send that data to the DD storagepool, we keep it on the tape drives. DD has to do its own CLEAN process once a week, which is kinda like a reclamation on the TSM server. So we just let the TSM server do it's normal reclamation on the DD storagepools, but knowing that the data will not actually re-appear on the DD until after it does it's clean. Ben -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Loon, EJ van - SPLXM Sent: Monday, June 15, 2009 9:31 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Data Domain - questions from a meeting with DD Hi Rick! Quite a coincident: I just spoke to DataDomain too! 1) The Dutch guys also confirmed that compressed data is dedupable, but it depends how compression is done. If it's file level compression, you can expect that a copy of that file is similar and thus dedupable. A compressed ntkernel.dll from system A is similar to a compressed ntkernel.dll from system B. 2) and 3) The Dutch DD guys stated that most TSM users with a DataDomain appliance use a storagepool type=disk! Normally you don't want to store your data in a diskpool, because it becomes heavily fragmented overtime, but since the DD appliance does backend defragmentation, front-end fragmentation on you diskpool this will have no negative effect on performance. I found it a very interesting option, since the appliance does in-band and on-the-fly deduplication, compression and defragmentation and has the option to duplicate the de-duplicated data to a remote location. Kind regards, Eric van Loon KLM Royal Dutch Airlines -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Rhodes Sent: maandag 15 juni 2009 15:29 To: ADSM-L@VM.MARIST.EDU Subject: Data Domain - questions from a meeting with DD Hello, The other day we had a meeting with Data Domain - just the normal vendor update about their products. While informative, several comments were made by DD systems that I thought I like your opinions on. (note: we are a all tape shop, and probably will be until the next hardware refresh in several years) Interesting comments DD made: 1) de-dup of compressed data. We told them that much of our backup load comes from Oracle backups that are already compressed via Unix compress utility then pushed to TSM. We asked what effect this would have on de-dup. They said we could expect a 4x-5x de-dup ratio on this compress data. I had always thought that compressed data would not de-dup. 2) NFS mounted storage pools. They said that the vast majority of their installations, including TSM installations, use NFS (some OpenStorage for NetBackup, and a little CIFS) for backups to disk. In other words, VTL emulation is a very small percentage of their installations. If you have DD hdwr, are you using NFS or file devices, or, a vtl interface? Is anyone using NFS for a storage pool? 3) DD reclaim of scratch tapes I asked about when/how DD will reclaim no longer used space. In other words, when a scratch tape passed reuse-delay. He said that DD had some kind of interface to backup softw such that it knows when data is no longer needed and reclaimes that space. I know TSM v5.5 has a feature to clear a vol when no longer used with a vtl, but DD was saying they had a special interface to accomplish this. Does DD have a special interface to TSM? Thanks Rick - The information contained in this message is intended only for the personal and confidential use of the recipient(s) named above. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error
Re: Internal compression?
No, this isn't internal compression. If you look at the sched log for the clients, do you see and failed messages? For instance, let's say you have copy serialization set to one of the shared options (sharedynamic, sharestatic). If the client starts to move the file, then it opens, then it closes and the client starts again, all the data that is moved (whether successful or not) is counted toward the GB transferred number. I've seen this account for the difference between what you think was backed up based on session roll up stats and what's migrated or copied during daily processing. Just an idea. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Shawn Drew Sent: Thursday, June 11, 2009 12:14 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Internal compression? I noticed some strange data. I never paid attention to it before, but now that I noticed it, I find it is the same across all of our TSM instances. Is there any reason that STGPOOL backups and Migrations would only process about half the quantity of data that is reported to be backed up? They are all AIX/TSM 5.4.4 and ALL nodes have Compression = no at the server level - Quantity of data backed up last night on one small system was 140 GB. (I manually confirmed this by adding up all the Total number of bytes transferred numbers from the actlog) select sum(bytes) from summary where activity='BACKUP' and cast((current_timestamp-END_TIME)days as integer)1 - Storage pool backups, which run after the client backups only transferred 68GB (confirmed from adding up the ANR0986I messages from the actlog) select sum(bytes) from summary where activity='STGPOOL BACKUP' and cast((current_timestamp-END_TIME)days as integer)1 (And yes, I did verify we are backing up all the primary pools that the backup data could have possibly landed) - Migration also only transferred 68GB select sum(bytes) from summary where activity='MIGRATION' and cast((current_timestamp-END_TIME)days as integer)1 Run these select statements on your server and see if you see the same thing. Regards, Shawn Shawn Drew This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc.
Re: Internal compression?
Yoo hoo! Me and Richard agree! I love when that happens. And I got there first. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Richard Sims Sent: Thursday, June 11, 2009 12:33 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Internal compression? On Jun 11, 2009, at 2:14 PM, Shawn Drew wrote: Is there any reason that STGPOOL backups and Migrations would only process about half the quantity of data that is reported to be backed up? One commonly encountered reason is retries, evidenced in the full client log. (It's a product deficiency that the summary statistics don't report the retries in any way, so you have to pore over the full log.) Richard Sims
Re: Performance question
I would guess that it won't go any faster via iSCSI. Perhaps it might be slower due to one more protocol conversion on each end. You can move about one quarter TB/hour over a GigE. So that's 1200 hours to move all that data... That's a long time! How about deploying another TSM server local to that data and doing the backup there and removing the tapes for DR? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Christian Svensson Sent: Wednesday, May 27, 2009 5:26 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Performance question Hi all *SMers I got a new challenge in a front of me. Every quarter will I get a large package with 10 000 new files where the total size is 300 TB of Data (each file is average size 30GB large). This is static data that will not replace any other files or modified. The problem is that I can not do any LAN-Free backup over SAN because of the data and TSM is on different locations (20 Miles between the sites). The link between the data and my TSM Server is black fiber but I only have 1 GBit limited speed on this link. Why I don't know... I don't have any time limit how long time it takes to backup/restore this files, even if it takes 1 hour or 30 days. The customer don't care. But as TSM Admin I want to do this backup/restore as fast as possible. I was thinking of to do a LAN-Free backup but via iSCSI so I sending larger blocks between server and tape but I don't know if that will give me any performance benefit then backup as normal incremental backup over normal TCP/IP network. My question is, will it go faster to do a backup via iSCSI then regular LAN Backup? Best Regards Christian Svensson Cell: +46-70-325 1577 E-mail: christian.svens...@cristie.se Skype: cristie.christian.svensson
TSM Sales Engineer needed
Folks, http://www.storserver.com/main.cfm?menu=1detail=jobs/job_CSE.htm Describes positions. I need someone in Chicago, Southern California and Oregon/Washington. Resumes to me if you have or know of somebody with interest. Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com
Re: Changing volume status with active data to SCRATCH
Del volume discarddata=yes will do it. A pretty big hammer and you will not be able to get the data back without restoring the TSM database. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Mario Behring Sent: Tuesday, May 12, 2009 3:55 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Changing volume status with active data to SCRATCH Hi list, How can I change the status of a particular tape in the LTO Library to SCRATCH considering that there is still not expired data on it? Is there any way to expire this data or tell TSM to ignore it and change the tape to SCRATCH? Thanks Mario
Re: LTO for long term archiving
Here here. I'll reiterate: the data must be stored in the lowest common denominator format. No databases, no strange and wondrous binary formats. Simple text is the best. That's why microfiche is still used! I would be very concerned about medical data: it is in so many different formats and one has to wonder if our elected officials and their beaurocratic minions in the various regulatory agencies are aware of these facts and the problem they are going to have in the future. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kauffman, Tom Sent: Wednesday, May 06, 2009 7:38 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving Soapbox time. The media is not important - any sane retention policy will require that the information be copied at least annually, with enough copies for redundancy. BUT - to be able to PROCESS the data 25 years from now imposes additional requirements. First and formost -- the data will NOT be in any format except unloaded flat-file; ASCII, UTF-8, or UTF-16 encoding. You will NOT be able to process proprietary data formats 25 years from now. Look at the Domesday Book - the version written on parchment or vellum in 1086 is still readable. The BBC digitized it in 1986 and found it almost impossible to find systems to read the digitized version in 2002 - 16 years later. You're trying for 25 years. Tom Kauffman NIBCO, Inc -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Evans, Bill Sent: Tuesday, May 05, 2009 5:34 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: LTO for long term archiving I truly doubt that archiving drives, servers and tapes for 25 years each time the technology updates will let you read the tapes because the drive and server will probably not even boot up and run. You will have to update the data every two LTO cycles or so. LTO will read two generations back and 25 years from now we will be on LTO14 or 15, so LTO4 is toast. Or, more probably, we will be storing into some kind of flash drive at Petabyte capacities. I think that Blu-Ray DVD will meet your 25 year mark without having to retrieve and update to new media. I know that my 1986 CD's (those not seriously scratched or warped from laying on the dash ) still work on today's systems. Properly stored DVD's would need to have players stored also, but, these are mechanically simpler than LTO drives and servers and would most likely still run. They are also much cheaper, so having a new DVD in storage every 5 years is no big expense. The bigger issue is where and how do you keep track of all of this? I think TSM's HSM is probably capable, however, I'm not real comfortable recommending it. We have had several years of problems running HSM on Solaris and have finally turned it off. What is needed is a good archiving tool that can keep an updated DB of content and storage location that users can browse. We recently restored a power point file written by Office version (?) on an OS 9 Mac. This could NOT be read by Office XP, 2003, 2007 (PC) or Office 2004 or 2008 (Mac). We had to find an old Mac OS/9 that still had a copy of Office 2000, read it, write it back to the 2000 version .ppt file before any 2004-2009 software could read it. If it had been 18 years instead of 9 years, then we never would have been able to read it at all, that old OS 9 Mac would never have been saved. This will happen more and more as our programs become more complex and require significant changes in the file formats. So the real problem is not just how to archive the data for 25 years, it's how to archive the applications for 25 years so we can access that data! Actually, stone tablets are, so far, the best archive media... Thanks, Bill Evans Research Computing Support FRED HUTCHINSON CANCER RESEARCH CENTER 206.667.4194 ~ bev...@fhcrc.org -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Tuesday, May 05, 2009 1:42 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving I like the implication, but I'm pretty sure somebody actually thought being able to read the information would have been a good idea. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Remco Post Sent: Tuesday, May 05, 2009 2:39 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving I do agree, having the tapedrives around _could_ be important. I know of at least one environment that was able to produce the media that stores the data
Re: LTO for long term archiving
To me the problem is having the drives around and more importantly, the interfaces to the drives. I think that probably the best bet is to plan on archiving a TSM server with a drive along with the media periodically. Snap off the last database backup, restore it on the to be archived server (a good test in itself), and store the whole kit together. If one needs to retrieve an archive, fire up the archived server, query the database to determine what tape is required, get it, retrieve the data and put the whole mess away. The other way to do this would be to migrate the archived data to new tape media as you march through time. I like this approach as that will have the double advantage of refreshing and verifying the data on those tapes. One could shelve the media in the archive pools and do this on a very controller basis when the media changes. Lots of data movement potentially, but it would become a nicely verified process that your auditors could look at to help ensure compliance. It's one thing to say we're doing and quite another to show we're doing it. Having the archive data more readily retrievable has obvious benefits as well. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebschman, George J. Sent: Tuesday, May 05, 2009 2:25 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving Does anyone have 25 year old tape media or tape drives around? Will you stil be able to use LTOx media in 25 years? -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Thomas Denier Sent: Tuesday, May 05, 2009 4:11 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] LTO for long term archiving I work for a large hospital. I have been asked to investigate possible configurations for archiving something between a few hundred terabytes and a petabyte of data for 25 years. This would be clinical records that we need to keep in case of a malpractice suit. The retention period is 25 years because there are two ways we can get sued for alleged malpractice involving a pediatric patient. The parents or guardians have a seven year window of opportunity to file suit, starting at the time of the alleged malpractice. The patient has a seven year window of opportunity, starting at his or her 18th birthday. In principle, the retention period should vary depending on patient age, but nobody I have talked to so far thinks it is practical to sort records in this way; they want a uniform retention period that covers the worst case scenario (a patient allegedly harmed as a newborn suing just before the end of his or her seven year window). As far as I can tell, the most expensive part of such a configuration is the media, and LTO media will cost about a third as much as the most economical MagStar media (extended length 3592 volumes read and written with TS1130 drives). With the sort of workload described above I don't expect any difficulty staying within the recommended limit on the number of times an LTO volume passes over the tape heads. Are there any other reasons to be nervous about using LTO for long term archives? IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error, please notify the author by replying to this message and then kindly delete the message. Thank you.
Re: LTO for long term archiving
I like the implication, but I'm pretty sure somebody actually thought being able to read the information would have been a good idea. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Remco Post Sent: Tuesday, May 05, 2009 2:39 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving I do agree, having the tapedrives around _could_ be important. I know of at least one environment that was able to produce the media that stores the data, but no drives. But then again, they only had to retain the data, not the infra to access it. On May 5, 2009, at 22:35 , Kelly Lipp wrote: To me the problem is having the drives around and more importantly, the interfaces to the drives. I think that probably the best bet is to plan on archiving a TSM server with a drive along with the media periodically. Snap off the last database backup, restore it on the to be archived server (a good test in itself), and store the whole kit together. If one needs to retrieve an archive, fire up the archived server, query the database to determine what tape is required, get it, retrieve the data and put the whole mess away. The other way to do this would be to migrate the archived data to new tape media as you march through time. I like this approach as that will have the double advantage of refreshing and verifying the data on those tapes. One could shelve the media in the archive pools and do this on a very controller basis when the media changes. Lots of data movement potentially, but it would become a nicely verified process that your auditors could look at to help ensure compliance. It's one thing to say we're doing and quite another to show we're doing it. Having the archive data more readily retrievable has obvious benefits as well. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebschman, George J. Sent: Tuesday, May 05, 2009 2:25 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] LTO for long term archiving Does anyone have 25 year old tape media or tape drives around? Will you stil be able to use LTOx media in 25 years? -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Thomas Denier Sent: Tuesday, May 05, 2009 4:11 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] LTO for long term archiving I work for a large hospital. I have been asked to investigate possible configurations for archiving something between a few hundred terabytes and a petabyte of data for 25 years. This would be clinical records that we need to keep in case of a malpractice suit. The retention period is 25 years because there are two ways we can get sued for alleged malpractice involving a pediatric patient. The parents or guardians have a seven year window of opportunity to file suit, starting at the time of the alleged malpractice. The patient has a seven year window of opportunity, starting at his or her 18th birthday. In principle, the retention period should vary depending on patient age, but nobody I have talked to so far thinks it is practical to sort records in this way; they want a uniform retention period that covers the worst case scenario (a patient allegedly harmed as a newborn suing just before the end of his or her seven year window). As far as I can tell, the most expensive part of such a configuration is the media, and LTO media will cost about a third as much as the most economical MagStar media (extended length 3592 volumes read and written with TS1130 drives). With the sort of workload described above I don't expect any difficulty staying within the recommended limit on the number of times an LTO volume passes over the tape heads. Are there any other reasons to be nervous about using LTO for long term archives? IMPORTANT: E-mail sent through the Internet is not secure. Legg Mason therefore recommends that you do not send any confidential or sensitive information to us via electronic mail, including social security numbers, account numbers, or personal identification numbers. Delivery, and or timely delivery of Internet mail is not guaranteed. Legg Mason therefore recommends that you do not send time sensitive or action-oriented messages to us via electronic mail. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error, please notify the author by replying to this message and then kindly delete the message. Thank you. -- Met
Re: Copypool using more tapes then primary tapepool
Do you manually kick reclamation on the copy pool? I noticed that the reclamation threshold on that pool is set to 100%. It isn't unusual for reclamation processing to be somewhat off among pools. Generally, the copy pool will be slightly larger since reclamation isn't as aggressive and you will always have more partially empty tapes in that pool (if they are being removed from the library) than you will have in the primary pool. 6-9 tapes I wouldn't worry. Hundreds? Then something is probably wrong. That said, you clearly need something better to do! Isn't it amazing how easy it is to run a TSM environment? The stuff that used to be hard is easy so you can start focus on other things! Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Larry Peifer Sent: Friday, April 17, 2009 2:39 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Copypool using more tapes then primary tapepool Why are we using more tapes in the copypool library vs the primary tape library? There is a 6 - 9 tape difference between the copypool and the primary tape pool. We average ~500 GB per tape so that's 1.5 - 4.5 TB of data. It doesn't seem like there should be that much of a discrepancy. There is both backup data and archive data mixed on the tapes and the DbBackups are taken into account. We have 2 identically configured IBM 3584 tape libraries. On a daily basis our disk pools are migrated (migrate stgpool diskpool lo=0) to the primary tape pool. Then a daily schedule (backup stgpool tapepool6 tapepool7 maxprocess=4) is run to keep everything equal between the 2 tape libraries. Daily expiration and reclamation processes finish fine. Schedules report successful completion daily. Running TSM Server 5.4 with AIX 5.3 on p520 server. LTO2 tapes with HW compression Storage Pool configurations: Storage Pool Name: DISKPOOL Storage Pool Type: Primary Device Class Name: DISK Estimated Capacity: 2,400 G Space Trigger Util: 0.4 Pct Util: 0.4 Pct Migr: 0.4 Pct Logical: 100.0 High Mig Pct: 90 Low Mig Pct: 70 Migration Delay: 0 Migration Continue: Yes Migration Processes: 4 Reclamation Processes: Next Storage Pool: TAPEPOOL6 Reclaim Storage Pool: Maximum Size Threshold: No Limit Access: Read/Write Description: Main Disk Storage Pool Overflow Location: Cache Migrated Files?: No Collocate?: Reclamation Threshold: Offsite Reclamation Limit: Maximum Scratch Volumes Allowed: Number of Scratch Volumes Used: Delay Period for Volume Reuse: Migration in Progress?: No Amount Migrated (MB): 1,235,496.70 Elapsed Migration Time (seconds): 9,284 Reclamation in Progress?: Last Update by (administrator): admin Last Update Date/Time: 08/24/07 09:50:37 Storage Pool Data Format: Native Copy Storage Pool(s): Active Data Pool(s): Continue Copy on Error?: Yes CRC Data: No Reclamation Type: Overwrite Data when Deleted: Storage Pool Name: TAPEPOOL6 Storage Pool Type: Primary Device Class Name: LTOCLASS6 Estimated Capacity: 121,841 G Space Trigger Util: Pct Util: 32.9 Pct Migr: 47.0 Pct Logical: 99.3 High Mig Pct: 90 Low Mig Pct: 70 Migration Delay: 0 Migration Continue: Yes Migration Processes: 2 Reclamation Processes: 2 Next Storage Pool: Reclaim Storage Pool: Maximum Size Threshold: No Limit Access: Read/Write Description: Primary Sequential Tape Overflow Location: Cache Migrated Files?: Collocate?: No Reclamation Threshold: 100 Offsite Reclamation Limit: Maximum Scratch Volumes Allowed: 300 Number of Scratch Volumes Used: 152 Delay Period for Volume Reuse: 3 Day(s) Migration in Progress?: No Amount Migrated (MB): 0.00 Elapsed Migration Time (seconds): 0 Reclamation in Progress?: No Last Update by (administrator): admin Last Update Date/Time: 04/07/09 14:06:34 Storage Pool Data Format: Native Copy Storage Pool(s): Active Data Pool(s): Continue Copy on Error?: Yes CRC Data: Yes Reclamation Type: Threshold Overwrite Data when Deleted: Storage Pool Name: TAPEPOOL7 Storage Pool Type: Copy Device Class Name: LTOCLASS7 Estimated Capacity: 120,330 G Space Trigger Util: Pct Util: 32.3 Pct Migr: Pct Logical: 99.3 High Mig Pct: Low Mig Pct: Migration Delay: Migration Continue: Yes Migration Processes: Reclamation Processes: 2 Next Storage Pool: Reclaim Storage Pool: Maximum Size Threshold: Access: Read/Write Description: Copy Pool Overflow Location: Cache Migrated Files?: Collocate?: No Reclamation Threshold: 100 Offsite Reclamation Limit: No Limit Maximum Scratch Volumes Allowed: 300 Number of Scratch Volumes Used: 157 Delay Period for Volume Reuse: 3 Day(s) Migration in Progress?: Amount Migrated (MB): Elapsed Migration Time (seconds): Reclamation in Progress?: Yes Last Update by (administrator): admin Last Update Date/Time: 12/14/07 13:56:37 Storage Pool Data Format: Native Copy Storage Pool(s): Active Data Pool(s): Continue Copy on Error?: CRC Data: No Reclamation Type
Re: TSM database information
I would create a second TSM instance on your server and restore the appropriate database backup to that instance. Then you can issue the queries you need to determine if you have the tapes or if they were over written since the backup. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of RAYMOND J RAMIREZ RAMIREZ Sent: Tuesday, April 07, 2009 12:07 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] TSM database information Hello to all, I have a special situation. In February I moved an AIX client from our old TSM 5.2 server to add it to our newer TSM 5.4 server, then I deleted the file spaces that belonged to this client and removed the client from the old server. Now the users want to recover old files that was on the AIX client before I moved it and deleted the file spaces. I know I can restore the database to the point before the deletion, but I need to be sure that the files can be recovered before attempting this. Everything was backed up on IBM 3590 cartridges in a large IBM library, and I have a tape management report that identifies all the TSM database tapes as available, as all of the data tapes, too. But I also need to know which of the 4,000+ data tapes has the files I need. I need a method to read the TSM database tape file without restoring it, and in reading it, I wish to know which are the data tapes with the files I need. If the tapes are available (since reclamation and reuse could have destroyed the original files), then I will restore the database tape, and run TSM to recover the requested files. But if most or all of the tapes were reused (the tape management system can verify this), then I can be sure that the data is lost and I would not have to do the TSM database restore. It may sound confusing, but it is like knowing if there is fish in a lake before travelling toward the lake to catch fish.I am open to any and all suggestions and recommendations. Raymond J. Ramirez, P.E. Distributed Systems Supervisor ITS Operations and Infrastructure
Re: Backup files to empty tape
I assume you are using scratch tapes in your library for the pool where this data is heading. By default, TSM will choose one of the filling scratch tapes first. To make sure it uses a brand new tape, set all the filling tapes in that pool in that library to ReadOnly. Then start the backup to that pool. TSM will be forced to take another tape from scratch. Note the volser of the tape being used. When the backup is done, check that tape out and set the readonly tapes back to readwrite. There are variations on the theme but this will work. Make sure nothing else is going on during this or you might get other data on that tape. Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Mario Behring Sent: Wednesday, April 01, 2009 12:23 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Backup files to empty tape Hi list, I have to backup some files from a TSM node to an empty tape and send this tape away (check out from the library). What is the best approach to perform this task? Thanks Mario
Re: Tape Drive SN issue
I assume these message occur when you try to define the path? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Huebner,Andy,FORT WORTH,IT Sent: Monday, March 30, 2009 9:50 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Tape Drive SN issue We have a three drives that generate these messages and we are unable to use the drives because the path will not stay on-line. We have not been able to find a solution to this problem. 03/30/2009 04:30:41 ANR8963E Unable to find path to match the serial number defined for drive L1R3DR1 in library IBM3494A . (SESSION: 139365) 03/30/2009 04:30:41 ANR8873E The path from source TSMSERVER1 to destination L1R3DR1 (/dev/rmt13) is taken offline. (SESSION: 139365) These are 3590-E drives; the other 21 do not have this problem. The 24 drives are in one 3494 library, the library is shared by 2 TSM servers, 1 TSM server is the library manager. There are also 8 AS/400 LPARs that use the library. There have not been any problems reported by the AS/400. The problem started a few months ago with no known changes. The AIX error logs do not show any problems, the fiber switch also does not show any errors. IBM has checked the drives and they do not show any errors. These drives are on different HBA's and different FC switches, both of which have working drives. AIX has been restarted, due to other reasons, and that did not fix the problem. The drives have been reset and power cycled. We are running AIX 5.3 TSM 5.4.2.0. What have we missed, what could be looked at next? Does ANR8963E have a hidden meaning? Andy Huebner This e-mail (including any attachments) is confidential and may be legally privileged. If you are not an intended recipient or an authorized representative of an intended recipient, you are prohibited from using, copying or distributing the information in this e-mail or its attachments. If you have received this e-mail in error, please notify the sender immediately by return e-mail and delete all copies of this message and any attachments. Thank you.
Re: define path problems
You use the IBM TSM driver for HP drives, not the IBM drivers or the HP drivers... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alexander Lazarevich Sent: Thursday, March 26, 2009 1:39 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] define path problems Well, we have HP drives that came with the Overland Neo 4100 library. Any idea where to get the HP drivers for LTO2 drives? I've looked on HP's website and can only find the Windows drivers, not TSM. It also seems odd that these are identical drivers to the ones that were replaced, shouldn't the same TSM driver we've been using for the last 5 years still work for the replacement drives? Thanks for the help! Alex On Thu, 26 Mar 2009, Dennis, Melburn (IT Solutions US) wrote: It looks like you need to reinstall the drivers for the new drives. Your OS is seeing the drives, but TSM needs specific drivers for them to be associated as LTO class devices. If they are IBM drives, you can go the IBM ftp site and download the latest drivers (ftp://service.boulder.ibm.com/storage/devdrvr/Windows/Win2000/Latest/). Once you have updated the drivers, redefine the drives (delete paths, drives first), and it should show up as LTO. Mel Dennis Systems Engineer II Siemens IT Solutions and Services, Inc. Energy Data Center Operations 4400 Alafaya Trail MC Q1-108 Orlando, FL 32826 Tel.: 407-736-2360 Mob: 321-356-9366 Fax: 407-243-0260 mailto:melburn.den...@siemens.com www.usa.siemens.com/it-solutions -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alexander Lazarevich Sent: Thursday, March 26, 2009 12:33 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] define path problems I'm having another problem and could use more help. I've got lots of tapes in the library with lots of data on them, and they do seem to be defined and ready to go: tsm: ITG-TSMq libv Library Name Volume Name Status Owner Last Use Home Device Element Type --- -- - - --- - LB6.0.0.5 ITG002L2 PrivateITG-TSMData 45 LB6.0.0.5 ITG003L2 PrivateITG-TSMData 48 LB6.0.0.5 ITG004L2 PrivateITG-TSMData 53 but for some reason they are just unavailable to the server and I can't figure out why. One clue is that since I replaced the drives (delete drives, update library path, define drives, define drive paths) the new drives are showing up as a new device type GENERICTAPE1: tsm: ITG-TSMq drive Library Name Drive Name Device Type On-Line --- --- LB6.0.0.5MT1.0.0.4GENERICTAPE Yes LB6.0.0.5MT2.0.0.4GENERICTAPE Yes Before the drives were replaced, these showed up as LTOCLASS1 device types. Here's the device type I have: tsm: ITG-TSMq dev DeviceDeviceStorage DeviceFormatEst/Max Mount Class Access Pool TypeCapacity Limit Name StrategyCount (MB) - ----- - -- -- DISK Random 3 LTOCLASS1 Sequential 1LTO DRIVE DRIVES So my guess is I need to reassociate those tapes with device type LTO1CLASS, but I don't see any way to do this. Any idea if I'm on the right track? I appreciate the help, Alex On Thu, 26 Mar 2009, Alexander Lazarevich wrote: Thank you Mel. Alex On Thu, 26 Mar 2009, Dennis, Melburn (IT Solutions US) wrote: ANR8466E command: Invalid update request for drive drive name in library library name. Explanation: An invalid update request has been made for the given drive. This can occur if a new device name is given and the characteristics of the device do not match the characteristics of the original device. System action: The server does not process the command. User response: If a different type of drive has been installed, the old drive definition must be deleted with a DELETE DRIVE operation, and a new drive must be defined. The UPDATE DRIVE command cannot be used in this case. As it says in the error description, you must not only delete the drive paths, but the drive definitions themselves. Deleting the drive path does not remove the previous drives information, and when you try to redifine the path, its expecting to connect to that old drive, which obviously no longer exists. Mel Dennis Systems Engineer II Siemens IT Solutions and Services, Inc. Energy Data Center Operations 4400 Alafaya Trail MC Q1-108 Orlando, FL 32826 Tel.: 407-736-2360 Mob: 321-356-9366 Fax: 407
Re: define path problems
When you put the new drives in the library, Device Manager found them. You will need to right click on each drive and update the driver to use the TSM driver. The HP driver probably bound to the drive automatically after the reboot. You must have the IBM TSM Device Driver controlling these drives. Once you have the correctly done, you will have a mt device in the TSM Device Driver view. Update the drive and path definitions and TSM and you should be fine. Call me at the number below if you struggle and I'll help you. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alexander Lazarevich Sent: Thursday, March 26, 2009 1:39 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] define path problems Well, we have HP drives that came with the Overland Neo 4100 library. Any idea where to get the HP drivers for LTO2 drives? I've looked on HP's website and can only find the Windows drivers, not TSM. It also seems odd that these are identical drivers to the ones that were replaced, shouldn't the same TSM driver we've been using for the last 5 years still work for the replacement drives? Thanks for the help! Alex On Thu, 26 Mar 2009, Dennis, Melburn (IT Solutions US) wrote: It looks like you need to reinstall the drivers for the new drives. Your OS is seeing the drives, but TSM needs specific drivers for them to be associated as LTO class devices. If they are IBM drives, you can go the IBM ftp site and download the latest drivers (ftp://service.boulder.ibm.com/storage/devdrvr/Windows/Win2000/Latest/). Once you have updated the drivers, redefine the drives (delete paths, drives first), and it should show up as LTO. Mel Dennis Systems Engineer II Siemens IT Solutions and Services, Inc. Energy Data Center Operations 4400 Alafaya Trail MC Q1-108 Orlando, FL 32826 Tel.: 407-736-2360 Mob: 321-356-9366 Fax: 407-243-0260 mailto:melburn.den...@siemens.com www.usa.siemens.com/it-solutions -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alexander Lazarevich Sent: Thursday, March 26, 2009 12:33 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] define path problems I'm having another problem and could use more help. I've got lots of tapes in the library with lots of data on them, and they do seem to be defined and ready to go: tsm: ITG-TSMq libv Library Name Volume Name Status Owner Last Use Home Device Element Type --- -- - - --- - LB6.0.0.5 ITG002L2 PrivateITG-TSMData 45 LB6.0.0.5 ITG003L2 PrivateITG-TSMData 48 LB6.0.0.5 ITG004L2 PrivateITG-TSMData 53 but for some reason they are just unavailable to the server and I can't figure out why. One clue is that since I replaced the drives (delete drives, update library path, define drives, define drive paths) the new drives are showing up as a new device type GENERICTAPE1: tsm: ITG-TSMq drive Library Name Drive Name Device Type On-Line --- --- LB6.0.0.5MT1.0.0.4GENERICTAPE Yes LB6.0.0.5MT2.0.0.4GENERICTAPE Yes Before the drives were replaced, these showed up as LTOCLASS1 device types. Here's the device type I have: tsm: ITG-TSMq dev DeviceDeviceStorage DeviceFormatEst/Max Mount Class Access Pool TypeCapacity Limit Name StrategyCount (MB) - ----- - -- -- DISK Random 3 LTOCLASS1 Sequential 1LTO DRIVE DRIVES So my guess is I need to reassociate those tapes with device type LTO1CLASS, but I don't see any way to do this. Any idea if I'm on the right track? I appreciate the help, Alex On Thu, 26 Mar 2009, Alexander Lazarevich wrote: Thank you Mel. Alex On Thu, 26 Mar 2009, Dennis, Melburn (IT Solutions US) wrote: ANR8466E command: Invalid update request for drive drive name in library library name. Explanation: An invalid update request has been made for the given drive. This can occur if a new device name is given and the characteristics of the device do not match the characteristics of the original device. System action: The server does not process the command. User response: If a different type of drive has been installed, the old drive definition must be deleted with a DELETE DRIVE operation, and a new drive must be defined. The UPDATE DRIVE command cannot be used in this case. As it says in the error description, you must not only delete
Re: VTL and Dedup ( TS7569G)
Yes, that is correct. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Hart, Charles A Sent: Monday, March 23, 2009 10:02 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) I imagine the 1400MBS is for the Clustered version? -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Sunday, March 22, 2009 12:36 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) The white paper I listed in my response to this thread was written by ESG. They tested the TS7560 and obtained on the order of 1400MB/sec. And Charles is correct: it is an x86 box, actually the IBM x3850 which has, perhaps, the best architecture in the class. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Hart, Charles A Sent: Saturday, March 21, 2009 8:32 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) It works well if you understand your data and how you can push to it with in reason before you deploy another. The IBM product likes more CPU cores ... Understand these are x86 boxes... We see up to 500MBS Writes to one of our VTL's that ingests Exchange Backups via the TSM TDP. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of W. Curtis Preston Sent: Saturday, March 21, 2009 12:24 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Funny. I would say that BAD dedupe is the enemy of throughput. There IS good dedupe. I've seen it. It hurts neither backup, nor restore performance. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Clark, Robert A Sent: Friday, March 20, 2009 4:44 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Dedupe is an errand boy, sent by the storage industry, to collect a bill. Dedupe is the enemy of throughput. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of W. Curtis Preston Sent: Friday, March 20, 2009 4:25 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Why do you hate all things dedupe? -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Friday, March 20, 2009 10:12 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Funny, but I was researching the TS7650 yesterday and found this article on the IBM website. Pretty good detail about the product in a non-TSM environment. ftp://service.boulder.ibm.com/storage/tape/ts7650g_esg_validation.pdf And then this on in the TSM environment. I think this one might have been written by somebody somewhat less familiar with TSM than we would be. Seemed a little heavy handed about TSM. ftp://ftp.software.ibm.com/common/ssi/sa/wh/n/tsw03043usen/TSW03043USEN. PDF My overall impression, and I hate all things de-dup, was this is a pretty good product offering. I'm sure it's way expensive but understand there are some follow on products coming that will address the lower end of this market. Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alex Paschal Sent: Friday, March 20, 2009 11:03 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Sabar. I couldn't find a TS7569G via Google, but on the TS7650G, also a deduping VTL, after data goes through the factoring (dedup) algorithm it is run through a compression algorithm. You probably won't see much deduplication, but on the first backup you should see a decrease in size similar to the decrease you would see from the compression on a tape drive. Regards, Alex -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Sabar Martin Hasiholan Panggabean Sent: Thursday, March 19, 2009 5:10 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Does anyone here has been implementing or know how Dedup works in TSM using TS7569G ? Let say I have 100 TB of data and backup to this VTL. On the 1st attempt of backup / full backup, will this data size decrease on the VTL BR, Martin P This message (including any attachments) is intended only for the use of the individual or entity to which it is addressed and may contain information that is non-public, proprietary, privileged, confidential, and exempt from disclosure under applicable law or may constitute as attorney work product
Re: VTL and Dedup ( TS7569G)
What Robert said. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of W. Curtis Preston Sent: Saturday, March 21, 2009 11:24 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Funny. I would say that BAD dedupe is the enemy of throughput. There IS good dedupe. I've seen it. It hurts neither backup, nor restore performance. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Clark, Robert A Sent: Friday, March 20, 2009 4:44 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Dedupe is an errand boy, sent by the storage industry, to collect a bill. Dedupe is the enemy of throughput. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of W. Curtis Preston Sent: Friday, March 20, 2009 4:25 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Why do you hate all things dedupe? -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Friday, March 20, 2009 10:12 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Funny, but I was researching the TS7650 yesterday and found this article on the IBM website. Pretty good detail about the product in a non-TSM environment. ftp://service.boulder.ibm.com/storage/tape/ts7650g_esg_validation.pdf And then this on in the TSM environment. I think this one might have been written by somebody somewhat less familiar with TSM than we would be. Seemed a little heavy handed about TSM. ftp://ftp.software.ibm.com/common/ssi/sa/wh/n/tsw03043usen/TSW03043USEN. PDF My overall impression, and I hate all things de-dup, was this is a pretty good product offering. I'm sure it's way expensive but understand there are some follow on products coming that will address the lower end of this market. Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alex Paschal Sent: Friday, March 20, 2009 11:03 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Sabar. I couldn't find a TS7569G via Google, but on the TS7650G, also a deduping VTL, after data goes through the factoring (dedup) algorithm it is run through a compression algorithm. You probably won't see much deduplication, but on the first backup you should see a decrease in size similar to the decrease you would see from the compression on a tape drive. Regards, Alex -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Sabar Martin Hasiholan Panggabean Sent: Thursday, March 19, 2009 5:10 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Does anyone here has been implementing or know how Dedup works in TSM using TS7569G ? Let say I have 100 TB of data and backup to this VTL. On the 1st attempt of backup / full backup, will this data size decrease on the VTL BR, Martin P This message (including any attachments) is intended only for the use of the individual or entity to which it is addressed and may contain information that is non-public, proprietary, privileged, confidential, and exempt from disclosure under applicable law or may constitute as attorney work product. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, notify us immediately by telephone and (i) destroy this message if a facsimile or (ii) delete this message immediately if this is an electronic communication. Thank you. DISCLAIMER: This message is intended for the sole use of the addressee, and may contain information that is privileged, confidential and exempt from disclosure under applicable law. If you are not the addressee you are hereby notified that you may not use, copy, disclose, or distribute to anyone the message or any information contained in the message. If you have received this message in error, please immediately advise the sender by reply email and delete this message.
Re: VTL and Dedup ( TS7569G)
Funny, but I was researching the TS7650 yesterday and found this article on the IBM website. Pretty good detail about the product in a non-TSM environment. ftp://service.boulder.ibm.com/storage/tape/ts7650g_esg_validation.pdf And then this on in the TSM environment. I think this one might have been written by somebody somewhat less familiar with TSM than we would be. Seemed a little heavy handed about TSM. ftp://ftp.software.ibm.com/common/ssi/sa/wh/n/tsw03043usen/TSW03043USEN.PDF My overall impression, and I hate all things de-dup, was this is a pretty good product offering. I'm sure it's way expensive but understand there are some follow on products coming that will address the lower end of this market. Thanks, Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Alex Paschal Sent: Friday, March 20, 2009 11:03 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Sabar. I couldn't find a TS7569G via Google, but on the TS7650G, also a deduping VTL, after data goes through the factoring (dedup) algorithm it is run through a compression algorithm. You probably won't see much deduplication, but on the first backup you should see a decrease in size similar to the decrease you would see from the compression on a tape drive. Regards, Alex -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Sabar Martin Hasiholan Panggabean Sent: Thursday, March 19, 2009 5:10 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] VTL and Dedup ( TS7569G) Hi, Does anyone here has been implementing or know how Dedup works in TSM using TS7569G ? Let say I have 100 TB of data and backup to this VTL. On the 1st attempt of backup / full backup, will this data size decrease on the VTL BR, Martin P This message (including any attachments) is intended only for the use of the individual or entity to which it is addressed and may contain information that is non-public, proprietary, privileged, confidential, and exempt from disclosure under applicable law or may constitute as attorney work product. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, notify us immediately by telephone and (i) destroy this message if a facsimile or (ii) delete this message immediately if this is an electronic communication. Thank you.
Re: Server platform comparison
Hashed many times over the past several months. Might check the archives. Should be based on knowledge within the organization. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Mark Devine Sent: Friday, March 20, 2009 3:38 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Server platform comparison I haven't googled it just yet, but I though you folks may have references as to the best TSM server platform. Currently targeted for AIX, but the server consolidation crew is throwing around HP Linux. Any URL's would be greatly appreciated. TIA
Re: network data transfer rate aggregate data transfer rate
Network data transfer rate is the rate when the client is actually sending the data. Usually very close to the wire speed. Aggregate is amount of data transferred divided by the backup's elapsed time. Not just the time the client is actually moving data on the wire. If you don't see the network rate very high (in the xx,xxx range) it can indicate a network problem. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Arthur Poon Sent: Thursday, March 19, 2009 10:04 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] network data transfer rate aggregate data transfer rate I have a question about aggregate data transfer rate. Why it is always a lot slower than the network data transfer rate? For example: ANE4961I (Session: 18137, Node: MFDCITRIX) Total number of bytes transferred: 13.61 GB (SESSION: 18137) 03/18/09 15:37:37 ANE4963I (Session: 18137, Node: MFDCITRIX) Data transfer time: 914.17 sec (SESSION: 18137) 03/18/09 15:37:37 ANE4966I (Session: 18137, Node: MFDCITRIX) Network data transfer rate: 15,618.11 KB/sec (SESSION: 18137) 03/18/09 15:37:37 ANE4967I (Session: 18137, Node: MFDCITRIX) Aggregate data transfer rate: 2,220.91 KB/sec (SESSION: 18137) 03/18/09 15:37:37 ANE4968I (Session: 18137, Node: MFDCITRIX) Objects compressed by: 14% (SESSION: 18137) AP
Re: Mixing LTO2 and LTO4 drives and media in the same library
LTO4 drives will read, but not write LTO2 media. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Howard Coles Sent: Thursday, March 19, 2009 1:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Mixing LTO2 and LTO4 drives and media in the same library We do this with a mixed LTO3 and LTO2 drives. We did not partition the library. However, I would highly suggest putting the smaller format higher in the drive number list. DRIVE01 = LTO2 . . . DRIVE10 =LTO4. That will keep the LTO4 drives available for LTO4 tapes, if LTO4 drives will read and write LTO2 tapes. See Ya' Howard -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of David McClelland Sent: Thursday, March 19, 2009 9:13 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Mixing LTO2 and LTO4 drives and media in the same library Hi Team, I've a question about mixing LTO drives and media in the same library - LTOn and LTOn+2. * Windows TSM 5.4.3.2 (soon to be migrated to a 5.5.1.1 server running on AIX) * IBM 3584 single frame library, presently with 4 x IBM LTO2 drives * Currently no library clients or storage agents. My client is looking to migrate from 4 x LTO2 drives to 8 x LTO4 drives in their 3584 library (as well as performing some other TSM Server platform migration activities). The LTO drive generation-spanning capabilities are such that LTO4 drives can read from but not write to LTO2 media; that LTO2 drives can read/write to LTO2 media, but will have nothing to do with LTO4 media. Now, for an interim period, it has been suggested that the library be configured with 5 x LTO4 drives and 3 x LTO2 drives. TSM Server and the IBM 3584 Tape Library are both stated to support mixed media and tape drives between these LTO generations. However, the idea behind this is that during this interim period, new incoming backup data should be written to LTO4 media by the LTO4 drives (as well as some existing LTO2 data being moved/consolidated onto LTO4 media), and that offsite copies only continue to be generated to LTO2 media using the LTO2 drives (various reasons for this, including maintaining media compatibility for DR purposes). If possible, largely I think for flexibility, the preference is to achieve this within a single library partition. It's this last bit that I'm having trouble getting my head around. Obviously, a new device class (FORMAT=ULTRIUM4C) will be added to the TSM Server to make use of the LTO4 drives (making sure that the current LTO2 device class is locked down to FORMAT=ULTRIUM2C), as will a new set of LTO4 storage pools pointing to this new device class. However, with both of these drive types being located in a single library partition I don't quite see how they'll be able to make sure that only LTO2 scratch tapes get mounted into LTO2 drives and that LTO4 drives don't go trying to write to LTO2 media. Is it simply a case of manually allocating the LTO2 scratch media to the LTO2 copy storagepool (and setting its MAXSCRATCH to 0), and ensuring that LTO4 is the only media in the general scratch pool? Will TSM ensure that, given we're explicit in our device class definitions, when a backup stgpool operation requests a mount of an LTO2 scratch volume (or indeed an LTO2 filling volume) to perform a write it'll only mount one into one of the LTO2 drives, and similar for LTO4 media and drives? I suspect not... Is anyone running with a similar configuration to this (if it's possible)? Of course, the other blindingly obvious option is to partition the 3584, creating one library partition for the 3 x LTO2 drives and enough slots to manage the offsite requirement, and another for the 5 x LTO4 drives with the remainder of the carts. The client doesn't have ALMS for quick flexible library partitioning (I don't know about the cost of this feature, but I expect that they'd be unwilling to consider it for only an interim period) so we'd have to use more rigid non-ALMS partitioning method. This option appeals to me in that it's pretty simple to get my head around, but perhaps requires a little more work to perform the library repartitioning. Any thoughts gratefully received! Cheers, David Mc, London, UK No virus found in this outgoing message. Checked by AVG. Version: 7.5.524 / Virus Database: 270.11.18/2009 - Release Date: 18/03/2009 07:17
Re: Calculating Change in Backup Environment
First, you aren't really doing what you think you're doing: providing the ability to restore data back to any point in the previous 90 days. You get seven days for a file that changes every day. Steve is right: should probably do something like verexists=unlimited and retextra=21. That way if you do a backup more than once a day you can still get 21 days worth of restores. This boils down to how many files really change 21 times in 21 days. Most files won't do that. Large things, like SQL and Exchange databases will. But you wouldn't really want to restore your SQL database back 21 days in any event. How would you get back to scratch? You couldn't. So keeping these types of things for shorter periods makes more sense. Anecdotally, my experience has been that increasing/reducing the numbers does not generally result in a ton more storage consumed/saved as most files are created and then never changed (or deleted). They get backed up the one time and then live until they are deleted and retonly is reached. I would definitely get to the 21/21 soon, though, as your user's expectations are probably different from what you're providing! Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Steven Harris Sent: Wednesday, March 18, 2009 3:24 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Calculating Change in Backup Environment Dennis, FWIW, I don't like a N days/N versions strategy. It can have unexpected side effects. N Days/unlimited versions has more predictable behaviour and also is more efficient at expiry time. Regards Steve. Steven Harris TSM Admin, Sydney Australia Dennis, Melburn (IT Solutions US) To melburn.den...@s ADSM-L@VM.MARIST.EDU IEMENS.COMcc Sent by: ADSM: Dist Stor Subject Manager [ADSM-L] Calculating Change in ads...@vm.marist Backup Environment .EDU 18/03/2009 11:46 PM Please respond to ADSM: Dist Stor Manager ads...@vm.marist .EDU Currently, our backup environment employs a 90-day / 7 revision backup policy. My customer has come to me to find out how much our backup data would grow or shrink if we went to a 21-day / 21-revision backup policy. Have any of you out there experience requests like this before, and if so, how were you able to get a reasonable guestimate of this. Mel Dennis Systems Engineer II Siemens IT Solutions and Services, Inc. Energy Data Center Operations 4400 Alafaya Trail MC Q1-108 Orlando, FL 32826 Tel.: 407-736-2360 Mob: 321-356-9366 Fax: 407-243-0260 mailto:melburn.den...@siemens.com www.usa.siemens.com/it-solutions
Re: Two different retention policies for the same node
The more important issue regarding DR is the prioritization of application restore based on the business itself. As it turns out, after a disaster about 90% of the stuff that's backed up isn't necessary to run the business. And the ability to get the previous seven days doesn't help either. More important to have two different DR pools: one for important data and one for the rest. Then optimize the important data DR pool to ensure it can do what you need it to do: I must have application XYZ back up and available to users within 24 hours of the disaster to a point no further back than 48 hours. And it turns out that most business's ability to use recovered data within 24 hours is sketchy at best. Where are the people that use the data going to be? How will customers interact with them? All of these issues are actually about 100 times more difficult than restoring data. Yet few think of them! If you worry DR application by application and think about all aspects of using the data, the problem actually becomes simpler since there is less data to worry about the time frame to recover it is probably longer than you think. It's all about RPO and RTO. In our sales practice, I'm spending a lot of time consulting (during pre-sales so it's free) about DR issues. Bottom line: you need to have a very good plan, but since you will probably never execute (beyond testing), you probably shouldn't spend too much time/money on it. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Howard Coles Sent: Tuesday, March 17, 2009 2:47 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Two different retention policies for the same node There could be some serious issues with this. If you have an onsite volume that has 170 day old data, with 5 day old data and 80 day old data (due to reclamation), and the volume goes bad, all you'll be able to restore is the 5 day old data. However, this is a real challenge. I'd like to see the solution. It appears on the surface that this is a result of is the we want to keep it forever but can't afford the cost mentality. Champagne taste on Beer budget. See Ya' Howard -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Michael Green Sent: Tuesday, March 17, 2009 2:57 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Two different retention policies for the same node I've been asked to provide a DR/BAckup solution that seems to contradict TSM methodology, but I've decided I'll throw this in here anyway. Given the following retention policy: RETE=180 RETO=180 VERE=NOL VERD=NOL (180 days, no version limit) I've been asked to find a way to keep offsite only 7 days worth of data (on deduped disk or somthng like that), both active and inactive. So that it would allow us to restore complete system image from any day within last week. Doable (without resorting to double backups under different MCs)? -- Warm regards, Michael Green
Re: TSM Library Manager
I've seen something similar in the non sharing environment if the element number to drive relationship is whacked. For instance, you suggest that the drive at /rmt0 is element 256 when in reality it's 257. The library is instructed to mount a tape in element 256 and then TSM looks for tape in /rmt0. Verify that the paths and element numbers actually line-up correctly. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Morris.Marshael Sent: Tuesday, March 10, 2009 10:24 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] TSM Library Manager I have setup a library manager in TSM. This server is TSM 5.5.2 I am setting up a library client and it also has TSM 5.5.2. The TSM library manager is using the drives with no problem. I have setup the paths for the library client and I'm having problems with the client. The server, when requested, will mount a tape from the client and change ownership to the client name. Within the client activity log I get: ANR0408I Session 15 started for server LIBRARYSERVER (AIX-RS/6000) (Tcp/Ip) for library sharing. ANR0409I Session 15 ended for server LIBRARYSERVER (AIX-RS/6000). ANR8779E Unable to open drive rmt2, error number=2. ANR0409I Session 3 ended for server LIBRARYSERVER (AIX-RS/6000). ANR1404W Scratch volume mount request denied - mount failed. Librarymanager tapes from AIX: lsdev -Cc tape rmt0 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt1 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt2 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt3 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt4 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt5 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) smc0 Available 07-08-02 IBM 3584 Library Medium Changer (FCP) client tapes from AIX: lsdev -Cc tape rmt0 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt1 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt2 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt3 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt4 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt5 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) smc0 Available 07-08-01 IBM 3584 Library Medium Changer (FCP) I have checked the serial numbers of the drives on the server to the client and I believe that they are correct. Can someone help me in figuring out what I have missed or what might be the problem? Thanks, Marshael mccg.org email firewall made the following annotation CONFIDENTIALITY NOTICE: The information transmitted in this e-mail message, including any attachments, is for the sole use of the intended recipient(s) or entity to which it is addressed and may contain confidential, privileged and/or proprietary information. Any unauthorized review, retransmission, use, disclosure, dissemination or other use of,or taking any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you are not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, or other use of this message or its attachments is strictly prohibited. If you have received this message in error, please notify the sender immediately by reply e-mail, or by calling (478) 633-7272, and destroy the original message, attachments and all copies thereof on all computers and in any other form. Thank you. The Medical Center Of Central Georgia. http://www.mccg.org/ 03/10/09, 12:24:19
Re: How do I override compression setting?
How about exclude.compression on that filespace? exclude.compression Excludes files from compression processing if the compression option is set to yes. This option applies to backups and archives. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Schneider, John Sent: Tuesday, March 10, 2009 10:11 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] How do I override compression setting? Greetings, I have a specific NFS filesystem on a TSM AIX client that I need to backup without compression because the files are very large and already compressed, so it is pointless to compress them. But compression is yes in the dsm.sys file because I want to compress the other filesystems. The TSM client definition for compression is set to client, so presumably the client can choose to either compress or not. Here is the dsmsched.log of the schedule. Note that I turn off compression in the schedule: 03/10/09 10:00:38 03/10/09 10:00:38 Schedule Name: @1002 03/10/09 10:00:38 Action:Incremental 03/10/09 10:00:38 Objects: /mnt/mksysb 03/10/09 10:00:38 Options: -subdir=y -compression=n 03/10/09 10:00:38 Server Window Start: 09:57:56 on 03/10/09 03/10/09 10:00:38 03/10/09 10:00:38 Executing scheduled command now. 03/10/09 10:00:38 --- SCHEDULEREC OBJECT BEGIN @1002 03/10/09 09:57:56 03/10/09 10:00:38 Incremental backup of volume '/mnt/mksysb' 03/10/09 10:00:45 ANS1898I * Processed 500 files * 03/10/09 10:00:50 ANS1898I * Processed 1,000 files * Buncha lines skipped... 03/10/09 10:02:12 ANS1898I * Processed 5,500 files * 03/10/09 10:07:02 Normal File-- 2,489,241,600 /mnt/mksysb/osba ckup/eeyore/eeyore-03082009 ANS1360I Compressed Data Grew 03/10/09 10:13:05 Normal File-- 2,489,241,600 /mnt/mksysb/osba ckup/eeyore/eeyore-03082009 [Sent] 03/10/09 10:13:05 Successful incremental backup of '/mnt/mksysb' 03/10/09 10:13:05 --- SCHEDULEREC STATUS BEGIN 03/10/09 10:13:05 Total number of objects inspected:5,624 03/10/09 10:13:05 Total number of objects backed up:1 03/10/09 10:13:05 Total number of objects updated: 0 03/10/09 10:13:05 Total number of objects rebound: 0 03/10/09 10:13:05 Total number of objects deleted: 0 03/10/09 10:13:05 Total number of objects expired: 0 03/10/09 10:13:05 Total number of objects failed: 0 03/10/09 10:13:05 Total number of bytes transferred: 3.93 GB 03/10/09 10:13:05 Data transfer time:8.86 sec Why am I still getting ANS1360I Compressed Data Grew when the options for the schedule are -compression=n? It shouldn't be trying to compress, should it? We are running TSM client 5.4.2.0, and TSM server 5.4.3.0 on AIX, in case that matters. Best Regards, John D. Schneider Lead Systems Administrator - Storage Sisters of Mercy Health Systems 3637 South Geyer Road St. Louis, MO 63127 Phone: 314-364-3150 Cell: 314-750-8721 Email: john.schnei...@mercy.net This e-mail contains information which (a) may be PROPRIETARY IN NATURE OR OTHERWISE PROTECTED BY LAW FROM DISCLOSURE, and (b) is intended only for the use of the addressee(s) named above. If you are not the addressee, or the person responsible for delivering this to the addressee(s), you are notified that reading, copying or distributing this e-mail is prohibited. If you have received this e-mail in error, please contact the sender immediately.
Re: TSM Library Manager
Can you see that the library is indeed hanging a tape in some drive? And if so, is it indeed the correct drive? Visual inspection of what you think is happening is what I'm suggesting. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Morris.Marshael Sent: Tuesday, March 10, 2009 11:11 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM Library Manager I checked the drives to see what element numbers they were pointing to and checked the library to see what element they were showing for the drives, all are in sync. Thanks, Marshael -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Tuesday, March 10, 2009 1:04 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: TSM Library Manager I've seen something similar in the non sharing environment if the element number to drive relationship is whacked. For instance, you suggest that the drive at /rmt0 is element 256 when in reality it's 257. The library is instructed to mount a tape in element 256 and then TSM looks for tape in /rmt0. Verify that the paths and element numbers actually line-up correctly. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Morris.Marshael Sent: Tuesday, March 10, 2009 10:24 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] TSM Library Manager I have setup a library manager in TSM. This server is TSM 5.5.2 I am setting up a library client and it also has TSM 5.5.2. The TSM library manager is using the drives with no problem. I have setup the paths for the library client and I'm having problems with the client. The server, when requested, will mount a tape from the client and change ownership to the client name. Within the client activity log I get: ANR0408I Session 15 started for server LIBRARYSERVER (AIX-RS/6000) (Tcp/Ip) for library sharing. ANR0409I Session 15 ended for server LIBRARYSERVER (AIX-RS/6000). ANR8779E Unable to open drive rmt2, error number=2. ANR0409I Session 3 ended for server LIBRARYSERVER (AIX-RS/6000). ANR1404W Scratch volume mount request denied - mount failed. Librarymanager tapes from AIX: lsdev -Cc tape rmt0 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt1 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt2 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt3 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt4 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) rmt5 Available 07-08-02 IBM 3580 Ultrium Tape Drive (FCP) smc0 Available 07-08-02 IBM 3584 Library Medium Changer (FCP) client tapes from AIX: lsdev -Cc tape rmt0 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt1 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt2 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt3 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt4 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) rmt5 Available 07-08-01 IBM 3580 Ultrium Tape Drive (FCP) smc0 Available 07-08-01 IBM 3584 Library Medium Changer (FCP) I have checked the serial numbers of the drives on the server to the client and I believe that they are correct. Can someone help me in figuring out what I have missed or what might be the problem? Thanks, Marshael mccg.org email firewall made the following annotation CONFIDENTIALITY NOTICE: The information transmitted in this e-mail message, including any attachments, is for the sole use of the intended recipient(s) or entity to which it is addressed and may contain confidential, privileged and/or proprietary information. Any unauthorized review, retransmission, use, disclosure, dissemination or other use of,or taking any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you are not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, or other use of this message or its attachments is strictly prohibited. If you have received this message in error, please notify the sender immediately by reply e-mail, or by calling (478) 633-7272, and destroy the original message, attachments and all copies thereof on all computers and in any other form. Thank you. The Medical Center Of Central Georgia. http://www.mccg.org/ 03/10/09, 12:24:19
Re: How do I override compression setting?
That is the question: does the compress=no on the command line override the compress yes in the dsm.sys? Apparently not. The docs state, though, that the exclude.compression will override the dsm.sys. Might be you only option. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Schneider, John Sent: Tuesday, March 10, 2009 11:29 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] How do I override compression setting? Kelly, It is good to know an option like that exists, thanks. But that is not really my question. What is wrong with what I am doing? I don't see any mistakes in my configuration. When I create a schedule and put -compression=n in the options for the schedule, the TSM client should turn compression off for that schedule, right? I shouldn't have to put a special exclude in, should I? Best Regards, John D. Schneider Phone: 314-364-3150 Cell: 314-750-8721 Email: john.schnei...@mercy.net -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Tuesday, March 10, 2009 12:07 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] How do I override compression setting? How about exclude.compression on that filespace? exclude.compression Excludes files from compression processing if the compression option is set to yes. This option applies to backups and archives. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Schneider, John Sent: Tuesday, March 10, 2009 10:11 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] How do I override compression setting? Greetings, I have a specific NFS filesystem on a TSM AIX client that I need to backup without compression because the files are very large and already compressed, so it is pointless to compress them. But compression is yes in the dsm.sys file because I want to compress the other filesystems. The TSM client definition for compression is set to client, so presumably the client can choose to either compress or not. Here is the dsmsched.log of the schedule. Note that I turn off compression in the schedule: 03/10/09 10:00:38 03/10/09 10:00:38 Schedule Name: @1002 03/10/09 10:00:38 Action:Incremental 03/10/09 10:00:38 Objects: /mnt/mksysb 03/10/09 10:00:38 Options: -subdir=y -compression=n 03/10/09 10:00:38 Server Window Start: 09:57:56 on 03/10/09 03/10/09 10:00:38 03/10/09 10:00:38 Executing scheduled command now. 03/10/09 10:00:38 --- SCHEDULEREC OBJECT BEGIN @1002 03/10/09 09:57:56 03/10/09 10:00:38 Incremental backup of volume '/mnt/mksysb' 03/10/09 10:00:45 ANS1898I * Processed 500 files * 03/10/09 10:00:50 ANS1898I * Processed 1,000 files * Buncha lines skipped... 03/10/09 10:02:12 ANS1898I * Processed 5,500 files * 03/10/09 10:07:02 Normal File-- 2,489,241,600 /mnt/mksysb/osba ckup/eeyore/eeyore-03082009 ANS1360I Compressed Data Grew 03/10/09 10:13:05 Normal File-- 2,489,241,600 /mnt/mksysb/osba ckup/eeyore/eeyore-03082009 [Sent] 03/10/09 10:13:05 Successful incremental backup of '/mnt/mksysb' 03/10/09 10:13:05 --- SCHEDULEREC STATUS BEGIN 03/10/09 10:13:05 Total number of objects inspected:5,624 03/10/09 10:13:05 Total number of objects backed up:1 03/10/09 10:13:05 Total number of objects updated: 0 03/10/09 10:13:05 Total number of objects rebound: 0 03/10/09 10:13:05 Total number of objects deleted: 0 03/10/09 10:13:05 Total number of objects expired: 0 03/10/09 10:13:05 Total number of objects failed: 0 03/10/09 10:13:05 Total number of bytes transferred: 3.93 GB 03/10/09 10:13:05 Data transfer time:8.86 sec Why am I still getting ANS1360I Compressed Data Grew when the options for the schedule are -compression=n? It shouldn't be trying to compress, should it? We are running TSM client 5.4.2.0, and TSM server 5.4.3.0 on AIX, in case that matters. Best Regards, John D. Schneider Lead Systems Administrator - Storage Sisters of Mercy Health Systems 3637 South Geyer Road St. Louis, MO 63127 Phone: 314-364-3150 Cell: 314-750-8721 Email: john.schnei...@mercy.net This e-mail contains information which (a) may be PROPRIETARY IN NATURE OR OTHERWISE PROTECTED BY LAW FROM DISCLOSURE, and (b) is intended only for the use of the addressee(s) named above. If you are not the addressee, or the person responsible for delivering this to the addressee(s), you
Re: TSM Library Manager
And always consider what the truly smart guys have to say first when reading the list! Thanks Paul. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Tuesday, March 10, 2009 1:41 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] TSM Library Manager At 03:17 PM 3/10/2009, Schneider, John wrote: Greetings, I have a solution to the problem of library sharing and different rmt names. We have a similar, but different solution. We have a script we use to rename AIX tape devices to a predictable name, based on the last 4 characters of the WWN. After running it on each AIX system, the device names will be the same on each system. After running tsmchrmt rmt0, the device will have a name similar to: rmt.f0c6.0.0 where f0c6 are the last 4 chars of the device's WWN. I don't recall where we got the seed for this script, but here is our version, which we call tsmchrmt: #!/bin/sh if [ $# != 1 ] then echo must specify 1 rmt device name as an argument. exit 4 fi d=$1 WWN=`/usr/sbin/lsattr -El $d -a ww_name|cut -f2 -d |cut -c15-` LUN=`/usr/sbin/lsattr -El $d -a lun_id|cut -f2 -d |cut -c3` root=`echo $d|cut -c1-3` new_name=$root.$WWN.$LUN.0 let j=0 while [[ -e /dev/$new_name ]] do let j=j+1 new_name=$root.$WWN.$LUN.$j done /usr/sbin/chdev -l $d -a new_name=$new_name ..Paul -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Preferred TSM Platform
I have to disagree with that. We routinely run multiple (up to six, and the only reason it's only six is we don't have any more to test so perhaps more will run) LTO4 drives as fast as they want to run using and IBM x3850 processor. We run four at a time using an x3650. The buses are PCI-E, drives are either SAS or FC. Would that box run six or eight of the IBM fancy pants drives? I don't know, haven't ever seen it tried. For most sites, Windows and the crummy little hardware it runs on will be just fine. For you big fellas, not so much. If you are in the gotta push 3-6TB/day Windows will work. For you 10TB/day folks, maybe not. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kauffman, Tom Sent: Thursday, February 26, 2009 7:39 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform Yup. It boils down to Wanda's statement: I/O, I/O, it's all about I/O --- Wanda Prather If you can do the work with LTO-1 or -2 drives, or DLT-7000, or similar speed/capacity, then Windows will work. When you get into high-speed/high-capacity drives the Intel/AMD architecture comes unglued. A single LTO-4 drive will use ALL the I/O bandwith of a PCI or PCI-x bus, and a significant chunk of a PCIe(1) bus. The challenge becomes one of finding a suitable X86 server with multiple PCIe busses in the design. IBM has the GX series I/O modules with two PCIe busses each for the P6 architecture that allows for significant I/O bandwith expansion -- for a cost, of course. Tom Kauffman NIBCO, Inc -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Wednesday, February 25, 2009 4:14 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: Preferred TSM Platform I love it when somebody quotes me! Somebody is listening. I had this discussion with one of our customers yesterday. It really does/should boil down to the OS experience you have on hand. Does the AIX platform have more capacity/performance than the best Windows platform? I'm guessing it probably does. But at what cost? And then more importantly: do either of the platforms have enough for you? If both do, then pick the one that makes more sense based on your OS experience. And remember: one can always divide and conquer. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Allen S. Rout Sent: Wednesday, February 25, 2009 1:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform On Wed, 25 Feb 2009 15:57:37 +0100, Henrik Vahlstedt s...@statoilhydro.com said: Time to quote Kelly... So to me it's either AIX or Windows (yes, you can do a lot of TSM on Windows once you get past the bigotry!). Choose whichever one you have the most experience with. gollum Ac! It BURNS usss, naty windowsss /gollum - Allen S. Rout - Prefers AIX for this. *koff* :) CONFIDENTIALITY NOTICE: This email and any attachments are for the exclusive and confidential use of the intended recipient. If you are not the intended recipient, please do not read, distribute or take action in reliance upon this message. If you have received this in error, please notify us immediately by return email and promptly delete this message and its attachments from your computer system. We do not waive attorney-client or work product privilege by the transmission of this message.
Re: Preferred TSM Platform
Sure, but if you don't have any AIX expertise and have to buy/rent that, the cost goes up significantly. I'm no huge fan of Windows, but almost everybody has some of that expertise while AIX expertise is not available in most shops. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Orville Lantto Sent: Thursday, February 26, 2009 9:15 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform Those high end Windows boxes are priced similarly to equivalent (or better) AIX boxes. Check the benchmarks before deciding. Orville L. Lantto From: Kauffman, Tom Sent: Thu 2/26/2009 08:38 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform Yup. It boils down to Wanda's statement: I/O, I/O, it's all about I/O --- Wanda Prather If you can do the work with LTO-1 or -2 drives, or DLT-7000, or similar speed/capacity, then Windows will work. When you get into high-speed/high-capacity drives the Intel/AMD architecture comes unglued. A single LTO-4 drive will use ALL the I/O bandwith of a PCI or PCI-x bus, and a significant chunk of a PCIe(1) bus. The challenge becomes one of finding a suitable X86 server with multiple PCIe busses in the design. IBM has the GX series I/O modules with two PCIe busses each for the P6 architecture that allows for significant I/O bandwith expansion -- for a cost, of course. Tom Kauffman NIBCO, Inc -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Wednesday, February 25, 2009 4:14 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: Preferred TSM Platform I love it when somebody quotes me! Somebody is listening. I had this discussion with one of our customers yesterday. It really does/should boil down to the OS experience you have on hand. Does the AIX platform have more capacity/performance than the best Windows platform? I'm guessing it probably does. But at what cost? And then more importantly: do either of the platforms have enough for you? If both do, then pick the one that makes more sense based on your OS experience. And remember: one can always divide and conquer. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Allen S. Rout Sent: Wednesday, February 25, 2009 1:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform On Wed, 25 Feb 2009 15:57:37 +0100, Henrik Vahlstedt s...@statoilhydro.com said: Time to quote Kelly... So to me it's either AIX or Windows (yes, you can do a lot of TSM on Windows once you get past the bigotry!). Choose whichever one you have the most experience with. gollum Ac! It BURNS usss, naty windowsss /gollum - Allen S. Rout - Prefers AIX for this. *koff* :) CONFIDENTIALITY NOTICE: This email and any attachments are for the exclusive and confidential use of the intended recipient. If you are not the intended recipient, please do not read, distribute or take action in reliance upon this message. If you have received this in error, please notify us immediately by return email and promptly delete this message and its attachments from your computer system. We do not waive attorney-client or work product privilege by the transmission of this message. This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.
Re: Preferred TSM Platform
We have had a couple of customers over the years running TSM on Solaris. I must echo Mike's comments. As Solaris would optimistically finish third in IBMs race for resources there will necessarily be fewer resources both on the development and support side. If/when there are problems they will be solved more slowly than on the Windows or AIX. I guess I would enter the TSM on Solaris world with caution. That said, I have found that if you are a very good Solaris person, the issues are much easier to solve as you can often walk the IBM resource through the problem. But it will take more of your time if there is a problem. The most prevalent issues we have seen are integrating with libraries and drives as you would expect. Perhaps if you stay with IBM tape products these problems would be less? Who knows. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of De Gasperis, Mike Sent: Thursday, February 26, 2009 9:09 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform We're primarily a Solaris based TSM shop, our backup server platforms are T2000's and T5220's currently which seem to be very good at handling the I/O of the newer T1A B drives along with the speeds of LTO's and what not. Most of our servers are loaded up with dual port 4Gb Emulex cards usually eight total HBA ports per server. Network wise we use the onboard 4 Gb ports and usually a dual port Gb card and Etherchannel/trunking to give us a large pipe for backup traffic. Speed wise the machines are great for an enterprise solution, price wise I think they're fantastic as well. The only issues we seem to run in to is IBM Solaris pointing fingers at each other when there are complicated bugs encountered that can be resolved via simple queries to get to the root of the problem. We're primarily using SAN based storage SUN/EMC arrays along with EDL's, disk suite management is usually done with Veritas for us though mpxio is always an option. For any disk we use in TSM we typically use raw volumes and not formatted file systems. I think the preferred platform is still AIX as TSM just seems to perform better on it with less of these odd bugs we see from time to time. The new Sun servers though are a great buy performance wise and really do handle these newer tape drive speeds well. Most of these new servers we're using we can't even get the CPU usage to go above 40% yet with 400-500 clients on them. -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Sergio Fuentes Sent: Thursday, February 26, 2009 10:58 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform We're actually considering a new platform for future TSM servers simply because we're not an AIX shop anymore (TSM being the lone holdout). We're not a very good windows shop either, and our strength is really in Solaris and/or Linux, technically speaking. When I compare the hardware and LVM features for Solaris with those of Linux, I can see the benefits of Solaris. But this listserv group has me second-guessing myself since I have yet to hear from someone with a Solaris-based TSM infrastructure. (I would stick with AIX if I could, but you know... politics). Solaris 10 and the built-in features of ZFS alone have kind of swayed me towards Solaris. It's the only native LVM-based filesystem that I think can compete with what I'm used to, namely JFS2. As for hardware, Sun offers some pretty hefty I/O-centric boxes, with a hefty pricetag. But the pricey p650 that we're on now has lasted almost 7 years, is still very stable and not breaking much of a sweat. Still, the range of servers that Sun offers (which I don't see in the Dell world) is another advantage. Any thoughts from anyone running a TSM server on Solaris? We could use the insight since I believe we'll be rolling out a development environment on Solaris as a proof-of-concept. Anyone familiar with DB2 performance on Solaris? Thanks! SF Jim Zajkowski wrote: On Feb 25, 2009, at 10:09 AM, Strand, Neil B. wrote: consider Solaris Actually I'm considering replacing my Linux TSM server with Solaris - either SPARC or x86 - predominately because Solaris has a fast TCP/IP stack, ZFS, and fewer driver issues than on Linux. Has anyone also moved from Linux to Solaris? --Jim
Re: Preferred TSM Platform
IBM x3850, dual quad core processors, 16GB, 7 PCI-E slots with four 2.5 15K SAS 73GB drives (have to use external storage on this guy), around $14K. Can have up to four quad core procs, 256GB memory. This is one screaming dude when used with TSM. And it's IBM... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Orville Lantto Sent: Thursday, February 26, 2009 11:19 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform This level of performance is pretty near the bottom of the pSeries range, but a comparable would be a pSeries 520 which could be had for this a similar price. I just checked and a basic 520 is list priced below $12,000. Orville Lantto | Consultant -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Zoltan Forray/AC/VCU Sent: Thursday, February 26, 2009 11:17 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform I haven't checked AIX prices lately, but my last Dell 2900 with 8-1TB drives (for the LZ) and 2-500GB mirrored drives (OS and DB) with 8GB RAM cost under $11K. As I mentioned, the RH Linux license is around $50. I think it has 2-PCIe slots for the HBA's and 2-GIGe NICS. Orville Lantto orville.lan...@glasshouse.com Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 02/26/2009 11:17 AM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject Re: [ADSM-L] Preferred TSM Platform Those high end Windows boxes are priced similarly to equivalent (or better) AIX boxes. Check the benchmarks before deciding. Orville L. Lantto From: Kauffman, Tom Sent: Thu 2/26/2009 08:38 To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform Yup. It boils down to Wanda's statement: I/O, I/O, it's all about I/O --- Wanda Prather If you can do the work with LTO-1 or -2 drives, or DLT-7000, or similar speed/capacity, then Windows will work. When you get into high-speed/high-capacity drives the Intel/AMD architecture comes unglued. A single LTO-4 drive will use ALL the I/O bandwith of a PCI or PCI-x bus, and a significant chunk of a PCIe(1) bus. The challenge becomes one of finding a suitable X86 server with multiple PCIe busses in the design. IBM has the GX series I/O modules with two PCIe busses each for the P6 architecture that allows for significant I/O bandwith expansion -- for a cost, of course. Tom Kauffman NIBCO, Inc -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Wednesday, February 25, 2009 4:14 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: Preferred TSM Platform I love it when somebody quotes me! Somebody is listening. I had this discussion with one of our customers yesterday. It really does/should boil down to the OS experience you have on hand. Does the AIX platform have more capacity/performance than the best Windows platform? I'm guessing it probably does. But at what cost? And then more importantly: do either of the platforms have enough for you? If both do, then pick the one that makes more sense based on your OS experience. And remember: one can always divide and conquer. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Allen S. Rout Sent: Wednesday, February 25, 2009 1:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform On Wed, 25 Feb 2009 15:57:37 +0100, Henrik Vahlstedt s...@statoilhydro.com said: Time to quote Kelly... So to me it's either AIX or Windows (yes, you can do a lot of TSM on Windows once you get past the bigotry!). Choose whichever one you have the most experience with. gollum Ac! It BURNS usss, naty windowsss /gollum - Allen S. Rout - Prefers AIX for this. *koff* :) CONFIDENTIALITY NOTICE: This email and any attachments are for the exclusive and confidential use of the intended recipient. If you are not the intended recipient, please do not read, distribute or take action in reliance upon this message. If you have received this in error, please notify us immediately by return email and promptly delete this message and its attachments from your computer system. We do not waive attorney-client or work product privilege by the transmission of this message. This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate
Re: Changing Storage Pool Status
Nothing until reclamation happens on the volumes. Then they will be reclaimed appropriate to the collocation type selected. You can force the issue by doing move data operations on the volumes. Newly arriving data from clients will be placed onto tape according to the collocation type. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Lepre, James Sent: Thursday, February 26, 2009 1:06 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Changing Storage Pool Status Hello Everyone, My question is if I have a Pool that is collocated, and I change it to Collocation by group what happens to the data already in the pool. Thank you James Lepre Senior Server Specialist Solix Inc 100 South Jefferson Road Whippany NJ 07981 Phone 1-973-581-5362 Cell 1-973-223-1921 --- Confidentiality Notice: The information in this e-mail and any attachments thereto is intended for the named recipient(s) only. This e-mail, including any attachments, may contain information that is privileged and confidential and subject to legal restrictions and penalties regarding its unauthorized disclosure or other use. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or the taking of any action or inaction in reliance on the contents of this e-mail and any of its attachments is STRICTLY PROHIBITED. If you have received this e-mail in error, please immediately notify the sender via return e-mail; delete this e-mail and all attachments from your e-mail system and your computer system and network; and destroy any paper copies you may have in your possession. Thank you for your cooperation.
Re: Preferred TSM Platform
I love it when somebody quotes me! Somebody is listening. I had this discussion with one of our customers yesterday. It really does/should boil down to the OS experience you have on hand. Does the AIX platform have more capacity/performance than the best Windows platform? I'm guessing it probably does. But at what cost? And then more importantly: do either of the platforms have enough for you? If both do, then pick the one that makes more sense based on your OS experience. And remember: one can always divide and conquer. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Allen S. Rout Sent: Wednesday, February 25, 2009 1:52 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Preferred TSM Platform On Wed, 25 Feb 2009 15:57:37 +0100, Henrik Vahlstedt s...@statoilhydro.com said: Time to quote Kelly... So to me it's either AIX or Windows (yes, you can do a lot of TSM on Windows once you get past the bigotry!). Choose whichever one you have the most experience with. gollum Ac! It BURNS usss, naty windowsss /gollum - Allen S. Rout - Prefers AIX for this. *koff* :)
Re: backups direct to tape
Interesting. I wonder why that is? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Fred Johanson Sent: Tuesday, February 24, 2009 7:22 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape In our experience, when we turned on simultaneous writes for the Lanfree stgpool, the backups went to the network. Fred Johanson TSM Administrator University of Chicago 773-702-8464 -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Monday, February 23, 2009 5:44 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape I don't think the comment that simultaneous writes won't work for LAN Free. The simultaneous write thing happens at the storage pool level (copystg= parameter). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Monday, February 23, 2009 4:31 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape Also, are these RMAN backups? If so, they use the API. I'm not sure if simultaneous write works for the API. Anyone know? Also, RMAN supports multiple tape streams. We are setup to use 4 parallel streams. We used to use physical tape, but now the backups just go to serial disk volumes (devclass=file). The DBAs here are happy with that. ..Paul At 06:14 PM 2/23/2009, Fred Johanson wrote: The big gotcha here is that you cannot use the simultaneous write feature if the backup is going lanfree. From: ADSM: Dist Stor Manager [ads...@vm.marist.edu] On Behalf Of Kelly Lipp [l...@storserver.com] Sent: Monday, February 23, 2009 1:28 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape Gill, I will start by asking: What do you think happens? The cool thing about TSM is it will usually do what you think it will do. If a tape error occurs, the current transaction (whatever it happens to be) will fail. What that means will vary. In the case of a client/agent backup, a write error on a tape will cause TSM to mount another tape and restart the failed transaction. Data integrity is the hallmark of the product. Direct to tape backups are actually cool. In fact, you can write two tapes simultaneously: the onlinepool and the drpool volumes. This avoids having to migrate data from disk to tape and having to back that data up. In the case of a 500GB database that will reduce the amount of data flowing through your server from 1.5TB to 500GB. Huge savings. And the write to two tapes happens at the same speed as to one. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Monday, February 23, 2009 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backups direct to tape I have a question that I hope those using TSM to back up direct to tape can answer. We don't, and never have, done this here and I'm not looking to implement it either. I am really looking for information as to what happens to the backup is a tape error occurs. (I just know I'm going to get this question when the dba's who have convinced everyone netbackup and direct to tape is so great) I am interested too if the TSM agents have the ability to compensate for this compared to a standard client backup. Will the agent take this into consideration and continue the backup with a different tape or will it die? The same question applies to the standard backup. Thanks for your help. Geoff Gill TSM Administrator PeopleSoft Sr. Systems Administrator SAIC M/S-G1b (858)826-4062 (office) (858)412-9883 (blackberry) Email: geoffrey.l.g...@saic.com -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: backups direct to tape
Gill, I will start by asking: What do you think happens? The cool thing about TSM is it will usually do what you think it will do. If a tape error occurs, the current transaction (whatever it happens to be) will fail. What that means will vary. In the case of a client/agent backup, a write error on a tape will cause TSM to mount another tape and restart the failed transaction. Data integrity is the hallmark of the product. Direct to tape backups are actually cool. In fact, you can write two tapes simultaneously: the onlinepool and the drpool volumes. This avoids having to migrate data from disk to tape and having to back that data up. In the case of a 500GB database that will reduce the amount of data flowing through your server from 1.5TB to 500GB. Huge savings. And the write to two tapes happens at the same speed as to one. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Monday, February 23, 2009 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backups direct to tape I have a question that I hope those using TSM to back up direct to tape can answer. We don't, and never have, done this here and I'm not looking to implement it either. I am really looking for information as to what happens to the backup is a tape error occurs. (I just know I'm going to get this question when the dba's who have convinced everyone netbackup and direct to tape is so great) I am interested too if the TSM agents have the ability to compensate for this compared to a standard client backup. Will the agent take this into consideration and continue the backup with a different tape or will it die? The same question applies to the standard backup. Thanks for your help. Geoff Gill TSM Administrator PeopleSoft Sr. Systems Administrator SAIC M/S-G1b (858)826-4062 (office) (858)412-9883 (blackberry) Email: geoffrey.l.g...@saic.com
Re: backups direct to tape
And moving it from the old to TSM is virtually impossible. It would require a restore of the data and then an archive of it into TSM. As you state, figuring what data that is is the issue and not easily (if at all) resolvable. Write off the 3000 tapes, keep the last of the catalogs around and cross your fingers to nobody will ask for data from them. I think undergoing a project to move that old data is a complete waste of time. But once on TSM, you'll never have that stupid problem again! Simply continue to move your archive data to the latest tape technology as your implementation moves. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Monday, February 23, 2009 1:25 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape Thanks for the response Kelly. I actually thought it worked the way you explained but I did not want to say something without confirmation since I wasn't exactly sure. I didn't, however, know that you could write both pools at the same time. So this is just another nail I hope to put in the netbackup coffin. What I have seen with netbackup here is if a tape error occurs the backup stops, period. What I have to do now is put together a short slide show on how we will deal with moving 125 nodes to TSM and how to deal with the data in netbackup we need moved in to TSM. At this point unfortunately I doubt anyone can tell me anything about the data in netbackup. Oh sure, I can see the expiration date for tapes but people around here want to keep 'everything forever' so finding the real data we need to keep from those 3000 tapes I think is going to be challenging. The short term, 2 week and 6 month tapes, in my opinion could expire on their own since it's going to be some time before everything is moved. Having both clients on the same machine is not an issue in my opinion. It's the long term data that we need to find, the REAL long term data, and move it in to TSM. Finding it is the hard part, moving it is the easy part. Geoff Gill TSM Administrator PeopleSoft Sr. Systems Administrator SAIC M/S-G1b (858)826-4062 (office) (858)412-9883 (blackberry) Email: geoffrey.l.g...@saic.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Monday, February 23, 2009 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: backups direct to tape Gill, I will start by asking: What do you think happens? The cool thing about TSM is it will usually do what you think it will do. If a tape error occurs, the current transaction (whatever it happens to be) will fail. What that means will vary. In the case of a client/agent backup, a write error on a tape will cause TSM to mount another tape and restart the failed transaction. Data integrity is the hallmark of the product. Direct to tape backups are actually cool. In fact, you can write two tapes simultaneously: the onlinepool and the drpool volumes. This avoids having to migrate data from disk to tape and having to back that data up. In the case of a 500GB database that will reduce the amount of data flowing through your server from 1.5TB to 500GB. Huge savings. And the write to two tapes happens at the same speed as to one. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Monday, February 23, 2009 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backups direct to tape I have a question that I hope those using TSM to back up direct to tape can answer. We don't, and never have, done this here and I'm not looking to implement it either. I am really looking for information as to what happens to the backup is a tape error occurs. (I just know I'm going to get this question when the dba's who have convinced everyone netbackup and direct to tape is so great) I am interested too if the TSM agents have the ability to compensate for this compared to a standard client backup. Will the agent take this into consideration and continue the backup with a different tape or will it die? The same question applies to the standard backup. Thanks for your help. Geoff Gill TSM Administrator PeopleSoft Sr. Systems Administrator SAIC M/S-G1b (858)826-4062 (office) (858)412-9883 (blackberry) Email: geoffrey.l.g...@saic.com
Re: backups direct to tape
I don't think the comment that simultaneous writes won't work for LAN Free. The simultaneous write thing happens at the storage pool level (copystg= parameter). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Paul Zarnowski Sent: Monday, February 23, 2009 4:31 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape Also, are these RMAN backups? If so, they use the API. I'm not sure if simultaneous write works for the API. Anyone know? Also, RMAN supports multiple tape streams. We are setup to use 4 parallel streams. We used to use physical tape, but now the backups just go to serial disk volumes (devclass=file). The DBAs here are happy with that. ..Paul At 06:14 PM 2/23/2009, Fred Johanson wrote: The big gotcha here is that you cannot use the simultaneous write feature if the backup is going lanfree. From: ADSM: Dist Stor Manager [ads...@vm.marist.edu] On Behalf Of Kelly Lipp [l...@storserver.com] Sent: Monday, February 23, 2009 1:28 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] backups direct to tape Gill, I will start by asking: What do you think happens? The cool thing about TSM is it will usually do what you think it will do. If a tape error occurs, the current transaction (whatever it happens to be) will fail. What that means will vary. In the case of a client/agent backup, a write error on a tape will cause TSM to mount another tape and restart the failed transaction. Data integrity is the hallmark of the product. Direct to tape backups are actually cool. In fact, you can write two tapes simultaneously: the onlinepool and the drpool volumes. This avoids having to migrate data from disk to tape and having to back that data up. In the case of a 500GB database that will reduce the amount of data flowing through your server from 1.5TB to 500GB. Huge savings. And the write to two tapes happens at the same speed as to one. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Gill, Geoffrey L. Sent: Monday, February 23, 2009 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] backups direct to tape I have a question that I hope those using TSM to back up direct to tape can answer. We don't, and never have, done this here and I'm not looking to implement it either. I am really looking for information as to what happens to the backup is a tape error occurs. (I just know I'm going to get this question when the dba's who have convinced everyone netbackup and direct to tape is so great) I am interested too if the TSM agents have the ability to compensate for this compared to a standard client backup. Will the agent take this into consideration and continue the backup with a different tape or will it die? The same question applies to the standard backup. Thanks for your help. Geoff Gill TSM Administrator PeopleSoft Sr. Systems Administrator SAIC M/S-G1b (858)826-4062 (office) (858)412-9883 (blackberry) Email: geoffrey.l.g...@saic.com -- Paul ZarnowskiPh: 607-255-4757 Manager, Storage Services Fx: 607-255-8521 719 Rhodes Hall, Ithaca, NY 14853-3801Em: p...@cornell.edu
Re: Defining FILE Device class
Get rid of the double quote before the /filedev1 and the space after the , and you should be fine. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Abdullah, Md-Zaini B BSP-IMI/231 Sent: Wednesday, February 11, 2009 8:35 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Defining FILE Device class Hi TSM expoertise, I have an issue when creating a FILE device with miltiple directories locating in different filesystems. Please advise. from AIX --- r...@bspsap11 # df -k|grep /file /dev/fstlv_filedev1 103546880 1031263201%4 1% /filedev1 /dev/fstlv_filedev2 103546880 1031263201%4 1% /filedev2 From TSM server - TSMdefine devclass FILE_DEV_HR devtype=file directory=/filedev1, /filedev2 SHAREd=yes MOUNTLimit=2 MAXCAPacity=40G ANR8366E DEFINE DEVCLASS: Invalid value for DIRECTORY parameter. ANS8001I Return code 3. Md.Zaini.Abdullah Seria Head Office Jalan Utara, Panaga, Seria KB3534, Brunei Darussalam Tel: +673-3-37 3538 Email: md-zaini.abdul...@shell.com Internet: http://www.shell.com
Re: Data Retention Settings: Unintended Consequences
Boy, it sure looks like you'll have three versions forever on this one. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nick Laflamme Sent: Wednesday, January 28, 2009 12:30 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Data Rention Settings: Unintended Consequences A client's TSM server has the following copygroup settings: tsm: SERVER1q copygroup * active f=d Policy Domain Name: OPEN_SYSTEM_ENVIRONMENT Policy Set Name: ACTIVE Mgmt Class Name: BACKUP_SHARK Copy Group Name: STANDARD Copy Group Type: Backup Versions Data Exists: 3 Versions Data Deleted: 3 Retain Extra Versions: No Limit Retain Only Version: 91 snip Because Versions Data Deleted is more than 1, and because Retain Extra Versions is set to No Limit, am I correct in deducing that TSM will keep three copies of deleted files forever, because Retain Only Version will never become relevant? Or does TSM use Retain Only Version for all versions (copies) once there isn't an active file? (That's not what the doc says, of course.) Thanks, Nick
Linux GFS File system backup using SAN Agent
Folks, Anybody tried this? Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com
Re: I'm missing something somewhere -- part 2
Is cache=yes on that pool? Are you talking about percent utilized or percent migratable? I'm reasonably sure you know the difference but thought I'd confirm... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kauffman, Tom Sent: Monday, January 12, 2009 1:50 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] I'm missing something somewhere -- part 2 I'm looking at 'first causes' for my off-site copy imbalance in my primary archive pool - and I've run into something 'interesting'. Some background - The archive pool (called ARCHIVEPOOL) has 5 disk volumes, all at 8 GB. Max file size is 5 GB. Migration threshold is 60%. Maxproc is 2. So when we reach 60% or above in the pool, two processes kick off. And we migrate all the way down to 0%. But what I see in the log looks like this: ANR0984I Process 4758 for MIGRATION started in the BACKGROUND at 02:26:20. ANR1000I Migration process 4758 started for storage pool ARCHIVEPOOL automatically, highMig=60, lowMig=0, duration=No. ANR0513I Process 4758 opened output volume 444019L4. ANR1001I Migration process 4758 ended for storage pool ARCHIVEPOOL. ANR0986I Process 4758 for MIGRATION running in the BACKGROUND processed 256 items for a total of 7,094,272 bytes with a completion state of SUCCESS at 02:26:22. ANR0984I Process 4759 for MIGRATION started in the BACKGROUND at 02:26:23. ANR1000I Migration process 4759 started for storage pool ARCHIVEPOOL automatically, highMig=60, lowMig=0, duration=No. ANR0513I Process 4759 opened output volume 444019L4. ANR1001I Migration process 4759 ended for storage pool ARCHIVEPOOL. ANR0986I Process 4759 for MIGRATION running in the BACKGROUND processed 256 items for a total of 8,732,672 bytes with a completion state of SUCCESS at 02:26:25. ANR0984I Process 4760 for MIGRATION started in the BACKGROUND at 02:26:27. ANR1000I Migration process 4760 started for storage pool ARCHIVEPOOL automatically, highMig=60, lowMig=0, duration=No. ANR0513I Process 4760 opened output volume 444019L4. ANR1001I Migration process 4760 ended for storage pool ARCHIVEPOOL. For several hours. For some reason the pool is over 60% full; migration kicks in and 15.7 MB migrates; the migration ends; and the pool is STILL over 60% full. Why?? More importantly, how do I fix this? Thanks - Tom CONFIDENTIALITY NOTICE: This email and any attachments are for the exclusive and confidential use of the intended recipient. If you are not the intended recipient, please do not read, distribute or take action in reliance upon this message. If you have received this in error, please notify us immediately by return email and promptly delete this message and its attachments from your computer system. We do not waive attorney-client or work product privilege by the transmission of this message.
Re: Best way to use TSM to move 2Tb of data
If you assume a file create rate of about 100,000/hour then you are looking at a 20 hour restore if all else goes well. You might squeeze more file creates out of your new server, but who really knows? If you assume a 200 GB/hour transfer rate and use image instead, you can cut the restore time in half. You can't improve the file create rate by using multiple streams. In fact, that actually reduces the rate. I'm still advocating the image route. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nicholas Rodolfich Sent: Thursday, December 11, 2008 3:08 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data It is ~2,000,000 individual files after hours.
Re: Best way to use TSM to move 2Tb of data
Amen to Dwight's comment! And can you imagine a filespace with 10M files? I shudder... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Dwight Cook Sent: Friday, December 12, 2008 9:54 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data I've seen restores of 1+M files take days due to the delays associated with general system over head (creating directory entries, etc...) and by days I mean 5-7+. And so again, I'll mention... Just because you CAN put a million or more files on a single drive doesn't mean it's a good idea! Dwight -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Friday, December 12, 2008 10:47 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data If you assume a file create rate of about 100,000/hour then you are looking at a 20 hour restore if all else goes well. You might squeeze more file creates out of your new server, but who really knows? If you assume a 200 GB/hour transfer rate and use image instead, you can cut the restore time in half. You can't improve the file create rate by using multiple streams. In fact, that actually reduces the rate. I'm still advocating the image route. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nicholas Rodolfich Sent: Thursday, December 11, 2008 3:08 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data It is ~2,000,000 individual files after hours.
Re: Best way to use TSM to move 2Tb of data
AH. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Fred Johanson Sent: Friday, December 12, 2008 10:34 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data You mean like this? CRONUSXBkup /mnt/ide0 2 OFFSITEPO- 10,222,45 3,529,279 3,526,778 OL1 .74 .09 Or its departmental companion? ATHENSXBkup /mnt/ide0 1 OFFSITEPO- 5,011,898 3,380,065 3,355,847 OL .04 .95 Or another department that has 10 boxes like this? OIABkup \\oia\d$ 4 TAPEPOOL 3,947,878 19,719,65 19,714,64 8.00 4.36 Or the user who did a network mount of a Time Server to her desktop? By the time I got back from vacation she had created a filespace with 36M files. Fred Johanson TSM Administrator University of Chicago 773-702-8464 -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Friday, December 12, 2008 11:17 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data Amen to Dwight's comment! And can you imagine a filespace with 10M files? I shudder... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Dwight Cook Sent: Friday, December 12, 2008 9:54 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data I've seen restores of 1+M files take days due to the delays associated with general system over head (creating directory entries, etc...) and by days I mean 5-7+. And so again, I'll mention... Just because you CAN put a million or more files on a single drive doesn't mean it's a good idea! Dwight -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Friday, December 12, 2008 10:47 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data If you assume a file create rate of about 100,000/hour then you are looking at a 20 hour restore if all else goes well. You might squeeze more file creates out of your new server, but who really knows? If you assume a 200 GB/hour transfer rate and use image instead, you can cut the restore time in half. You can't improve the file create rate by using multiple streams. In fact, that actually reduces the rate. I'm still advocating the image route. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nicholas Rodolfich Sent: Thursday, December 11, 2008 3:08 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data It is ~2,000,000 individual files after hours.
Re: Best way to use TSM to move 2Tb of data
How about an image backup? Eliminate the small file issues on backup and restore... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Dwight Cook Sent: Thursday, December 11, 2008 1:45 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data Is it lots of little files (I know, silly question with it being a windows file server). Also, how long is over night? Is that compressed client data or is it file space data? Is that a backup or archive? What is your network? 100 Mb/sec fast Ethernet? Gig Ethernet? Teamed NIC's? I'd run archives...if you have multiple high level directories that will let you get away with it, I'd run one archive command against each high level directory to try to have between 4 10 client sessions established with the TSM server. Run client compression if you have enough processing power on your windows box. (what ever your network/NIC max through put is... it will typically move 3 times more data in the same period of time if you have enough power to compress down the data without slowing down your entire process) Anyway, as you can see... a few variables involved that you didn't mention... Dwight -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nicholas Rodolfich Sent: Thursday, December 11, 2008 1:36 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Best way to use TSM to move 2Tb of data Hello All, I have a client that needs to move a 2Tb volume on their Windows file server. Their TSM sever is also on Windows w/TSM v5.3.4. What is the fastest/best way to accomplish this. We tested an archive but it only got 340Gb overnight. They don't have enough disk based pool space to land it there so we will have to use tape. I suggest kicking up the resourceutilization parameter on the client to enable multiple sessions but I am hoping you guys/gals can teach me something. Thanks for all your help!! Nicholas=
Re: Best way to use TSM to move 2Tb of data
I think the advantage is on the restore: you won't have to create a gazillion little files which is actually the bottleneck (typically) in Windows. The backup will be limited to one stream, but that will be faster too, on the order of what a GiGE network can optimally do: 200-300GB/hour. I think end-to-end, using image will be quicker. Perhaps a little slower on the backup, but remarkably quicker on the restore. Now, if the number of files is less than a million, say, that might not be so. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Dwight Cook Sent: Thursday, December 11, 2008 1:56 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data That will limit you to a single session performing an image backup (won't it??? I don't use image backup) Windows clients have become better at pumping data but lately but a single session still won't come near maxing out a NIC. We have some multiple TB SAP data bases on windows servers (I know just kill me now, please) and we can, with teamed NIC's that are 100 Mb/sec fast Ethernet (so team can push 200 Mb/sec) anyway, we can shove 55 GB/hr of compressed client data with about 4 or 5 sessions so that is around 150 GB of file space per hour. (we can only get windows clients running at 35 GB/hr best with a single 100 Mb/sec fast Ethernet nic) -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Kelly Lipp Sent: Thursday, December 11, 2008 2:47 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data How about an image backup? Eliminate the small file issues on backup and restore... Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Dwight Cook Sent: Thursday, December 11, 2008 1:45 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Best way to use TSM to move 2Tb of data Is it lots of little files (I know, silly question with it being a windows file server). Also, how long is over night? Is that compressed client data or is it file space data? Is that a backup or archive? What is your network? 100 Mb/sec fast Ethernet? Gig Ethernet? Teamed NIC's? I'd run archives...if you have multiple high level directories that will let you get away with it, I'd run one archive command against each high level directory to try to have between 4 10 client sessions established with the TSM server. Run client compression if you have enough processing power on your windows box. (what ever your network/NIC max through put is... it will typically move 3 times more data in the same period of time if you have enough power to compress down the data without slowing down your entire process) Anyway, as you can see... a few variables involved that you didn't mention... Dwight -Original Message- From: ADSM: Dist Stor Manager [mailto:ads...@vm.marist.edu] On Behalf Of Nicholas Rodolfich Sent: Thursday, December 11, 2008 1:36 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Best way to use TSM to move 2Tb of data Hello All, I have a client that needs to move a 2Tb volume on their Windows file server. Their TSM sever is also on Windows w/TSM v5.3.4. What is the fastest/best way to accomplish this. We tested an archive but it only got 340Gb overnight. They don't have enough disk based pool space to land it there so we will have to use tape. I suggest kicking up the resourceutilization parameter on the client to enable multiple sessions but I am hoping you guys/gals can teach me something. Thanks for all your help!! Nicholas=
Re: Server Platform Upgrade
If I were going to be in the x86 family, I would move into the larger platforms with more processors and more PCI-E slots. So the HP DL580 I think would be your best bet. That will reduce the 5x cost benefit somewhat but provide you with more flexibility. Consider the IBM x3850 M2 server. These are great boxes and IBM will continue to love and support you. And perhaps a bit cheaper than HP. Similar to my comment about Windows vs. Unix below, if you have a ton of HP stuff in your site now, go with HP. Linux or Windows then. Hmmm. I'm not a Linux guy so I would choose windows. If most of your clients are windows and most of your internal IT knowledge is Windows, stay Windows. If you have the Linux expertise, then use that. Won't make much difference to TSM (I know, I know, Windows for I/O sucks and all that, which BTW is contrary to what I know and believe...). You can probably buy two x86 and split the load and still save money over the mongo AIX box. I'd rather do that. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 x7105 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Sam Sheppard Sent: Thursday, December 04, 2008 11:51 AM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] Server Platform Upgrade We are currently running 3 TSM servers at Version 5.5, two on z/OS and the third on a Solaris 10 box. We have been tasked to combine these into one. The current Solaris system is on a Sun V240 w/8GB memory: Around 400 clients (combined) including a 3TB Exchange system, and a fairly large SAP implemtation in development on MS SQL Server, several Oracle boxes, and a large number of Windows servers. Total database size (Solaris) is around 80GB, expiration runs about two hours. z/OS databases total 50GB. 8 TS1120 tape drives in a 3494 ATL. 1.2TB array for storage pools and database on the Solaris server which we are currently trying to separate. Total daily backup volume is around 1TB with additional weekly backups of 9TB. I am the TSM guy and the z/OS systems programmer and as such don't really have a feel for hardware sizing or configuration on the Unix side of things and so have to rely on our Unix guys. I suggested that AIX would be the preferred platform for this implementation with another Solaris box having the advantage of not requiring converting the exist- ing one. They came up with the following options with their favorite being the x86(HP) with Linux because it is much cheaper and they claim would be more powerful. The AIX and Sun configurations are similar in price at around 5 times the x86. Thought, comments, considerations? What are people using for disk storage pools and would the internal drives on these boxes be adequate? I'm also in the process of freeing one array on our ESS800 (Shark) for fiber connection to this configuration. Here are the proposed options: IBM Power 550 ExpressHP DL380 G5 Up to 8 cores and 256GB RAM Up to 8 cores and 64GB RAM Six 300GB internal SAS 15k drivesEight 146GB internal SAS 15k drives Three PCIe and two PCI-X slots Four PCIe slots Dual port 10 GB Ethernet card 4 GB fiber channel cards - x2 Thanks Sam Sheppard San Diego Data Processing Corp. (858)-581-9668
Re: Waiting for access to input volume
I don't think that's true. The purpose of the new code was to eliminate the problem with sequential volumes. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Petrullo, Michael G. Sent: Tuesday, November 25, 2008 11:35 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Waiting for access to input volume Kelly, The multi-access read only applies to random access volumes, ie. Disk. Shawn, This is a flaw of TSM. If a process is using/holding a volume another process will not be able to use it until it completes. I have ran into this issue when running an export that was using a volume and then a migration tried to access the same volume. The migration wasn't able to get access to the volume until the export finished. I've even gone to the extent of calling IBM about this issue and they informed me that it is not even in the works for future versions of TSM. Hope this clears things up a little. Regards, Mike -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Kelly Lipp Sent: Wednesday, November 19, 2008 5:46 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Waiting for access to input volume I haven't been following the entire thread, but multi-access read is now available on certain TSM volumes. It's a V5.5 feature. May be time to upgrade... May be time to read the release notes at any rate to see if your problem is addressed by this new feature. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Shawn Drew Sent: Wednesday, November 19, 2008 3:33 PM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] Waiting for access to input volume It's intermittent slow backups that can sometimes occupy the virtual tape like this. I would like to make sure that a single slow backup doesn't hold up the whole Life Cycle. I would prefer to have the single Backup Storage pool process fail and move to the next step in the housekeeping script than having to cancel the backup manually when I happen to notice this happening. If I can't find a time-out, then I will setup a monitor script to cancel if the wait time gets too large) Any ideas? Regards, Shawn Shawn Drew Internet [EMAIL PROTECTED] Sent by: ADSM-L@VM.MARIST.EDU 11/19/2008 04:01 PM Please respond to ADSM-L@VM.MARIST.EDU To ADSM-L cc Subject Re: [ADSM-L] Waiting for access to input volume Perhaps a stuck restore has the volume? Check Q RESTORE Shawn Drew [EMAIL PROTECTED] Sent by: ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU 11/19/2008 03:45 PM Please respond to ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU To ADSM-L@VM.MARIST.EDU cc Subject [ADSM-L] Waiting for access to input volume Does anyone know how to set the time-out for this? The device class Mount Wait doesn't seem to apply 14 Backup Storage Pool Primary Pool VTL_C1, Copy Pool VTL_C2, Files Backed Up: 7, Bytes Backed Up: 17,871,357,954, Unreadable Files: 0, Unreadable Bytes: 0. Current Physical File (bytes): 10,866,872 Waiting for access to input volume K00459L3 (27271 seconds). Current output volume: W00472L3. Regards, Shawn Shawn Drew This message and any attachments (the message) is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. BNP PARIBAS (and its subsidiaries) shall (will) not therefore be liable for the message if modified. Please note that certain functions and services for BNP Paribas may be performed by BNP Paribas RCC, Inc. IMPORTANT: E-mail sent through the Internet is not secure and timely delivery of Internet mail is not guaranteed. Legg Mason therefore, recommends that you do not send any action-oriented or time-sensitive information to us via electronic mail, or any confidential or sensitive information including: social security numbers, account numbers, or personal identification numbers. This message is intended for the addressee only and may contain privileged or confidential information. Unless you are the intended recipient, you may not use, copy or disclose to anyone any information contained in this message. If you have received this message in error, please notify the author by replying to this message and then kindly delete the message. Thank you.
Re: IBM 3200 Tape Library SAS adapters
SAS is very different from SCSI so all bets are off when comparing the two. An x4 4 lane SAS adapter will support 4 SAS drives. You will need an interposer plugged into the SAS card which splits the four lanes onto four cables. Then plug each drives cable into the interposer and you are completely fanned out. One lane per drive (each lane is 3Gb/sec transfer rate, well fast enough for one LTO4 tape drive). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Mark Stapleton Sent: Thursday, November 20, 2008 10:53 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM 3200 Tape Library SAS adapters No docs, but you want as many ports as possible for maximum throughput. Officially, you can put three drives per HBA port (through a switch, not Y cables). I tell customers that a maximum of two drives per HBA port will make them a LOT happier; three LTO4 drives could saturate the connection, and that's a bad thing with tape. -- Mark Stapleton ([EMAIL PROTECTED]) CDW Berbee System engineer 7145 Boone Avenue North, Suite 140 Brooklyn Park MN 55428-1511 763-592-5963 www.berbee.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Larry Peifer Sent: Thursday, November 20, 2008 12:49 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] IBM 3200 Tape Library SAS adapters We are in the process of purchasing an IBM 3200 Tape Library with 4 - LTO4 half-height SAS drives. Our IBM sales rep and tech support rep are telling us that we need 4 - 5912 HBAs, one for each drive, to support this configuration in our IBM 9117-MMA (p570) host. From our review of the tech specs it looks like the 5912 Adapter (PCI-X DDR Dual -x4 SAS Adapter) should be able to support all four SAS tape drives from one card using 2- YO cables in the same way it can support 4 disk drives. Can anyone point us to someone or some docs that can give us more information. Larry Peifer San Onofre Nuclear Generating Station AIX System Admin TSM system Admin
Re: IBM 3200 Tape Library SAS adapters
I'm going with yes, you can and it should perform just fine. I, however, don't have any firsthand experience with that config. Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Larry Peifer Sent: Thursday, November 20, 2008 11:28 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM 3200 Tape Library SAS adapters So if I understand Kelly correctly, I can have 1 - 5912 going to 4 lto4 hh drives running at 3Gb/sec per drive and not oversaturate the bus or adapter when all four drives are active. Do you know anyone actually running with this configuration with a p570? Kelly Lipp [EMAIL PROTECTED] COM To Sent by: ADSM: ADSM-L@VM.MARIST.EDU Dist Stor cc Manager [EMAIL PROTECTED] Subject .EDU Re: [ADSM-L] IBM 3200 Tape Library SAS adapters 11/20/2008 09:57 AM Please respond to ADSM: Dist Stor Manager [EMAIL PROTECTED] .EDU SAS is very different from SCSI so all bets are off when comparing the two. An x4 4 lane SAS adapter will support 4 SAS drives. You will need an interposer plugged into the SAS card which splits the four lanes onto four cables. Then plug each drives cable into the interposer and you are completely fanned out. One lane per drive (each lane is 3Gb/sec transfer rate, well fast enough for one LTO4 tape drive). Kelly Lipp CTO STORServer, Inc. 485-B Elkton Drive Colorado Springs, CO 80907 719-266-8777 www.storserver.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Mark Stapleton Sent: Thursday, November 20, 2008 10:53 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: [ADSM-L] IBM 3200 Tape Library SAS adapters No docs, but you want as many ports as possible for maximum throughput. Officially, you can put three drives per HBA port (through a switch, not Y cables). I tell customers that a maximum of two drives per HBA port will make them a LOT happier; three LTO4 drives could saturate the connection, and that's a bad thing with tape. -- Mark Stapleton ([EMAIL PROTECTED]) CDW Berbee System engineer 7145 Boone Avenue North, Suite 140 Brooklyn Park MN 55428-1511 763-592-5963 www.berbee.com -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of Larry Peifer Sent: Thursday, November 20, 2008 12:49 PM To: ADSM-L@VM.MARIST.EDU Subject: [ADSM-L] IBM 3200 Tape Library SAS adapters We are in the process of purchasing an IBM 3200 Tape Library with 4 - LTO4 half-height SAS drives. Our IBM sales rep and tech support rep are telling us that we need 4 - 5912 HBAs, one for each drive, to support this configuration in our IBM 9117-MMA (p570) host. From our review of the tech specs it looks like the 5912 Adapter (PCI-X DDR Dual -x4 SAS Adapter) should be able to support all four SAS tape drives from one card using 2- YO cables in the same way it can support 4 disk drives. Can anyone point us to someone or some docs that can give us more information. Larry Peifer San Onofre Nuclear Generating Station AIX System Admin TSM system Admin
