v6.3.5 hung db2??

2015-02-13 Thread Rhodes, Richard L.
Two days ago we upgrade one of our TSM instances to v6.3.5 (from v6.3.4). This is our first v6.3.5 instance. It runs on a AIX server. Last night at 19:32 it looks like DB2 went into some kind of a loop. The instance became unresponsive. Dsmadmc cmds hung (didn't error, just hung). Dsmserv

FW: v6.3.5 hung db2??

2015-02-13 Thread Rhodes, Richard L.
Now this is really weird. TSM came up after we rebooted. But it threw a bunch of ANR msgs, then QUIT LOGGING. It seems to be running - I go onto a server and did a incr bkup, but nothing is logging in the actlog. 02/13/15 10:00:22 ANRD_2891663292 GetDomainByNodeId(pmcache.c:2645)

DEVCLASS=FILE - what am I missing

2015-02-13 Thread Zoltan Forray
Up until recently, I have always used DEVCLASS=DISK for disk storage and always preformatted/allocated the disk volumes into multiple chunks to all for multi-I/O benefits. When I recently stood-up a new server, I decided to try DEVCLASS=FILE for disk-based storage/incoming backups. I thought I

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Andrew Raibeck
Hi Rick, Off-hand I am not sure what the problem is, I think it would be a good idea to open a PMR if you have not already done so. Best regards, - Andy Andrew Raibeck | Tivoli Storage Manager Level 3 Technical Lead

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Rhodes, Richard L.
Yea. I opened a Sev 1. Thanks! Rick -Original Message- From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On Behalf Of Andrew Raibeck Sent: Friday, February 13, 2015 10:57 AM To: ADSM-L@VM.MARIST.EDU Subject: Re: FW: v6.3.5 hung db2?? Hi Rick, Off-hand I am not sure

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Rainer Tammer
Hello, please keep us posted. I will have to go from 6.3.4-300 to a higher version because of the NDMP dump 2TB overwrite problem... Bye Rainer On 13.02.2015 17:05, Rhodes, Richard L. wrote: Yea. I opened a Sev 1. Thanks! Rick -Original Message- From: ADSM: Dist Stor

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Paul Zarnowski
At 12:12 PM 2/13/2015, Zoltan Forray wrote: Well, last night became a disaster. Backups failing all over because it couldn't allocate any more files and also would not automatically shift to use the nextpool which is defined as a tape pool. Alas, TSM doesn't automatically roll over when the

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Nick Laflamme
FILE allows deduplication; DISK doesn't. My impression after some experimenting is that FILE wasn't meant to replace DISK; it was solely meant to replace tape device classes. We didn't need to, so those experiments quietly ended. On Fri, Feb 13, 2015 at 12:30 PM, Zoltan Forray zfor...@vcu.edu

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Prather, Wanda
Probably just bad luck…. When I set up FILE pools for customers, I usually have to tweak them a couple of times to get the sizing right, depends on the load, the number of concurrent sessions, etc. Been there, done that, got the scars. Assumptions you should change: • Unlike a disk

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Gee, Norman
I front end several of my file device class pool with a disk device pool which will migrate to the file device pool and control the number of filling volumes. Volume that are in filling status will not reclaim, but you can manually move the data. -Original Message- From: ADSM: Dist

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Steven Langdale
As the number of vols is damn close to Max scratch. I'd say you ran out. No thing in actlog? Bung it up and see what happens. On Fri, 13 Feb 2015 17:15 Zoltan Forray zfor...@vcu.edu wrote: Up until recently, I have always used DEVCLASS=DISK for disk storage and always preformatted/allocated

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Zoltan Forray
Sure there were plenty of errors: 2/12/2015 10:00:23 PM ANR0522W Transaction failed for session 2639 for node RO-CVS (Linux x86-64) - no space available in storage pool BACKUPPOOL and all successor pools. 2/12/2015 10:04:20 PM ANR0522W Transaction failed for session 2648 for node RDO3 (WinNT) -

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Zoltan Forray
WOW - I didn't realize that. Thanks for pointing that out. Won't automatically go to nextstgpool, didn't automatically reclaim? So, what is the advantage/benefit of DEVCLASS=FILE? Sounds like time to go back to DEVCLASS=DISK On Fri, Feb 13, 2015 at 1:22 PM, Paul Zarnowski p...@cornell.edu

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Diego Fidalgo
Hi all, I'll like to remove my UserID in this group, can I do that? Regards, Diego F. On Feb 13, 2015 1:54 PM, Rainer Tammer t...@spg.schulergroup.com wrote: Hello, please keep us posted. I will have to go from 6.3.4-300 to a higher version because of the NDMP dump 2TB overwrite

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Rhodes, Richard L.
Working with some good support folks! Looks like we hit this: http://www-01.ibm.com/support/docview.wss?crawler=1uid=swg1IT06126 The v6.3.5 and v7.1.0 caused a bug in the rc.dsmserv startup script. The result is that db2 was running on limited memory - 32MB in our case. This was the

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Zoltan Forray
Thanks for all the replies. Pretty much confirms that FILE isn't for me. We don't do dedupe and there are a lot of manual/monitoring processes involved (I have enough to do with 8-TSM servers I manage - don't need more). Now to migrate 7TB of disk to tape so I can switch back to using DISK

Re: FW: v6.3.5 hung db2??

2015-02-13 Thread Mitchell, Ruth Slovik
Rick, Thank you for letting us know about this. It would be interesting to know if related messages were captured in the db2diag.log when this started to manifest itself. Best, Ruth U of I, Urbana, IL -Original Message- From: ADSM: Dist Stor Manager [mailto:ADSM-L@VM.MARIST.EDU] On

Re: DEVCLASS=FILE - what am I missing

2015-02-13 Thread Zoltan Forray
Thanks for the detailed explanations / experiences / suggestions. I greatly appreciate and will store away in case I ever try this again. Yes we do have lots of clients backing up at once - we easily hit 40-simultaneous sessions thus the reason for the high number. NUMOPENVOLSALLOWED is set to