ANS5092S - Server out of Data Storage Space; but storage pool only 1.9% utilised!

[falcon73]

I've 10 Windows servers to be backup using baclient (v5.3), and tdp for mail & database (v5.2.1). TSM server (v5.3) is sitting on AIX v5.2. Initially after implemented and configured it work accordingly for a month.



Suddenly an all my Windows file system backup failed with the above mentioned error message. When I checked my storage pool on the AIX, it shown only 1.9% utilised.



What could be the cause of this error?



Please advise ASAP.



Thank you.

 

[sgabriel62]

How much space is available for your DB recovery logs. Do a q log and send us your results. You may have to expand this area.



Let us know





Steven

 

[Harry_Redl]

Hi,

what about your stgpool structure? Are you backing up to the disk or to the tape? Have you got any volume in your pools in READWRITE access mode? Any in 'Filling' state?



Hope it helps



Harry

 

[falcon73]

hi Steven,



Here is the "q log" information as per your request:-



Space Capacity Extension Reduction Size Usable Pages Util Pct

(MB) (MB) (MB) (MB) (bytes) Pages Util

--------- -------- --------- --------- ------- --------- --------- ----- -----

4,000 4,000 0 3,996 4,096 1,023,488 254 0.0 0.3



Hope this relevant in response to your request.



Hi Harry, we are backup up to disk stgpool, and this how it look when I "q stgpool"



BACKUPPOOL DISK 205 G 52.2 52.2 80 30 DAILYPOOL

DAILYPOOL LTODEV 16,794 G 3.1 26.8 99 99

DAILYPOOL- LTODEV 0.0 M 0.0

OFF

DAILYPOOL2 LTODEV 0.0 M 0.0 0.0 90 70

DBBACKUPPO LTODEV 0.0 M 0.0 0.0 90 70

OL

HRPOOL LTODEV 4,506 G 0.1 9.1 90 70

HRPOOL-OFF LTODEV 0.0 M 0.0

MONTHLYPOOL LTODEV 0.0 M 0.0 0.0 90 70

MONTHLYPOO LTODEV 0.0 M 0.0

L-OFF

WEEKLYPOOL LTODEV 44,646 G 1.8 8.3 90 70

WEEKLYPOOL LTODEV 0.0 M 0.0

-OFF



Your last question, I'm lost, unable to answer it!



Thanks.

Falcon73

 

[dmdbt]

Hi falcon73,



look at your tape-pool

be sure the amount of scratch volumes is not restricted

and the "Delay Period for Volume Reuse " is not too high



look at your disk-pool:

be sure the Definition "Next Storage Pool" is correct.







dmdbt

 

[falcon73]

Hi dmdbt,



Our tape-pool is not restricted and I dont think there is a delay period for volume reuse.



The definition "next storage pool" is confirmed correct.



What could be the actual problem I'm having???



Please advise ASAP as i missed 4 days of my windows file system backup.



Thanks.



Falcon73

 

[may]

can you do a



q stg xxxx f=d ==> on the Stgpool wich is targeted by the backup



and a



q stg xxxx f=d ==> on the "next Stgpool "

 

[Harry_Redl]

Hi,

so involved pools are Backuppool (Disk, 205G, 52%) and DailyPool (LTO) (if it is correctly set as "Next" pool for Backuppool)

TSM can store your data on volumes marked as "ReadWrite" only. So check vour volumes in both pools for these volumes



q vol stgp=Backuppool acc=readwrite

q vol stgp=dailypool acc=readw



Moreover, tapes in your Dailypool can be in Full state. It means the tape is at the end and no data can be appended. It does not mean that the tape holds XY GB of active data. So you can have all tapes Full and yet your stgpool utilization can be 2%.



Oposite of this state is state "Filling" - it means tape you can append to.



q vol stgp dailypool status=filling



For successfull backup you need

1) enough space in Readwrite volumes in Backuppool

AND/OR

2) enough space in Readwrite volumes in Dailypool is state "Filling" (Full tapes can be in ReadW state as well but you cannot write to them ....)



Hope it helps



Harry



P.S. What is the amount of data you are backing up?

 

[falcon73]

Hi Harry,



My incremental is 80-90GB on daily and weekly full backup is about 1.2TB.



Anyway, your recommendation is very useful. Hope it does identify the root of the problem.



Thanks anyway, keep you posted of the latest development.



Falcon73

 

[robbenleaf]

We just resolved a similar issue: we had consolidated some disk storage pools a few weeks ago, and a few servers started getting only partial backups, which would fail with the ANS5092S message.



It turns out that some of the data was directed to use a management class that didn't get updated to reflect the new storage pool structure; the storage pool that it was trying to use no longer exists. We updated the copy group in that management class to use the new disk pool and activated the policy set, and things look much better now.



So you may want to check all of your backup copy groups to make sure they are using the disk pools you think they are.

 

[Liron19]

We just resolved a similar issue: we had consolidated some disk storage pools a few weeks ago, and a few servers started getting only partial backups, which would fail with the ANS5092S message.



It turns out that some of the data was directed to use a management class that didn't get updated to reflect the new storage pool structure; the storage pool that it was trying to use no longer exists. We updated the copy group in that management class to use the new disk pool and activated the policy set, and things look much better now.



So you may want to check all of your backup copy groups to make sure they are using the disk pools you think they are.

---------------
Hi, I'm having the same problem- and was reading your post- can you explain to a newbie in this how to check if backup copy groups are using the disk pools they are supposed to use?
It started with the TSM not backing up any files, then a technition opend the TSM and placed a tape in the upper slot. (there was another tape in the second slot, but the first slot was empty).
then the TSM started to backup a few files (hardly enough though)..
now it has come to a situation where the TSM is backing up maybe 10 files of every server. I don't know what's wrong... I know the tapes are supposed to over-ride the information whenever it's full, but have no idea how to fix this.
the idea you suggested makes sence to me, but I'm having problems understading, since English is my second language. Plus I'm really new to this TSM and backup buissness.
hope I didn't confuse anyone with my question, thanks and have a nice day.
Liron19.

 

[BBB]

"q copy f=d" shows the destination storage pools for your copygroups.

It sounds more like to me that you've run out of space in one of your storage pools (q stg f=d), or have run out of scratch tapes (q vol - look for scratch tapes) or some other newbie problem.

 

[sgabriel62]

Falcon
Hey it does appear that you have a client or management class that is going to a storage pool that is non-existent or labled read only. Lets be a bit more specific - where are your mail clients going to? Where are your DB clients going to? Where are your standard backups going to? Do you have your TDP agents correctly configured outside of defaults. Are they going straight to tape or lanfree configurable? Lets review the activity log alot better as well - lets isolate clients to determine which is creating the error. Im not suspecting all three types are causing the issue - Id expect one of the TDP Agents
Send us more info
If this matter is resolved sorry for my late insert - please share the result

 

[Liron19]

"q copy f=d" shows the destination storage pools for your copygroups.

It sounds more like to me that you've run out of space in one of your storage pools (q stg f=d), or have run out of scratch tapes (q vol - look for scratch tapes) or some other newbie problem.

I will check those commnads and reply with the answer- but since my tsm and servers are on a LAN, I need to type the resaults of everything and then post them here..it will be a while- but I'll try all of those.

 

[Liron19]

thanks anyways...

hello all,
I just called IBM technitions (yet again...)
and this time it just might work..
I had to update the path to my drives, which were also offline..
as I thought, ttly newbie mistake (more like n00b )
anyways, hope it will last long..
on some of my servers one drive is still offline, so I tried this command and it says one of my parameters are missing:
update path tsm (drivename) srctype=server desttype=drive online=yes.
is there something missing?!

 

[balirake]

The resolution is simple. just update maxscratch of your Storage pool
update stg <STGPOOLNAME> maxscratch=<NEW_SCRATCH_COUNT>