UNAVAILABLE Tape and Space Reclamation

[RandomAdmin]

Hello,

I know title is not that clear but this is a twofold issue and not sure how I should approach it, I will start from the Space Reclamation issue.

If I issue a q proc I see Space Reclamation running for one specific volume (input volume A output volume B) this would be fine if it would actually run, when I issue the q proc what I get is something like:

Volume A01045 (storage pool Pool_TL01), Moved
Files: 0, Moved Bytes: 0, Unreadable Files: 0,
Unreadable Bytes: 0. Current Physical File
(bytes): 362,310,008,240 Current input volume:
A01045. Current output volume(s): A01100.

And it stays like this forever, I've even left it run overnight with no real advancement, if I do cancel the process TSM just restarts it and I'm again at square 1. Apparently no error exist with this specific tape the only thing I see it is marked as FULL.

Other issue is with unavailable tapes, this morning while checking the daily report I see two unavailable tapes I've already tried to run and audit library and inventory from the library itself to no avail here's what I see in the report:

VOLUME_NAME ACCESS STATUS
A01105 UNAVAILABLE FULL
A01116 UNAVAILABLE FULL

Should I wait for TSM to expire inventory in these tapes? Or do something.

Apologies for the silly questions but I'm still moving my first steps and while I try to do most of the stuff on my own with the help of the forum and documentation I could not really find anything about this one.

Thanks in advance!

 

[marclant]

For reclamation to work on volumes, the access must be read/write.

Are those volumes checked in the library? You can check with the "query libvolume" command. If they are checked in, then you can update the access to read only:

update volume A01105 access=readonly

Repeat for other unavailable volumes.

Then retry reclamation.

 

[RandomAdmin]

Thanks marclant,

I've updated the volume to RO but seems reclamation is not going through anyhow, what I've found in the actlog which could be related is :

05/29/2015 14:14:09      ANR8356E Incorrect volume A01117 was mounted instead of
                          volume A01116 in library TL01. (PROCESS: 343)
05/29/2015 14:14:37      ANR1081E Space reclamation is ended for volume A01116.
                          Storage media is inaccessible. (PROCESS: 343)
05/29/2015 14:14:37      ANR0985I Process 343 for SPACE RECLAMATION running in the
                          BACKGROUND completed with completion state FAILURE at
                          14:14:37. (PROCESS: 343)

Not sure this is of any help but I'm running a new audit library.

It is worth mentioning last week we suffered a DB2 failure which forced me to restore the DB which went through smoothly and at the time I ran an audit with the fix parameter.

Not sure if the above is relevant but being ignorant I give as many details as possible.

If there is any further action I can take or you have any idea I would be more than glad to hear

 

[marclant]

Ah. When you restored the DB, if it was a point in time, the library inventory at the time of the backup would not necessarily be the same.

Doing an audit library should fix that.
ANR8356E will put volumes unavailable though, so you may need to update any volumes that are unavailable to RO after the audit completes.

 

[RandomAdmin]

Thanks again marclant,

apparently, despite having run the audit, this is the case and now audit library is even failing. any idea how further proceed? I would open a PMR with IBM if only this server was not TSM 6.2... (yeah I know it is EOL but it's a long story why we can't update this).

Thanks again.

 

[marclant]

Audit will fail if other processes or sessions are trying to use the drives. Make sure there is no tape mounted before starting the audit library "query mount". If there are no mounts, try the audit again. If the audit fails again, show us the errors.

 

[RandomAdmin]

Ok I had to do some ninja search in the actlog (which is quite a feat when you don't know what you're doing) and found couple of entries which could help:

05/29/2015 15:02:38      ANR8379I Mount point in device class 3592_TL01 is waiting
                          for the volume mount to complete, status: WAITING FOR
                          VOLUME. (SESSION: 5532)
05/29/2015 15:02:38      ANR8376I Mount point reserved in device class 3592_TL01,
                          status: RESERVED. (SESSION: 5532)
05/29/2015 15:02:38      ANR8334I         4 matches found. (SESSION: 5532)
05/29/2015 15:02:56      ANR8914I Drive DRIVE_3_TL02 (mt2.0.0.6) in library IZHTL02
                          needs to be cleaned. (SESSION: 5481, PROCESS: 340)
05/29/2015 15:02:56      ANR8914I Drive DRIVE_3_TL02 (mt2.0.0.6) in library IZHTL02
                          needs to be cleaned. (SESSION: 5481, PROCESS: 340)
05/29/2015 15:03:00      ANR8468I 3592 volume A00003 dismounted from drive
                          DRIVE_3_TL02 (mt2.0.0.6) in library IZHTL02. (SESSION:
                          5481, PROCESS: 340)
05/29/2015 15:08:26      ANR8912E Unable to verify the label of volume from
                          slot-element 1092 in drive DRIVE_2_TL01 (mt1.0.0.8) in
                          library IZHTL01. (SESSION: 5532, PROCESS: 346)
05/29/2015 15:09:40      ANR8460E AUDIT LIBRARY process for library IZHTL01 failed.
                          (SESSION: 5532, PROCESS: 346)
05/29/2015 15:09:40      ANR0985I Process 346 for AUDIT LIBRARY running in the
                          BACKGROUND completed with completion state FAILURE at

05/29/2015 15:09:40      ANR8460E AUDIT LIBRARY process for library IZHTL01 failed.
                          (SESSION: 5532, PROCESS: 346)
05/29/2015 15:09:40      ANR0985I Process 346 for AUDIT LIBRARY running in the
                          BACKGROUND completed with completion state FAILURE at
more...   (<ENTER> to continue, 'C' to cancel)

                          15:09:40. (SESSION: 5532, PROCESS: 346)

I think the ANR8912E Unable to verify the label in volume bla bla has something to do with the failure, as I write I have a new audit going on but while this runs I keep seeing processes for space reclaim popping up for the same volumes I've posted above.

Do you think that stopping TSM and do an inventory on library would be of any help?

 

[marclant]

The can't read label is a by-product of this:

05/29/2015 15:02:56      ANR8914I Drive DRIVE_3_TL02 (mt2.0.0.6) in library IZHTL02
                          needs to be cleaned. (SESSION: 5481, PROCESS: 340)
05/29/2015 15:02:56      ANR8914I Drive DRIVE_3_TL02 (mt2.0.0.6) in library IZHTL02
                          needs to be cleaned. (SESSION: 5481, PROCESS: 340)

If your library is setup for automatic cleaning, make sure you have a cleaning tape checked in.

If not, manually clean the drive with a cleaning tape. HELP CLEAN DRIVE for syntax.

 

[RandomAdmin]

Thanks a lot I missed those lines, as a matter of fact I'm still waiting for cleaning tapes for that library but guess I will not get them before end of June... Well not much I can do to help this apart from wondering while I get all these problematic customers

Thanks for your help Marclant really helped me!