Volumes Showing as Full while 0.0% used; no Output in q content <volume>

[kertvaros]

OS: RHEL 6.10
Spectrum Protect Server: 8.1.8.100

We converted deduplication pool of type FILE into directory container pool; all apparently went OK:
>q stg
Session established with server TSM_SDCPRIM02: Linux/x86_64
Server Version 8, Release 1, Level 8.000
Server date/time: 03/05/2020 15:26:59 Last access: 03/05/2020 14:36:35


Storage Device Storage Estimated Pct Pct High Low Next Storage Pool
Pool Name Class Name Type Capacity Util Migr Mig Mig Pct



----------- ---------- --------- ---------- ----- ----- ---- --- -----------
DCPOOL DIRECTORY 237,298 G 41.8
DDPOOL SMDSFILE DEVCLASS 229 G 0.0 0.0 90 70

However, three volumes show as full although 0.0% used:

Protect: TSM_SDCPRIM02>q vol stg=ddpool status=full
Session established with server TSM_SDCPRIM02: Linux/x86_64
Server Version 8, Release 1, Level 8.000
Server date/time: 03/06/2020 11:19:30 Last access: 03/05/2020 15:26:59


Volume Name Storage Device Estimate- Pct Volume
Pool Name Class Name d Capaci- Util Status
ty
------------------------ ----------- ---------- --------- ----- --------
/tsm/tsm_sdcprim02/smds- DDPOOL SMDSFILE 73.7 G 0.0 Full
1d07/ddvol053
/tsm/tsm_sdcprim02/smds- DDPOOL SMDSFILE 71.9 G 0.0 Full
1d16/ddvol023
/tsm/tsm_sdcprim02/smds- DDPOOL SMDSFILE 72.7 G 0.0 Full
2d07/ddvol00D

Apparently there is no content to show on either of them:

Protect: TSM_SDCPRIM02>q content /tsm/tsm_sdcprim02/smds1d07/ddvol053
ANR2034E QUERY CONTENT: No match found using this criteria.
ANS8001I Return code 11.

Protect: TSM_SDCPRIM02>q content /tsm/tsm_sdcprim02/smds2d07/ddvol00D
ANR2034E QUERY CONTENT: No match found using this criteria.
ANS8001I Return code 11.

Ran AUDIT VOL with FIX=YES option but to no avail:

03/06/2020 11:27:38 ANR0984I Process 224 for AUDIT VOLUME (REPAIR) started in
the BACKGROUND at 11:27:38 AM. (SESSION: 810873,
PROCESS: 224)
03/06/2020 11:27:38 ANR2312I Audit Volume (Repair) process started for volume
/tsm/tsm_sdcprim02/smds1d07/ddvol053 (process ID 224).
(SESSION: 810873, PROCESS: 224)
03/06/2020 11:27:38 ANR4132I Audit volume process ended for volume
/tsm/tsm_sdcprim02/smds1d07/ddvol053; 15 files
inspected, 0 damaged files deleted, 0 damaged files
marked as damaged, 0 files previously marked as damaged
reset to undamaged, 0 objects updated. (SESSION:
810873, PROCESS: 224)
03/06/2020 11:27:38 ANR0987I Process 224 for AUDIT VOLUME (REPAIR) running in
the BACKGROUND processed 15 items with a completion
state of SUCCESS at 11:27:38 AM. (SESSION: 810873,

Volume still shows as full but no data:

Protect: TSM_SDCPRIM02>q VOL /tsm/tsm_sdcprim02/smds1d07/ddvol053

Volume Name Storage Device Estimate- Pct Volume
Pool Name Class Name d Capaci- Util Status
ty
------------------------ ----------- ---------- --------- ----- --------
/tsm/tsm_sdcprim02/smds- DDPOOL SMDSFILE 73.7 G 0.0 Full


How can these few volume safely be turned into Volume Status EMPTY so that we can proceed with volume deletion and re-purposing the storage?

Thanks!

Kerti

 

[marclant]

On a sequential volume, full means that at one time, Spectrum Protect wrote all the way to the end of it. As data expires, or is moved off like in your case with a convert, the space utilization goes down. However, because it was once written to the end, it remains full. The reason is that with sequential volumes, you can only write new data at the end of the existing data, once you reach the end of the volume, it's full.

Now, to be empty, there needs to be absolute 0% utilized. 0.0 shown in the output is rounded, it just means less than 0.05%. It's odd that Q CONTENT shows nothing, but the audit inspected 15 files.

If you don't care about the 15 files inspected, you can do a delete volume with discarddata=yes. But I'm guessing that without knowing what files these are, it's a hard call to make. Might be best to open a case with IBM.