Confusion Regarding Check Out of Tapes and Checking Back In Again

[saivan]

Greetings, we just got a new TSM and TS4300 system in place and I'm the lucky guy who gets to learn it. However, it's a steep learning curve, and I know next to nothing.

We have 7 tapes to start with. 1 for data, 2 for copy pool, 2 for db backup and 1 scratch.

I checked out 3 tapes last week according to DRM procedures outlined in documentation, but when going back to check back in, I'm very confused. First off the dbbackups stopped running, saying there was no scratch available.

This is what I have here (000002M8 is not even on this list):

Protect: SP1>q drm

Volume Name State Last Update Automated
Date/Time LibName
---------------- ----------------- ------------------- ----------------
000003M8 Mountable 08/12/2019 06:14:51 LIBR1
000006M8 Mountable 08/22/2019 07:55:04 LIBR1
000007M8 Mountable 08/14/2019 18:01:36 LIBR1
000005M8 Mountable 08/13/2019 18:01:25 LIBR1
000004M8 Vault 08/13/2019 16:39:26
000001M8 Vault 08/13/2019 16:39:26

I put the three tapes in the I/O slot today and moved them back into the main drawer and then tried to figure out how to get them back into service, but am stumped. I was able to get one tape to "mountable" state by issuing this command: update volume 000003M8 access=readwrite , but it didn't work for 4M8 and 1M8.

As of now:
SP1> q volume
Volume Name Storage Pool Name Device Class Name Estimated Capacity Pct Util Volume Status
000002M8 LTOPOOL LTOTAPE 20.464 T 22.508 Filling
000003M8 LTOCOPYPOOL LTOTAPE 20.464 T 9.069 Filling
000006M8 LTOCOPYPOOL LTOTAPE 20.464 T 13.438 Filling

SP1> q libvolume
Library Name Volume Name Status Owner Last Use Home Element Device Type
LIBR1 000001M8 Private SP1 1,001 LTO
LIBR1 000002M8 Private SP1 Data 1,004 LTO
LIBR1 000003M8 Private SP1 1,003 LTO
LIBR1 000004M8 Private SP1 1,002 LTO
LIBR1 000005M8 Private SP1 DbBackup 1,008 LTO
LIBR1 000006M8 Private SP1 Data 1,007 LTO
LIBR1 000007M8 Private SP1 DbBackup 1,006 LTO

Any assistance would be greatly appreciated, as I'm fairly certain I'm doing this totally wrong.

 

[moon-buddy]

First of all, welcome to TSM or SP as it is now called! I know the steep learning curve and really, you should be going to TSM or SP Basic Admin Training. TSM or SP is not an intuitive backup software to learn.

If you must do this on your own, read the rest of this post.

Before you drive into the big black hole, please read and understand what DRM or Disaster Recovery Module is about and how it works. Read this (I am presumming you are running TSM under Windoze):

https://www.ibm.com/support/knowledgecenter/SSEQVQ_8.1.0/srv.reference/b_srv_admin_ref_windows2.pdf

DRM generally runs by:

- Retention period
- Type of stgpool
- how tapes are checked out

If the retention period has not been satisfied, checking back in tapes and making it readwrite will not do you good. TSM knows it has not satisifed the retention (resuse) period for it to be useable as scratch. See "help q drmstatus" from the CLI.

The system complained about "no scratch tapes" because there really isn't any. Scratch tapes are always required. My policy with scratch tapes is to have at least double the amount of tapes that I am sending offsite on a daily basis.

Also, a real DRM operation requires that DB backups be sent offsite daily. I still see that you have these inside the library when you did "q libvol". DB backups are from DRM retention or resuse rules. So, "q drmstatus" will tell you how long the DB tapes will be offsite before TSM will recall them. I track DB backups to tape manually so I know when these becomes available as scratch.

As you may have found out, or guessed, you can change the retention or reuse period.

It would be a long write to say what needs to be done for DRM. Truly, DRM is one other type of Operation within TSM.

Again, if you must study TSM on your own, read the link I sent you. In all honesty, you are better of by enrolling in a Basic TSM/SP Admin course followed by an Advanced TSM/SP course.

 

[saivan]

Thank you for your response. I will continue to educate myself on this, and I have also acquired some additional help from experienced admins.