Blocked slot in Tape Library is not reported to the GUI

[mfo]

Hi, I just stumbled upon this bug in Veeam B&R V7R2 which might be interesting to some of you:
Support Case is 00491239.

We have a Quantum SuperLoader 3 Tape Library which has 16 slots and a LTO5 SAS drive.

When a tape A is in the LTO drive of the loader and I block its previous place in the library with a new tape B, the tape A will never leave the LTO drive:

Code: Select all

[16.12.2013 09:47:17] <01> Info     [MediaPool] Searching next tape medium to write to in media pool 'weekly', 16 total, 5 avaliable, 1 free, startNewMediaSet 'False'
[16.12.2013 09:47:17] <01> Info     [MediaPool] Last modified tape medium found '926', is Full 'False'
[16.12.2013 09:47:17] <01> Info     Tape medium '926' is successfully locked
[16.12.2013 09:47:17] <01> Info     [TapeLoader] Moving tape 926 to Drive 0 (Tape0)
[16.12.2013 09:47:17] <01> Error    Tape not exchanged. (Veeam.Tape.Core.TapeMediumNotExchangedException)
[16.12.2013 09:47:17] <01> Error       at Veeam.Tape.Core.AutomatedTapeMediumExchanger.ExchangeTapeMediumsForBackup(Boolean updateMediaInfo)
[16.12.2013 09:47:17] <01> Error       at Veeam.Tape.Core.TapeBackupTaskPerformerBase.BackupStartSession(ITapeBackupObject task)
[16.12.2013 09:47:17] <01> Error       at Veeam.Tape.Core.VmTapeBackupTaskPerformer.Backup()
[16.12.2013 09:47:17] <01> Error    Tape 923 requires return to original slot 14, but it is already filled. Move tape manually, and retry operation. (System.Exception)

Tape Backup Jobs still show up as successful, but only have a duration of 1-2 min, so no data is written. The issue with the blocking tape is not propagated to the Veeam B&R GUI or to email reports. On a quick look at the reports, the jobs seem to have run without problems although no backups have been written to tape. I noticed this error one week after its first appearance, so this could have some bad consequences.

In my opinion, this error should either pop up in the GUI/email report so the user can fix this issue or B&R should put the tape to a different slot in the autoloader. With barcodes on the tapes, there should be no need to store the tape at its original slot.

Best regards,
Christoph

[Dima P.]

Hello Christoph,

Could you, please, clarify do you use barcode functionality for your tape library? I believe, we confirmed this to be an issue related only to blind libraries.

[mfo]

I checked the webinterface of my tape library and barcodes are enabled. Also Veeam B&R identifies the barcodes of the tapes and their slots.
So I assume that the barcode scanner works fine.

I just read this post here:
http://forums.veeam.com/viewtopic.php?f=29&t=19388
I think that was the same problem but in that case the error was reported correctly. In my case, no error showed up at all.

In my tape job options, the setting "Eject media once the job finishes" was disabled, as I assumed this would put a tape to the mailslot rather than a library slot.
Enabling this would reduce the risk of blocking a slot, which is fine with me. I normally don't change tapes during the night when backup is runnuing.

Nevertheless, the error should still be propagated to the GUI and not disappear in some hidden log file. I have no idea why this worked in the other case and not here.

Best regards,
Christoph