Hi,
Last year we upgraded to 7.5.0.4 + deduplication, since then the backup system is not completely stable.
Randomly our windows backups started to fail with error statuses:
- Media write error (84)
- File write failed (14)
- File read failed (13)
In the beginning only a few of them failed and as a workaround I created for them a special policy without the accelerator and they started to finish successfully again.
BUT this Friday all full windows backups that never failed before failed at once. Again as a workaround I re-launched them without the accelerator and finished ok.
This Friday all backups failed with error statuses 13 & 14. Inside the Detailed Status tab another status error can be seen:
"
03/03/2013 11:27:15 - Info bpfis(pid=0) done. status: 1542: An existing snapshot is no longer valid and cannot be mounted for subsequent operations
“
And the the full log says something related to the snapshot not being deleted:
"
03/03/2013 11:24:06 - Info bpbkar(pid=4844) NOT using change journal data for <D:\>: unable to validate change journal usage <reason=snapshot needed to flush open files to databases>
03/03/2013 11:25:47 - Error bptm(pid=7258) image copy failed: error 2060018: file not found
03/03/2013 11:25:47 - Info bpbkar(pid=4844) accelerator sent 270046720 bytes out of 5562443264 bytes to server, optimization 95.1%
03/03/2013 11:25:47 - Info bpbkar(pid=4844) bpbkar waited 30 times for empty buffer, delayed 5773 times.
03/03/2013 11:26:50 - Info ochlp-bak002(pid=7258) StorageServer=PureDisk:ochlp-bak002; Report=PDDO Stats for (ochlp-bak002): scanned: 5360324 KB, CR sent: 11367 KB, CR sent over FC: 0 KB, dedup: 99.8%, cache hits: 0 (0.0%)
03/03/2013 11:26:53 - Error bptm(pid=7258) media manager terminated by parent process
03/03/2013 11:26:53 - Error bpbrm(pid=5326) could not send server status message
03/03/2013 11:26:58 - Info bpbkar(pid=4844) done. status: 14: file write failed
03/03/2013 11:26:58 - end writing; write time: 00:06:49
03/03/2013 11:27:03 - Info bpbrm(pid=9186) Starting delete snapshot processing
03/03/2013 11:27:03 - Info bpfis(pid=0) Snapshot will not be deleted
file write failed(14)
03/03/2013 11:27:15 - Info bpfis(pid=280) Backup started
03/03/2013 11:27:15 - Critical bpbrm(pid=9186) from client svpdom11: cannot open C:\Program Files\Veritas\NetBackup\online_util\fi_cntl\bpfis.fim.svpdom11_1362305533.1.0
03/03/2013 11:27:15 - Info bpfis(pid=280) done. status: 1542
03/03/2013 11:27:15 - end operation
03/03/2013 11:27:15 - Info bpfis(pid=0) done. status: 1542: An existing snapshot is no longer valid and cannot be mounted for subsequent operations
"
I am sure that the problem has something to do with the accelerator as backup jobs only fail when the accelerator is enabled. No matter if is the client or the server which deduplcates the data, change journal enabled, etc.
My Solaris or Linux clients work perfectly without problems.
My environment:
- 1 Master 7.5.0.4 Linux
- 1 Media 7.5.0.4 Linux
- Clients Windows/Solaris/linux 7.5.0.4
Thank for your time
Best Regards