07-30-2012 06:41 AM
Backups are failing with Error: 13 in NetBackup.
Master: 6.5.6 - HP-UX
Client: 5.1.4 - Win 2K3
Backups is failing on C Drive with Error: 13,
Jul 27, 2012 1:13:01 PM - estimated 2926564 kbytes needed
Jul 27, 2012 2:56:26 PM - connecting
Jul 27, 2012 2:56:27 PM - connected; connect time: 0:00:00
Jul 27, 2012 2:56:27 PM - begin writing
Jul 27, 2012 4:57:25 PM - Error bpbrm (pid=XXXX) socket read failed: errno = 52 - Stream ioctl timeout
Jul 27, 2012 11:34:18 PM - end writing; write time: 8:37:51
file read failed (13)
Nothing found in BPBKAR logs...
It is getting stuck at around 2 GB, when i skipped the file it was showing as Current File in Activity Monitor & re-tried backup, it again got stuck at around 2 GB on some other file.
Please help.
07-30-2012 06:59 AM
You do have some issues there - I will cover one thing that may help later but your biggest issue is at the start of your thread:
Master: 6.5.6 - HP-UX
Client: 5.1.4 - Win 2K3
NBU 6.5 goes end of life in October - NBU 5.1 went end of life a while ago.
As a first step the advice has to be to upgrade everything to 7.1.0.4 and then consider 7.5
In the mean time it sounds from what you say that a lot of files are getting skipped and these may be actually causing timeouts
Change the Client Read and Client Connect timeouts on the Media Servers to 1800 or 3600 to start with.
Next - look at what is being skipped and assess why - if they are things like SQL databases then exclude them so the the process of trying to back up locked files does not create your timeouts
So exclude skipped files, increase your timeouts on the media servers but most importantly upgrade!
Hope this helps
07-30-2012 07:42 AM
In addition to Mark's excellent post - if you don't upgrade everything right away, then at least upgrade the Windows client to 6.5.6 TODAY!
Ensure that 'Checkpoint Restart' in the policy is enabled so that job can at least be restarted from the failure point.
Please share details about backup job - multistream or single stream? Only C-drive or all drives? Other drives successful?
Is there a firewall between client and server? We often see that firewall times out, causing NBU job to fail.
Is the master also the media server?
Which version of HP-UX on the master?
What kind of transfer rate is seen for this backup? If you say the job fails after about 2 Gb, it means throughput is real slow as we can see backup is failing after 2 hours on the media server but then it takes another 4.5 hours for the job to actually fail?
What Timeouts are in place?
After upgrading client, check that bpbkar log on client is still in place. Also verify that bpbrm and bptm logs exist on the media server. This will tell up to when data and metadata is received from the client.
07-30-2012 08:22 AM
I have tried excluding the file, Activity Monitor CURRENT FILE was pointing to, my Backup is getting stuck/failing at around 2GB of data writting always, ir-respective of the file at that moment.
I am checking from server end DISK CHECK Utility, as other Drives (D, E, F, G, H, & I) are backedup successfully. Will update you guys once the output comes.
07-30-2012 08:44 AM
OK - always running a problems report against this client over a week or two - it will list in this ever file that it had to skip which can help with exclusions - especially as if you save the report to a text file you can copy and paste out of it when doing the exclusions