solaris client backup failing with error code 13 for /device mount point only

[wannawin]

client : Netbackup-solaris 10- 6.5.3

Master: Solaris-netbackup6.5.6

Please find attached detail logs

[Marianne]

Please exclude /devices from backups. 

There is no point in trying to backup device files.

They get automatically added during OS install and whenever new devices are added and reconfiguration reboot is done or when devfsadm command is run.

[wannawin]

Thanks marianne.

But the higher management not allowed to exclude this from backup rotation. so to resolve this is the only option.

[jim_dalton]

New management!

[wannawin]

Hi Jim,

Yup, i told him what marianne mentioned above but they said .. we are not going to  exclude this from backup rotation and we have to resolve this issue.

[Andy_Welburn]

Prior to the Oracle Solaris 10 OS, /devices was an on-disk filesystem composed of subdirectories and files. Beginning with the Oracle Solaris 10 OS, /devices is a virtual filesystem that creates these subdirectories and special files on demand.

The devfs filesystem manages a name  space  of  all  devices under  the Solaris operating environment and is mounted during boot on the /devices name space.The /devices name space is dynamic and reflects the  current state  of  accessible  devices  under  the Solaris operating environment. The names of all attached device instances  are present under /devices.

The content under /devices is under the exclusive control of the devfs filesystem and cannot be changed.

devfsadmd(1M) is the daemon  version  of  devfsadm(1M).  The daemon  is  started during system startup and is responsible for handling both reconfiguration boot processing and updating /dev and /devices in response to dynamic reconfiguration event notifications from the kernel.

Taken from various man page & Oracle documentation sources - with this in mind I wonder what your management is hoping to get from a backup of /devices on a Solaris 10 system?

[Nicolai]

Are we sure its /devices ?

Timer expired could also be 1000 of small files in a Oracle audit directory - Configure CLIENT_READ_TIMEOUT to 1900

01/30/2014 05:43:48 - Error bpbrm (pid=21611) socket read failed: errno = 62 - Timer expired

[jim_dalton]

Nicolai has a point...I've yet to see the connection between /devices and the logfile.Maybe I missed something.

Try backing up /devices and note what you end up with. It works so you can keep em happy on that front, but it does appear your issue is not in fact related to /devices.

Jim

[Andy_Welburn]

Ditto. Saves me typing the same!

Maybe this will help:

Usage specifics of the bpbkar_path_tr touch file to enable enhanced debug logging of the bpbkar process
http://www.symantec.com/business/support/index?page=content&id=TECH31513

Beware the caveat:

Note: The use of this touch file will cause large amount of logging in the bpbkar logs, so ensure this is turned off by removing this touch file once the troubleshooting process is finished.

Or maybe the info is already in the logs?

[wannawin]

client read time out changed and backup is in progress but no byte count reported from last 1 hr. 01/30/2014 16:29:19 - requesting resource 9940B_9840B_HCART_HCART2_MIXED 01/30/2014 16:29:19 - requesting resource p02inf03.NBU_CLIENT.MAXJOBS.r02jbweb08 01/30/2014 16:29:19 - requesting resource p02inf03.NBU_POLICY.MAXJOBS.FIS_OPT1_OCDC_UNIX 01/30/2014 16:29:20 - granted resource p02inf03.NBU_CLIENT.MAXJOBS.r02jbweb08 01/30/2014 16:29:20 - granted resource p02inf03.NBU_POLICY.MAXJOBS.FIS_OPT1_OCDC_UNIX 01/30/2014 16:29:20 - granted resource L81603 01/30/2014 16:29:20 - granted resource OC9940B_a1p02d11 01/30/2014 16:29:20 - granted resource OC_Stg_hcart2_p02danbs06 01/30/2014 16:29:20 - estimated 0 kbytes needed 01/30/2014 16:29:31 - connecting 01/30/2014 16:29:31 - connected; connect time: 0:00:00 01/30/2014 16:29:31 - begin writing

[wannawin]

failed again 01/30/2014 15:56:44 - requesting resource 9940B_9840B_HCART_HCART2_MIXED 01/30/2014 15:56:44 - requesting resource p02inf03.NBU_CLIENT.MAXJOBS.r02jbweb08 01/30/2014 15:56:44 - requesting resource p02inf03.NBU_POLICY.MAXJOBS.FIS_OPT1_OCDC_UNIX 01/30/2014 15:56:44 - granted resource p02inf03.NBU_CLIENT.MAXJOBS.r02jbweb08 01/30/2014 15:56:44 - granted resource p02inf03.NBU_POLICY.MAXJOBS.FIS_OPT1_OCDC_UNIX 01/30/2014 15:56:44 - granted resource L81603 01/30/2014 15:56:44 - granted resource OC9940B_a1p02d11 01/30/2014 15:56:44 - granted resource OC_Stg_hcart2_p02danbs06 01/30/2014 15:56:45 - estimated 0 kbytes needed 01/30/2014 15:56:54 - connecting 01/30/2014 15:56:54 - connected; connect time: 0:00:00 01/30/2014 15:56:54 - begin writing 01/30/2014 18:36:55 - Error bpbrm (pid=11765) socket read failed: errno = 62 - Timer expired 01/30/2014 18:37:02 - end writing; write time: 2:40:08 file read failed (13) under bpbkar bash-3.2$ tail -f log.013014 12:47:13.883 [2730] <4> bpbkar: INF - setenv BPEND_TIMEOUT=300 12:47:13.883 [2730] <4> bpbkar: INF - setenv RESTARTED=1 12:47:13.883 [2730] <4> bpbkar: INF - setenv BACKUPID=r02jbweb08_1391097404 12:47:13.883 [2730] <4> bpbkar: INF - setenv UNIXBACKUPTIME=1391097404 12:47:13.884 [2730] <4> bpbkar: INF - setenv BACKUPTIME=Thu Jan 30 09:56:44 2014 12:47:13.885 [2730] <4> bpbkar: INF - BACKUP START 12:47:13.885 [2730] <4> bpbkar: INF - Estimate:-1 -1 12:47:13.887 [2730] <2> bpbkar add_to_filelist: starting sizeof(filelistrec) <68> 12:47:13.888 [2730] <4> bpbkar: INF - Processing /devices

[Marianne]

Suggestion:

Log a call with Oracle. 

Ask them their recommendation with regards to backup of /devices.

Please tell us what they say....

[jim_dalton · Accepted solution]

Oracle may suggest to not bother, however as noted earlier I did it, it works, took a minute or two, so we need more info to address your issue...maybe theres something misconfigured.

Other suggestion: upgrade, you are way behind so maybe what you see has been resolved in a later release. My successful backup was on 7601.

Jim

[Marianne]

What happens when you use OS tar command to backup /devices to /dev/null?

e.g. 

tar cvf /dev/null /devices

Let us know....