BRRestore failed exit 5

[sheriff_e47]

Dear All,

I want to perform my first ever restore SAP using BRRestore. But it got failed: 

May 4, 2015 4:03:11 PM - begin Restore
May 4, 2015 4:03:12 PM - media needed: 000053
May 4, 2015 4:03:12 PM - media needed: 000244
May 4, 2015 4:03:12 PM - restoring from image qas01ec_1430690331
May 4, 2015 4:15:26 PM - Error bpbrm (pid=16468) bpcd on qas01ec exited with status 23: socket read failed
May 4, 2015 4:15:26 PM - Error bpbrm (pid=16468) client restore EXIT STATUS 23: socket read failed
May 4, 2015 4:19:14 PM - restored from image qas01ec_1430690331; restore time: 0:16:02
May 4, 2015 4:19:15 PM - end Restore; elapsed time 0:16:04
the restore failed to recover the requested files  (5)

I have uploaded the bpcd log here. Please help to solve this. Please let me know if you need some detail information /logs. 

FYI:

client: qas01ec

media: myseccbi

master: backup

Thank you very much

[RamNagalla]

17:38:57.361 [28845] <8> bpcd peer_hostname: gethostbyaddr failed : HOST_NOT_FOUND (1)
17:38:57.361 [28845] <16> bpcd peer_hostname: gethostbyaddr failed to return peer host, herrno = 1
17:38:57.361 [28845] <16> bpcd main: Couldn't get peer hostname

it is not resolving the IP 10.1.7.164 .

make sure this IP is in client hosts file or in DNS to resovle.

PS:-  you are still using netbackup 6.5, which is out of support years ago.. please try to upgarde it to 7.X, 7.6.1 is the latest avaliable.

bpcd main: <---- NetBackup 6.5 0 ------------initiated

[Marianne]

The bpcd log does not cover the same period as the Job details.

Job details shows timestamp starting at 4:03:11 PM - 16:03:11 up to 16:19:15.

bpcd log starts at 16:53.

Or is there a time difference between the client and master server?
If so, how much of a time difference?

We see in bpcd a connection attempt from an IP address that the client cannot resolve to a hostname:

17:09:06.351 [691] <2> logconnections: BPCD ACCEPT FROM 10.1.7.164 .....
...
17:09:06.354 [691] <8> bpcd peer_hostname: gethostbyaddr failed : HOST_NOT_FOUND (1)

Where does this IP address (10.1.7.164) fit in the picture?

Please also share the steps that were followed on the client to start the restore. 
This message seems 'odd':

Unexpected working directory: /root

[sheriff_e47]

Hi,

Here the IP:

client: 10.1.7.23

media: 10.1.7.170

master: 10.1.7.171

And for 10.1.7.164 is IP of another media server (mysprt01ec). I don't know why it use mysprt01ec not myeccbi.

bash-3.00# cat /usr/openv/netbackup/bp.conf 
SERVER = Backup
SERVER = myssmrsldepxi
SERVER = mystrn
SERVER = mysides
SERVER = myseccbi
CLIENT_NAME = qas01ec
REQUIRED_INTERFACE = qas01ec
MEDIA_SERVER = myseccbi
bash-3.00# 

[Marianne]

You will need bprd log on the master server as it seems that the restore instruction is sent to the '10.1.7.164' media server. 
There could be a FORCE_RESTORE..... entry in bp.conf on the master server.

If bprd log folder on the master server does not exist, you will need to create it, restart NBU, then restart NBU.
While creating/checking bprd log folder on the master, check bp.conf for FORCE_RESTORE... entry.

If no such entry in bp.conf, copy bprd log to bprd.txt after next failure and upload as File attachment.

[sheriff_e47]

Hi Marianne,

Thanks for the info. I found that it apparently has FORCE_RESTORE to mysprt01ec in bp.conf in master server. 

I have replaced it to it original media server and it's working normal now. I will keep monitoring it first untill it finish the restore and update here the result.

#FORCE_RESTORE_MEDIA_SERVER = myseccbi mysprt01ec <------ here's the cause
FORCE_RESTORE_MEDIA_SERVER = myseccbi myseccbi

Thanks a lot :)

[Marianne · Accepted solution]

Simply delete the entry.

There is no need for FORCE_RESTORE... entry if you want the backup media server to perform the restore. 
This is the NBU default for restore from tape.

You only need the entry if you want to force the restore via another media server.