RMAN backup jobs are failing with child process killed by signal (27)

[smsaccet1]

We have been getting RMAN backup failures with child process killed by signal from multiple oracle servers. only a few jobs are failing not all jobs. RMAN backups are triggered through RMAN template,we are not using OIP... we are using HPP(Sepaton)as OST device. All our backups are backed up to HPP Storage server and replicated to other site through Netbackup AIR. there is no issue with replication. I have attached bptm and bpbrm logs. can anyone help to fixing this issue.

Master: Netbackup appliance 7.7.2

RMAN client(Media server) - 7.7.2 

I also created a case with Veritas and Hitachi...They have been unable to fix this yet.

Job detailed status:

03/31/2017 18:23:11 - Info bpbrm (pid=7917) reading file list for client
03/31/2017 18:23:11 - Info bpbrm (pid=7917) listening for client connection
03/31/2017 18:23:11 - Info bpbrm (pid=7917) INF - Client read timeout = 300
03/31/2017 18:23:11 - connected; connect time: 0:00:00
03/31/2017 18:23:12 - Info bpbrm (pid=7917) accepted connection from client
03/31/2017 18:23:12 - Info dbclient (pid=3395) Backup started
03/31/2017 18:23:12 - Info bpbrm (pid=7917) bptm pid: 7954
03/31/2017 18:23:12 - Info bptm (pid=7954) start
03/31/2017 18:23:12 - Info bptm (pid=7954) using 262144 data buffer size
03/31/2017 18:23:12 - Info bptm (pid=7954) using 30 data buffers
03/31/2017 18:23:12 - Info bptm (pid=7954) start backup
03/31/2017 18:25:11 - begin writing
03/31/2017 18:25:17 - Info dbclient (pid=3395) dbclient(pid=3395) wrote first buffer(size=262144)
03/31/2017 18:29:59 - end writing; write time: 0:04:48
03/31/2017 18:30:00 - Error bpbrm (pid=7917) child killed by signal (139)
03/31/2017 18:30:00 - Info dbclient (pid=3395) done. status: 27: child process killed by signal
child process killed by signal (27)

[Marianne]

I can only view .txt logs on my mobile phone. Will have a look at your logs on Monday.

Do you have bphdb and dbclient logs as well?
Please upload as well (preferably .txt format).

Have you tried to increase Client Read Timeout to 1800 on media server? 300 (5 min) is not enough for db backups.

[smsaccet1]

bphdb logs are not logged on the bphdb folder. I have uploaded the bpclient logs here. I had to upload as zip file as it exceeds 10 MB limit.

[Marianne]

Veritas Support engineers have the time and tools to sift though large high level logs.
I trust that they will be able to help you.

[Michael_G_Ander]

Have you talked with the Oracle DBA, in my experience they can often help with errors like this.

https://vox.veritas.com/t5/NetBackup/General-database-backup-error-troubleshooting/m-p/675381 is the steps I have used to troubleshoot database backups