cancel
Showing results for 
Search instead for 
Did you mean: 

Unable to backup Oracle database

Fortune_KN
Level 3
Partner

Good morning,

I am trying to backup my Oracle database with netbackup 7.6.1 since few days without success.

the status below:

05/23/2016 10:53:33 - Info nbjm (pid=1356) starting backup job (jobid=28711) for client 172.21.5.48, policy Oracle_test, schedule Full
05/23/2016 10:53:33 - Info nbjm (pid=1356) requesting MEDIA_SERVER_WITH_ATTRIBUTES resources from RB for backup job (jobid=28711, request id:{36045CE4-20CC-11E6-BDBA-9262310CA726})
05/23/2016 10:53:33 - requesting resource NetBackup_DD_FRV_DP-stu
05/23/2016 10:53:33 - requesting resource fr_master.NBU_CLIENT.MAXJOBS.172.21.5.48
05/23/2016 10:53:33 - requesting resource fr_master.NBU_POLICY.MAXJOBS.Oracle_test
05/23/2016 10:53:33 - granted resource  fr_master.NBU_CLIENT.MAXJOBS.172.21.5.48
05/23/2016 10:53:33 - granted resource  fr_master.NBU_POLICY.MAXJOBS.Oracle_test
05/23/2016 10:53:33 - granted resource  NetBackup_DD_FRV_DP-stu
05/23/2016 10:53:33 - estimated 0 kbytes needed
05/23/2016 10:53:33 - Info nbjm (pid=1356) started backup (backupid=172.21.5.48_1463997213) job for client 172.21.5.48, policy Oracle_test, schedule Full on storage unit NetBackup_DD_FRV_DP-stu
05/23/2016 10:53:33 - started process bpbrm (pid=13378)
05/23/2016 10:53:34 - Info bpbrm (pid=13378) 172.21.5.48 is the host to backup data from
05/23/2016 10:53:34 - Info bpbrm (pid=13378) reading file list for client
05/23/2016 10:53:34 - Info bpbrm (pid=13378) starting bphdb on client
05/23/2016 10:53:34 - Info bphdb (pid=10599) Backup started
05/23/2016 10:53:34 - Info bphdb (pid=10599) Processing /usr/openv/netbackup/ext/db_ext/oracle/samples/rman/hot_database_backupOryx1_BP1.sh
05/23/2016 10:53:34 - Info bphdb (pid=10599) Waiting for the child status
05/23/2016 10:53:34 - connecting
05/23/2016 10:53:34 - connected; connect time: 0:00:00
05/23/2016 10:53:35 - Error bpbrm (pid=13378) from client 172.21.5.48: ERR - Script exited with status = 2 <none of the requested files were backed up>
05/23/2016 10:53:35 - Error bpbrm (pid=13378) from client 172.21.5.48: ERR - bphdb exit status = 6: the backup failed to back up the requested files
05/23/2016 10:53:36 - Info bphdb (pid=10599) done. status: 6: the backup failed to back up the requested files
05/23/2016 10:53:36 - end writing
the backup failed to back up the requested files  (6)
 

the problem seems to come from the script but i have no idea!

 

thank you for the help!

 

3 ACCEPTED SOLUTIONS

Accepted Solutions

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Please do not use IP address for client backups!

For Oracle backups, you need the following:

Successful forward and reverse name lookup between the master server and and the client.

Successful forward and reverse name lookup between the media server and and the client.

If all servers and clients are not on DNS, then add hosts entries and test forward and reverse name lookup in all directions using bpclntcmd.
(Remember to clear NBU host cache after adding hosts entries)
See :
How to verify name resolution for NetBackup (tm) systems, using the "bpclntcmd" command : http://www.veritas.com/docs/000026124 

When and how to clear the NetBackup host cache : http://www.veritas.com/docs/000008194

Ensure the same hostnames that are added to hosts files are the exact same names used in NBU config  - client name in Policy and SERVER names in bp.conf.
1st SERVER entry in Client's bp.conf must always be the master.

Confirm that port 1556 is open in all directions between client and master and client and media server.
Check that OS firewall is disabled or allowing port 1556 in all directions.

Once you have fixed all of the above, let us know the outcome.

We may need more if successful comms does not resolve the problem.

 

 

View solution in original post

Michael_G_Ander
Level 6
Certified

This post contains some step for troubleshooting database backup

https://www.veritas.com/community/forums/general-database-backup-error-troubleshooting

As I see this: Script exited with status = 2, I would check bphdb stdout and stderr files first for indication of why the script fails

 

The standard questions: Have you checked: 1) What has changed. 2) The manual 3) If there are any tech notes or VOX posts regarding the issue

View solution in original post

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

The Client is using the wrong interface to connect to the master server.

Please use the hostname associated with 172.21.5.48 IP in client's bp.conf as CLIENT_NAME.
Also use this hostname in NBU Oracle policy in Clients tab.
You may also want to hard-code the hostname (associated with 172.21.5.48 IP) as NB_ORA_CLIENT  variable in the rman script.

The above will ensure that the client uses the correct interface when connecting to the master.

Ensure that the master can resolve 172.21.5.48 to the hostname in Oracle policy.

View solution in original post

8 REPLIES 8

Michal_Mikulik1
Moderator
Moderator
Partner    VIP    Accredited Certified

Hello,

so it was working before? If yes, then the cause won't be in the script if it is unchanged.

My quick tips are:

- database is stopped

- database is out of ARCHIVELG mode now

- sysdba password has been changed

- ...

Enable dbclient debug client log (see NetBackup Oracle Admin Guide) and it should tell more. Or simply consult your Oracle DBA.

Regards

Michal

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Please do not use IP address for client backups!

For Oracle backups, you need the following:

Successful forward and reverse name lookup between the master server and and the client.

Successful forward and reverse name lookup between the media server and and the client.

If all servers and clients are not on DNS, then add hosts entries and test forward and reverse name lookup in all directions using bpclntcmd.
(Remember to clear NBU host cache after adding hosts entries)
See :
How to verify name resolution for NetBackup (tm) systems, using the "bpclntcmd" command : http://www.veritas.com/docs/000026124 

When and how to clear the NetBackup host cache : http://www.veritas.com/docs/000008194

Ensure the same hostnames that are added to hosts files are the exact same names used in NBU config  - client name in Policy and SERVER names in bp.conf.
1st SERVER entry in Client's bp.conf must always be the master.

Confirm that port 1556 is open in all directions between client and master and client and media server.
Check that OS firewall is disabled or allowing port 1556 in all directions.

Once you have fixed all of the above, let us know the outcome.

We may need more if successful comms does not resolve the problem.

 

 

Fortune_KN
Level 3
Partner

Thank you! ASAP i have the result i will be back here

Fortune_KN
Level 3
Partner

hello,

@Michal: it hasn't worked before i am just setting this. there is a hot backup, and the database is in Archivelog

@ Marianne

everything is okay except (i think) the result of  "bpclncmd -pn" on the client :

expecting response from server fr_master
172.21.5.8 *NULL* 172.21.5.8 42522

172.21.5.8 is an Ip@ of the client, but which is used for other purpose.
 

You will find as attachment the dbclient log and the rman log. 

The Rman log specifies the following error : "Server Status:  the specified policy is not of the correct client type" , but i have the Policy time is "Oracle", and the name of the client is the same as the name in the policy. 

by the way, i am using Netbackup 7.6.1 on a RHEL 6.2 master and media,  and the client is a  Solaris 9 with Netbackup 7.5 client 

Michael_G_Ander
Level 6
Certified

This post contains some step for troubleshooting database backup

https://www.veritas.com/community/forums/general-database-backup-error-troubleshooting

As I see this: Script exited with status = 2, I would check bphdb stdout and stderr files first for indication of why the script fails

 

The standard questions: Have you checked: 1) What has changed. 2) The manual 3) If there are any tech notes or VOX posts regarding the issue

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

The Client is using the wrong interface to connect to the master server.

Please use the hostname associated with 172.21.5.48 IP in client's bp.conf as CLIENT_NAME.
Also use this hostname in NBU Oracle policy in Clients tab.
You may also want to hard-code the hostname (associated with 172.21.5.48 IP) as NB_ORA_CLIENT  variable in the rman script.

The above will ensure that the client uses the correct interface when connecting to the master.

Ensure that the master can resolve 172.21.5.48 to the hostname in Oracle policy.

Nicolai
Moderator
Moderator
Partner    VIP   

Move on and start using Netbackup Oracle Intelligent Policies yes

It is really easier than using scripts, it took me less than 10 minutes to set up Oracle Intelligent Policies first time I tried it.

Name resolution must be in place, else nothing work in Netbackup.

 

Fortune_KN
Level 3
Partner
Hello all, Thank you so much for the help. Marianne and Michael have been very helpful on this case. The problem was coming from both a wrong interface usage and the script. It is okay now. Thank you so much @Nicolai, you are right. But since my Os is Solaris 9 I can't use netbackup 7.6 client, and then no oracle intelligent policy. Thank you all.