Start a Conversation

Unsolved

This post is more than 5 years old

3217

September 24th, 2015 05:00

Oracle RMAN incremental backup doesnt finishing (NMDA1.6)

Dear Networker Experts!

Our customer using Networker 7.6SP5.build1160 to back up Oracle database.

Oracle version is:  11

We use NMDA 1.6 version.

Client version is: 7.6sp5.

The full backup is running fine.

When we run an incremental backup

(Rman parameter backup level=1,type=cumulative, or cumulative incremental backup)

the backup doesnt completed, two saveset is still remaining and group doesnt completed successfully.

The group shows only two remaining savesets:

- RMAN:\DB_name ... and

- save

The client side the save command and nsrdasv command also running.

However i recognize in the logs that rman commands are successfully completed and finished.

Other parameters:

- RMAN: Filesperset: doesnt change, default: 1

- RMAN: Section size: doesnt change, default: 0

- RMAN: Backup as compressed backupset

- NW: group inactivity timeout is 0, and we checked also with 30, doesnt works

What can be wrong? Why is the group doesnt finish?

Please help us to fix this problem. Thanks in advance.

BR:

Pal Szabo

2 Intern

 • 

14.3K Posts

September 24th, 2015 14:00

I think at the time of NW 7.6 NMDA 1.6 didn't exist yet... of course, you know very well you are running ancient versions by current standards.  Anyway, try couple of things - try backup without compression and see if that makes any difference.  I never used level backup with RMAN as it is pain to restore so I am not aware of anything specific that would explain why level backup hangs, but you should check alert log on Oracle side and also via SQL check status of RMAN jobs to see on what procedure is waiting.  This could be very well Oracle issue too.

2 Intern

 • 

202 Posts

September 28th, 2015 06:00

thanks a lot we try your suggestions

3 Posts

September 29th, 2015 00:00

Dear Experts,

Thanks for your suggestions! We have tried it but the problem still exist. On the other hand our Oracle support (database expert) says that RMAN log is just fine and it reports that the backup was successful, but the NetWorker job keeps running.
It seems to me that the NetWorker jobs don't realize that the RMAN script is finished and returned with success.
Pal suggested us to uninstall the old NetWorker (client side) and NMDA and install the latest... we've made the changes but the problem does still occur.

Client side versions:

lgtonmda-8.2.1-1

lgtoclnt-8.2.1-1

Any other suggestions are highly appreciated!

Thanks in advance!

Best regards,

Jozsef

2 Intern

 • 

14.3K Posts

September 29th, 2015 08:00

Can you post your RMAN script (skip connect string)?

3 Posts

October 11th, 2015 16:00

Here's the full backup config from the client's Apps & Modules tab:

(I only changed the DB_INSTANCE name to DBNAME_STD because of security reasons.)

BACKUP_LOG_OPT:"Back up all logs"; BKUP_DB:"true"; BKUP_FORMAT:"%d_%U"; BKUP_LEVEL:"1"; BKUP_TYPE:"CUMULATIVE"; CAT_INSTANCE:"OMS"; CAT_USER:"rman"; DB_INSTANCE:"DBNAME_STD"; DB_USER:"RMANUSER"; DELETE_LOGS:"true"; DELETE_LOGS_ON_TIMESTAMP:"false"; INCLUDEARCH:"true"; INCLUDECF:"true"; NUM_CHANNELS:"32"; ORACLE_DELETE_EXPIRED_BACKUPS:"false"; ORACLE_FRA_DISK_BACKUPS:"false"; ORACLE_HOME:"/u01/app/oracle/product/11.2.0.3/dbhome_1"; TNS_ADMIN:"/u01/app/oracle/product/11.2.0.3/dbhome_1/network/admin"; name:"RMAN"; type:"NSR Oracle RMAN";

RMAN_SCRIPT:"RUN {

ALLOCATE CHANNEL CH1 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH2 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH3 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH4 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH5 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH6 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH7 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH8 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH9 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH10 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH11 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH12 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH13 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH14 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH15 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH16 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH17 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH18 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH19 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH20 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH21 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH22 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH23 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH24 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH25 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH26 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH27 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH28 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH29 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH30 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH31 TYPE 'SBT_TAPE'\;

ALLOCATE CHANNEL CH32 TYPE 'SBT_TAPE'\;

BACKUP

INCREMENTAL  LEVEL 1  CUMULATIVE

FORMAT '%d_%U'

DATABASE

INCLUDE CURRENT CONTROLFILE

PLUS ARCHIVELOG DELETE INPUT

\;

RELEASE CHANNEL CH1\;

RELEASE CHANNEL CH2\;

RELEASE CHANNEL CH3\;

RELEASE CHANNEL CH4\;

RELEASE CHANNEL CH5\;

RELEASE CHANNEL CH6\;

RELEASE CHANNEL CH7\;

RELEASE CHANNEL CH8\;

RELEASE CHANNEL CH9\;

RELEASE CHANNEL CH10\;

RELEASE CHANNEL CH11\;

RELEASE CHANNEL CH12\;

RELEASE CHANNEL CH13\;

RELEASE CHANNEL CH14\;

RELEASE CHANNEL CH15\;

RELEASE CHANNEL CH16\;

RELEASE CHANNEL CH17\;

RELEASE CHANNEL CH18\;

RELEASE CHANNEL CH19\;

RELEASE CHANNEL CH20\;

RELEASE CHANNEL CH21\;

RELEASE CHANNEL CH22\;

RELEASE CHANNEL CH23\;

RELEASE CHANNEL CH24\;

RELEASE CHANNEL CH25\;

RELEASE CHANNEL CH26\;

RELEASE CHANNEL CH27\;

RELEASE CHANNEL CH28\;

RELEASE CHANNEL CH29\;

RELEASE CHANNEL CH30\;

RELEASE CHANNEL CH31\;

RELEASE CHANNEL CH32\;

}"; RMAN_SCRIPT_TEMPLATE:"true"; name:"RMAN Script"; type:"NSR Oracle RMAN Script";

NLS_LANG:"AMERICAN_AMERICA.EE8ISO8859P2"; name:"nsrnmo"; type:"NSR Oracle nsrnmo";

ARCHIVE_LOG_ONLY_FLOW:"false"; ORACLE_DISK_BACKUP_FLOW:"false"; SNAPSHOT_BACKUP_LOGS:"false"; SNAPSHOT_BACKUP_ONLY:"true"; name:"Oracle Internal Options"; type:"NSR Oracle Internal Options";

Kind Regards,

Jozsef

2 Intern

 • 

14.3K Posts

October 11th, 2015 17:00

Try to decrease number of channels and see if that makes any difference.  Also, try to test it by running RMAN script from client.

3 Posts

October 12th, 2015 07:00

Our database expert has run incremental backup from RMAN and it was totally fine. We've tested it with 8 CHs, with 16 CHs and with 32 CHs as well without any problem.

The problem only appears if we use the NetWorker and in that case the RMAN log says that everything is okay, only the NetWorker thinks that something went wrong. We have a valid incremental backup if we ask the RMAN but the NetWorker job somehow do not want to accept it and just runs onwards...

2 Intern

 • 

14.3K Posts

October 13th, 2015 06:00

The way you use it looks like wizard make thing.  Instead, try more classical way (which does not include application information field).  Also, you use 7.6 on client which is not supported and I don't think that version was around when NMDA 1.6 was released - so you need to fix that as well to get compatibility right.

5 Practitioner

 • 

274.2K Posts

November 11th, 2015 00:00

Hi,,,

 

Can you please answer the following questions:

1. When you run RMAN full backup; what backup level you use level full or incremental level 0?

If you run level full; then you need to change it to level 0; because level full can’t be used as a base for inc backup.

2. What is the duration of full / incremental level 0 backup?

3. Is “BCT”, Block Change Tracking, enabled in your oracle DB?

If the BCT is not enabled; and for the full backup you use incremental level 0, then incremental level 1 backup duration will be almost the same as “incremental level 0” because RMAN will read the entire DB files like the full backup, but it will write less amount of data to the backup volume & you might feel like the backup is not progressing for long time.

Regards,

No Events found!

Top