RMAN full backup is getting hanged after 12 hours backup

Is is a stale session on NW side or also on Oracle side (you can check via SQL query in Oracle)?  12h session sounds awfully long - is this single session or there are couple of sessions already finished in RMAN log up to that time?

You can do several things:

- try to establish if this break at the same time (like 98% or time wise exactly after xxx seconds/minutes)

- you can try NMDA 1.5

- you can try scheduled backup on your side and not crontab

- change some parameters like filesperset or number of allocated channels just to see if this does any difference

DBAs could also try to run RMAN in debug, but given that it lasts so long, I assume they will need to allocate loads of space to make sure they capture the output.

They say nothing from there side ,  NMDA version is 1.2  on Oracle DB  and networker version is 8.0.1 .

Files part parameter   , discussed in morning only as of now it is 20 they are going to change to 10 for next week .

DBA schedule them via crontab , only this system only

Hi  Crvelin,

Thanks for update , It is not a single session , it has multiple sessions already completed , they say( DB team ) 98% completed  .

As said by DBA  they cant see any stale sessions on oracle , but on networker  I can see

Well, if DBAs say 98% is completed then there must be stale session seen on Oracle side too.  Which NMDA, Oracle version and NW do you use?  Does this happen also if you change filesperset parameter to more or less (depending on current setting)?  Do you run backup or do DBAs initiate it? Is this error specific to single system or you have seen it elsewhere?

Any parameters needs to be changed in RMAN  for fine tuning  , this were advised by EMC engineer.

PARMS='ENV=(NSR_SERVER=ABC,NSR_CLIENT=XYZ,NSR_DATA_VOLUME_POOL=ASD,NSR_MMDB_RETRY_TIME=30,NSR_FXBUSY_RETRIES=600,NSR_NO_BUSY_ERRORS=FALSE,NSR_END_ERROR_IGNORE=TRUE, NSR_DEBUG_LEVEL=3,NSR_DPRINTF=TRUE)' MAXOPENFILES 1;

Yes  earlier it used  to break exactly at same  time , after that found issue with License.

So we have increased License .

Now it got broken at 1 Hour after .

NMDA version  can change , but as of now No EMC  engineer has  raised any bugs on same , we will try this option as Last .

Scheduling this backup  is  again is completely differs , as  till last month it was running very much successful.

Files per set  is changed  to 10  now .

With Debug mode need to check on same .

Above all are there any parameters to be set in SN or Juke box like idle time out's..any thing which needs to  be done   on networker server  .

If you suspect timeout, it will be on network side most likely.  You can set something like

NSR_KEEPALIVE_WAIT=3500

export NSR_KEEPLIVE_WAIT

in /nsr/nsrrc, but most likely this is not your issue from what you have said.