cancel
Showing results for 
Search instead for 
Did you mean: 

Error code 25s logged - very strange behaviour

MartinR1
Level 3

We have a very starange issue that happens in our environment. We use version 7.7.3 writing to MSDP dedup poools.

Every day say day 1 at 13:00 we will get a cluster of error code 25s on several clients and several dedup pools it can be any job - we would get say between 20 and 50 code 25 errors. These code 25 errors only ever happen on Application backups - SQL, Rman Oracle and SAP. UNIX OS Windows OS and VMware backups are never ever affected.

On day 2 the same will happen but this time at 13:30, day 3 at 14:01, day 4 at 14:30 etc, always the next day but +30 minutes (give or take a few minutes) we have engaged support and are working through things but so far have drawn a blank. The clients are all on 7.7.3.

If anyone has any ideas on this I would be most grateful as we are not making much headway at all - if the jobs are re run they will all complete. But every day we are having to deal with many auto tickets due to these database application type backup failures.

Many thanks

 

3 REPLIES 3

NBU35
Level 6

Hi 

Try one more thing when you face this issue. initiate a manual backup of any standard or windows policy. check how long it takes to appear in activity monitor. if takes longer than usual, then we have same issue.

We are on same version of NBU and it happens in my infra as well. Happens because at that time nbpem does not respond in timely manner. restarting nbpem then fix the issue.

Majorly it occurs when you have some issue in your dns and nbpem takes long to resolve client names.

 

Pleas let me know if you execute a backup manually using policy from GUI during that period, how long that job takes to appear in activity monitor ? If takes longer than usual  then high possibility that we have same issue. It is issue with response time of NBPEM. restarting NBPEM fix issue for that time. It generally happens when something is wrong at your DNS. Just a piece of advice, analyze nbpem logs for the period of issue. it might point out to longer response time.

We are at same version of NBU, we face same issue multiple times, it is a curse because it cause failure of MY BCV backups.

 

Thank you for this although we use host files for resolution and not DNS - but you have given us an avenue for further investigation.

Did you escalate this with Veritas support? What did they say?

Thank you for your response.