We have a hot catalog policy which runs on 3
different schedules. A daily full (every morning at 6am), day incremental
(every hour from 7am to 6pm) and night incremental (every 3 hours from 6pm to
6am). 2 weekends ago the catalog failed with an error code 2 (Error
Validating NBDB backup in /usr/openv/db/staging). It hasn’t
happened since this incident 2 weekends ago
We were able to get the catalog to run again
successfully by restarting NBU after a reboot of our master server but our
question is why does the catalog suddenly fail in the first place and after
this failure, why does a simple re-boot get it all working again ?
We had similar issues soon after upgrading to 6.5 where we were advised to
delete our existing catalog policy, restart NBU and then create a new one. In
that case automated catalogs would not run but we could manually kick them
off without any problems.
1st and 2nd level support have
been unable to work out the issue based on logs we have sent them. They
are trying to find out where NBU logs information on this validation so they
can increase the logging level to try and see why the validation is sporadically
failing. They have had me run a validation command against the database
which came back without error.
Anyone seen anything similar while I wait for
backline to talk to the development team ?
|