EV index backup is inconsistent

[Nate_D1]

I am having issues with my EV Indexes backup. I have NBU7 and EV9, I have an EV policy type backing up my EV monitoring db, fingerprint db and indexes. When the Index job kicks off, it spawns off 10 jobs, and I usually have 4-6 of them fail, and the rest pass. They fail with status 196 "client backup was not attempted because the backup window closed(196).  My guess is that there are too many other jobs going on during its backup window, and so its not able to grab a seat. However, the ONLY other jobs that NBU has during that timeframe (early AM) are the other 2 EV jobs.

This environment was inhereted, so I am trying to figure out how to whip it into shape a bit. Im not sure if this is a scheduling issue, a priority issue, or a problem with EV, or maybe a best practice backup issue. I have read through the EV best practices backup pdf, but didnt help me much here.

Thank You

Nate

[AndrewB]

nate, are you using the EV agent for NBU? how are you setting EV into backup mode?

[Nate_D1]

I was under the impression that the Enterprise-Vault policy type makes EV goes into backup mode automatically during the backup process (quiescence).

[AndrewB · Accepted solution]

ok so it looks like you are using the agent. i would compare the application logs on your EV server(s) and see if you can correlate the events where backup mode is being set/cleared to the events in your NBU logs for the jobs.