Splunk design question

[naomibn]

Hello everyone,

I have a requirement where I have three servers in PROD and three in DR. UF agents are installed on all these servers.

All the logs to be monitored will be stored in a NFS. The scenario is like

UF01,UF02,UF03 (PROD) ----- >> monitoring same files in NFS
UF04,UF05,UF06 (DR) ----- >> monitoring same files in NFS

At one point of time there should be only one UF monitoring all these files. If UF01 goes down, either UF02 or UF03 should monitor the NFS.

Is there some kind of a watch mechanism available anywhere.

Thanks,
Nao

[naomibn · Accepted solution]

We tested it. We can utilize the fishbucket folder for this.

[jplumsdaine22]

UFs cannot share state with one another. If you monitor the same NFS path with three different UFs you will get triplicated results.

Where are the logs being generated? It would be better to run the UF on the host that is generating the logs, that way you don't need to deal with any NFS issues.