Hello. We are trying to upgrade our Splunk UFs from 8.1.9 to 9.0.1. We use a configuration management tool (Puppet) to upgrade our UFs. This has worked fine until the leap to 9.0.1.
We are experiencing a timeout during a kvstore migration. This appears to be a known issue, and I employed the solution found here.
The solution (or workaround) appears simple: Edit server.conf and add this:
[kvstore]
disabled = true
And that works.
However, I'm concerned about the consequences of this fix. I'm not familiar with the kvstore or how it works, etc. Is there any harm in using this workaround?
Thank you in advance.
Hi
kvstore is used only on SH layers, so there is no harm to disable it on all other nodes. Actually this is the best practices!
r. Ismo