Yash Bhutwala

I think a count meets the minimum requirements for an operator like us to know when not to do maintenance on a VC and that's suitable for now, but obviously,...

@rolfyone any update on this?

Unfortunately, no, we had to cleanup the pvcs and restart from checkpoint so we didn't continue to miss attestations and lose rewards.

See another set of errors in another node here: ``` time="2022-08-30 15:09:28" level=info msg="Processing block batch of size 199 starting from 0x8042c314... 4586753/4587345 - estimated time remaining 59s" blocksPerSecond=9.9 peers=46...

Maybe this is an issue when restoring from checkpoint sync?

No, this error does not occur right away when starting a node with checkpoint sync. I shared our node's flags before, but here it is again. We are using infura...

Ok noted, but in this case, the first error I sent you here was before any restarts or shutdowns: https://github.com/prysmaticlabs/prysm/issues/11327#issuecomment-1231810731

@prestonvanloon I emailed you the beaconchain.db specifically from: https://github.com/prysmaticlabs/prysm/issues/11327#issuecomment-1231810731

thanks @potuz, for this error, @prestonvanloon mentioned that it may be related to an improper shutdown, i.e.: https://github.com/prysmaticlabs/prysm/issues/11327#issuecomment-1231843716 and also said: "Later this week, I'll try some experimentation with SIGKILL...

Btw I just checked and we use a terminationGracePeriodSeconds of 180 seconds (this is what was set when I posted that error; more than your recommendation of 120 seconds)