Error - commit is for a block we do not know about

Hi all,

We are getting the following error in the logs of our validator:

10:21AM INF commit is for a block we do not know about; set ProposalBlock=nil commit=C109A5A72B5520E99824A3E274C74EE25E0A304E8DCC0BCCF15FDF9C115CDF3B commit_round=0 height=12750921 module=consensus proposal=
10:26AM INF commit is for a block we do not know about; set ProposalBlock=nil commit=A508BDDD2D92A5D734D9A242A210BA778FF4ECA5C062D6FBAA6DCEA98217E47E commit_round=0 height=12750972 module=consensus proposal=

It appears after we see a missed block in the explorer.

Is this normal? Have you seen this in your logs? Any insights or pointers will be greatly appreciated.

Thanks!

Have you tried dropping this error to the Discord? There is a lot of cosmos devs there

Yes with no result. Thanks for suggestion though

Showed it to a devOps. They didnt say much. Just that its hard to say =)

Possibly because it missed the block, the node takes time to catch up with the set so to speak and after it resynchs it spits this out. Maybe the node is too heavy, maybe the storage is too slow, etc. Wild guesses here

Yeah, the error is too generic. Our validator is on-premises with enterprise grade hardware and nvme disks and has plenty of resources. Sentries are in AWS. Network connectivity also looks healthy. So, we are stuck…

If you manage to crack it, let us know too please

I also tryed toreduce missed blocks on my node. And setting

peer_gossip_sleep_duration to 2ms
helped (I still see some missed blocks, but less than before)

Hi, were you able to figure out how to solve this? Stuck on this issue too, I tried reducing peer_gossip_sleep_duration to 2ms but no luck yet

Hi, that error is a little bit normal.

You will frequently see that error when your node is almost synced. Is that where you are seeing it?