Consistency check failed with 2 inconsistent replicas

I lost part of the nodes when upgrading the cluster, and I don’t know how much. Now the cluster will automatically stop frequently because of consistency issues. I have five nodes in my cluster, and I randomly close one node in the previous method to test if it has a problem. CockroachDB stops in all four nodes.

Hi @chaihuo

Sorry for missing this earlier, somehow it was miscategorized and not visible on the main forum.

It’d be helpful to know what your replication factor is, and the results of the problem ranges report. You can find the report here: http://<adminurl>/#/reports/problemranges. Finally, what do you see in the logs during the period when the cluster is not available?

Hey @chaihuo,

Have you run into any consistency check issues after your initial report of it? I know this is an older issue, but just following up on something we’ve seen where we see a similar error.

Let us know!