Failed to synchronize config to other nodes

Couchbase Server Kubernetes
1

Hello! I’m currently in the progress of deploying couchbase community edition on, and testing failure handing currently. I’ve set up a system where nodes will failover themselves when the old pod is killed/restarted (tried with recovery, but the node kept being unreachable). It works, if just one pod dies at a time. But if i kill 2 pods, i’ll get this error message: Failed to synchronize config to other nodes

Curl request to failover server is
curl -s -v -X POST -u [username]:[password] [address]:8091/controller/failOver -d otpNode=ns_1@[ip] -d allowUnsafe=true

Any reason why this would happen?

2

Hi @Magnus, Can you provide following information?

  • Couchbase Server version
  • Kubernetes version
  • number of replicas
  • number of pods in the cluster

Also check out Couchbase Autonomous Operator since its full-fledge product we’ve built for running Couchbase Server on Kubernetes.

Thanks!

3

Hi anil, I ended up figuring it out (turns out the problem was when i had multiple nodes, i had to fail them all over, instead of one at a time).

I have checked out the couchbase autonomous operator, but I can’t really justify getting the enterprise edition of CB, yet.