I’ve been using the Couchbase Java SDK and am frequently encountering errors such as this:

Error Type: net.spy.memcached.OperationTimeoutException : [N/A]
Error Messages: Timeout waiting for value: waited 2,500 ms. Node status: Connection Status { 10.xx.xx.xx/10.xx.xx.xx:11210 active: false, authed: true, last read: 391,748 ms ago }

My application code is connecting using the external URL of the CB server, however the error is reporting the internal IP of the CB node.

There does not seem to be any pattern as to why/when these errors occur.

Does anyone have any insight as to how to resolve?


timeouts can mean all and everything. You should try to correlate it to something else that’s happening in your environment to get you closer to pinpointing it:

  • GC behavior
  • Network traffic and latency peaks
  • Unwanted server utilization

Once you’ve ruled out all those three, you should go in and do more profiling to see what’s actually holding them up occasinally. If you see a period of timeouts when you are doing rebalance it could be that your cluster is undersized.

Let me know if you have more infos on your env to help troubleshoot it down.