We are using Couchbase 4.1.0 in the cluster with 4 nodes and in our application, NodeJS client SDK 2.0.8.
Everything works fine, but the issue take place when any node goes down (failover) in the Couchbase cluster.
Our observation is, when any node goes down in cluster, we did not able to get some documents (We believe that may be the documents that were on that down node/server).
And when we restart our application (means it will make new Couchbase communication object), then only it starts working again.
We believe that here client SDK is not getting updated map somehow and that is the core issue behind it.
Can someone please help us in this regard ?
During a failover situation, there will be a period of time where document operations may timeout, is this what you are observing?
Thanks for your reply.
We are not talking about “during” failover. We know that during failover we have timeouts.
But our problems is, we did not able to access some documents even after node failed over.
This problem will continue until we restart our application and node sdk will make complete new Couchbase communication object.
Can you plz help us in this situation?
Would you be able to try out the newest version of the Node.js SDK (2.2.4) and let me know if you are still able to reproduce the issue? There have been quite a few bug fixes in the interim versions that may help with the issues you are seeing.