Intermittent 'A timeout occured after 30000ms selecting a server using CompositeServerSelector' errors when writing to GridFS

I have a 3 shard with 3 nodes in each replica set. I am getting

'A timeout occured after 30000ms selecting a server using CompositeServerSelector{ Selectors = WritableServerSelector, LatencyLimitingServerSelector{ AllowedLatencyRange = 00:00:00.0150000 } }. Client view of cluster state is { ClusterId : “2”, ConnectionMode : “Automatic”, Type : “Unknown”, State : “Disconnected”, Servers : [{ ServerId: “{ ClusterId : 2, EndPoint : “Unspecified/localhost:27015” }”, EndPoint: “Unspecified/localhost:27015”, State: “Disconnected”, Type: “Unknown” }] }

error intermittently (happens once in few days, frequency has increased to once a day at last couple of days).

Below is the stacktrace from the error in all the cases where this error happened.

   at MongoDB.Driver.Core.Clusters.Cluster.ThrowTimeoutException(IServerSelector selector, ClusterDescription description)
   at MongoDB.Driver.Core.Clusters.Cluster.WaitForDescriptionChangedHelper.HandleCompletedTask(Task completedTask)
   at MongoDB.Driver.Core.Clusters.Cluster.WaitForDescriptionChanged(IServerSelector selector, ClusterDescription description, Task descriptionChangedTask, TimeSpan timeout, CancellationToken cancellationToken)
   at MongoDB.Driver.Core.Clusters.Cluster.SelectServer(IServerSelector selector, CancellationToken cancellationToken)
   at MongoDB.Driver.GridFS.GridFSBucket`1.GetSingleServerReadWriteBinding(CancellationToken cancellationToken)
   at MongoDB.Driver.GridFS.GridFSBucket`1.OpenUploadStream(TFileId id, String filename, GridFSUploadOptions options, CancellationToken cancellationToken)
   at MongoDB.Driver.GridFS.GridFSBucket`1.UploadFromStream(TFileId id, String filename, Stream source, GridFSUploadOptions options, CancellationToken cancellationToken)
   at MongoDB.Driver.GridFS.GridFSBucket`1.UploadFromBytes(TFileId id, String filename, Byte[] source, GridFSUploadOptions options, CancellationToken cancellationToken)

I would appreciate if anyone can help me understand what could be the reason for these errors.

did you find any solution?
i am having same problem

Same problem here, since a few weeks I’m getting these errors. After initial deployment of the cluster everything worked fine

No. We added retries error as workaround.