How to Refresh Config Metadata for "Time Monotonicity Violation"

Ops and Admin Sharding
1

After performing a restore using the PBM solution, we are encountering the following error, and the mongos process is restarting every 5 minutes.

Here is my thought process:

  • The error and restarts seem to occur because the chunk version cached in mongos is ahead of the one in the config server.
  • Even after mongos re-synchronizes with the config server’s chunk information, the same error continues.
  • Would the issue be resolved if we refresh the config server’s chunk metadata based on the current shard information?

Is there a way to update or refresh the config server’s chunk metadata based on the current state of the shards?

Also, are there any other possible workarounds or solutions?

Thank you.

below is sample of this error:

{“t”:{“$date”:“2025-07-04T08:40:57.078+09:00”},“s”:“F”, “c”:“CONTROL”, “id”:6384300, “svc”:“R”, “ctx”:“CatalogCache-2”,“msg”:“Writing fatal message”,“attr”:{“message”:“DBException::toString(): Location6493100: Time monotonicity violation: lookup time { chunkVersion: { e: ObjectId(‘683017aec543eb7413a60cf5’), t: Timestamp(1747982254, 15), v: Timestamp(1, 0) }, forcedRefreshSequenceNum: 21, epochDisambiguatingSequenceNum: 40 } which is less than the earliest expected timeInStore { chunkVersion: { e: ObjectId(‘6859017b4346eda458a63659’), t: Timestamp(1750663547, 14), v: Timestamp(1, 0) }, forcedRefreshSequenceNum: 21, epochDisambiguatingSequenceNum: 21 }.\nActual exception type: mongo::error_details::throwExceptionForStatus(mongo::Status const&)::NonspecificAssertionException\n\n”}}
{“t”:{“$date”:“2025-07-04T08:40:57.078+09:00”},“s”:“F”, “c”:“CONTROL”, “id”:6384300, “svc”:“R”, “ctx”:“CatalogCache-2”,“msg”:“Writing fatal message”,“attr”:{“message”:“\n”}}