Our environment is a 2 node Shard of 5 node replica set with a 3 node replica set config servers and 6 mongos. We happened to change the hostname(short to FQDN) of 1 of the config server and we restarted the config server. Once the config server came up, we saw all the 6 mongos servers restarted(with 1 second interval). So, would like to know if this is how the mongo works or is there something I am missing?

Regards, Kenny

  • 1
    Restarting config servers should not restart the mongos instances. What specific version of MongoDB server (x.y.z) are you using and how are you starting and stopping your instances? The restart behaviour sounds like something that might happen with automation tooling like Chef/Puppet/Ansible. If you check the logs on your mongos there is normally an indication of what signal the process received in order to trigger a restart. Try grepping the MongoDB logs for signalProcessingThread. – Stennie Jul 18 '19 at 10:41
  • @Stennie Thanks for checking. We are using Mongodb 3.4 version and regarding the log, the log level is set to default and we only get basic information. In the log I found "error encountered while cleaning up distributed ping for <server> caused by InterruptedAtShutdown" and then it shuts down with code 0 !! – Kenny Jul 18 '19 at 14:25
  • What specific version of 3.4.x is this? Ideally you should be using the latest minor release which is currently 3.4.21. – Stennie Jul 18 '19 at 19:50
  • The version is 3.4.10 and we have planned for a upgrade soon.. we still have the issue and we are unable to identify it !! – Kenny Jul 25 '19 at 14:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.