Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-36045

unexpected "re-evaluating sync source"

XMLWordPrintableJSON

    • Type: Icon: Bug Bug
    • Resolution: Done
    • Priority: Icon: Major - P3 Major - P3
    • None
    • Affects Version/s: 3.2.10
    • Component/s: Replication
    • Labels:
      None
    • Server Triage
    • ALL
      1. Create a MongoDB replicaset
      2. Do some write requests to the replicaset
      3. Wait a long time ( bigger than 30 seconds )
      4. Do some write requests again, and the secondary nodes may re-evaluate sync source 

      I got this warning in the log file:

      "Fri Jun 29 22:00:02.585 I REPL [ReplicationExecutor] re-evaluating sync source because our current sync source's most recent OpTime is (term: 2, timestamp: Jun 29 21:50:07:1) which is more than 30s behind member xxx.xxx.xxx.xxx:7005 whose most recent OpTime is (term: 2, timestamp: Jun 29 22:00:01:1)"

      The write majority request is hung up and timeout when the two secondaries are both doing "re-evaluating sync source"

      The current sync source is primary node, so I think its "OpTime" should not behind the secondary nodes.

       

            Assignee:
            backlog-server-triage [HELP ONLY] Backlog - Triage Team
            Reporter:
            zhenyipeng zhenyipeng
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

              Created:
              Updated:
              Resolved: