Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-85871

Don't signal oplog waiters under ReplicationCoordinator mutex

    XMLWordPrintableJSON

Details

    • Icon: Improvement Improvement
    • Resolution: Unresolved
    • Icon: Major - P3 Major - P3
    • None
    • None
    • None
    • None
    • Replication

    Description

      Something I noticed while analyzing work done under the ReplicationCoordinator mutex is that we signal all tailable oplog waiters while advancing the committed snapshot, which seems unnecessary and costly to do while holding the mutex. The CappedInsertNotifier already has its own mutex.

      Attachments

        Activity

          People

            backlog-server-repl Backlog - Replication Team
            louis.williams@mongodb.com Louis Williams
            Votes:
            0 Vote for this issue
            Watchers:
            7 Start watching this issue

            Dates

              Created:
              Updated: