Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-21937

Calculate write concern and distlock timeouts for talking to the config servers relative to the electionTimeout for the config replica set

    XMLWordPrintableJSON

Details

    • Icon: Bug Bug
    • Resolution: Won't Fix
    • Icon: Major - P3 Major - P3
    • None
    • None
    • Replication, Sharding
    • None
    • Sharding
    • ALL

    Description

      As part of SERVER-21050 we increased the distlock and write concern timeout for config server operations so that they were actually long enough to allow proper retries in the event of config server failure with the default election timeout of 10 seconds. For people on fast reliable networks who turn down their election timeout, however, these values may be too long and lead to slower error detection and retry logic execution. We should calculate these timeouts in term of the election timeout, or at least allow these timeouts to be independently controlled so that people can adjust them down at the same time that they adjust down their election timeout.

      Attachments

        Activity

          People

            backlog-server-sharding [DO NOT USE] Backlog - Sharding Team
            spencer@mongodb.com Spencer Brody (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: