Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-38647

backup_restore_rolling.js can fail due to stepdown

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Major - P3
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 3.6.11, 4.0.6, 4.1.7
    • Component/s: Replication
    • Labels:
      None
    • Operating System:
      ALL
    • Backport Requested:
      v4.0, v3.6
    • Sprint:
      Repl 2019-01-14
    • Linked BF Score:
      18

      Description

      If test machine load is high enough to delay heartbeats by over 10 seconds, a replica can call for an election and cause the primary to step down. BackupRestoreTest doesn't anticipate this and can fail. In one observed case, it fails when calling listDatabases on a replica it thought was still the primary - this results in "not master" and aborts the test.

      Let's increase the electionTimeoutMillis when setting up the replica set to prevent elections.

        Attachments

          Activity

            People

            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: