Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-61454

Change retry policy when txns are rolled back for eviction

    • Type: Icon: Improvement Improvement
    • Resolution: Duplicate
    • Priority: Icon: Major - P3 Major - P3
    • None
    • Affects Version/s: None
    • Component/s: Stability
    • None
    • Storage Execution

      When WiredTiger is under a certain extreme level of dirty cache pressure, it will roll back a candidate longest-running transaction and output a message oldest pinned transaction ID rolled back for eviction

      Today, when this occurs, it is translated into a WriteConflictException, which retries the operation almost immediately. This can result in a situation, particularly where CPU is scarce, where the system live-locks retrying operations forever and never making progress, if the operations themselves are CPU-intensive.
      Ideally, for this specific type of rollback, the retry loop would pause a (relatively long, say multiple seconds) amount of time before retrying.

            Assignee:
            backlog-server-execution [DO NOT USE] Backlog - Storage Execution Team
            Reporter:
            milkie@mongodb.com Eric Milkie
            Votes:
            0 Vote for this issue
            Watchers:
            14 Start watching this issue

              Created:
              Updated:
              Resolved: