Uploaded image for project: 'WiredTiger'
  1. WiredTiger
  2. WT-10715

Automatically try previous checkpoints when opening a checkpoint cursor

    • Type: Icon: Task Task
    • Resolution: Unresolved
    • Priority: Icon: Major - P3 Major - P3
    • None
    • Affects Version/s: None
    • Component/s: Checkpoints
    • Labels:
    • 13
    • StorEng - Defined Pipeline

      After WT-10545, we are now returning an error message when the checkpoint cursor hangs due to mismatching information.

      This ticket suggests an alternative solution: if a previous checkpoint exists, try to open the checkpoint cursor on it. Try this until the list of checkpoints is exhausted. If none succeeds, an error can be returned.

      There are specific scenarios to consider:

      • When a named checkpoint is given to the checkpoint cursor, it should succeed.
      • When an unnamed checkpoint is given:
        • If it is an internal call, this should succeed.
        • If it is an external call, the name should be WiredTigerCheckpoint (see WT-10637). This means the latest checkpoint has been requested. If this one cannot be opened for any reason, try a previous one.

            Assignee:
            backlog-server-storage-engines [DO NOT USE] Backlog - Storage Engines Team
            Reporter:
            etienne.petrel@mongodb.com Etienne Petrel
            Alexander Gorrod, Vamsi Boyapati
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated: