[WT] Add block_truncate_max to bound the page-cache invalidation stall caused by a single large ftruncate()

ExportXMLWordPrintableJSON

    • Type: Task
    • Resolution: Unresolved
    • Priority: Major - P3
    • None
    • Affects Version/s: None
    • Component/s: None
    • None
    • Storage Engines
    • 18.467
    • None
    • None

      "a checkpoint returning a large amount of space with one ftruncate() call" blocks every request that needs to read from the same file, and showing that rate-limiting a single truncate with block_truncate_max reduces the stall to a negligible value.

        1. image-2026-09-22-15-25-26-840.png
          image-2026-09-22-15-25-26-840.png
          220 kB
        2. image-2026-09-22-15-29-10-337.png
          image-2026-09-22-15-29-10-337.png
          336 kB
        3. image-2026-09-22-15-30-23-898.png
          image-2026-09-22-15-30-23-898.png
          300 kB
        4. image-2026-09-22-15-38-20-667.png
          image-2026-09-22-15-38-20-667.png
          204 kB

            Assignee:
            Unassigned
            Reporter:
            y yz (EXT)
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated: