Skip to content
Snippets Groups Projects
  1. May 06, 2022
  2. May 05, 2022
  3. Apr 29, 2022
  4. Apr 28, 2022
  5. Apr 22, 2022
  6. Apr 19, 2022
  7. Apr 11, 2022
  8. Apr 01, 2022
  9. Mar 28, 2022
  10. Mar 24, 2022
  11. Mar 17, 2022
  12. Mar 15, 2022
  13. Mar 14, 2022
  14. Mar 10, 2022
  15. Mar 04, 2022
  16. Feb 24, 2022
  17. Feb 10, 2022
  18. Feb 04, 2022
  19. Jan 26, 2022
  20. Jan 20, 2022
  21. Jan 17, 2022
  22. Jan 11, 2022
  23. Jan 07, 2022
  24. Jan 06, 2022
  25. Dec 13, 2021
    • mvelosob's avatar
      Avoid large buffer reservations with RAO (#1054) · 5e602c9f
      mvelosob authored
      Refactor the RecallTaskInjector to limit the number of tasks passed
      at once to the TapeReadSingleThread for a tape with RAO and to reserve
      disk space in smaller batchs
      
      In every call of RecallTaskInjector::synchronousFetch the injector
      will try to pop jobs from the queue so that it holds an ammount of
      work equal to the limit of files or bytes imposed by the RAO
      implementation or by the value given by the
      BulkRequestRecallMaxBytes and BulkRequestRecallMaxFiles config
      options in /etc/cta/cta-taped.conf.
      
      In every call of RecallTaskInjector::injectBulkRecalls the
      RecallTaskInjector will inject a set of tasks limited by the
      BulkRequestRecallMaxBytes and BulkRequestRecallMaxFiles config
      options in /etc/cta/cta-taped.conf to the TapeReadSingleThread
      and DiskWriteThreadPool.
      
      The disk space reservation is done once for every job batch (instead of
      all the disk space being reserved when the jobs are popped)
      
      This prevents tapeservers with RAO from reserving a large amount
      of the disk buffer upfront, which would cause the buffer to fill quickly
      by a few drives, which cannot fill it fast enough
      5e602c9f
  26. Dec 02, 2021
  27. Nov 30, 2021
    • mvelosob's avatar
      Remove support for activity fair share scheduling (#1077) · 09adf3cc
      mvelosob authored
      Remove all the code associated to the activity weight fair share
      scheduling feature:
      
      - Remove catalogue functions interacting with the ACTIVITY_WEIGHTS table
      - Remove code related to activity weights from the scheduler.
      - Deprecate and mark as optional objectstore schema fields for activity
        weights (shcema should be backwards and forwards compatible).
      - Store only activity information in the drive state. In the catalogue
        methods, default the weight values to 1.0 (avoids having to perform a
      backwards incompatible catalogue upgrade)
      09adf3cc
  28. Nov 24, 2021
  29. Nov 18, 2021
  30. Nov 17, 2021
  31. Nov 15, 2021
  32. Nov 12, 2021
  33. Nov 11, 2021
Loading