An error occurred fetching the project authors.
  1. 24 Aug, 2021 2 commits
  2. 19 Aug, 2021 1 commit
  3. 12 Aug, 2021 1 commit
    • Alex Kalderimis's avatar
      Add retry mechanism for proxied Jira installations · e5d90d20
      Alex Kalderimis authored
      This adds a queue for proxied JiraConnect installation event requests.
      Failures cause new attempts to be enqueued, up to three times. This
      ensures that we can retry, without tying up a worker thread, and
      allowing the attempts to be spread out over time, since if a remote
      server is experiencing downtime, we would anticipate immediate retries
      would be less likely to succeed.
      e5d90d20
  4. 10 Aug, 2021 1 commit
    • Simon Tomlinson's avatar
      Prune database partitions older than a retention period · c8b84ced
      Simon Tomlinson authored
      When a monthly-partitioned model is specified with
      `partition_by :some_column, retain_for: 2.months`
      this feature will detach partitions that are more than two months in the
      past. A week later, a second worker will drop the tables that were
      previously detached.
      
      Changelog: added
      c8b84ced
  5. 29 Jul, 2021 1 commit
  6. 28 Jul, 2021 1 commit
  7. 27 Jul, 2021 1 commit
  8. 26 Jul, 2021 1 commit
  9. 23 Jul, 2021 1 commit
  10. 21 Jul, 2021 1 commit
    • Sean McGivern's avatar
      Tag workers that need their own queue · 044051a2
      Sean McGivern authored
      Some workers can't be re-routed to a queue shared with other workers
      because they - or another worker - depends on specific information about
      that queue. For now, we'll tag those so we know which workers they are.
      044051a2
  11. 15 Jul, 2021 1 commit
  12. 13 Jul, 2021 3 commits
  13. 09 Jul, 2021 1 commit
  14. 08 Jul, 2021 2 commits
  15. 06 Jul, 2021 1 commit
    • Shinya Maeda's avatar
      Define deduplication strategy in Resource Group Worker · d26139b3
      Shinya Maeda authored
      Currently, the AssignResourceFromResourceGroupWorker
      doesn't have any deduplication strategies, so that
      multiple jobs with the same argument could be scheduled
      simulteniously, which is suspected as a cause of race condition.
      This commit adds `until_executed` deduplication strategy
      to prevent multiple jobs from running on the same
      resource group.
      
      Changelog: fixed
      d26139b3
  16. 02 Jul, 2021 1 commit
  17. 01 Jul, 2021 1 commit
  18. 30 Jun, 2021 1 commit
  19. 22 Jun, 2021 1 commit
  20. 16 Jun, 2021 1 commit
  21. 15 Jun, 2021 1 commit
  22. 14 Jun, 2021 1 commit
  23. 10 Jun, 2021 1 commit
  24. 09 Jun, 2021 3 commits
  25. 08 Jun, 2021 1 commit
  26. 04 Jun, 2021 1 commit
  27. 27 May, 2021 3 commits
  28. 26 May, 2021 1 commit
  29. 25 May, 2021 2 commits
  30. 24 May, 2021 1 commit
  31. 21 May, 2021 1 commit