Uploaded image for project: 'Nuxeo Platform'
  1. Nuxeo Platform
  2. NXP-30182

Increase Bulk Status TTL on completion with error

    Details

    • Type: Bug
    • Status: Resolved
    • Priority: Minor
    • Resolution: Fixed
    • Affects Version/s: 10.10, 11.4
    • Fix Version/s: 10.10-HF43, 11.5, 2021.2
    • Component/s: Bulk, Streams
    • Release Notes Summary:
      The bulk status is kept for 24h in case of error.
    • Team:
      PLATFORM
    • Sprint:
      nxplatform #29
    • Story Points:
      2

      Description

      Once a Bulk Command is completed, its definition and its state which is stored in the KV store have a TTL of 1h applied.

      Now suppose the following scenario, a bulk command is started, the scroller is running and reaches the threshold of 100k document ids where it sends records downstream for processing, the scroller continues and encounters an error like a cursor timeout, the command is then put in completed in error with a TTL of 1h.

      The downstream records are processed, if it takes more than 1h after the error the state and the command will not be available and raises failure in downstream computations.

      For this reason and also because in case of a problem it is better to keep information longer for debugging purposes the TTL in case of error should be increased to 12h.

        Attachments

          Activity

            People

            • Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                PagerDuty

                Error rendering 'com.pagerduty.jira-server-plugin:PagerDuty'. Please contact your Jira administrators.