Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-91511

Slow stop causes processor to get stuck in "stopping" state

    • Type: Icon: Bug Bug
    • Resolution: Fixed
    • Priority: Icon: Blocker - P1 Blocker - P1
    • 8.1.0-rc0
    • Affects Version/s: None
    • Component/s: None
    • None
    • Atlas Streams
    • Fully Compatible
    • ALL
    • Sprint 51, Sprint 52

      A user initiated a stop at 6/13/24 10:05:28.992 PM.

      The stop request from Agent<->mongostream failed due to timeout at 6/13/24 10:08:29.013 PM. The user gets a timeout error back.

      The executor did not finish shutting down until 6/13/24 10:09:28.930 PM.

      This ultimately leaves the processing in a "stopping" state on mongostream, and subsequent stop requests from the customer fail with "Stream processor is already being stopped".

      We should investigate why the stop took so long in a separate ticket. In this ticket we should fix the stop flow to fully stop and remove the processor from mongostream, even if the stop request fails due to timeout.

      https://splunk.corp.mongodb.com/en-US/app/cloud/search?earliest=-4h%40m&latest=now&q=search%20index%3Dmhouse%20666b6c777cdf4ab6c88ede9c%20(c%3DSTREAMS%20OR%20c%3D*)%20host%3Dstreams-spp-sp10-66c6cf9657-zgvk5&display.page.search.mode=smart&dispatch.sample_ratio=1&display.page.search.tab=events&display.general.type=events&sid=1718320488.1909710

            Assignee:
            sandeep.dhoot@mongodb.com Sandeep Dhoot
            Reporter:
            matthew.normyle@mongodb.com Matthew Normyle
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved: