Details

    • Type: Bug
    • Status: Resolved
    • Priority: High
    • Resolution: Cannot Reproduce
    • Affects Version/s: Marathon 1.4.6
    • Fix Version/s: None
    • Component/s: Health & Readiness
    • Labels:

      Description

      Issue
      Command health checks do not seem to be working properly. The first health check when a deployment of the new app is running, it executes the command and the instances are healthy. However, when observing the health check debug info (first success, last success, last failure, etc.), they do not appear to be executed after that first time.

      Reproduce
      1. Add a new app to Marathon (with multiple processes, watching the process the docker container livelyhood does not rely on) with a health check like following:

      {
       "gracePeriodSeconds": 120,
       "intervalSeconds": 10,
       "timeoutSeconds": 5,
       "maxConsecutiveFailures": 2,
       "delaySeconds": 15,
       "command": { 
         "value": "ps ax | grep -v grep | grep $process | wc -l" 
       },
       "protocol": "COMMAND"
       }

      2. Wait for the instance to be marked as healthy
      3. Observe the health check debug info
      4. To be sure it's not being executed, kill the process you're monitoring

      General info
      *Marathon version:* 1.4.6
      *Framework Id:* 6af3e03a-9431-4d81-af17-03de6267670a-0002

      This isssue has been created automatically from Marathon GitHub Issue 5488.

        Attachments

          Activity

            People

            • Assignee:
              Unassigned
              Reporter:
              marathon-bot Marathon Bot
              Team:
              Orchestration Team
              Watchers:
              Marathon Bot, Yordi Pauptit
            • Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: