Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-43363

mongod --bind_ip_all zombie process and failing service start

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Critical - P2
    • Resolution: Done
    • Affects Version/s: 4.2.0
    • Fix Version/s: None
    • Component/s: Admin, Stability, Usability
    • Labels:
      None
    • Operating System:
      ALL
    • Steps To Reproduce:
      Hide

      No steps, seems to have happened randomly.

      Show
      No steps, seems to have happened randomly.

      Description

      I've run into this issue today on Ubuntu 18.04 using 4.2.0 mongod CE:

      1. `sudo service mongod start` fails
      2. looking into the log `/var/log/mongodb/mongod.log` shows that it failed to start because another process is listening (Socket in use) error code 48
      3. doing a `ps wuax | grep mongo` I can see that a process of mongod is already running as `mongod --bind_ip_all` user 999
      4. killing it with `kill -9 $PID` doesn't do anything, the process respawns. using pkill or htop to send different kill commands also fail, the process stays there/respawns with new pid.
      5. terminating it with `sudo service mongod stop` doesn't do anything, it still stays running
      6. connecting to it via mongo cli simply hangs-up at the connecting part.

       

      This happened without any abnormal termination or closure of the platform.

      Short of uninstalling and re-installing, I have no other way to use mongod at this point?

        Attachments

          Activity

            People

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

              Dates

              • Created:
                Updated:
                Resolved: