Uploaded image for project: 'Crowd Data Center'
  1. Crowd Data Center
  2. CWD-4318

Allow Crowd startup scripts to run in the foreground

    • Icon: Suggestion Suggestion
    • Resolution: Fixed
    • 3.0.0
    • None
    • None
    • Our product teams collect and evaluate feedback from a number of different sources. To learn more about how we use customer feedback in the planning process, check out our new feature policy.

      The Crowd start-up scripts should have an option to run in the foreground.

      Workaround

      Modify the script under <crowd-install-directory>/apache-tomcat/bin/startup.sh, changing the last line from:
      exec "$PRGDIR"/"$EXECUTABLE" start "$@"
      To
      exec "$PRGDIR"/"$EXECUTABLE" run "$@"

      Once you've done this change, you can now use the ./start-crowd.sh script to start Crowd in foreground.

            [CWD-4318] Allow Crowd startup scripts to run in the foreground

            That's great! Any idea when 3.0 will be released? I still see 2.12.0 on the downloads page - https://www.atlassian.com/software/crowd/downloads/binary/atlassian-crowd-2.12.0.tar.gz

            Jason @ FPC added a comment - That's great! Any idea when 3.0 will be released? I still see 2.12.0 on the downloads page - https://www.atlassian.com/software/crowd/downloads/binary/atlassian-crowd-2.12.0.tar.gz

            Starting with 3.0, you can now do:

            ./start_crowd.sh -fg

            or

            start_crowd.bat /fg
            

            to run Crowd in the foreground.

            Lukasz Pater added a comment - Starting with 3.0, you can now do: ./start_crowd.sh -fg or start_crowd.bat /fg to run Crowd in the foreground.

            Hello Jason,

            I did some comparison on the Crowd startup.sh script with the one from Confluence and I see that to run the Confluence application in foreground, Confluence uses the catalina.sh file, specifying the run command along with $@ to say that this should run in foreground:
            ./catalina.sh run $@

            So in this case, we can do the same with Crowd. However, to change the start-crowd.sh script to run in foreground we just need to modify the script under <crowd-install-directory>/apache-tomcat/bin/startup.sh, changing the last line from:
            exec "$PRGDIR"/"$EXECUTABLE" start "$@"
            To
            exec "$PRGDIR"/"$EXECUTABLE" run "$@"

            Once you've done this change, you can now use the ./start-crowd.sh script to start Crowd in foreground.

            Best regards,

            Eduardo Mallmann
            Atlassian Support

            Eduardo Mallmann (Inactive) added a comment - Hello Jason, I did some comparison on the Crowd startup.sh script with the one from Confluence and I see that to run the Confluence application in foreground, Confluence uses the catalina.sh file, specifying the run command along with $@ to say that this should run in foreground: ./catalina.sh run $@ So in this case, we can do the same with Crowd. However, to change the start-crowd.sh script to run in foreground we just need to modify the script under <crowd-install-directory>/apache-tomcat/bin/startup.sh , changing the last line from: exec "$PRGDIR"/"$EXECUTABLE" start "$@" To exec "$PRGDIR"/"$EXECUTABLE" run "$@" Once you've done this change, you can now use the ./start-crowd.sh script to start Crowd in foreground. Best regards, Eduardo Mallmann Atlassian Support

            Hello,

            Is it possible to have an Atlassian engineer review the startup scripts for JIRA and Crowd, and advise what manual updates we should make to the Crowd script so that it can support running in the foreground (and let init on the host properly run start/stop/status)?

            Thanks,

            Jason @ FPC added a comment - Hello, Is it possible to have an Atlassian engineer review the startup scripts for JIRA and Crowd, and advise what manual updates we should make to the Crowd script so that it can support running in the foreground (and let init on the host properly run start/stop/status)? Thanks,

            joe added a comment -

            I agree that it would be good to have this consistent with the other products. As this is a change to the existing behaviour, I'm filing it as an improvement for now.

            joe added a comment - I agree that it would be good to have this consistent with the other products. As this is a change to the existing behaviour, I'm filing it as an improvement for now.

            This ticket is a spin off of https://support.atlassian.com/servicedesk/customer/portal/16/CWDSUP-10441. It is mostly incorrect, namely:

            • This is not a new feature, Crucible and Crowd have a bug
            • This is not minor, it affects all systems running Crucible/Crowd as a service - init systems like sysv and upstart, even the init scripts are unable to properly manage the service. The admin is required to use `ps` to find the Crucible/Crowd process and cannot use existing system utilities to manage the process as a service.
            • This bug is limited to Crowd and Crucible - JIRA, Stash Bamboo, Confluence all have startup scripts which support the `-fg` flag.

            Jason @ FPC added a comment - This ticket is a spin off of https://support.atlassian.com/servicedesk/customer/portal/16/CWDSUP-10441 . It is mostly incorrect, namely: This is not a new feature, Crucible and Crowd have a bug This is not minor, it affects all systems running Crucible/Crowd as a service - init systems like sysv and upstart, even the init scripts are unable to properly manage the service. The admin is required to use `ps` to find the Crucible/Crowd process and cannot use existing system utilities to manage the process as a service. This bug is limited to Crowd and Crucible - JIRA, Stash Bamboo, Confluence all have startup scripts which support the `-fg` flag.

              Unassigned Unassigned
              rrudnicki Renato Rudnicki (Inactive)
              Votes:
              4 Vote for this issue
              Watchers:
              6 Start watching this issue

                Created:
                Updated:
                Resolved: