Comment 2 for bug 2064368

Reviewed: https://review.opendev.org/c/starlingx/distcloud/+/917825
Committed: https://opendev.org/starlingx/distcloud/commit/7ce8b728696b7d336f646d52de99811c9d93d416
Submitter: "Zuul (22348)"
Branch: master

commit 7ce8b728696b7d336f646d52de99811c9d93d416
Author: Li Zhu <email address hidden>
Date: Tue Apr 30 22:43:30 2024 -0400

    Fix up confirm_stop functions of DC OCF scripts

    The DC OCF scripts were not updated over the switch to Debian
    in StarlingX 8.0. As a result, it could lead to orphan processes
    over the service restart or controller swact. The orphan processes
    consume resources and perform duplicate/obsolete tasks (e.g.
    auditing the same subclouds as the corresponding worker processes)
    until their work queues are empty.

    This commit fixes up the pgrep option to restore the functionality
    of the confirm_stop function of the OCF script. Processes that
    fail to be terminated will get killed.

    Test Plan:
      - Deploy a small DC system. Verify that all DC services can
        be started, stopped and restarted by SM.
      - Deploy a large DC system with many subclouds. Reduce the
        thread_pool_size of dcmanager-audit-worker. Let the system
        soak for a couple of hours. Restart the service in the
        middle of the audit cycle. Verify that dcmanager-audit-worker
        sevice was successfully restarted and there are no orphan
        processes.

    Closes-Bug: 2064368
    Change-Id: Ie5cbc89cde374e32d4e0a3799a9f8833c071d206
    Signed-off-by: Tee Ngo <email address hidden>