yahoo-eng-team team mailing list archive

[OpenStack Infra <2039038@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.opendev.org/c/openstack/neutron/+/897959
Committed: https://opendev.org/openstack/neutron/commit/b0107bf40a4510c299a30cb9ad8e03646b3cebf9
Submitter: "Zuul (22348)"
Branch:    master

commit b0107bf40a4510c299a30cb9ad8e03646b3cebf9
Author: Rodolfo Alonso Hernandez <ralonsoh@xxxxxxxxxx>
Date:   Tue Oct 10 22:59:23 2023 +0000

    Don't respawn a process that has been stopped
    
    The ``AsyncProcess`` has a watchdog process executed by
    ``_watch_process`` method. This method detects if the process has
    been stopped and if a respawn interval is defined, the method revives
    the process. However if the ``stop`` command has been triggered, the
    process should not be respawned again.
    
    Closes-Bug: #2039038
    Change-Id: I3afad9346ef426960fd3e2a7480f94eae88f9de4


** Changed in: neutron
       Status: In Progress => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/2039038

Title:
  [fullstack] The "dhclient" kill command can take more than 90 seconds

Status in neutron:
  Fix Released

Bug description:
  The "dhcpclient" spawned in the fullstack tests have a respawn
  interval configured. Because of that, when the process is stopped, the
  ``AsyncProcess._handle_process_error`` method will restart it again,
  as seen in the fullstack test logs. For example:
  https://paste.opendev.org/show/bG1K9ke8gcRsbjZMDB70/

  The ``AsyncProcess._kill_process_and_wait`` has an active wait until
  the process is killed, slowing down the test tear down process and
  increasing the fullstack CI job time.

  This bug is related to
  https://bugs.launchpad.net/neutron/+bug/2033651.

To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/2039038/+subscriptions