stop_nova() leaves running nova processes

[Adam Gandelman]

As part of deployment, lib/ironic needs to stop + start nova services. It uses the stop_nova() and start_nova() functions defined in lib/nova. It seems stop_nova() does not reliably stop all nova services. The subsequent nova_start() call leaves "address already in use" errors in logs of services that failed to stop. This is being picked up by Tempest jobs log checkers as an non-whitelisted error:

2014-04-07 22:38:49.193 |
2014-04-07 22:38:49.193 | Log File Has Errors: n-obj
2014-04-07 22:38:49.194 | *** Not Whitelisted *** 2014-04-07 22:30:24.783 15020 ERROR nova.wsgi [-] Could not bind to 0.0.0.0:3333
2014-04-07 22:38:49.195 | *** Not Whitelisted *** 2014-04-07 22:30:24.789 15020 CRITICAL nova [-] error: [Errno 98] Address already in use
2014-04-07 22:38:49.195 | Logs have errors

Looks like pid files are not accurate:

ubuntu@precise:~/devstack$ ps aux | grep nova-obj
ubuntu 31468 2.0 0.4 124128 38440 pts/25 S+ 17:57 0:00 /usr/bin/python /usr/local/bin/nova-objectstore --config-file /etc/nova/nova.conf
ubuntu@precise:~/devstack$ cat /opt/stack/status/stack/n-obj.pid
31467

[Sean Dague (sdague)]

This devstack bug was last updated over 180 days ago, as devstack
is a fast moving project and we'd like to get the tracker down to
currently actionable bugs, this is getting marked as Invalid. If the
issue still exists, please feel free to reopen it.