pool.waitall needed in dhcp-agent sync_state

[Ed Bak]

While debugging an issue in the dhcp-agent, I noticed that it was difficult to determine when sync_state completes. I would like to add the following pool.waitall and LOG message to the sync_state method.

for network in active_networks:
    pool.spawn_n(self.safe_configure_dhcp_for_network, network)
pool.waitall()
LOG.info(_('Synchronizing state complete'))

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/59863

[Nachi Ueno (nati-ueno)]

The ease of debugging is only reason for this bug report?

[Ed Bak (ed-bak2)]

And to ensure that the last pool of threads are complete before another sync_state begins.

[Jian Wen (wenjianhn)]

Ed, I think you need to add commet #3 to the commit msg of your patch.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/59863
Committed: http://github.com/openstack/neutron/commit/9f6ff7e312e6d6ddfc520bcfa8a4f40cf5806845
Submitter: Jenkins
Branch: master

commit 9f6ff7e312e6d6ddfc520bcfa8a4f40cf5806845
Author: Ed Bak <email address hidden>
Date: Tue Dec 3 23:00:23 2013 +0000

    Change to improve dhcp-agent sync_state

    Added a pool.waitall to dhcp-agent sync_state
    to ensure that the last pool of threads are complete
    before another sync_state begins. Added another
    log message to aid in debugging. Also added the
    test_sync_state_waitall unit test.

    Change-Id: I13d79a212672c4086e3082783743cb1ef2b48cfb
    Closes-Bug: #1257514

[OpenStack Infra (hudson-openstack)]

Related fix proposed to branch: master
Review: https://review.openstack.org/60541

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/60541
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=5d75bf762f02466648d961b8c5fa7a65de6e709d
Submitter: Jenkins
Branch: master

commit 5d75bf762f02466648d961b8c5fa7a65de6e709d
Author: armando-migliaccio <email address hidden>
Date: Fri Dec 6 07:31:53 2013 -0800

    Switch to using spawn to properly treat errors during sync_state

    Changes f1b9ac5a and 9f6ff7e3 were aimed at improving the robustness
    of the sync_state method. However both missed switching from spawn_n
    to spawn to properly deal with raised exceptions.

    Change-Id: Ieda4a5ecca63de67d8a9757962e7e3383fd9c9ea
    Related-bug: #1257514

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/havana
Review: https://review.openstack.org/98635

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/98635
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=3c5a11c900c3ce9b9c23aecd64808a49cd7035bd
Submitter: Jenkins
Branch: stable/havana

commit 3c5a11c900c3ce9b9c23aecd64808a49cd7035bd
Author: Ed Bak <email address hidden>
Date: Tue Dec 3 23:00:23 2013 +0000

    Change to improve dhcp-agent sync_state

    Added a pool.waitall to dhcp-agent sync_state
    to ensure that the last pool of threads are complete
    before another sync_state begins. Added another
    log message to aid in debugging. Also added the
    test_sync_state_waitall unit test.

    Change-Id: I13d79a212672c4086e3082783743cb1ef2b48cfb
    Closes-Bug: #1257514
    (cherry picked from commit 9f6ff7e312e6d6ddfc520bcfa8a4f40cf5806845)