Journal entries can get stuck forever causing busy wait

[Mike Kolesnik]

As the journal handling is now written the journal thread changes the oldest journal row status to processing and gets to process it.
If the thread or process for some reason dies, say simple kill -9, this row get stuck in 'processing' state forever.

If a later journal row is created for the same resource, or a resource dependant on the resource in the first row, the journal thread will refrain from handling it each time and put it back in the queue. This will lead to a busy wait since the journal thread is running in a "while True" construct.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/304168

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/304168
Committed: https://git.openstack.org/cgit/openstack/networking-odl/commit/?id=153bc13cb761ce370de850252a011505be09fea1
Submitter: Jenkins
Branch: master

commit 153bc13cb761ce370de850252a011505be09fea1
Author: Arvind Somya <email address hidden>
Date: Fri Apr 8 19:00:35 2016 -0400

    Journal entries can get stuck forever causing busy wait

    This patches fixes the bug caused by a dead neutron process that leaves
    a journal row in processing state. A new cleanup process runs at set
    maintenance intervals that attempts to reset these rows back to the
    pending state.

    Change-Id: Ib1ec21ff27491e077f5bfe94e7b06391a505c39d
    Closes-bug: #1563297

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/networking-odl 3.0.0 release.