db sync is happening in step 4 and can cause deployment failures

[Giulio Fidente]

Apparently the ::mysql classes we include in step 2 for the OpenStack projects are only meant to create the empty database and the owner, they do not perform the db sync which happens instead in step 4.

This causes sporadic failures when the non-master nodes try to start a service before the db sync is finished.

[Steven Hardy (shardy)]

Fix proposed https://review.openstack.org/#/c/308505/

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/308505
Committed: https://git.openstack.org/cgit/openstack/tripleo-heat-templates/commit/?id=efadbd85a444851a83e3ebed42b4dab36e9c784f
Submitter: Jenkins
Branch: master

commit efadbd85a444851a83e3ebed42b4dab36e9c784f
Author: Giulio Fidente <email address hidden>
Date: Fri Apr 22 19:19:33 2016 +0200

    Run sync in step 3 when $sync_db

    This will configure the openstack services and run the initial
    db sync in step 3 (instead of step 4) for the node for which
    $sync_db is true.

    Closes-Bug: #1572952
    Change-Id: I29012ee0a8b281e4472353ee7d9d44912e8a9b6c

[Doug Hellmann (doug-hellmann)]

This issue was fixed in the openstack/tripleo-heat-templates 5.0.0.0b1 development milestone.