Workflow executed more than once when using cron-trigger with multiple engines

[Moshe Elisha]

Scenario:

1. Start multiple Mistral engines
2. Create a cron-trigger that executes a workflow.
3. The workflow might be executed multiple times for each execution time.

Logs are attached.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/243234

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/243234
Committed: https://git.openstack.org/cgit/openstack/mistral/commit/?id=7f9c0c62419888201e129282ba9d548d5fffdff5
Submitter: Jenkins
Branch: master

commit 7f9c0c62419888201e129282ba9d548d5fffdff5
Author: Moshe Elisha <email address hidden>
Date: Mon Nov 9 17:20:35 2015 +0000

    Ensure only one WF execution for every CT cycle

    Make sure that only one engine starts a workflow for every cron
    trigger cycle by checking the row count of the DML statements used
    to update the cron trigger in the DB:
    * Modify the update cron trigger to include a WHERE statement that
      verifies that no other engine already updated the cron trigger.
    * Change the sqlachemy method used for deleting the cron trigger to be
      able to retrieve the row count of the deleted rows

    Start the workflow execution only if the current engine was able to
    update the cron trigger DB row

    Change-Id: I9d3bec51844de218a46cf6ff53bedf9a00069e30
    Closes-Bug: #1513548

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/liberty
Review: https://review.openstack.org/250206

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/250206
Committed: https://git.openstack.org/cgit/openstack/mistral/commit/?id=b2aa9810ff9f8b3f160b4fc2e66c33cb103160a3
Submitter: Jenkins
Branch: stable/liberty

commit b2aa9810ff9f8b3f160b4fc2e66c33cb103160a3
Author: Moshe Elisha <email address hidden>
Date: Mon Nov 9 17:20:35 2015 +0000

    Ensure only one WF execution for every CT cycle

    Make sure that only one engine starts a workflow for every cron
    trigger cycle by checking the row count of the DML statements used
    to update the cron trigger in the DB:
    * Modify the update cron trigger to include a WHERE statement that
      verifies that no other engine already updated the cron trigger.
    * Change the sqlachemy method used for deleting the cron trigger to be
      able to retrieve the row count of the deleted rows

    Start the workflow execution only if the current engine was able to
    update the cron trigger DB row

    Change-Id: I9d3bec51844de218a46cf6ff53bedf9a00069e30
    Closes-Bug: #1513548
    (cherry picked from commit 7f9c0c62419888201e129282ba9d548d5fffdff5)

[Thierry Carrez (ttx)]

This issue was fixed in the openstack/mistral 2.0.0.0b1 development milestone.

[Doug Hellmann (doug-hellmann)]

This issue was fixed in the openstack/mistral 1.0.1 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/mistral 1.0.1 release.