yahoo-eng-team team mailing list archive

[OpenStack Infra <1520567@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.openstack.org/250790
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=4dbf1decf91f208b15805983fb4d6de5ccafe74a
Submitter: Jenkins
Branch:    master

commit 4dbf1decf91f208b15805983fb4d6de5ccafe74a
Author: Pavel Kholkin <pkholkin@xxxxxxxxxxxx>
Date:   Fri Nov 27 15:34:34 2015 +0300

    bug and tests in 'instance_info_cache'
    
    Added unit tests for 'instance_info_cache' db section,
    fixed incorrect behaviour with ignoring exception that relates
    to introducing oslo.db enginefacade.
    
    Closes-Bug: #1520563
    Closes-Bug: #1520567
    
    Co-Authored-By: Sergey Nikitin <snikitin@xxxxxxxxxxxx>
    
    Change-Id: I35ec688d7b176e0cb0b9ce6bd0c241b6631422c5


** Changed in: nova
       Status: In Progress => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1520567

Title:
  Incorrect pass exception with oslo_db.enginefacade

Status in OpenStack Compute (nova):
  Fix Released

Bug description:
  After ignore and pass exception the following exception could happen:

  sqlalchemy.exc.InvalidRequestError: This Session's transaction has
  been rolled back due to a previous exception during flush. To begin a
  new transaction with this Session, first issue Session.rollback().
  Original exception was: (sqlite3.IntegrityError) UNIQUE constraint
  failed: instance_info_caches.instance_uuid

  We are in a transaction, we can't have an integrity failure in a
  transaction, and then just continue working, we need to use a
  savepoint.

  The problem could be reproduced e.g. in instance_info_cache_update but
  also could happen in other similar places where we pass an db
  exception.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1520567/+subscriptions