vmware-nsx/neutron/scheduler
armando-migliaccio cf69c5b4c3 Fix session's InvalidRequestError because of nested rollback
This patch addresses the issue by removing the extra nesting
(which is effectively redundant).

The longer story about this patch is the following:

Race conditions may cause DBDuplicateEntry exceptions that
require a transaction to be rollbacked back, and yet make
the whole operation succeed. A classic example is what has
been solved in commit fbc6b99. If the rollback is done in a
nested transaction, the above mentioned exception is raised.

To address the problem, we could use savepoints by means of
sqlalchemy's begin_nested(); Even though this approach is
preferable, it causes quite a bit of changes in the unit
tests (because of sqlite); it may also require that certain
DBMS, or certain DB backend configurations, support savepoints.

In the end, the simpler approach was chosen.

Closes-bug: #1354072

Change-Id: Ic9393319e55f71d681124bd3052e939724ebab6b
2014-08-08 16:01:49 -07:00
..
__init__.py Remove the useless vim modelines 2014-06-21 15:07:31 +08:00
dhcp_agent_scheduler.py Switch to using of oslo.db 2014-06-30 14:07:59 +03:00
l3_agent_scheduler.py Fix session's InvalidRequestError because of nested rollback 2014-08-08 16:01:49 -07:00