[OVSDB-100] Flows failed to get programmed in a very random fashion Created: 19/Sep/14  Updated: 19/Oct/17  Resolved: 12/Jan/16

Status: Resolved
Project: ovsdb
Component/s: openstack.net-virt-providers
Affects Version/s: unspecified
Fix Version/s: None

Type: Bug
Reporter: Madhu Venugopal Assignee: Unassigned
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Operating System: All
Platform: All


External issue ID: 1972

 Description   

There are a few cases, identified such as lack of exception handling in event processing and others which might cause random failures in flow programming.



 Comments   
Comment by Dave Tucker [ 23/Sep/14 ]

https://git.opendaylight.org/gerrit/11353

Comment by Flavio Fernandes [ 25/Oct/14 ]

A temporary work around to this bug was put in place by
adding a Thread.sleep(500):

https://github.com/opendaylight/ovsdb/commit/e2506003510932187bbac87e0d5b6bbd3c4d57a2

try

{ commitFuture.get(); // TODO: Make it async (See CONTROLLER-624) logger.debug("Transaction success for write of Flow "+flowBuilder.getFlowName()); Thread.sleep(500); <====== here }

catch (Exception e)

{ logger.error(e.getMessage(), e); modification.cancel(); }

while that helped, a better fix should be implemented, so
flow programming can scale.

Generated at Wed Feb 07 20:35:30 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.