Reviewed: https://review.openstack.org/367180 Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=12420c1585abadc741440dcd48a5540ef85357db Submitter: Jenkins Branch: master
commit 12420c1585abadc741440dcd48a5540ef85357db Author: Kevin Benton <[email protected]> Date: Wed Sep 7 18:51:58 2016 -0700 Mark quota operations as retriable This decorates the quota system operations with the retry decorators. This will help significantly with the bug this marks as closed since operations in the quota engine after commit should no longer trigger retries of the full API operation. The logic to find the args in the decorator had to be adjusted to deal with functions already decorated. This just uses the getargspec function from pecan that deals with decorated functions. Partial-Bug: #1612798 Closes-Bug: #1596075 Change-Id: Ib786117dcea08af75551770ea4c30d460382b829 ** Changed in: neutron Status: In Progress => Fix Released -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1596075 Title: Neutron confused about overlapping subnet creation Status in neutron: Fix Released Bug description: If I use heat to create a set of networks, each with its own subnet, sometimes I will get an error from Neutron that there is an overlap. There is an example heat template attached. In my environment, this will fail about 1 or 2 times in 10. The heat template does: for X in seq 0 10 create network X assign subnet 172.16.X.0/24 to network X One of the stacks will get an error on one of the subnets, indicating: Requested Subnet With Cidr: 172.16.7.0/24 For Network: 73f99807-0b2a- 49d3-8d11-B7bd74d02f4d Overlaps With Another Subnet. This is in turn coming from _validate_subnet_cidr() in db/ipam_backend_mixin.py. What is happening is that the 'new_subnet_cidr' passed in is matching against itself. Since I have 'allow_overlapping_ips=true', it does: subnet_list = network.subnets but, in the failure case, this subnet is already there. In the 'no failure case' it is not yet there. I have 3 heat API servers, and 72 heat workers (3 servers). The api servers are round-robin load balanced I just (2016-06-24) reproduced this against the master/head from github (so midway between mitaka and newton). I'm not sure if there is some missing locking, but for sure this is a race condition. I have ipam_driver unset If i comment out the exception in _validate_subnet_cidr() then it works OK. To manage notifications about this bug go to: https://bugs.launchpad.net/neutron/+bug/1596075/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : [email protected] Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp
