Hi, I would like to request a FFE for neutron: https://review.openstack.org/#/c/71859/. This patch is a workaround for a bug that an external network can become invisible to non-admin users after restarting neutron-server. (https://bugs.launchpad.net/neutron/+bug/1254555). Once an external network becomes invisible, non-admin user cannot connect a router to the outside and cannot allocate new floating IPs. It break usability much. According to the analysis of the cause, more neutron agents we have, more likely this bug occurs. This means it usually happens in production environments (when neutron-server is restarted). This patch was merged in the master two month ago and we confirmed this patch works well. I believe this patch is row risk and fixes a user-visible issue. Thanks, Akihiro On Fri, Jan 31, 2014 at 7:33 PM, Alan Pevec <apevec@gmail.com> wrote:
Hi all,
I'm opening the thread to discuss proposed exceptions for the stable/havana, either breaking the freeze or breaking the backport rules. I've two for now:
* https://review.openstack.org/69884 - Keystone: significant performance improvement, includes db migrations which are verboten in general, so my first reaaction was -2. But upon closer look, I think this is acceptable since even without running migration code will work after upgrade. Migrations are sequential to last in havana (034) so 035-036 could be treated like "reserved for Havana backports" like Nova did in https://github.com/openstack/nova/commit/ab2c467da951071a8aac4eb6ca032371c69... Of course that means no more db migrations in Keystone Havana!
* https://review.openstack.org/70016 - Horizon: session data type change - I'm not a Django expert, so I'd like input for Horizon team to verify Kieran's claims that it won't break existing sessions.
Cheers, Alan
_______________________________________________ Openstack-stable-maint mailing list Openstack-stable-maint@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-stable-maint