Open Stack

> On 24 Oct 2015, at 21:25, Doug Wiegley <dougwig at parksidesoftware.com> wrote:
> 
> 
>> On Oct 19, 2015, at 5:33 PM, Ihar Hrachyshka <ihrachys at redhat.com> wrote:
>> 
>>> On 16 Oct 2015, at 10:50, Takashi Yamamoto <yamamoto at midokura.com> wrote:
>>> 
>>> if i move fwaas tests from neutron to neutron-fwaas, [1]
>>> is there easy way to run them together with the rest of neutron api tests
>>> for gate-neutron-dsvm-api job?
>> 
>> Before we jump in to reflect current gating status-quo, do we have a definite answer why we want to keep the gate coupling in place? Can’t we leave the fwaas api job for fwaas repo only? Do we think core is unstable enough to justify additional coupling? Is core bad at handling backwards compatibility when it comes to (re)moving code that is unneeded for the core repo?
> 
> Core is absolutely dreadful at backwards compatibility (I’m talking recent history here.) The co-gate is necessary to guard against that at the moment.

Interesting. Any details? Is it just a matter of some dumb code move/rename, or is it something really architectural, like single API manager for all services?

I wonder what we may do to make the difference for eventual breakage of the tie. Should we start collecting post-mortem details on breakage cases to start understanding where we lag?

Ihar
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 455 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20151027/016e0c20/attachment.pgp>