| *** MarkAtwood has quit IRC | 00:14 | |
| *** thingee has quit IRC | 00:39 | |
| *** thingee has joined #openstack-defcore | 01:42 | |
| *** woodster_ has quit IRC | 07:18 | |
| *** hogepodge has quit IRC | 07:52 | |
| *** hogepodge has joined #openstack-defcore | 07:54 | |
| *** pilgrimstack has quit IRC | 07:56 | |
| *** pilgrimstack has joined #openstack-defcore | 08:01 | |
| johnthetubaguy | markvoelker: hogepodge: we have tried to be very clear about the state of Nova proxy APIs, we wrote this in kilo: http://docs.openstack.org/developer/nova/project_scope.html#no-more-api-proxies | 08:25 |
|---|---|---|
| johnthetubaguy | well, at the very end of kilo | 08:26 |
| *** markvoelker has quit IRC | 08:27 | |
| *** markvoelker has joined #openstack-defcore | 09:27 | |
| *** markvoelker has quit IRC | 09:32 | |
| -openstackstatus- NOTICE: OVH servers are down, we are working to solve it. This will cause that jobs queue is processed slowly, please have patience. | 10:36 | |
| *** pilgrimstack has quit IRC | 11:15 | |
| *** pilgrimstack has joined #openstack-defcore | 11:16 | |
| *** MarkAtwood has joined #openstack-defcore | 11:38 | |
| *** MarkAtwood has quit IRC | 11:48 | |
| *** markvoelker has joined #openstack-defcore | 12:17 | |
| markvoelker | johnthetubaguy: yes, and thanks for that--I believe we actually cited that document a couple of times during discussion in the past cycle. | 12:20 |
| markvoelker | It is always helpful when projects signal intent clearly. | 12:21 |
| johnthetubaguy | cool, glad that is helping | 12:21 |
| -openstackstatus- NOTICE: OVH provider is enabled again, please wait for the job queue to be processed | 12:47 | |
| *** pilgrimstack has quit IRC | 13:06 | |
| *** pilgrimstack has joined #openstack-defcore | 13:17 | |
| *** mfisher_ora has joined #openstack-defcore | 14:18 | |
| openstackgerrit | Chris Hoge proposed openstack/defcore: Added cinder scoring for Mitaka release cycle https://review.openstack.org/290271 | 14:49 |
| hogepodge | johnthetubaguy: do you think glance v2 will be supported by the proxy in Newton? | 14:56 |
| johnthetubaguy | hogepodge: we are having a summit session joint with glance to try work out the way forward | 14:57 |
| hogepodge | It makes me sad that the answer to that isn't 'yes'. | 14:57 |
| johnthetubaguy | hogepodge: the problem is v2 can't actually let us implement v1 on top of it, after digging deeper | 14:58 |
| johnthetubaguy | the deeper we dig, the more problems we are finding, although it feels like we are getting to the bottom of the pile, but we did say that before | 14:58 |
| johnthetubaguy | we really want to just delete the proxy, if that were an option, it might turn out to be the best approach | 14:59 |
| hogepodge | what's the timeline on that? | 14:59 |
| johnthetubaguy | basically it turns out v1 and v2 have made different choices on if the cases of image property names should be respected, and its hard to hide that | 14:59 |
| hogepodge | if nova chose to delete it | 14:59 |
| johnthetubaguy | hogepodge: unsure at this point, we have been telling people "never" in the past, but thats ideal doesn't seem possible because of glance wanting to delete v1 | 15:02 |
| johnthetubaguy | hogepodge: we will know more after the summit session, and we have discussed the issues through | 15:02 |
| johnthetubaguy | before we just couldn't find anyone to work on the problem, now we have folks stepping up to help, which is great, but we are finding problems that we need to resolve | 15:03 |
| hogepodge | johnthetubaguy: I'll try to attend the session, if it doesn't conflict with other defcore or qa committments. I also want to sit on on the neutron get-me-a-network discussions | 15:04 |
| johnthetubaguy | so its clear, if a user wants to list their images they *should* be doing that with glance v2's image list, the problem is what to do about the older APIs, until this point Nova has tried very hard to never let an old API request that used to work fail at some point in the future | 15:06 |
| johnthetubaguy | hence we are still on the first version of our API, effectively | 15:07 |
| hogepodge | johnthetubaguy: we have the problem where the interoperability standard depends on the nova proxy apis. we're starting to require project specific apis, so it's time for us to think about how to start depending on the proxies too | 15:15 |
| johnthetubaguy | hogepodge: I wouldn't recommend anyone uses the proxy API, if that helps decide that. | 15:15 |
| hogepodge | johnthetubaguy: so we were depending on that "never let an old API request" policy, and glance v1->v2 proved rough | 15:16 |
| hogepodge | johnthetubaguy: we didn't have much of a choice before, as the only api candidates came from compute and objectstore. but now we're looking at adding index to the latest guidelines http://git.openstack.org/cgit/openstack/defcore/tree/2016.01.json#n1399 | 15:17 |
| johnthetubaguy | hogepodge: hmm, I hadn't appreciated glance wasn't on the list | 15:18 |
| hogepodge | http://git.openstack.org/cgit/openstack/defcore/tree/next.json#n59 | 15:18 |
| johnthetubaguy | I am glance v2 CRUD is totally the way forward, so thats all good, regardless of that the Nova image proxy does | 15:19 |
| hogepodge | johnthetubaguy: in part the working group was starting small to test the waters and make the standard easier to meet. | 15:19 |
| johnthetubaguy | hogepodge: understood, I just forgot glance was excluded at that point, makes sense | 15:19 |
| johnthetubaguy | so keeping read-only compatibility, should be OK-ish, that is a possible compromise option. | 15:20 |
| johnthetubaguy | for Nova's image proxy API thingy | 15:21 |
| johnthetubaguy | oh wait, you have full crud, I didn't know the compute proxy API supported that, one sec... | 15:22 |
| hogepodge | kind of have full crud. because of compatibility issues between the different deployed apis I don't think we'll ever have tested CRUD operations on images | 15:23 |
| hogepodge | the glance task api is basically a manifestation of an anti-interoperability pattern | 15:24 |
| johnthetubaguy | hogepodge: totally, although I think they understand that now, which is good | 15:25 |
| johnthetubaguy | well, as a community, I think we are all more aware of that now I guess, which is good | 15:25 |
| johnthetubaguy | hogepodge: if we did drop the images proxy, I guess the minimum would be "wait till no valid certification" includes the nova image proxy API? | 15:27 |
| johnthetubaguy | honestly, that feels too soon to me, but just trying thinking through my "if we have to drop it" position really | 15:29 |
| hogepodge | johnthetubaguy: a guideline is good for a year, so if the proxy is in .next (and it will be unless something changes in the next two days) it won't be until 2017.08 that defcore would drop it entirely | 15:35 |
| hogepodge | johnthetubaguy: it could be dropped partially as early as 2017.01 | 15:35 |
| *** openstackgerrit has quit IRC | 15:48 | |
| *** openstackgerrit has joined #openstack-defcore | 15:48 | |
| *** pilgrimstack has quit IRC | 16:13 | |
| *** catherineD|2 has joined #openstack-defcore | 17:02 | |
| *** catherineD has quit IRC | 17:02 | |
| *** luzC- has joined #openstack-defcore | 17:12 | |
| *** luzC- has quit IRC | 17:18 | |
| *** lcazares has quit IRC | 17:19 | |
| *** amit213 has quit IRC | 17:19 | |
| *** kbaikov has quit IRC | 17:19 | |
| *** amit213 has joined #openstack-defcore | 17:22 | |
| *** kbaikov has joined #openstack-defcore | 17:22 | |
| *** pilgrimstack has joined #openstack-defcore | 18:11 | |
| *** pilgrimstack has quit IRC | 18:14 | |
| *** pilgrimstack has joined #openstack-defcore | 18:15 | |
| *** woodster_ has joined #openstack-defcore | 18:34 | |
| *** edmondsw has joined #openstack-defcore | 19:07 | |
| *** pilgrimstack has quit IRC | 19:17 | |
| openstackgerrit | Merged openstack/defcore: Added cinder scoring for Mitaka release cycle https://review.openstack.org/290271 | 19:30 |
| openstackgerrit | Merged openstack/defcore: Flag tests requiring multiple tenants and users https://review.openstack.org/253138 | 19:43 |
| openstackgerrit | Chris Hoge proposed openstack/defcore: Flag tests scheduled for removal from Tempest https://review.openstack.org/289071 | 19:48 |
| hogepodge | markvoelker: eglute: rebased https://review.openstack.org/#/c/289071/ | 19:49 |
| eglute | thank you hogepodge! | 19:49 |
| eglute | will take a look at it today | 19:49 |
| *** dhellmann has left #openstack-defcore | 20:01 | |
| *** cjvolzka has quit IRC | 20:38 | |
| *** galstrom_zzz is now known as galstrom | 20:39 | |
| *** mfisher_ora has quit IRC | 20:56 | |
| *** edmondsw has quit IRC | 21:08 | |
| *** cjvolzka has joined #openstack-defcore | 21:58 | |
| *** pilgrimstack has joined #openstack-defcore | 22:09 | |
| *** galstrom is now known as galstrom_zzz | 22:20 | |
| *** pilgrimstack has quit IRC | 22:27 | |
| openstackgerrit | Chris Hoge proposed openstack/defcore: Flag tests scheduled for removal from Tempest https://review.openstack.org/289071 | 23:06 |
| *** markvoelker has quit IRC | 23:20 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!