| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2019-06-06 | |||
| 13:14:57 | efried | thank you sir | |
| 13:15:15 | dtantsur | having the CI (on the way to being) green is appreciated :) | |
| 13:15:34 | efried | of course | |
| 13:35:36 | openstackgerrit | Eric Fried proposed openstack/openstacksdk master: Handle oslo.config exceptions in from_conf https://review.opendev.org/663439 | |
| 13:35:42 | efried | dtantsur: That should do it ^ | |
| 13:35:55 | efried | mordred: ^ | |
| 13:43:44 | mordred | efried: oh good - I was just looking at that but hadn't yet figured out what was up with the add_service calls :) | |
| 13:45:26 | mordred | Shrews: if you're bored this morning, ^^ that has 2x+2 but one is from me and I wrote some of the patch, so might be worth having another set of eyes on | |
| 13:45:44 | mordred | Shrews: otoh - efried wrote most of it, so you don't have to worry about much terrible code from me :) | |
| 13:48:18 | efried | mordred, dtantsur: I'll add consolidation of the log message to the fixups catch-all I'm working on now. | |
| 13:48:39 | mordred | coolio | |
| 13:49:45 | mordred | dtantsur: if you have another second, https://review.opendev.org/#/c/662698 by gtema could use another review - it's the underlying base for the rest of his stack | |
| 13:52:41 | dtantsur | lgtm | |
| 13:59:12 | mordred | woot! | |
| 14:33:01 | openstackgerrit | Eric Fried proposed openstack/openstacksdk master: Minor fixups from from_conf changes https://review.opendev.org/663685 | |
| 14:33:28 | efried | dtantsur, mordred: fixups ^ | |
| 14:34:49 | efried | mordred: You can abandon https://review.opendev.org/#/c/663435/ now, yah? | |
| 14:35:04 | mordred | yup! done | |
| 14:35:27 | mordred | efried: oh - piddle. we didn't add a release note for the service disabling. let me add one real quick | |
| 14:36:46 | efried | mordred: In terms of "here's a new feature you can use"? Cause otherwise, who would notice? | |
| 14:38:14 | mordred | yeah- well, there's a new behavior if you set has_foo in your config file | |
| 14:38:38 | mordred | (also, I think we should cut a release as soon as these changes land, and it would be a release without any release notes to indicate what's in it otherwise :) ) | |
| 14:40:02 | efried | heh | |
| 14:40:07 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add release note for new disable service functionality https://review.opendev.org/663690 | |
| 14:41:02 | efried | mordred: comment inline... | |
| 14:41:18 | mordred | yup. good call | |
| 14:44:00 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add release note for new disable service functionality https://review.opendev.org/663690 | |
| 14:45:47 | efried | whapow, big ol +1 | |
| 14:46:24 | efried | https://www.youtube.com/watch?v=ggXbzjnffAo | |
| 15:07:53 | mordred | this couch is cut in half | |
| 15:08:32 | openstackgerrit | Merged openstack/openstacksdk master: Support skipping unknown QP https://review.opendev.org/662698 | |
| 16:00:15 | elmiko | API SIG office hour is now open! | |
| 16:00:23 | edleafe | Damn, you beat me again | |
| 16:00:30 | elmiko | i even waited that time XD | |
| 16:00:31 | edleafe | Was just going to press Enter | |
| 16:00:43 | elmiko | i figured you would beat me to it lol | |
| 16:04:32 | elmiko | just as a followup to the discussion that happened last week, i proposed a patch to keystoneauth that fixes the issue with api-sig suggested error formatting | |
| 16:04:39 | elmiko | that patch has been merged into the master | |
| 16:05:15 | elmiko | there was a question about backporting from the folks who ran into this originally, but i think that is better left to the keystoneauth team | |
| 16:05:32 | edleafe | Yeah, that patch looked sane | |
| 16:06:30 | elmiko | =) | |
| 16:08:49 | edleafe | We'll try to handle the load ourselves :) | |
| 16:12:20 | elmiko | hehe | |
| 16:35:19 | dtroyer | stephenfin: if you're around I would love to have you take one last look at https://review.opendev.org/#/c/659382/ (—live-migration) before merging it. This is the last thing before cutting a 3.19 release for OSC. | |
| 16:35:52 | dtroyer | mordred: ^^^ if you feel like looking at Something Completely Different(TM) | |
| 16:36:51 | stephenfin | dtroyer: Won't have time for a full review but I've made my peace with that approach. I've already fought with argparse too much, heh | |
| 16:44:01 | edleafe | elmiko: Gotta run out a little early. Try to hold the fort while I'm gone. | |
| 16:44:21 | dtroyer | stephenfin: ok, thanks. I just wanted to make sure there were no outstanding concerns | |
| 16:44:31 | elmiko | edleafe: ack, take care o/ | |
| 16:45:44 | stephenfin | dtroyer: Not from me. My concern was with the duplication in 'migration --live-migration' but I haven't found a way to avoid that that doesn't involve rewriting argparse | |
| 16:45:57 | stephenfin | which is a hole I went far too deep down | |
| 16:47:16 | dtroyer | stephenfin: maybe in a year or so we can bring —live back as an alias :) it would be a forced break for anyone attempting to give it an argument from the old syntax. thanks again | |
| 16:58:35 | mordred | dtroyer: lgtm! | |
| 16:58:55 | dtroyer | mordred: Shit it! Thanks | |
| 16:59:01 | dtroyer | ooooooo!!!!!!! | |
| 16:59:03 | dtroyer | SHIP | |
| 16:59:04 | dtroyer | SHIP | |
| 16:59:05 | dtroyer | SHIP | |
| 16:59:45 | mordred | hahahaha | |
| 17:18:09 | openstackgerrit | Dean Troyer proposed openstack/osc-lib master: Add FakeModule from OSC https://review.opendev.org/660230 | |
| 17:18:36 | efried | mordred: lower constraints job is not happy, but I can't fathom why. | |
| 17:20:52 | efried | os-service-types, I'll betcha. | |
| 17:21:33 | mordred | efried: WEIRD | |
| 17:22:12 | efried | mordred: nah, we changed Connection's has_service to use CloudConfig's has_service, which is different in that it defaults to looking up the service type in ost. | |
| 17:22:28 | efried | mordred: Is it okay to bump the ost l-c to match the u-c? | |
| 17:22:40 | mordred | absolutely | |
| 17:22:57 | mordred | are we required to have a lower-constraints entry for every library? | |
| 17:22:58 | efried | ight, suitable for a ninja approve if I patch that in right quick? | |
| 17:23:03 | mordred | yup | |
| 17:23:17 | efried | I think you're required to have l-c for anything listed in requirements or test-requirements, yeah. | |
| 17:23:43 | mordred | nod. well - makes sense I guess - be sure to bump the requirements lower bound too, obvs | |
| 17:23:44 | efried | I know we're keeping placement up to date with latest, forcefully. | |
| 17:25:24 | openstackgerrit | Eric Fried proposed openstack/openstacksdk master: Minor fixups from from_conf changes https://review.opendev.org/663685 | |
| 17:25:24 | openstackgerrit | Eric Fried proposed openstack/openstacksdk master: Handle oslo.config exceptions in from_conf https://review.opendev.org/663439 | |
| 17:26:40 | efried | mordred: ^, top rebased, will require only +W. Not sure if your reno patch will need rebase+W | |
| 17:27:28 | efried | btw, ost 1.7.0 is out, not sure if we should keep sdk in sync | |
| 17:28:06 | efried | fwiw philosophy taken in placement is "we should always be pinning to the latest os-traits and os-resource-classes". So we actually put in tests that fail as soon as a new release hits upper-constraints. | |
| 17:28:29 | efried | e.g. https://review.opendev.org/#/c/661132/ | |
| 17:30:07 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add release note for new disable service functionality https://review.opendev.org/663690 | |
| 17:30:15 | mordred | efried: all three re-+A'd | |
| 17:30:36 | mordred | efried: we probably shoudl do the same - I like that as an idea | |
| 17:30:44 | mordred | sdk should always have latest ost | |
| 17:33:41 | efried | mordred: okay, I'll propose a canary patch | |
| 17:40:49 | openstackgerrit | Merged openstack/python-openstackclient master: Add changes-before attribute to server list https://review.opendev.org/657373 | |
| 17:47:44 | openstackgerrit | Eric Fried proposed openstack/openstacksdk master: Pin to latest os-service-types https://review.opendev.org/663733 | |
| 17:47:49 | efried | mordred: ^ | |
| 17:47:56 | efried | I didn't know where to put it | |
| 17:48:04 | efried | I didn't want to make a whole new test module for it | |
| 17:48:16 | efried | If you want me to do so, or have a better suggestion for where to put it, lmk. | |
| 18:14:38 | mordred | efried: I think that looks great | |
| 18:44:22 | openstackgerrit | Merged openstack/shade master: Cap sphinx for py2 to match global requirements https://review.opendev.org/663545 | |
| 18:44:34 | efried | I've bumped the nova side to use 0.30.0, which *should* make the bottom patch and the placement swapout go green. | |
| 18:48:14 | mordred | efried: hopefully we'll get you an 0.31 soon | |
| 18:48:31 | efried | mordred: That'll be neat... but I'm not sure I actually need it. | |
| 18:48:50 | efried | oh, never mind | |
| 18:49:03 | efried | was gonna say, nova will be well-behaved and only use Connection for things it's supposed to. | |
| 18:49:10 | efried | But that doesn't help the case where conf is misconfigured | |
| 18:49:25 | efried | though that's a pretty cornery corner. | |
| 18:50:05 | efried | Because we still have to pass in the session, which only exists if we had ksa session opts, which we pull in at the same time as adapter opts. So the only busted path is interface+valid_interfaces I think. | |
| 18:54:43 | mordred | nod | |