Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2019-05-16
11:09:23 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Remove deprecated stuff - Image https://review.opendev.org/659431
11:09:23 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Remove deprecated stuff - Identity https://review.opendev.org/659430
11:09:24 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Remove deprecated stuff - Volume https://review.opendev.org/659399
11:09:24 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Remove deprecated stuff - Network https://review.opendev.org/659400
11:44:37 stingrayza hey folks. is anybody available to take a look at https://review.opendev.org/659366 ? I'd appreciate it greatly - would help me to get unstuck...
12:44:45 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Aggregate functional test tweak https://review.opendev.org/659552
16:00:06 edleafe The API-SIG Office Hour has begun. C'mon in, the water's fine!
16:00:12 elmiko \o/
16:01:23 elmiko lol
16:01:52 gtema well, it's end of Thursday here for me
16:08:41 dtantsur o/
16:08:58 dtantsur what about we merge some guidelines? ;)
16:09:12 elmiko sounds good to me, what did you have in mind?
16:09:24 dtantsur the two that we have frozen: version discovery and mDNS
16:09:32 gtema yeah, EACH SERVICE MUST DELIVER VER DISCOVERY DOCUMENT :D
16:09:47 dtantsur I haven't checked the comments on https://review.opendev.org/#/c/459405/ in depth though
16:10:15 dtantsur https://review.opendev.org/#/c/651222/ did not raise any objections in a month
16:10:51 elmiko the second one looks like it's ready for merge
16:11:01 edleafe I think we can merge https://review.opendev.org/#/c/651222/
16:11:08 edleafe Looks unanimous
16:11:08 elmiko although the first has no objections, seems like it would be easy to add morgan's suggestions
16:11:13 elmiko ++
16:11:20 edleafe Done
16:12:09 edleafe Yeah, we should do one more revision on https://review.opendev.org/#/c/459405. Once that's pushed, I can re-freeze it
16:12:46 dtantsur I can do it.. one day..
16:12:54 dtantsur if somebody has free time (LOL), go ahead with the updates
16:13:15 edleafe dtantsur: Weren't you just saying you wanted something to do with all your free time?
16:13:38 dtantsur I'm enjoying fixing ironic CI 24/7 :(
16:14:27 openstackgerrit Colleen Murphy proposed openstack/api-sig master: Update liaison for keystone https://review.opendev.org/659613
16:15:57 elmiko ouch
16:19:28 dtantsur speaking of which, I'm exhausted, so I'll call it a day a bit earlier.. see you
16:19:43 elmiko rest easy dmitry o/
16:20:19 dtantsur|afk I have to move some furniture today, otherwise I will :D
16:22:57 elmiko yikes!
16:25:40 edleafe I would update, but I really don't know the answer to the formatting question on https://review.opendev.org/#/c/459405/19/guidelines/consuming-catalog.rst@293
16:26:01 edleafe too many levels :)
16:27:46 elmiko hmm, that's a good question and i'm not sure the correct answer
16:31:18 edleafe We can wait for when dtantsur|afk has copious free time.
16:31:40 elmiko i wonder if mordred might comment there too, he certainly understands the intent better than most
16:32:03 edleafe I was going to suggest that, but was afraid to summon the spirit of mordred
16:32:14 elmiko i mean, by the reading of the rule it seems like morgan's assumption is correct re the v3 v3.6 example
16:32:17 elmiko haha, fair
16:33:07 edleafe Yeah, but I don't trust my reading of that rule. Maybe that's a sign that a bit more explanation is needed
16:33:28 elmiko i share your concerns, i felt very similar
16:34:14 elmiko i guess i will add that to the comments, at the least it helps expose an area where more clarity is needed
16:58:53 mordred elmiko: what did I do?
17:01:13 elmiko mordred: oh nothing in specific, morgan had brought up a question on the version discovery pr and we thought you might be able to shed some light
17:01:21 mordred oh jeez. what a fun edgecase
17:01:27 elmiko exactly
17:01:35 mordred my first response is "don't do that"
17:01:42 elmiko i just don't know the problem space well enough to provide a solid answer
17:01:45 elmiko lol
17:02:08 mordred I don't know that I've ever encountered a vX.Y url with a service that doesn't have a working discovery document
17:02:36 elmiko any guidance, or opinion, you could share on the pr would help =)
17:02:53 mordred so while we do account for crazy things - most of them are tied to crazy things people have actually done - not crazy things people might have done
17:02:58 mordred I will leave a comment
17:03:15 elmiko yeah, and the speculation business is a dangerous one
17:03:17 elmiko thanks
17:03:20 mordred ++
17:11:40 dustinc dtantsur|afk or anyone else able to answer a question about node.patch for baremetal...the ironicclient equivalent would raise an exception if a node is locked when you try to patch it, but the docstring for it does not how any exceptions. I dug in a bit and did not see any possible exception sources (other than parent class, but only if patch is not enabled, so not relevant for node) can someone confirm?
17:11:52 dustinc implemented here: https://review.opendev.org/#/c/647730/
17:12:13 dustinc **the docstring for sdk**
19:09:56 openstackgerrit Merged openstack/python-openstackclient master: Update sphinx requirement. https://review.opendev.org/659279
#openstack-sdks - 2019-05-17
00:20:52 openstackgerrit Dean Troyer proposed openstack/python-openstackclient master: Remove code migrated to osc-lib long ago https://review.opendev.org/659693
04:35:35 openstackgerrit Merged openstack/python-openstackclient master: Aggregate functional test tweak https://review.opendev.org/659552
07:32:51 dtantsur dustinc: I'm not sure I entirely get the question.. you'll get HTTP CONFLICT if the node is locked, both ironicclient and openstacksdk will retry it for a bit
07:48:58 openstackgerrit Dmitry Tantsur proposed openstack/api-sig master: Update Sphinx requirements and actually use Python 3 for docs https://review.opendev.org/659739
07:49:07 dtantsur elmiko, edleafe, pretty urgent ^^^
07:49:19 dtantsur gtema: ^^
07:53:21 gtema dtantsur: great, was wanting to do that as well. Thanks
10:39:51 openstackgerrit Surya Seetharaman proposed openstack/python-openstackclient master: Microversion 2.73: Support adding the reason behind a server lock https://review.opendev.org/659124
10:41:42 tssurya dtroyer: addressed your comments ^ hope it looks ok now
11:19:14 dtroyer tssurya: thank you… it looks like you still have a problem with requirements though… novaclient 14.0 is not the common minimum yet. you may have to check that version too unless it gets raised globally
11:20:14 tssurya dtroyer: so its not enough I bump it here: https://review.opendev.org/#/c/659124/4/lower-constraints.txt ?
11:20:23 tssurya sorry for the stupid questions (haven't done this before)
11:21:48 tssurya apart from that I also relealised some problems with the logic (https://review.opendev.org/#/c/659124/4/openstackclient/compute/v2/server.py@1217) I'll work on that before pinging you
11:24:10 dtroyer tssurya: We can allow an older version than the global minimum, we can not go the other way. I would prefer to work with as wide a range of dependencies as possible
11:25:12 tssurya dtroyer: ah ok can you point me to the global min requirements ?
11:25:27 tssurya not that I will change that, I'll do the check like you stated then
11:26:06 dtroyer all of the requirements are in https://opendev.org/openstack/requirements
11:27:02 tssurya thanks
11:33:08 dtantsur dtroyer, tssurya, IIRC minimum requirements are no longer tracked in the central repository
11:33:33 tssurya dtantsur, dtroyer: yea I was also looking as to where the min requirements are
11:33:33 dtantsur each project can bump requirements.txt + lower-constraints.txt, as long as it does not contradict global requirements
11:33:53 tssurya I only see max possible ones: https://opendev.org/openstack/requirements/src/branch/master/upper-constraints.txt#L221
11:34:00 dtantsur right
11:34:29 tssurya cool thanks :)
12:29:56 mordred elmiko: responded on https://review.opendev.org/#/c/459405 - I was misreading that section yesterday, I have a better answer today
12:30:43 openstackgerrit Surya Seetharaman proposed openstack/python-openstackclient master: Microversion 2.73: Support adding the reason behind a server lock https://review.opendev.org/659124
12:31:13 edleafe dtantsur: lloks good. Fast-approved
12:31:17 edleafe lloks*
12:31:25 edleafe heh
12:31:28 edleafe looks
12:37:32 openstackgerrit Merged openstack/api-sig master: Update Sphinx requirements and actually use Python 3 for docs https://review.opendev.org/659739
12:41:09 dtantsur thx edleafe!
12:51:09 openstackgerrit Merged openstack/api-sig master: Add guideline on DNS-based service discovery https://review.opendev.org/651222
12:53:01 dtantsur \o/
13:07:52 brtknr mordred: I found a bug in the os_stack module I submitted months ago, which appears to fail when using new Ansible with openstacksdk older than 0.28.0, which I've fixed in this PR: https://github.com/ansible/ansible/pull/56575
13:10:49 elmiko mordred: ack, thanks!
13:16:17 stingrayza Hi folks, would anybody mind casting an eye on https://review.opendev.org/659366/ please. (It's required for https://github.com/ansible/ansible/pull/56453)

Earlier   Later