Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2019-08-07
13:08:40 openstackgerrit Loan Harrouin proposed openstack/openstacksdk master: Add get_actions for server resource https://review.opendev.org/675113
13:22:30 edleafe Who knew that our API work extended to snail mail? https://twitter.com/shawnacscott/status/1158813468704698368
13:24:26 dtantsur oh lol
13:39:20 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Initalize pool_executor so close works https://review.opendev.org/675120
13:39:20 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for global_request_id https://review.opendev.org/674807
13:39:21 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Fix discovery cache sharing https://review.opendev.org/675121
13:39:38 mordred dtantsur, efried: ^^ there we go - with discovery cache fixed and also with context manager fixed
13:46:21 Shrews mordred: do we have any existing tests in sdk that exercise the large object code? I'm not finding any
13:46:37 mordred Shrews: yes - becuase they sucked to write
13:46:45 Shrews :)
13:46:56 Shrews i must be looking in the wrong places then
13:48:09 mordred Shrews: test_create_dynamic_large_object and test_create_static_large_object in openstack/tests/unit/cloud/test_object.py
13:48:26 Shrews ah ha
13:48:33 mordred also test_object_segment_retry_failure
13:48:53 Shrews i forgot we still have mixed tests
13:52:47 mordred yeah. one of these days it's all going to be perfect and shiny
14:03:12 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for global_request_id https://review.opendev.org/674807
14:03:12 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Fix discovery cache sharing https://review.opendev.org/675121
14:04:02 mordred Shrews: if you have a sec, the first two in that stack are bugfixes I noticed while working on the third for efried
14:04:52 openstackgerrit Loan Harrouin proposed openstack/openstacksdk master: Add get_actions for server resource https://review.opendev.org/675113
14:05:19 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for global_request_id https://review.opendev.org/674807
14:05:30 mordred efried: you are quick at re-reviewing things
14:06:39 mordred dtantsur: if we do lift in the temp-url stuff, we should probably look to see what from the form_post method we can reuse. (we're using that form-post method in zuul log uploading to swift, so I'm pretty sure it's solid)
14:07:06 dtantsur makes sense, yeah
14:41:21 efried mordred: I just happened to be looking at my email at the time :)
14:47:06 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Rationalize endpoint_for and get_endpoint_from_catalog https://review.opendev.org/675130
14:47:38 mordred Shrews, dtantsur|bbl: ^^ that also adds some tests and docs for a method we added last week - and makes it make sense with the existing method that was already there
14:52:09 efried mordred: Attempting to use sdk auto passthrough proxy for cyborg gets Sundar this error: http://paste.openstack.org/show/755594/
14:52:16 efried is that expected?
14:55:31 mordred efried: does cyborg not publish a version discovery document?
14:56:19 efried I was certain they did
14:56:37 efried mordred: esp because they have a v1 that's dead, and are working up v2 now.
14:58:41 mordred hrm. do you have a link to code where that's happening? I'd like to see the value for accelerator_api_version - or more generally the clouds.yaml / config settings and also the discovery documents
14:59:35 mordred that error basically happens when a version has been requested and discovery can't find a major api version on the service
15:00:15 mordred in any cas e- let's dig in and figure out what's going on there
15:01:02 mordred efried: oh - I'm sorry - I misspoke - that's what happens when there is no configured api version
15:01:15 efried mordred: Okay. Sundar is Pacific so might be a bit before he's on. The conf is here http://paste.openstack.org/show/755593/
15:01:46 mordred efried: oh - blerg. this is dumb
15:01:47 efried mordred: Oh, so you mean he could put a version in the conf and/or constructor and be good to go?
15:01:59 mordred efried: I have NO IDEA why I thought the logic for this case would be reasonable
15:02:16 efried heh
15:05:21 mordred efried: fix coming. also - I think we can make a more substantial improvement - which is to make a base class and a subclass of the ServiceDescription object - one that we use for services we know about, and one for thigns that we're just creating becaues they're in service-types-authority
15:05:53 mordred because there's warnings we throw in the current case for "you're only going to get passthrough REST here" - but that's a warning if we have a supported Proxy version because that means something went wrong
15:06:09 mordred if we DON'T have one of those, then you're expecting a REST passthrough, so the warning is stupid
15:06:22 mordred but - there's just a dumb logic error for the cyborg case
15:06:32 efried mordred: So then we wouldn't actually need the placement passthrough you made earlier this week?
15:07:03 efried this feels like it's kinda in the same neighborhood
15:08:23 openstackgerrit David Shrewsbury proposed openstack/openstacksdk master: WIP: Retry SLO manifest upload https://review.opendev.org/675134
15:09:12 Shrews mordred: Want to check that I'm attacking this in the right way before I get too far? ^^^
15:10:01 mordred efried: it is in the same neighborhood - and perhaps
15:10:23 mordred Shrews: yeah - I thnik so
15:13:37 mordred efried: yah - I can make one small patch to make this whole thing much easier I think
15:13:45 efried nice
15:18:20 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Skip most service_description for unknown services https://review.opendev.org/675135
15:18:32 mordred efried: ^^ that patch might fix the placement and the cyborg issue
15:21:28 efried mordred: asking Sundar to try it. (He's in a mtg, may be an hour or so.)
15:23:13 mordred efried: ok. let him know I'm very keen to make sure his stuff is working here
15:24:29 efried will do, and thanks mordred
16:07:55 Shrews mordred: ooh, delete_autocreated_image_objects *almost* does what we need for failed upload cleanup. Maybe add an optional parameter to limit it to object name prefix?
16:12:44 Shrews oh, actually, don't think i can use that at the point i need it
17:42:13 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Skip most service_description for unknown services https://review.opendev.org/675135
18:34:20 openstackgerrit Monty Taylor proposed openstack/keystoneauth master: Add remove_service to token fixtures https://review.opendev.org/675174
18:45:58 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Validate that connect_as connects as the project https://review.opendev.org/675178
19:50:07 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: WIP Make a reusable fixture for mocking clouds https://review.opendev.org/675187
19:59:15 efried_rollin oo
20:02:03 mordred efried: that's VERY VERY early and is almost completely unuseful for you yet
20:02:21 efried but the train has left the station
20:02:24 mordred my plan is to do some refactoring inside of sdk so that we make the fixture and then we use that fixture in our tests
20:02:36 mordred so that, you know - you can count on it
20:02:53 mordred I thnik it's going to make sdk's tests better too
20:03:02 efried spect so
20:03:06 mordred because we have a bunch of wonky hacks going on right now
20:03:30 efried mordred: Have we done the thing yet where we run some form of nova dsvm CI on sdk patches?
20:03:45 efried you start mucking with endpoint methods, makes me nervous.
20:04:31 mordred don't be nervous - we never write bugs
20:04:37 mordred efried: and no - but we should
20:05:10 mordred efried: we have plenty of dsvm jobs going in sdk - but nothing running a tempest-based anything because up until recently there was no need to
20:08:47 kmalloc mordred: my TODO comments are.... long ago todo ;)
20:08:54 mordred kmalloc: :)
20:09:06 mordred kmalloc: well - check that WIP - I think you'll immediately get where I'm going with that and will hopefully like it
20:09:34 kmalloc mordred: oh that is a good direction
20:09:42 kmalloc didn't even take long to see it.
20:09:44 kmalloc good stuff
20:09:47 mordred \o/
20:12:35 kmalloc i'll review a bit more in depth knowing it's WIP
20:12:43 kmalloc but i think you're headed the right way for sure.
20:20:06 openstackgerrit David Shrewsbury proposed openstack/openstacksdk master: WIP: Retry SLO manifest upload https://review.opendev.org/675134
20:23:17 mordred Shrews: ^^ \o/
20:24:54 Shrews mordred: i'm completely boggled on the new test_slo_manifest_fail test and what requests to mock for the object cleanup phase
20:25:17 Shrews that's where it's barfing for me, atm
20:25:58 Shrews i know what calls i *think* it's going to make, but not the details
20:26:34 mordred hrm.
20:27:12 mordred well - it *should* fire a delete for each segment it sent an upload for, right?
20:27:18 mordred (in some order or another)
20:28:07 mordred oh - well - except - in this case you're going to return a list of segment objects as the return value from the list_objects call it's going to make
20:28:23 mordred so the json you return from list_object in your requests_mock should feed which segments it fires deletes for, yeah?
20:28:49 mordred (and you could probably make it a small list of segments so you don't go crazy)
20:29:23 Shrews i think that's right, just trying to figure out how to make that happen. thus far, it doesn't like my list objects uri i'm creating there
20:31:58 Shrews '{endpoint}/{container}?format=json&prefix={prefix}' <<-- that one
20:44:00 mordred Shrews: oh - so ... theres a thing with url parameters where you have to pass a thing

Earlier   Later