Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2018-07-26
16:07:40 odyssey4me mordred: it turns out that it's far more fun than anticipated - to replicate it you have to setup the compute endpoint with the wrong service backing it
16:08:09 odyssey4me in my test I put the placement service endpoints into the catalog, saying it was the compute service
16:08:11 odyssey4me https://gist.github.com/odyssey4me/fdfb2474bdd6fde4d1368af2af4044c6
16:08:45 odyssey4me while of course this is not someone one would expect someone to do, it is perhaps better to handle the error somewhat more elegantly
16:12:18 dtroyer odyssey4me: this sounds like it might be a good integration test for OSC. Those drop commands in the top and use requests-mock so we let the entire stack have a chance to get it wrong, but still wouldn't require a specially-configured could to test against...
16:27:58 cdent thanks elmiko, was rather distracted during all that
16:28:36 elmiko cdent: no worries =)
16:28:51 openstackgerrit Merged openstack/api-wg master: Add links to errors-example.json https://review.openstack.org/578369
16:28:52 openstackgerrit Merged openstack/api-wg master: Expand schema for error.codes to reflect reality https://review.openstack.org/580703
16:33:32 edleafe cdent, elmiko, dtantsur: https://etherpad.openstack.org/p/api-sig-newsletter is ready for your review
16:34:17 edleafe Oops - hold on a minute
16:34:56 edleafe ok, it's ready now. Forgot to paste the links
16:35:07 cdent lgtm
16:35:50 elmiko edleafe: maybe reference that notmyname suggested we look at the draft?
16:36:12 edleafe elmiko: sure. You wanna add that?
16:36:18 elmiko k
16:37:23 elmiko edleafe: how's that look?
16:37:37 edleafe lgtm
16:37:39 edleafe ship it?
16:37:45 elmiko +1
16:38:03 elmiko thanks again edleafe !
16:39:03 edleafe sent
16:39:18 elmiko \o/
19:10:20 mordred dtroyer, odyssey4me: yes to integration test!
19:11:20 mordred odyssey4me: and thanks for making that traceback ... it also looks like the base issue is in novaclient with an assumption on its part on how error payloads look. when we shift stuff over to sdk the error handler there is more permissive because it has to handle all of the services- so it's coded defensively to check and see what exists
19:11:35 mordred in any case, yes - I do think that error could be handled more elegantly
20:02:55 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Update create_object to handled chunked data https://review.openstack.org/585532
20:02:55 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add method for returning a raw response for an object https://review.openstack.org/586262
20:02:56 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for streaming object responses https://review.openstack.org/586250
22:10:35 openstackgerrit Merged openstack/os-service-types master: Add flag for returning unofficial types https://review.openstack.org/585411
#openstack-sdks - 2018-07-27
06:02:03 openstackgerrit Chen proposed openstack/python-openstackclient master: Compute: Add description support for server https://review.openstack.org/568549
06:22:00 openstackgerrit tianhui proposed openstack/python-openstackclient master: Compute: Add tag support for server add fixed ip https://review.openstack.org/581325
06:35:58 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Add --key-name and --key-unset option for server rebuild API. https://review.openstack.org/576422
06:56:32 openstackgerrit YAMAMOTO Takashi proposed openstack/osc-lib master: Replaced a few references to LP bugs with storyboard https://review.openstack.org/586453
07:54:46 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Support for microversions in base Resource https://review.openstack.org/583491
08:01:54 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: [WIP] Support for microversions in baremetal resources https://review.openstack.org/583532
08:23:47 openstackgerrit tianhui proposed openstack/python-openstackclient master: Compute: Add tag support for server add network https://review.openstack.org/581573
08:28:27 openstackgerrit tianhui proposed openstack/python-openstackclient master: Compute: Add tag support for server add port https://review.openstack.org/580977
08:35:53 openstackgerrit tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386
09:19:22 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Add --key-name and --key-unset option for server rebuild API. https://review.openstack.org/576422
10:51:42 dtantsur mordred: hey! can I get some reviews on https://review.openstack.org/#/c/583491/ please? this is an important step in microversioning support
11:43:42 mordred dtantsur: yes! sorry - was focused on stuff for the freeze
11:54:25 mordred dtantsur: lgtm. Shrews - you wanna look at https://review.openstack.org/#/c/583491 ?
11:54:38 dtantsur thanks!
11:55:33 Shrews sure. lemme finish coffeeing first
12:46:17 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add method for returning a raw response for an object https://review.openstack.org/586262
12:46:18 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for streaming object responses https://review.openstack.org/586250
13:00:26 openstackgerrit Fan Zhang proposed openstack/python-openstackclient master: Add --key-name and --key-unset option for server rebuild API. https://review.openstack.org/576422
13:47:20 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Support for microversions in baremetal resources https://review.openstack.org/583532
13:48:20 dtantsur mordred: how hard is it to add functional tests for ironic to the sdk?
13:50:26 mordred dtantsur: it shouldn't be too hard to add the tests - but I don't think we currently run against any devstacks configured for ironic (or any other deployments - doesn't have to be devstack)
13:50:38 mordred so we probably need to add a job
13:50:58 mordred right now our ironic coverages is basically through the shade layer and then through bifrost
13:51:04 mordred and the bifrost job
13:51:44 dtantsur mordred: but we don't have a bifrost job on openstacksdk itself, do we?
13:54:18 mordred oh - no we don't - we should add one
13:54:29 mordred seeing as how the ansible modules use openstacksdk now
13:57:52 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Run bifrost integration test jobs https://review.openstack.org/586564
13:57:56 mordred dtantsur: ^^
13:58:14 dtantsur mordred: looking already, thanks!
13:58:31 mordred dtantsur: we should probably update bifrost itself to just use openstacksdk directly and not via shade ... but maybe that's a step
13:59:15 dtantsur oh, right, I did not realize it also uses shade from python
14:00:17 mordred dtantsur: we added openstacksdk as a shade depend last cycle, so it should be a fairly easy switch
14:00:31 dtantsur right
14:01:09 dtantsur mordred: okay, I'll ponder creating a simple functional job for baremetal bits this afternoon. testing it locally does not scale well :)
14:06:03 dtantsur mordred: will I have to do something to use admin credentials?
14:09:54 mordred dtantsur: the functional test suite looks for clouds.yaml files already - and with a particular name
14:09:58 mordred dtantsur: TEST_CLOUD_NAME = os.getenv('OS_CLOUD', 'devstack-admin')
14:10:55 mordred dtantsur: oh - well - bother - the sdk-layer and shade-layer functional tests are still a little different
14:11:00 mordred that above is for the sdk layer ones
14:11:12 mordred for the shade layer (openstack/tests/functional/cloud) - it's
14:11:17 mordred self._demo_name = os.environ.get('OPENSTACKSDK_DEMO_CLOUD', 'devstack')
14:11:19 mordred self._op_name = os.environ.get(
14:11:21 mordred 'OPENSTACKSDK_OPERATOR_CLOUD', 'devstack-admin')
14:11:24 mordred but basically the same idea
14:11:53 mordred it's also worth looking at openstack/tests/functional/cloud/test_devstack.py ...
14:12:21 mordred we have a little system where you can set an env var in the job definition, like OPENSTACKSDK_HAS_MAGNUM: 1
14:12:24 dtantsur mordred: yeah, I'm looking at the SDK layer for now, since it's not covered at all
14:12:44 mordred and that test will check to make sure that the given service exists in the cloud
14:13:20 mordred then it's safe to put in lines like:
14:13:22 mordred if not self.user_cloud.has_service('volume'):
14:13:24 mordred self.skipTest('volume service not supported by cloud')
14:13:35 mordred into setUp
14:21:57 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: Add a simple baremetal functional job https://review.openstack.org/586574
14:22:06 dtantsur mordred: let's try this ^^^
14:26:31 mordred dtantsur: lgtm
14:26:39 mordred let's see what our friend zuul thinks
14:27:15 dtantsur yep :) I'm pretty sure I missed something on the first attempt
15:07:25 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Add support for streaming object responses https://review.openstack.org/586250
15:35:09 corvus mordred: what's the news on a release with fb8eed6f298e81a80fab614a20f4e61ad6f834bd in it?
15:35:22 corvus Update create_object to handled chunked data
15:37:46 mordred corvus: we just need to land it and then I can get a bugfix release landed
15:37:57 mordred if you wanna +A it (once zuul is happy again)
15:51:59 corvus mordred: +2
15:53:58 mordred corvus: hrm. should probably add a release note to that change since it'll be the only change in a bugfix release
15:54:12 mordred answering "why does this bugfix release exist" is usually a good thing to do
15:54:34 mordred k
15:54:36 mordred gah
15:54:38 corvus mordred: i would have thought the answer to that would be obvious! (because there was a bug!) ;)
15:54:43 mordred hehehe

Earlier   Later