| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-09-24 | |||
| 12:42:51 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Import rate limiting TaskManager from nodepool https://review.openstack.org/574285 | |
| 12:42:52 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add support for per-service rate limits https://review.openstack.org/604521 | |
| 14:17:33 | Horrorcat | mmands (e.g. openstack server list, which e.g. fails if the IdP does not offer a compute endpoint). | |
| 14:17:33 | Horrorcat | I think we're seeing a regression between osc-lib 1.9.0 to osc-lib 1.11.0. we are using keystone-to-keystone federation. when executing openstack endpoint list with an openrc configured for federation (using OS_SERVICE_PROVIDER), osc-lib 1.9.0 correctly queries the endpoints of the service provider. osc-lib 1.11.0 queries the endpoints of the identity provider instead. the same holds for other co | |
| 14:17:44 | Horrorcat | is this known/intentional or shall we report this as a bug? | |
| 14:17:52 | Horrorcat | also, is any additional information needed? | |
| 14:17:55 | Horrorcat | (cross-posted from #openstack) | |
| 14:21:11 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add support for per-service rate limits https://review.openstack.org/604521 | |
| 14:21:11 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Make RateLimitingTaskManager the TaskManager https://review.openstack.org/604520 | |
| 14:21:54 | mordred | Horrorcat: I'm not aware of that issue - so probably reporting it as a bug is a good idea | |
| 14:22:08 | Horrorcat | mordred, okay thanks | |
| 14:22:17 | mordred | kmalloc: ^^ I might need some help on that one, as I don't have good background with k2k stuffs | |
| 14:22:51 | kmalloc | mordred: k2k is pretty opaque :) | |
| 14:23:11 | kmalloc | mordred: I'll be able to dive into that later today (this afternoon) | |
| 14:23:27 | mordred | kmalloc: cool - thanks! | |
| 14:23:37 | kmalloc | mordred: knikolla is also a.good resource | |
| 14:23:40 | Horrorcat | I can provide a reproducer docker-compose-based setup, but only tomorrow (I’m in UTC+2) | |
| 14:23:47 | kmalloc | For all things federated. | |
| 14:23:59 | kmalloc | Horrorcat: yeah, no rush :) | |
| 14:49:00 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Restore timeout_scaling_factor https://review.openstack.org/604628 | |
| 15:15:24 | openstackgerrit | Matthias Lisin proposed openstack/openstacksdk master: openstackcloud.py: Implement FWaaS wrapper methods. https://review.openstack.org/604811 | |
| 17:13:06 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add support for per-service rate limits https://review.openstack.org/604521 | |
| 18:45:30 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Restore timeout_scaling_factor https://review.openstack.org/604628 | |
| 22:25:16 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: WIP Replace TaskManager with a semaphore and lock https://review.openstack.org/604926 | |
| #openstack-sdks - 2018-09-25 | |||
| 00:02:10 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: WIP Replace TaskManager with a semaphore and lock https://review.openstack.org/604926 | |
| 07:03:02 | openstackgerrit | Merged openstack/keystoneauth master: Cache root urls with and without trailing slashes https://review.openstack.org/604635 | |
| 07:11:32 | Horrorcat | mordred, I was about to report it as a bug, but I’m totally lost on the storyboard thing. Is there a primer on how to use it to report issues? | |
| 09:48:05 | openstackgerrit | Samuel de Medeiros Queiroz proposed openstack/openstacksdk master: Allow returning munch on create and get object https://review.openstack.org/597198 | |
| 10:08:27 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Set endpoint_override from endpoint with noauth https://review.openstack.org/604415 | |
| 10:08:27 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Remove profile https://review.openstack.org/604414 | |
| 10:08:28 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Remove api version default values https://review.openstack.org/600330 | |
| 10:08:28 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Use discovery instead of config to create proxies https://review.openstack.org/600292 | |
| 10:08:29 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Special case network and use it in openstack.cloud https://review.openstack.org/604645 | |
| 10:43:38 | openstackgerrit | Matthias Lisin proposed openstack/openstacksdk master: Support firewall service for SDK https://review.openstack.org/592303 | |
| 10:44:45 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: WIP Replace TaskManager with a semaphore and lock https://review.openstack.org/604926 | |
| 10:53:01 | openstackgerrit | Matthias Lisin proposed openstack/openstacksdk master: openstackcloud.py: Implement FWaaS wrapper methods. https://review.openstack.org/604811 | |
| 10:59:04 | openstackgerrit | Samuel de Medeiros Queiroz proposed openstack/openstacksdk master: Allow returning munch on create and get object https://review.openstack.org/597198 | |
| 11:24:17 | openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Reformat Adapter docstring https://review.openstack.org/605042 | |
| 11:24:18 | openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Add support for client-side rate limiting https://review.openstack.org/605043 | |
| 11:42:12 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Add NODE and HOST parameters in "server create" help text https://review.openstack.org/591944 | |
| 12:15:02 | samueldmq | morning sdk | |
| 12:15:18 | samueldmq | mordred: o/ | |
| 12:15:40 | samueldmq | mordred: I'm not getting what you're saying here https://review.openstack.org/#/c/596546 | |
| 12:19:15 | openstackgerrit | Monty Taylor proposed openstack-infra/shade master: Remove the task manager https://review.openstack.org/605052 | |
| 12:20:56 | mordred | morning samueldmq ! thanks for all the great patches you've been working on | |
| 12:21:39 | mordred | samueldmq: oh - what I'm saying is that we don't need get_container to return a munch of its contents - there is another call that does that already, it's list_objects | |
| 12:21:49 | samueldmq | mordred: my pleasure :-) | |
| 12:21:56 | mordred | (containers and objects are ... weird) | |
| 12:23:53 | mordred | so get_container returns the information about the container object itself - which is the stuff returned in the headers | |
| 12:23:58 | samueldmq | in the get_container too.. | |
| 12:24:02 | samueldmq | I just want to return 'bytes', 'count', 'last_modified', 'location', 'name', 'properties' | |
| 12:27:39 | mordred | yes - I think that's a great idea ... but I think you can normalize that from the headers that are already being fetched | |
| 12:28:23 | mordred | (and then obviously you'll need to leave the old ugly header names in if strict=False) | |
| 12:28:40 | mordred | samueldmq: storage_policy looks like another good one to have | |
| 12:32:18 | mordred | samueldmq: http://paste.openstack.org/show/730697/ there's a payload from a container we have for infra for logs | |
| 12:36:54 | openstackgerrit | Pavlo Shchelokovskyy proposed openstack/python-openstackclient master: Add --name-lookup-one-by-one option to server list https://review.openstack.org/568345 | |
| 12:41:36 | samueldmq | mordred: hmm sure, so perhaps I can leverage all that from the headers... | |
| 12:41:56 | samueldmq | mordred: so I'm doing /account/ to list containers and /account/container to list objects | |
| 12:42:21 | samueldmq | perhps I can do just HEAD on hte container and object and normalize what I get, instead of always listing everything to get one | |
| 12:42:51 | mordred | samueldmq: yes - I think for swift this is a good idea | |
| 12:43:31 | samueldmq | mordred: kk I've validated sdk against vexxhost, ecs, switchengines, ormuco, catalyst and devstack (pretty obvious but ok for my msc) | |
| 12:44:13 | mordred | samueldmq: neat! | |
| 12:44:18 | samueldmq | mordred: it's behaving all fine, except glance v1 which is entirely broken in sdk | |
| 12:44:36 | mordred | samueldmq: oh golly, really? | |
| 12:44:40 | samueldmq | we're passing info in the req body, when glance v1 expects things in the headers | |
| 12:44:50 | samueldmq | e.g disk_format | |
| 12:44:56 | samueldmq | mordred: yes :-) | |
| 12:45:22 | samueldmq | interesting part was to figure it out, since even docs for glance v1 are hard to find | |
| 12:45:40 | mordred | samueldmq: I see us setting headers in _update_image_properties_v1 ... | |
| 12:46:12 | samueldmq | mordred: but not for the upload using put v1 | |
| 12:46:45 | mordred | oh - you are so right! | |
| 12:46:46 | samueldmq | mordred: https://github.com/openstack/openstacksdk/blob/master/openstack/cloud/openstackcloud.py#L4831 | |
| 12:47:43 | mordred | yup | |
| 12:48:06 | samueldmq | mordred: that was a good catch | |
| 12:48:20 | mordred | ++ ... testing against real things is important :) | |
| 12:48:29 | mordred | samueldmq: did you find any clouds with glance v1 still? | |
| 12:48:40 | samueldmq | mordred: ah, besides all that, I have patches for normalizing network/subnet container and objects | |
| 12:48:59 | samueldmq | so they have a location and properties fields in it | |
| 12:49:16 | mordred | yes! I saw these and am excited by them (haven't fully looked at them yet, but will try to today) | |
| 12:49:18 | samueldmq | and look the same regarless what cloud is underneath | |
| 12:49:39 | samueldmq | mordred: they're not passing local tests yet, because unitt tests were not expecting too many fields | |
| 12:50:06 | samueldmq | normalize is important to everyhting... but for now I focused on these in my scope :( | |
| 12:50:29 | samueldmq | mordred: glance v1 was interesting.. nobody runs it by default, but some still have, e.g catalyst | |
| 12:50:44 | samueldmq | I had to set image_endpoint explicitly i nthe clouds.yml | |
| 12:50:53 | samueldmq | so it uses endpoint_override (iirc) | |
| 12:54:35 | mordred | well - good to hear nobody is using it by default :) | |
| 12:54:49 | samueldmq | hehe no ... | |
| 12:54:56 | mordred | samueldmq: so - when you're ready to really poke yourself in the eye ... | |
| 12:55:00 | samueldmq | btw I'll update vendor _support later on | |
| 12:56:27 | mordred | samueldmq: one of the things on the todo list, which is very similar to your normalize work | |
| 12:57:14 | mordred | samueldmq: is to get the shade layer to start using the underlying sdk layer - the sdk Resource objects are the equiv of the _normalize methods in the shade layer | |
| 12:57:39 | openstackgerrit | Merged openstack/openstacksdk master: Update the URL in doc https://review.openstack.org/604738 | |
| 12:57:40 | mordred | samueldmq: it's a very hand wavey project - tons of questions about what it wants to look like and whatnot | |
| 12:58:36 | samueldmq | mordred: well, yes but I think there might be a few differences between shade normalize and the resource objs | |
| 12:58:55 | mordred | and like - taking image for example - I think we might want to move the _v1 methods into openstack/image/v1/_proxy and the _v2 methods into openstack/image/v2/_proxy.py - and then make sure both proxy classes have a create_image method that takes the same arguments - and then we can just have the shade layer create_image method call self.image.create_image(*args, **kwargs) | |
| 12:58:56 | samueldmq | I don't remember an exemp out of my head but I think I've taken a look into that | |
| 12:59:02 | samueldmq | but definitely something to do | |
| 12:59:16 | mordred | samueldmq: yes - there definitely are | |
| 12:59:39 | samueldmq | mordred: nice (image eg.) | |
| 12:59:39 | mordred | samueldmq: it's also possible that the upload_image example I just wrote about might be a TERRIBLE way to go about doing things :) | |