| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2019-12-17 | |||
| 13:57:18 | gtema | exactly, that's why in stdin case I'm passing validate_checksum as False | |
| 13:58:03 | gtema | surely it's sad, but I have no clue how else can we workaround this | |
| 13:58:20 | mordred | oh - nod. yeah | |
| 13:58:38 | mordred | gtema: I think that's just life if you decide that for some reason you need to stream the image over stdin | |
| 13:58:46 | gtema | exactly | |
| 14:00:11 | gtema | so basically if we are in stdin we should not come into this calculation logic due to if isinstance(data, bytes) | |
| 14:01:25 | mordred | gtema: that patch looks great - definitely less scary than I was thinking it would be | |
| 14:01:29 | mordred | oh - wait | |
| 14:01:35 | mordred | you should add a release note for that perhaps | |
| 14:01:41 | gtema | me as well | |
| 14:01:44 | gtema | ok | |
| 14:02:02 | mordred | since that'll be a key feature osc depends on | |
| 14:03:13 | Shrews | mordred: will look when I return from the gym | |
| 14:03:19 | mordred | Shrews: ew. gym | |
| 14:05:15 | openstackgerrit | Artem Goncharov proposed openstack/openstacksdk master: Support uploading image from data and stdin https://review.opendev.org/698972 | |
| 14:05:20 | gtema | added RN only | |
| 14:07:31 | mordred | gtema: also - https://review.opendev.org/#/c/679914 looks good to me when you have a sec | |
| 14:07:49 | gtema | okay | |
| 14:12:08 | gtema | mordred: having a brief look there - in resources attributes are not in alphabetical order. While don't want to be neat-picky, do we want to get it this way? | |
| 14:13:44 | gtema | if it's ok for me - I can skip this comment | |
| 14:14:18 | gtema | I mean if ok for you | |
| 14:39:32 | mordred | gtema: I don't know that I care too much - I doubt a ton of people are going to be adding new attributes in parallel | |
| 14:39:47 | mordred | so I doubt we'll have too many merge conflicts | |
| 14:40:32 | gtema | it'S not about conflicts, it's more about keeping code clean. Especially if you look at image resources, where list of attrs take few pages - it'S really hard to orient | |
| 14:44:41 | mordred | nod. I'm also supportive of that comment if you want to make it | |
| 14:46:02 | gtema | Let us then +A this change with comment to order attributes. And in future really take care of this during review | |
| 14:46:11 | gtema | dtantsur - comments? | |
| 14:47:02 | mordred | gtema: ++ | |
| 14:49:30 | openstackgerrit | Vishakha Agarwal proposed openstack/keystoneauth master: [ussuri][goal] Drop python 2.7 support and testing https://review.opendev.org/699121 | |
| 15:54:56 | gtema | thanks Shrews | |
| 15:55:05 | Shrews | np | |
| 17:12:34 | openstackgerrit | Keith Berger proposed openstack/python-openstackclient master: Fix openstack server list --deleted --marker option https://review.opendev.org/691045 | |
| 20:02:31 | openstackgerrit | Matt Riedemann proposed openstack/python-openstackclient master: Fix openstack server list --deleted --marker option https://review.opendev.org/691045 | |
| 20:34:06 | openstackgerrit | Merged openstack/openstacksdk master: Support uploading image from data and stdin https://review.opendev.org/698972 | |
| 22:59:53 | openstackgerrit | Merged openstack/openstacksdk master: Stop supporting python2 https://review.opendev.org/699132 | |
| 23:36:26 | openstackgerrit | Merged openstack/openstacksdk master: Remove extra python2 test jobs https://review.opendev.org/699133 | |
| 23:47:43 | openstackgerrit | Merged openstack/openstacksdk master: Switch stable ansible job to 2.8 https://review.opendev.org/699134 | |
| #openstack-sdks - 2019-12-18 | |||
| 07:28:56 | openstackgerrit | wangfaxin proposed openstack/api-sig master: update reviewer url. https://review.opendev.org/699570 | |
| 10:23:03 | openstackgerrit | Bo Tran proposed openstack/openstacksdk master: Add return cluster_id when query actions list https://review.opendev.org/699650 | |
| 11:56:53 | sshnaidm | mordred, dtantsur Shrews pabelanger gtema odyssey4me and all others I'd like to have your opinions about the subject, in your time: http://lists.openstack.org/pipermail/openstack-discuss/2019-December/011597.html | |
| 11:58:41 | gtema | ack | |
| 13:48:00 | openstackgerrit | Merged openstack/cliff master: Add autoprogram_cliff_app_dist_name config opt https://review.opendev.org/692464 | |
| 18:26:19 | openstackgerrit | Merged openstack/openstacksdk master: Add dns_publish_fixed_ip attribute to subnets https://review.opendev.org/679833 | |
| 18:46:40 | openstackgerrit | David Shaughnessy proposed openstack/python-openstackclient master: [POC] Neutron Classifier CLI Mock up of Neutron Classifier CLI https://review.opendev.org/687256 | |
| 18:46:41 | openstackgerrit | David Shaughnessy proposed openstack/python-openstackclient master: Extension to the DSCP Marking Rule This extends the CLI to include the classification_group_id and the classification_priority value for deciding the relitive priority of the rules. https://review.opendev.org/699832 | |
| #openstack-sdks - 2019-12-19 | |||
| 02:17:19 | adriant | mordred, it appears that the SDK is busted when setting metadata on objectstorage container creation | |
| 02:17:46 | adriant | none of the header values are being marked as dirty, even though the resource does have them | |
| 02:18:19 | adriant | I have a feeling the issue is because of: https://github.com/openstack/openstacksdk/blob/master/openstack/object_store/v1/container.py#L116 | |
| 02:18:37 | adriant | any reason why you are setting "_synchronized=True" ? | |
| 02:18:45 | adriant | when by default, it never is? | |
| 02:20:24 | adriant | when I do: conn.object_store.create_container("test_5", read_ACL="stuff") | |
| 02:20:43 | adriant | I'd expect that read_ACL to be set, but it isn't, because while it does get added to resource headers, it isn't dirty | |
| 02:22:05 | adriant | yeah... based on your own docs. this is wrong: https://github.com/openstack/openstacksdk/blob/master/openstack/resource.py#L876-L880 | |
| 02:22:31 | adriant | _synchronized should be false | |
| 02:24:27 | adriant | Was that new function meant to actually be existing? because wouldn't the ID only be present when it already exists? | |
| 02:46:28 | openstackgerrit | Adrian Turjak proposed openstack/openstacksdk master: Add support for object storage container policies https://review.opendev.org/699905 | |
| 02:46:28 | openstackgerrit | Adrian Turjak proposed openstack/openstacksdk master: Fix bug in object storage container creation setting metadata https://review.opendev.org/699904 | |
| 02:47:45 | adriant | mordred: ^ have some more swift fixes for the SDK :P | |
| 02:48:13 | adriant | I feel like I find the most bugs in the SDK when dealing with swift because I'm the only person who tries to do swift things with the SDK... | |
| 03:22:17 | adriant | mordred: I added comments on those patches, the tests are failing because there is no way for the resource fields to be marked as: "pull from body, but ignore for update" because name/count etc can't be updated | |
| 03:22:43 | adriant | but if they are dirty, the resource code will include them in the body | |
| 06:45:05 | openstackgerrit | wangfaxin proposed openstack/js-openstack-lib master: Update review community page https://review.opendev.org/699919 | |
| 08:52:54 | openstackgerrit | Bram Verschueren proposed openstack/python-openstackclient master: Fix faulthy state argument choice https://review.opendev.org/699931 | |
| 09:59:34 | openstackgerrit | Bence Romsics proposed openstack/python-openstackclient master: Add command: router add/remove route --route https://review.opendev.org/674325 | |
| 12:40:55 | openstackgerrit | Bence Romsics proposed openstack/openstacksdk master: Handle HTTP errors in add/remove router interface calls https://review.opendev.org/687304 | |
| 13:47:56 | elmiko | dtantsur: i'm thinking about cancelling the next 2 api sig office hours, do you have any feelings about the matter? | |
| 13:54:30 | stephenfin | mordred: I want to do some work on closing the gap between novaclient and openstackclient next year, and I'm thinking I'll use openstacksdk to do that which has many of its own gaps that need closing | |
| 13:54:51 | stephenfin | mordred: Would it be a huge ask for you to provide a sample patch for one of nova's microversioned APIs that I can use as a blueprint? | |
| 13:55:06 | stephenfin | You or someone else involved with openstacksdk (I only know of you :)) | |
| 13:55:32 | gtema | stephenfin - welcome on board. I'm currently working on "repairing" image support in the OSC | |
| 13:55:43 | gtema | and was thinking also to go next for compute | |
| 13:55:53 | gtema | so let's work on this together | |
| 13:56:11 | stephenfin | Great! | |
| 13:56:58 | gtema | I recently made a patch for managing server tags in OSC using SDK, and the next step might be to switch server set/unset operations to use only SDK | |
| 13:57:21 | gtema | with that we can slowly switch parts of the compute, which make it easier | |
| 13:57:32 | stephenfin | gtema: I assume you're doing these repairs using openstacksdk, or are you using glanceclient? | |
| 13:57:46 | gtema | I am replacing glanceclient with SDK | |
| 13:57:57 | stephenfin | Cool | |
| 13:58:08 | stephenfin | The main issue I saw when I started looking at this was that there weren't an clear patterns around for microversioned APIs | |
| 13:58:32 | gtema | I would say it should be considered case by case | |
| 13:58:37 | stephenfin | we use decorators in novaclient but there doesn't seem to be an equivalent pattern for sdk | |
| 13:59:16 | gtema | in the image there is nothing like that. In the tagging of servers there is some first brief example for that, but it doesn't really enforce those | |
| 13:59:39 | stephenfin | so it sounds like we need to invent one? :) | |
| 13:59:59 | stephenfin | that's what I was hoping to drag mordred in for | |
| 14:00:05 | gtema | not really invent support, but perhaps invent pattern how to use it properly | |
| 14:00:11 | stephenfin | yeah, exactly | |
| 14:00:27 | gtema | actually AFAIK in the ironic world this should have been used a lot already | |
| 14:00:33 | stephenfin | I mean, I could just copy-paste the patterns from novaclient, but those aren't necessarily going to align with how everyone else is doing it | |
| 14:00:54 | stephenfin | I looked at baremetal, yeah. Support seemed pretty barebones still, unfortunately :( | |
| 14:01:01 | gtema | in SDK we have option just to specify specific microversion on the request level | |
| 14:01:06 | stephenfin | and wasn't consistent | |
| 14:01:21 | gtema | so I would say let's look to concrete cases | |
| 14:02:56 | stephenfin | Something as simple as migrating this https://github.com/openstack/python-novaclient/blob/eb98178ea88c77881f7c763b068408f769956eb2/novaclient/v2/servers.py#L217-L234 | |
| 14:03:45 | stephenfin | as an examples of an API that accepts different things or otherwise behaves differently depending on the microversion | |
| 14:04:32 | stephenfin | I'll give it some more thought though. You have a specific Gerrit tag for your image service rework, gtema? | |
| 14:04:33 | gtema | ok, so do you "need" to specify microversion in order to lock the server with the reason? | |
| 14:04:45 | gtema | osc-first | |
| 14:04:48 | stephenfin | You do, yeah | |
| 14:04:58 | gtema | https://review.opendev.org/#/c/650374/ | |
| 14:05:13 | stephenfin | If you make the request without the version, nova will default to version 2.1 and the schema check will fail | |