| gmaan | afaik, nova is the first one to add tempest coverager for RBAC at least for existing tests to move to new defaults. I know there are some rbac test in cinder and keystone tempest plugins but those are checking access positive and negative testing at DB layer | 00:24 |
|---|---|---|
| gmaan | at some point, we need to move all tempest tests to new defaults (least access role) which is in my todo list | 00:25 |
| *** mhen_ is now known as mhen | 01:43 | |
| opendevreview | Douglas Mendizábal proposed openstack/devstack master: Set password regex for security compliance https://review.opendev.org/c/openstack/devstack/+/956111 | 04:17 |
| opendevreview | Merged openstack/tempest master: Add new config option for nova policy available defaults https://review.opendev.org/c/openstack/tempest/+/953265 | 04:21 |
| gibi | it seem the openstacksdk-functional-devstack job is failing constantly and blocking the gate https://zuul.opendev.org/t/openstack/builds?job_name=openstacksdk-functional-devstack&skip=0 | 12:46 |
| gibi | openstack.exceptions.SDKException: Image creation failed: 'latin-1' codec can't encode character '\u2019' in position 662: ordinal not in range(256) | 12:46 |
| gibi | in openstack.tests.functional.image.v2.test_image.TestImage.test_tags | 12:46 |
| gibi | it seems it is started failing earier today | 12:48 |
| frickler | stephenfin: ^^ can you take a look? https://review.opendev.org/c/openstack/openstacksdk/+/910218 looks maybe related just from timing and touched code, though the job was passing on that change | 13:08 |
| stephenfin | looking | 13:09 |
| stephenfin | yeah that fails locally. wtf did it not fail in the gate... | 13:12 |
| stephenfin | gibi: frickler: apparently it's this change 😅 https://review.opendev.org/c/openstack/openstacksdk/+/956329 | 13:21 |
| stephenfin | ah, we consume CONTRIBUTING.rst as part of the image test. TIL. | 13:24 |
| frickler | fancy. also I spot a wrong project name in that change | 13:27 |
| gibi | stephenfin: thanks for taking a look | 13:30 |
| stephenfin | https://review.opendev.org/c/openstack/openstacksdk/+/956369 | 13:34 |
| frickler | thx. another point to note is that devstack declares all *.rst files as irrelevant, which is why the functional job wasn't run on the change. maybe using some dedicated data file might be better? | 14:16 |
| stephenfin | good point | 14:28 |
| gmaan | yeah. As this is running in many projects gatre, it make sense to run it in \sdk all changes, just for safer side. | 15:58 |
| opendevreview | Goutham Pacha Ravi proposed openstack/eslint-config-openstack master: Replace CLA instructions with DCO https://review.opendev.org/c/openstack/eslint-config-openstack/+/956391 | 17:47 |
| sean-k-mooney | gmaan: if you have time can you comment on https://review.opendev.org/c/openstack/watcher/+/956380/comment/3614ac2f_3a8bb840/ | 18:03 |
| sean-k-mooney | gmaan: correct me if im wrong but cant you make temest test detech the microverion of a given service and skip automaticly | 18:04 |
| sean-k-mooney | without needing to configure the microverion in the servies config section | 18:04 |
| sean-k-mooney | i.e. we shoudl be able to wrte temest test on master that wil skp on stble based on if that deployment supprot that microversion | 18:05 |
| sean-k-mooney | like this https://github.com/openstack/tempest/blob/master/tempest/api/compute/servers/test_servers.py#L321-L322 | 18:07 |
| gmaan | sean-k-mooney: yes, that is why we need to cap the test with min or max microversion we want to run those tests. and based on deployment available microversion test will be run or skipped | 18:07 |
| sean-k-mooney | ok looking at https://github.com/openstack/tempest/blob/master/tempest/api/compute/base.py#L49-L65 | 18:08 |
| gmaan | but for tempest to decide available microversionm, we need to configure the service min and max available microversion of partilular running env | 18:08 |
| sean-k-mooney | this is not auto discoverign form the api endpont | 18:08 |
| sean-k-mooney | its looking at the cofnig for the limits? | 18:08 |
| gmaan | we do not do auto discovery from service, it needs to be configured | 18:08 |
| sean-k-mooney | oh ok | 18:09 |
| sean-k-mooney | so dougs change makes sense then | 18:09 |
| sean-k-mooney | does devstack do this for use automaticlly on stable branches? | 18:09 |
| sean-k-mooney | i assume not we have to do it in the job or in the devstack plugin liek doug is propsoing | 18:09 |
| gmaan | I will check those but if he is just configuring the min/max service microversion then yes, for master it can be min=None and max=latest and for stable you can change max from latest->max-supported-version | 18:09 |
| gmaan | we do in devstack\ | 18:10 |
| sean-k-mooney | ah ok | 18:10 |
| sean-k-mooney | that explians it then anytime i have ran tempest on stable i have never needed to manually set the max verison in nova | 18:10 |
| gmaan | sean-k-mooney: this is how I do for stable branch when they are created https://review.opendev.org/c/openstack/devstack/+/945240 | 18:10 |
| gmaan | sean-k-mooney: I did for you always :) | 18:11 |
| sean-k-mooney | :) yep | 18:11 |
| sean-k-mooney | well that and the fact that i rarely run all of tempest locally jsut specific test is a factor | 18:11 |
| gmaan | we can do auto discvoery also but that is something I want to avoid in tempest. That can make mistake about wrongly configure the env or not detecting/testing the expected tests to run | 18:12 |
| sean-k-mooney | ya that fine | 18:13 |
| sean-k-mooney | i just tought this worked by 1 declering the requiremetns on the test | 18:13 |
| sean-k-mooney | which we do | 18:13 |
| sean-k-mooney | and 2 having the test check that via hiting hte microversion endpoint | 18:13 |
| sean-k-mooney | i tought the min/max wer optional overrieds | 18:14 |
| sean-k-mooney | rather then the primary way of contoling it | 18:14 |
| sean-k-mooney | thanks for explaining that | 18:14 |
| gmaan | yeah, test doing it is extrta work for every test to discover it from endpoint | 18:14 |
| sean-k-mooney | ya i kind of assumed it woudl be cached in the tempest clients but ya it makes sense the way it is | 18:15 |
| sean-k-mooney | or why its the way it si rather | 18:15 |
| gmaan | replied in gerrit | 18:19 |
| sean-k-mooney | thanks | 18:19 |
| sean-k-mooney | im going to call it a day, monday is a public holiday so ill chat to you on tuesday | 18:20 |
| sean-k-mooney | enjoy the weekend | 18:20 |
| gmaan | have a nice long weekend, same for me. Monday is public holiday here | 18:20 |
| -opendevstatus- NOTICE: All OpenDev IRC and Matrix bots will be offline briefly (including channel logging and meeting management) in order to move them to a newer server | 23:50 | |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!