| *** chkumar|off has quit IRC | 05:32 | |
| *** kopecmartin has joined #refstack | 06:23 | |
| *** tosky has joined #refstack | 07:37 | |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 08:24 |
|---|---|---|
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 09:22 |
| *** chandankumar has joined #refstack | 10:06 | |
| chandankumar | arxcruz: Hello | 10:06 |
| chandankumar | arxcruz: I left a comment here https://review.openstack.org/#/c/567820/ please have a look | 10:06 |
| chandankumar | arxcruz: I am using your patch as a depends -on refstack patch https://review.openstack.org/#/c/541273/ | 10:07 |
| arxcruz | chandankumar: regarding https://review.openstack.org/#/c/567820/ my concern was when people run it outside the environment, it was failing | 10:11 |
| arxcruz | chandankumar: because img_dir default is just etc | 10:12 |
| chandankumar | arxcruz: in refstack, we are running from outside the env, where there is no tempest workspace | 10:12 |
| arxcruz | so it gets in tempestconf code the $workspace ($home) and the etc | 10:12 |
| arxcruz | so it's /home/user/etc | 10:12 |
| arxcruz | and it fails to download the image because /home/user/etc doesn't exist | 10:12 |
| kopecmartin | chandankumar, you're cherypicking my patch, but it seems, you also need arxcruz's changes, right? so I'll make my patch dependent on arxcruz's patches | 10:26 |
| kopecmartin | if i get it right, it should solve the issue | 10:26 |
| arxcruz | yeah, i'm reviewing the code, and image_path should always have a value | 10:27 |
| arxcruz | https://review.openstack.org/#/c/562672/11/config_tempest/main.py line 345 | 10:27 |
| openstackgerrit | Chandan Kumar proposed openstack/python-tempestconf master: [DNM] Allow image creation by default for all users https://review.openstack.org/568252 | 10:28 |
| arxcruz | chandankumar: i think this will fail | 10:32 |
| chandankumar | arxcruz: which one? | 10:32 |
| chandankumar | arxcruz: DNM one? | 10:32 |
| arxcruz | you're removing an option from a function/method but not updating the other code that calls set_image_preferences | 10:32 |
| arxcruz | chandankumar: yup | 10:32 |
| chandankumar | arxcruz: yup sorry | 10:33 |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 10:34 |
| arxcruz | chandankumar: i'll not forgive you :P | 10:34 |
| openstackgerrit | Chandan Kumar proposed openstack/python-tempestconf master: [DNM] Allow image creation by default for all users https://review.openstack.org/568252 | 10:40 |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 10:42 |
| openstackgerrit | Chandan Kumar proposed openstack/python-tempestconf master: [DNM] Allow image creation by default for all users https://review.openstack.org/568252 | 10:55 |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 10:56 |
| openstackgerrit | Chandan Kumar proposed openstack/python-tempestconf master: [DNM] Allow image creation by default for all users https://review.openstack.org/568252 | 11:39 |
| kopecmartin | chandankumar, tosky, arxcruz , check the warning here https://docs.openstack.org/os-client-config/latest/index.html | 12:24 |
| kopecmartin | does it mean we should switch to the other module? | 12:24 |
| tosky | kopecmartin: will have, probably not "right now" | 12:29 |
| tosky | as usual, we should check various things: stability of the new API (since which version), version supported, etc | 12:30 |
| kopecmartin | tosky, ok, so should i create a card in our backlog to remember this? | 12:32 |
| tosky | kopecmartin: probably yes, especially if by "card" you mean "story on storyboard" | 12:34 |
| openstackgerrit | Chandan Kumar proposed openstack/python-tempestconf master: [DNM] Allow image creation by default for all users https://review.openstack.org/568252 | 12:41 |
| tosky | chandankumar: ^^ isn't it a copy of two other patches? | 12:46 |
| chandankumar | tosky: yup | 12:46 |
| chandankumar | tosky: too many dependent patches for making it working, I am merged two patches in my one patch | 12:46 |
| chandankumar | tosky: once things will work on refstack side, I will bring it back to normal | 12:46 |
| tosky | chandankumar: I disagree with the fact that it can't work with tons of dependent patches - it can | 12:49 |
| tosky | but *shrugs* | 12:49 |
| kopecmartin | tosky, chandankumar arxcruz these two tests are more failing then passing which blocks reviews.. can we blacklist them? https://paste.fedoraproject.org/paste/Vlxoe~GR-T8uBXHrqTAsCQ | 12:58 |
| tosky | kopecmartin: it seems a timeout issue; can you try to reduce the concurrency to 1? | 12:59 |
| kopecmartin | tosky, we can try that, I'll send a review | 13:01 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Reduce concurrency https://review.openstack.org/568286 | 13:04 |
| *** rmart04 has joined #refstack | 13:20 | |
| kopecmartin | chandankumar, I found a bug, i know why it fails on image_path, it's in my patch | 13:51 |
| chandankumar | kopecmartin: it is not taking defult | 13:52 |
| chandankumar | image name | 13:52 |
| kopecmartin | chandankumar, yes :/ , it seems None is not always None | 13:52 |
| kopecmartin | https://paste.fedoraproject.org/paste/26qBLjogP7iMWyPo~YuHQA | 13:52 |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 13:53 |
| chandankumar | tosky: does the microversion should also be sett by tempestconf tool? | 14:00 |
| tosky | chandankumar: I have no idea; can the python client discover the microversion? What are the guidelines followed by each upstream project which uses microversions? | 14:10 |
| kopecmartin | chandankumar, be aware, regarding your patch, if you specify an argument , the value you specify will be used, if you specify image_name to None, image_name is None .. so if you want to have default value, don't pass the argument at all | 14:12 |
| chandankumar | tosky: for compute for each api version changes tempest tests are added | 14:13 |
| chandankumar | tosky: https://docs.openstack.org/tempest/latest/microversion_testing.html#microversion-tests-implemented-in-tempest | 14:14 |
| tosky | chandankumar: those are the microversioning implemented, but then, what are the existing tests doing? Hardcoding the version used? | 14:16 |
| kopecmartin | tosky, setting concurrency to 1 didn't help :/ http://logs.openstack.org/86/568286/1/check/python-tempestconf-tempest-packstack-demo/f6b114f/job-output.txt.gz | 14:33 |
| tosky | kopecmartin: could it be a real bug? | 14:33 |
| kopecmartin | tosky, no idea, checking launchpad | 14:34 |
| tosky | and it is failing on packstack only, right? | 14:34 |
| kopecmartin | tosky, yes, i think so | 14:35 |
| tosky | is it there some timeout that can be tuned? | 14:36 |
| tosky | kopecmartin: stupid question: what is the status of ceilometer on both devstack and packstack deployments? | 14:38 |
| tosky | we may be a bit strict with resources | 14:38 |
| * kopecmartin checking | 14:38 | |
| kopecmartin | according to tempest.conf generated on packstack environment, ceilometer is Enabled | 14:39 |
| tosky | and on devstack? | 14:39 |
| tosky | it may be not be a problem on ceilometer itself, but it could be an issue with the resources on 8GiB | 14:40 |
| kopecmartin | there it's false | 14:40 |
| kopecmartin | not enabled | 14:40 |
| kopecmartin | ok, let's turn it off on packstack | 14:41 |
| kopecmartin | tosky, https://review.openstack.org/#/c/568306/ | 14:48 |
| tosky | kopecmartin: I guess we need a dependent patch to check | 14:49 |
| kopecmartin | working on that | 14:50 |
| tosky | kopecmartin: for example, the patch which reduces the concurrency could be retargeted | 14:50 |
| tosky | instead of creating a new one | 14:50 |
| tosky | patch recycling :) | 14:50 |
| kopecmartin | tosky, yeah .. if i'm not mistaken, I just need to add depends-on to commit message, right? | 14:51 |
| tosky | kopecmartin: yes | 14:51 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Reduce concurrency https://review.openstack.org/568286 | 14:52 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Make tempestconf easier to use as an library https://review.openstack.org/562672 | 14:52 |
| tosky | kopecmartin: you could also take https://review.openstack.org/568286 and restore the concurrency, and change something not so important somewhere else | 14:56 |
| chandankumar | tosky: I will find the more details tomorrow related to microversion and let you now | 14:58 |
| openstackgerrit | Chandan Kumar proposed openstack/refstack-client master: Generate tempest.conf automatically using refstack-client https://review.openstack.org/541273 | 15:14 |
| *** kopecmartin has quit IRC | 16:11 | |
| *** rmart04 has quit IRC | 16:33 | |
| *** tosky has quit IRC | 17:30 | |
| *** tosky has joined #refstack | 17:34 | |
| openstackgerrit | Luz Cazares proposed openstack/refstack-client master: Remove usage of tempest install_venv scripts https://review.openstack.org/564080 | 18:14 |
| *** Guest16323 is now known as mgagne | 18:40 | |
| *** mgagne has joined #refstack | 18:40 | |
| *** markvoelker_ has joined #refstack | 18:42 | |
| *** markvoelker has quit IRC | 18:43 | |
| *** markvoelker has joined #refstack | 18:49 | |
| *** markvoelker_ has quit IRC | 18:50 | |
| *** tosky has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!