openstackgerrit | Merged openstack/placement master: Adds py38 functional tests to gate https://review.opendev.org/751168 | 01:37 |
---|---|---|
*** zzzeek has quit IRC | 03:24 | |
*** zzzeek has joined #openstack-placement | 03:25 | |
*** evrardjp has quit IRC | 05:33 | |
*** evrardjp has joined #openstack-placement | 05:33 | |
*** spatel has joined #openstack-placement | 07:58 | |
*** spatel has quit IRC | 08:03 | |
*** spatel has joined #openstack-placement | 10:59 | |
*** spatel has quit IRC | 11:04 | |
*** spatel has joined #openstack-placement | 12:27 | |
*** spatel has quit IRC | 12:32 | |
openstackgerrit | Balazs Gibizer proposed openstack/placement master: Add functional-py3[89] tox targets https://review.opendev.org/760890 | 12:42 |
openstackgerrit | Balazs Gibizer proposed openstack/osc-placement master: Add functional-py39 tox target https://review.opendev.org/760912 | 12:44 |
*** cdent has joined #openstack-placement | 12:58 | |
*** spatel has joined #openstack-placement | 14:15 | |
*** martinkennelly has joined #openstack-placement | 17:47 | |
*** mugsie has quit IRC | 18:18 | |
*** mugsie has joined #openstack-placement | 18:21 | |
*** janice61 has quit IRC | 18:56 | |
*** janice61 has joined #openstack-placement | 19:03 | |
lbragstad | hey all, is there a way to tell gabbi to report captured logged during a functional run? i see the APIFixture is capturing stdout and logging, but i can't seem to find a way to tell gabbi to report those things during a failed test run | 19:12 |
lbragstad | captured logging* | 19:12 |
cdent | ahoy lbragstad | 19:50 |
cdent | Is this in the placement tests or somewhere else? | 19:51 |
lbragstad | yeah - specifically the placement functional API test built using gabbi | 19:51 |
lbragstad | tests* | 19:52 |
lbragstad | for context - i'm working on some tests for https://review.opendev.org/#/c/760235/ | 19:52 |
cdent | Hmmm. It’s supposed to be using the capture.Logging inner fixture for that (see in tests/functional/test_api.py | 19:52 |
cdent | and output.CaptureOutput | 19:52 |
cdent | if it’s not working there’s at least two possibilities: something changed … somewhere | 19:53 |
lbragstad | and after reading the contributor guide, it sounded like the functional api tests were the right spot for that sort of thing | 19:53 |
cdent | or the test is failing in too catastrophic a fashion | 19:53 |
cdent | or is that you’re after looking at the logs from within the test? | 19:54 |
lbragstad | not specifically - my tests aren't asserting anything in the logs, yet | 19:54 |
lbragstad | i'm just trying to use the logger to figure out the state of specific things during the request | 19:54 |
lbragstad | for example - http://paste.openstack.org/show/799708/ | 19:54 |
lbragstad | and since the tests appear to be capturing stdout and the logs, i figured it would be output during the test run | 19:55 |
cdent | does it make any difference if you switch to a higher log level? | 19:56 |
cdent | because of stestr you’ll only see the output if there’s a failure | 19:57 |
cdent | s/because/also because/ | 19:57 |
lbragstad | i'll try that quick | 19:57 |
lbragstad | cdent a higher log level doesn't appear to make a difference (i used debug, error, and critical) | 19:59 |
cdent | and this is on a failing test? | 20:00 |
lbragstad | the test is failing, yes | 20:00 |
lbragstad | this is what i have locally http://pastebin.test.redhat.com/915656 on top of 760235 | 20:02 |
cdent | try putting your log message somewhere other than in policy.py as the default fixutres may not be touching policy depending on which tests you’ve got going on | 20:04 |
*** artom has quit IRC | 20:04 | |
cdent | tests/functional/fixtures/gabbits.py has some stuff on which things get engaged for which fixtures | 20:05 |
cdent | ( can’t see that url ) | 20:05 |
lbragstad | oh - dang | 20:05 |
cdent | If you leave more info here, I’ll catch up later, but I’m a bit in the middle of something | 20:05 |
lbragstad | sorry - i'll repaste | 20:05 |
cdent | but would like to make sure this works, since it should… | 20:05 |
cdent | so can check back a bit later in the day | 20:06 |
lbragstad | http://paste.openstack.org/show/799709/ | 20:06 |
lbragstad | no worries - it's not a huge rush | 20:06 |
cdent | ✔ | 20:06 |
cdent | lbragstad: it’s because of the version of gabbi. In 2.0.4 or thereabouts I removed testtools and the the olsotest capture fixtures that do output capture expect that to be there (but don’t actually require it). Basically testtools has some “gather up the data at the end of the test” routines | 21:43 |
cdent | I removed testtools because it causes a dependency cycle in some contexts | 21:43 |
lbragstad | ahh | 21:44 |
lbragstad | interesting | 21:44 |
cdent | What I meant to do what include most of the relevant testtools functionality in the TestCase subclass that gabbi carries itself | 21:44 |
cdent | but apparently I left that bit out | 21:44 |
cdent | this is the pull request that remove the support: https://github.com/cdent/gabbi/pull/279 | 21:45 |
cdent | there’s a typo : Please where I meant “Pleas” | 21:45 |
cdent | I’ll make a bug and see about fixing it, soon. In the meantime if you drop down to gabbi<2 it should be fine | 21:46 |
lbragstad | ok - i'll give that a shot locally quick | 21:47 |
cdent | lemme know | 21:47 |
lbragstad | cdent excellent http://paste.openstack.org/show/799713/ | 21:50 |
cdent | yah, makes sense. I pinged you from the newly created issue | 21:51 |
lbragstad | cdent perfect, thank you for the help | 21:54 |
*** artom has joined #openstack-placement | 21:54 | |
cdent | lbragstad: keeping placement on <2 will probably be safe/fine. Most of the changes have been to remove python 2isms and make some minor tweaks here and there. The only change that could be relevant is this one https://github.com/cdent/gabbi/commit/c5b08a1fce4ee7be40aa2d2a2521fbe8c434d279 but in placement the headers are always the same so… | 21:55 |
cdent | But I’ll try to fix the bug in any case. In my copious free time and all that. | 21:55 |
lbragstad | :) | 21:55 |
lbragstad | yeah - i should be able to just use gabbi < 2 for investigative purposes, then i hop to get the tests working (which shouldn't be specific to gabbi < 2) | 21:56 |
lbragstad | s/hop/hope/ | 21:56 |
* cdent hops | 21:56 | |
*** artom has quit IRC | 22:02 | |
*** artom has joined #openstack-placement | 22:03 | |
*** spatel has quit IRC | 22:06 | |
*** cdent has quit IRC | 22:51 | |
*** spatel has joined #openstack-placement | 23:12 | |
*** spatel has quit IRC | 23:17 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!