| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-09-13 | |||
| 18:38:42 | mriedem | probably the walk migrations stuff | |
| 18:39:26 | mriedem | yeah there is a whitelist in filter_metadata_diff | |
| 18:39:56 | sean-k-mooney | the deprecation warnings at lesast are not coming form nova they are form sqlaclamate migrate | |
| 18:40:06 | sean-k-mooney | eventrulally we will need to move off that | |
| 18:40:08 | mriedem | yeah that's an old one | |
| 18:40:16 | mriedem | i disagree | |
| 18:40:55 | mriedem | just fix this https://bugs.launchpad.net/sqlalchemy-migrate/+bug/1814288 | |
| 18:40:56 | openstack | Launchpad bug 1814288 in sqlalchemy-migrate "DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead" [Medium,Confirmed] | |
| 18:41:25 | mriedem | sqlalchemy.migrate is opened by openstack, i'm a core, it's in maintenance mode, and changing nova to go from migrate to alembic is not worth the trouble | |
| 18:41:26 | sean-k-mooney | mriedem: didnt the maintainer say the did not want to maintain sqlalcamey migrate anymore | |
| 18:41:28 | mriedem | at least not while i'm around | |
| 18:41:36 | sean-k-mooney | oh ok | |
| 18:41:37 | mriedem | openstack has maintained it for like 5 years now | |
| 18:42:00 | mriedem | so patches welcome to fix https://bugs.launchpad.net/sqlalchemy-migrate/+bug/1814288 | |
| 18:42:01 | openstack | Launchpad bug 1814288 in sqlalchemy-migrate "DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead" [Medium,Confirmed] | |
| 18:42:15 | mriedem | migrate is probably due for a release in ussuri | |
| 18:42:57 | sean-k-mooney | well if i can squash some of those message simple then sure | |
| 18:43:45 | sean-k-mooney | thats the allcoate error right http://paste.openstack.org/show/775783/ | |
| 18:44:31 | sean-k-mooney | also http://paste.openstack.org/show/775785/ | |
| 18:45:56 | sean-k-mooney | the functional test seamed to pass but i know that the subunit thing is somewhat non determinisitc | |
| 18:51:05 | sean-k-mooney | huh when i open it in pycharm it sticts through getaragspec and tell you how to fix it | |
| 18:51:08 | openstackgerrit | melanie witt proposed openstack/nova-specs master: Amend "Configure max number of volumes to attach" spec https://review.opendev.org/682136 | |
| 18:51:42 | mriedem | i'm working on http://paste.openstack.org/show/775783/ | |
| 18:52:10 | sean-k-mooney | ok i was going to fix the migrate issue since it seam simpler | |
| 18:52:37 | mriedem | i can't quite figure out what's wrong with test_boot_reschedule_fill_provider_mapping_raises | |
| 18:52:38 | sean-k-mooney | also you sounded like you had an idea how to fix it | |
| 18:52:40 | mriedem | probably need gibi | |
| 18:52:59 | mriedem | using the SpawnIsSynchronous fixture in test_boot_reschedule_fill_provider_mapping_raises definitely makes it fail on that IndexError | |
| 18:53:02 | mriedem | i'm just not sure why yet | |
| 18:53:11 | mriedem | but i think it's because it's stubbing out fill_provider_mapping | |
| 18:53:51 | sean-k-mooney | ill open it and take a look but i dont know if ill be able to help i gues i can run it in a debugger and step in | |
| 18:56:15 | sean-k-mooney | or not... its one of those test where the debuger does not work | |
| 18:57:05 | artom | pysnoop | |
| 18:57:22 | sean-k-mooney | its the placmenet fixture | |
| 18:57:38 | sean-k-mooney | but i could use that i dont think it will help however | |
| 18:57:53 | mriedem | got it | |
| 19:08:54 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Use SpawnIsSynchronousFixture in reschedule functional tests https://review.opendev.org/682140 | |
| 19:08:55 | mriedem | sean-k-mooney: ^ | |
| 19:10:45 | sean-k-mooney | it seams to get rid fo the error | |
| 19:12:49 | sean-k-mooney | how is mock_fill in scope? | |
| 19:13:50 | sean-k-mooney | it refers to the mock you are creating in the with statemnet but you are using it in the stub fill provider mapping funciton | |
| 19:14:00 | mriedem | python magic | |
| 19:14:14 | mriedem | it's used in the function which isn't used until after the mock is setup | |
| 19:14:38 | mriedem | anyway, as for https://review.opendev.org/#/c/682133/ i'm going to do a full py26,func-py36,pep8 locally and if that's good clark is going to promote to the gate | |
| 19:14:47 | sean-k-mooney | right so its parsing the function signiture but not the body untill its used | |
| 19:14:53 | mriedem | yar | |
| 19:14:59 | sean-k-mooney | and by the time its use its inscope | |
| 19:14:59 | mriedem | that's something that always trips me up with python | |
| 19:15:17 | sean-k-mooney | that is slightly terrifying but ok | |
| 19:15:42 | artom | Wait, what? | |
| 19:16:07 | artom | The function can refer to its own mock? | |
| 19:16:11 | sean-k-mooney | dont look too closely https://review.opendev.org/#/c/682140/1/nova/tests/functional/test_servers.py@6808 | |
| 19:16:21 | sean-k-mooney | you might go insane | |
| 19:16:23 | artom | Too late, I clicked | |
| 19:16:42 | sean-k-mooney | but ya i get how it could work | |
| 19:16:43 | artom | There's gotta be a less funky clever way of doing that | |
| 19:17:11 | sean-k-mooney | i think there is a mock.wraps that does something similar | |
| 19:17:20 | artom | Yeah | |
| 19:17:26 | artom | Or even with side_effect | |
| 19:17:56 | artom | Something like side_effect=[original_fill] + itertools.repeat(raise) | |
| 19:17:59 | sean-k-mooney | it is a side effect | |
| 19:18:07 | sean-k-mooney | that is how the fuction is used | |
| 19:18:20 | sean-k-mooney | *stub_fuction | |
| 19:18:36 | artom | Right, I meant... we don't need the weirdly scoped function introspecting itself thing | |
| 19:18:38 | sean-k-mooney | with all that said what mriedem worte appeares to work | |
| 19:18:46 | artom | Not that it's not clever | |
| 19:18:59 | artom | It'll just cause headaches to whoever has to work with it next | |
| 19:26:16 | artom | o/ see ya Monday | |
| 19:27:44 | sean-k-mooney | oh i see what its doing its filling the provider mapping if tis not filled | |
| 19:28:25 | sean-k-mooney | and raise an exception if called again | |
| 20:51:31 | mriedem | sean-k-mooney: heh, of course now the libvirt logging patch failed in the gate b/c of the functional notifications patch | |
| 20:54:24 | sean-k-mooney | nothing else would do sure :) i has to keep up the tradtion of the rest | |
| 20:56:58 | mriedem | corvus is going to promote both | |
| 20:59:03 | sean-k-mooney | if they fail again we might need to squash them. the failures are not determisitic so we should be able to merge them on there own but we might not want to retry them that much | |
| 21:09:43 | efried | o/ | |
| 21:09:45 | efried | checking in | |
| 21:09:57 | efried | I assume we're holding rechecks until we can merge some of those gate fixes mriedem? | |
| 21:10:09 | efried | I've got eight or so lined up. | |
| 21:10:15 | efried | rechecks, not fixes | |
| 21:11:26 | mriedem | efried: yeah | |
| 21:11:31 | mriedem | corvus just promoted to top of gate | |
| 21:12:06 | efried | nice of him | |
| 21:12:37 | mriedem | i think i just bring out the best in people | |
| 21:12:39 | mriedem | you know? | |
| 21:12:50 | efried | threats of violence will do that | |
| 21:12:53 | mriedem | ha | |
| 21:13:34 | efried | do you want me to merge https://review.opendev.org/#/c/682140/ ? | |
| 21:13:43 | efried | despite artom hatin on it | |
| 21:14:04 | efried | even though you spelled stacktrace wrong | |
| 21:17:26 | sean-k-mooney | he did. unsurpisingly i did not notice | |
| 21:18:28 | mriedem | i got it correct in one spot | |
| 21:18:30 | mriedem | i was rushing | |
| 21:18:54 | mriedem | it's up to me if you want to approve it or let it sit, it's not the big blow up like the libvirt host capabilities one | |
| 21:20:02 | mriedem | i think both your alternative and artom's are equally gross | |
| 21:20:04 | sean-k-mooney | ya that one does not actully casuse the test to fail and the log mesage is like 20 lines vs 100s | |
| 21:20:18 | efried | how many times do we actually need to raise? | |
| 21:21:09 | sean-k-mooney | well once really. the stub raises if its called more then once | |
| 21:21:20 | sean-k-mooney | i doubt we will call it again if we raise | |
| 21:22:25 | sean-k-mooney | i +1'd it because a.) it works and b.) while it was surpising at first glance i understand how it works after looking at it for 30 seconds | |
| 21:22:54 | mriedem | you can't use wraps because we need to side effect the error in the 2nd call, | |