| *** code-R has joined #openstack-oslo | 00:01 | |
| *** spzala has joined #openstack-oslo | 00:02 | |
| *** snizar_ has quit IRC | 00:02 | |
| *** snizar has quit IRC | 00:02 | |
| *** code-R_ has joined #openstack-oslo | 00:03 | |
| *** code-R has quit IRC | 00:06 | |
| *** spzala has quit IRC | 00:07 | |
| *** mtanino has quit IRC | 00:18 | |
| *** hockeynut has joined #openstack-oslo | 00:22 | |
| *** thumpba_ has quit IRC | 00:39 | |
| *** thumpba has joined #openstack-oslo | 00:39 | |
| *** thumpba has quit IRC | 00:44 | |
| *** ihrachys has quit IRC | 00:50 | |
| *** ihrachys has joined #openstack-oslo | 00:51 | |
| *** hockeynut has quit IRC | 01:02 | |
| *** hockeynut has joined #openstack-oslo | 01:02 | |
| *** spzala has joined #openstack-oslo | 01:04 | |
| *** ihrachys has quit IRC | 01:04 | |
| *** dimtruck is now known as zz_dimtruck | 01:08 | |
| *** spzala has quit IRC | 01:09 | |
| *** code-R_ has quit IRC | 01:42 | |
| *** zz_dimtruck is now known as dimtruck | 01:49 | |
| *** yamahata has quit IRC | 02:20 | |
| openstackgerrit | Haifeng Yan proposed openstack/oslo.vmware: Fix TypeError:six.StringIO(resp.content) must be str or None, not bytes. https://review.openstack.org/351601 | 02:28 |
|---|---|---|
| *** spzala has joined #openstack-oslo | 02:29 | |
| *** spzala has quit IRC | 02:30 | |
| *** spzala has joined #openstack-oslo | 02:30 | |
| *** spzala has quit IRC | 02:44 | |
| *** spzala has joined #openstack-oslo | 02:44 | |
| *** spzala has quit IRC | 02:48 | |
| *** thumpba has joined #openstack-oslo | 03:19 | |
| *** jeckersb is now known as jeckersb_gone | 03:49 | |
| *** links has joined #openstack-oslo | 03:54 | |
| *** ozamiatin has joined #openstack-oslo | 04:00 | |
| *** hockeynut has quit IRC | 04:03 | |
| *** ozamiatin has quit IRC | 04:08 | |
| *** dimtruck is now known as zz_dimtruck | 04:10 | |
| *** zz_dimtruck is now known as dimtruck | 04:10 | |
| *** dimtruck is now known as zz_dimtruck | 04:19 | |
| *** amotoki has joined #openstack-oslo | 04:38 | |
| *** davidlenwell has quit IRC | 04:42 | |
| *** code-R has joined #openstack-oslo | 04:44 | |
| *** code-R_ has joined #openstack-oslo | 04:49 | |
| *** code-R has quit IRC | 04:52 | |
| *** davidlenwell has joined #openstack-oslo | 04:55 | |
| *** ozamiatin has joined #openstack-oslo | 05:00 | |
| *** itisha has quit IRC | 05:20 | |
| *** eezhova has joined #openstack-oslo | 05:22 | |
| *** code-R_ has quit IRC | 05:24 | |
| *** code-R has joined #openstack-oslo | 05:24 | |
| *** ihrachys has joined #openstack-oslo | 05:30 | |
| *** code-R has quit IRC | 05:30 | |
| *** gcb_ has joined #openstack-oslo | 05:35 | |
| *** thumpba has quit IRC | 05:40 | |
| *** rcernin has joined #openstack-oslo | 05:46 | |
| *** eezhova has quit IRC | 05:50 | |
| *** thumpba has joined #openstack-oslo | 05:51 | |
| *** thumpba has quit IRC | 05:56 | |
| *** ihrachys has quit IRC | 06:15 | |
| *** e0ne has joined #openstack-oslo | 06:30 | |
| *** code-R has joined #openstack-oslo | 06:38 | |
| *** eezhova has joined #openstack-oslo | 06:42 | |
| *** code-R has quit IRC | 06:43 | |
| *** amotoki has quit IRC | 06:50 | |
| *** code-R has joined #openstack-oslo | 06:51 | |
| *** thumpba has joined #openstack-oslo | 06:52 | |
| *** dtardivel has joined #openstack-oslo | 06:53 | |
| *** thumpba has quit IRC | 06:56 | |
| *** tesseract- has joined #openstack-oslo | 07:00 | |
| *** mhickey has joined #openstack-oslo | 07:01 | |
| *** lpetrut has joined #openstack-oslo | 07:04 | |
| *** code-R has quit IRC | 07:07 | |
| *** andymaier has joined #openstack-oslo | 07:26 | |
| *** lpetrut has quit IRC | 07:29 | |
| *** yamahata has joined #openstack-oslo | 07:29 | |
| *** jamielennox is now known as jamielennox|away | 07:45 | |
| *** e0ne has quit IRC | 07:51 | |
| *** thumpba has joined #openstack-oslo | 07:52 | |
| *** thumpba has quit IRC | 07:57 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** thumpba has joined #openstack-oslo | 08:01 | |
| *** zzzeek has joined #openstack-oslo | 08:01 | |
| *** amotoki has joined #openstack-oslo | 08:02 | |
| *** amotoki has quit IRC | 08:02 | |
| *** amotoki has joined #openstack-oslo | 08:03 | |
| *** ihrachys has joined #openstack-oslo | 08:05 | |
| *** thumpba has quit IRC | 08:06 | |
| *** amotoki_ has joined #openstack-oslo | 08:08 | |
| *** amotoki has quit IRC | 08:08 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.log: Default use_stderr to False https://review.openstack.org/351283 | 08:11 |
| *** lucas-dinner is now known as lucasagomes | 08:17 | |
| openstackgerrit | Merged openstack/oslo.log: Fixes unit tests on Windows https://review.openstack.org/349498 | 08:17 |
| *** e0ne has joined #openstack-oslo | 08:19 | |
| *** ihrachys has quit IRC | 08:29 | |
| *** dkehn has quit IRC | 08:32 | |
| *** dkehn_ has quit IRC | 08:32 | |
| *** ihrachys has joined #openstack-oslo | 08:34 | |
| *** Jaison has joined #openstack-oslo | 08:35 | |
| *** links has quit IRC | 08:37 | |
| *** dkehn has joined #openstack-oslo | 08:38 | |
| *** eezhova has quit IRC | 08:44 | |
| *** jamielennox|away is now known as jamielennox | 08:50 | |
| *** dkehn_ has joined #openstack-oslo | 08:52 | |
| *** ozamiatin has quit IRC | 08:55 | |
| *** ihrachys has quit IRC | 08:57 | |
| *** thumpba has joined #openstack-oslo | 09:02 | |
| *** ihrachys has joined #openstack-oslo | 09:03 | |
| *** thumpba has quit IRC | 09:08 | |
| *** ihrachys has quit IRC | 09:11 | |
| *** ihrachys has joined #openstack-oslo | 09:11 | |
| *** gus has quit IRC | 09:22 | |
| *** andymaier has quit IRC | 09:23 | |
| *** yamahata has quit IRC | 09:24 | |
| *** sambetts|afk is now known as sambetts | 09:26 | |
| *** lpetrut has joined #openstack-oslo | 09:27 | |
| *** ihrachys has quit IRC | 09:28 | |
| *** ihrachys has joined #openstack-oslo | 09:30 | |
| *** code-R has joined #openstack-oslo | 09:30 | |
| *** ozamiatin has joined #openstack-oslo | 09:37 | |
| *** jamielennox is now known as jamielennox|away | 09:43 | |
| *** spzala has joined #openstack-oslo | 09:45 | |
| *** spzala has quit IRC | 09:49 | |
| *** andymaier has joined #openstack-oslo | 09:53 | |
| openstackgerrit | Qiming Teng proposed openstack/oslo.messaging: Trivial: fix notify filter when data item is None https://review.openstack.org/329754 | 09:53 |
| *** lpetrut1 has joined #openstack-oslo | 09:57 | |
| *** openstack has joined #openstack-oslo | 10:16 | |
| *** dkehn has joined #openstack-oslo | 10:17 | |
| *** e0ne has quit IRC | 10:18 | |
| *** dkehn_ has joined #openstack-oslo | 10:22 | |
| *** sdague has joined #openstack-oslo | 10:27 | |
| *** boden has joined #openstack-oslo | 10:28 | |
| *** ihrachys has quit IRC | 10:30 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Make zmq_immediate configurable https://review.openstack.org/356370 | 10:40 |
| *** ociuhandu has quit IRC | 10:50 | |
| *** yamamoto has joined #openstack-oslo | 10:50 | |
| *** yamamoto has quit IRC | 10:56 | |
| *** yamamoto has joined #openstack-oslo | 10:57 | |
| *** andymaier has quit IRC | 11:00 | |
| *** yamamoto has quit IRC | 11:02 | |
| *** yamamoto has joined #openstack-oslo | 11:03 | |
| *** yamamoto has quit IRC | 11:07 | |
| *** yamamoto has joined #openstack-oslo | 11:07 | |
| *** yamamoto has quit IRC | 11:09 | |
| *** yamamoto has joined #openstack-oslo | 11:09 | |
| *** cdent has joined #openstack-oslo | 11:14 | |
| *** andymaier has joined #openstack-oslo | 11:19 | |
| *** ihrachys has joined #openstack-oslo | 11:22 | |
| *** eezhova has joined #openstack-oslo | 11:23 | |
| *** lucasagomes is now known as lucas-hungry | 11:23 | |
| *** ociuhandu has joined #openstack-oslo | 11:28 | |
| openstackgerrit | Gevorg Davoian proposed openstack/oslo.messaging: [zmq] Implement retries for unacknowledged CALLs https://review.openstack.org/353529 | 11:29 |
| *** ihrachys has quit IRC | 11:32 | |
| *** yamamoto has quit IRC | 11:40 | |
| *** yamamoto has joined #openstack-oslo | 11:46 | |
| *** ansmith has joined #openstack-oslo | 11:49 | |
| *** boden has quit IRC | 11:52 | |
| *** boden has joined #openstack-oslo | 11:58 | |
| *** kgiusti has joined #openstack-oslo | 12:02 | |
| *** andymaier has quit IRC | 12:02 | |
| *** code-R_ has joined #openstack-oslo | 12:03 | |
| *** sheeprine has quit IRC | 12:03 | |
| *** sheeprine has joined #openstack-oslo | 12:06 | |
| *** code-R has quit IRC | 12:06 | |
| *** mtanino has joined #openstack-oslo | 12:07 | |
| *** jamielennox|away is now known as jamielennox | 12:11 | |
| *** amotoki_ has quit IRC | 12:13 | |
| *** andymaier has joined #openstack-oslo | 12:14 | |
| *** jpena is now known as jpena|lunch | 12:15 | |
| *** yamamoto has quit IRC | 12:15 | |
| *** amotoki has joined #openstack-oslo | 12:19 | |
| *** lucas-hungry is now known as lucasagomes | 12:26 | |
| *** yamamoto has joined #openstack-oslo | 12:28 | |
| *** e0ne has joined #openstack-oslo | 12:28 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Make zmq_immediate configurable https://review.openstack.org/356370 | 12:31 |
| *** yamamoto has quit IRC | 12:32 | |
| *** gcb_ has quit IRC | 12:33 | |
| *** jecarey has quit IRC | 12:39 | |
| *** gcb_ has joined #openstack-oslo | 12:46 | |
| *** mtanino has quit IRC | 12:49 | |
| kgiusti | ozamiatin: hey, just fyi those jenkins failures are caused by issues with the osic cloud servers - https://bugs.launchpad.net/openstack-gate/+bug/1613749 | 12:51 |
| openstack | Launchpad bug 1613749 in OpenStack-Gate "Timeouts when requesting a glance image created with a remote image from git.o.o" [Undecided,New] | 12:51 |
| kgiusti | ozamiatin: infra was working on it yesterday - apparently not fixed yet. | 12:51 |
| ozamiatin | kgiusti: thanks! the problem is not on our side then :)) | 12:53 |
| *** jecarey has joined #openstack-oslo | 12:55 | |
| kgiusti | ozamiatin: nope. Likely due to a network connectivity issue. Hint: if you see traceroute failures in the console output (and it's an osic host) you're hitting the bug. See: http://logs.openstack.org/67/352667/1/check/gate-oslo.messaging-src-dsvm-full-zmq/c2f8b97/console.html#_2016-08-17_11_00_25_657860 | 12:55 |
| *** yamamoto has joined #openstack-oslo | 12:55 | |
| ozamiatin | kgiusti: aha, got it | 12:56 |
| kgiusti | ozamiatin: we can either just keep rechecking in hope that we avoid an osic system, or disable voting in all the tempest-related checks (that's risky IMHO). | 12:57 |
| ozamiatin | kgiusti: hope they will fix the infra soon | 12:58 |
| ozamiatin | kgiusti: switching tempest off is not a good idea, I was wrong proposing this https://review.openstack.org/#/c/355919/ | 12:59 |
| patchbot | ozamiatin: patch 355919 - openstack-infra/project-config - Remove n-net related gates (ABANDONED) | 12:59 |
| kgiusti | ozamiatin: yeah - agreed. | 12:59 |
| *** rlrossit has joined #openstack-oslo | 12:59 | |
| *** spzala_ has joined #openstack-oslo | 13:01 | |
| kgiusti | ozamiatin: http://lists.openstack.org/pipermail/openstack-dev/2016-August/101811.html | 13:02 |
| *** gordc has joined #openstack-oslo | 13:05 | |
| *** pradk has quit IRC | 13:06 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Host name and target in socket identity https://review.openstack.org/356451 | 13:07 |
| *** jpena|lunch is now known as jpena | 13:18 | |
| *** yamamoto has quit IRC | 13:29 | |
| *** eandersson_ has joined #openstack-oslo | 13:30 | |
| *** yamamoto has joined #openstack-oslo | 13:30 | |
| *** edmondsw has joined #openstack-oslo | 13:37 | |
| *** mtanino has joined #openstack-oslo | 13:37 | |
| kgiusti | ansmith: ping - I'm going to update the amqp1 feature branch to bring it up-to-date with master before we start the merge | 13:38 |
| kgiusti | ansmith: that may affect this https://review.openstack.org/#/c/348009/2 as it's against the feature branch | 13:40 |
| patchbot | kgiusti: patch 348009 - oslo.messaging (feature/amqp-dispatch-router) - [AMQP 1.0] Environment setup to support dispatch r... | 13:40 |
| ansmith | kgiusti: ok, thanks | 13:40 |
| kgiusti | ansmith: you'll likely need to re-issue the patch on master (if we get the branch merged first) | 13:40 |
| *** jeckersb_gone is now known as jeckersb | 13:44 | |
| *** Jaison has quit IRC | 13:46 | |
| *** thumpba has joined #openstack-oslo | 13:52 | |
| *** cdent has quit IRC | 13:53 | |
| *** ociuhandu has quit IRC | 14:01 | |
| *** amotoki has quit IRC | 14:05 | |
| *** Rockyg has joined #openstack-oslo | 14:07 | |
| *** pradk has joined #openstack-oslo | 14:08 | |
| *** tonytan4ever has joined #openstack-oslo | 14:10 | |
| *** amotoki has joined #openstack-oslo | 14:12 | |
| *** ociuhandu has joined #openstack-oslo | 14:12 | |
| *** ihrachys has joined #openstack-oslo | 14:14 | |
| *** spzala_ has quit IRC | 14:14 | |
| *** spzala has joined #openstack-oslo | 14:14 | |
| *** mpotdar_ has joined #openstack-oslo | 14:14 | |
| *** spzala has quit IRC | 14:17 | |
| *** spzala has joined #openstack-oslo | 14:17 | |
| *** jeckersb is now known as jeckersb_gone | 14:17 | |
| *** zz_dimtruck is now known as dimtruck | 14:19 | |
| *** jeckersb_gone is now known as jeckersb | 14:20 | |
| *** hockeynut has joined #openstack-oslo | 14:20 | |
| *** ihrachys has quit IRC | 14:21 | |
| *** ihrachys has joined #openstack-oslo | 14:21 | |
| *** ihrachys has quit IRC | 14:23 | |
| *** ihrachys has joined #openstack-oslo | 14:24 | |
| *** cfriesen has quit IRC | 14:24 | |
| *** hemanthm|afk is now known as hemanthm | 14:27 | |
| *** yamamoto has quit IRC | 14:27 | |
| *** yamahata has joined #openstack-oslo | 14:27 | |
| *** yamamoto has joined #openstack-oslo | 14:29 | |
| *** itisha has joined #openstack-oslo | 14:30 | |
| *** yamamoto has quit IRC | 14:34 | |
| openstackgerrit | Javier Peña proposed openstack/oslo.vmware: Make the mock call in MemoryCacheTest forward-compatible https://review.openstack.org/355851 | 14:40 |
| *** ihrachys has quit IRC | 14:47 | |
| *** ihrachys has joined #openstack-oslo | 14:49 | |
| *** ihrachys has quit IRC | 14:50 | |
| *** ihrachys has joined #openstack-oslo | 14:51 | |
| *** siva_krishnan has left #openstack-oslo | 14:55 | |
| *** openstackgerrit has quit IRC | 15:03 | |
| *** openstackgerrit has joined #openstack-oslo | 15:04 | |
| *** ozamiatin has quit IRC | 15:07 | |
| *** ihrachys has quit IRC | 15:08 | |
| *** ihrachys has joined #openstack-oslo | 15:09 | |
| *** amotoki has quit IRC | 15:10 | |
| *** ihrachys has quit IRC | 15:11 | |
| *** ihrachys has joined #openstack-oslo | 15:12 | |
| *** ihrachys has quit IRC | 15:16 | |
| *** ihrachys has joined #openstack-oslo | 15:16 | |
| *** ihrachys has quit IRC | 15:26 | |
| *** ihrachys has joined #openstack-oslo | 15:26 | |
| *** sdake_ has joined #openstack-oslo | 15:27 | |
| sdake_ | hey folks | 15:27 |
| sdake_ | dhellmann sent me this direction - looking for a little coaching on using pbr properly in openstack projects | 15:27 |
| sdake_ | if we install via pip, pbr shows the installed version, even if working out of the git repo | 15:28 |
| sdake_ | what we want is even if a version is installed via pip, the git version will be used instead of the system-wide installed version | 15:28 |
| sdake_ | all our code is correct in terms of using the right set of python code, ansible code, and docker code depending on these two circumstances | 15:29 |
| sdake_ | it seems like pbr is unable to prioritize git repos ahead of the systemwide settings | 15:29 |
| sdake_ | (or maybe it can and we don't know how to do so) | 15:29 |
| *** ihrachys has quit IRC | 15:30 | |
| *** ihrachys has joined #openstack-oslo | 15:30 | |
| *** yamamoto has joined #openstack-oslo | 15:32 | |
| *** ihrachys has quit IRC | 15:32 | |
| *** yamamoto has quit IRC | 15:37 | |
| *** mpotdar_ has quit IRC | 15:37 | |
| *** mpotdar_ has joined #openstack-oslo | 15:38 | |
| *** ihrachys has joined #openstack-oslo | 15:42 | |
| *** ozamiatin has joined #openstack-oslo | 15:43 | |
| *** thumpba has quit IRC | 15:44 | |
| *** ihrachys has quit IRC | 15:48 | |
| *** ihrachys has joined #openstack-oslo | 15:49 | |
| *** code-R_ has quit IRC | 15:49 | |
| *** ihrachys has quit IRC | 15:51 | |
| *** mhickey has quit IRC | 15:53 | |
| *** ihrachys has joined #openstack-oslo | 15:53 | |
| *** amotoki has joined #openstack-oslo | 15:56 | |
| *** ozamiatin has quit IRC | 15:57 | |
| *** rcernin has quit IRC | 15:58 | |
| *** sdake_ is now known as sdake | 16:00 | |
| *** sdake is now known as sdake_ | 16:00 | |
| *** snizar_ has joined #openstack-oslo | 16:03 | |
| *** snizar has joined #openstack-oslo | 16:03 | |
| *** andymaier has quit IRC | 16:03 | |
| *** dmellado is now known as dmellado|off | 16:05 | |
| *** ihrachys has quit IRC | 16:12 | |
| jroll | sdake_: to clarify, you want "pip install ." (in a git repo) to be used instead of the already-installed system package e.g. apt-get install nova? | 16:12 |
| sdake_ | jroll meeting atm, can i get back with you? | 16:13 |
| sdake_ | kolla team meeting in openstack-meeting-4 | 16:13 |
| jroll | sdake_: of course, disclaimer I'm not an expert :) | 16:13 |
| *** code-R has joined #openstack-oslo | 16:13 | |
| dhellmann | sdake_ : IIRC, pbr is going to use the setuptools metadata for a package if that's present, and it would be after you pip install. What issue is that causing for you? | 16:21 |
| sdake_ | dhellmann - team meeting atm - finishes in 40 mins - can I get back to you? | 16:21 |
| dhellmann | sdake_ : yes, I was just leaving my question here for you | 16:22 |
| *** ihrachys has joined #openstack-oslo | 16:25 | |
| *** ihrachys has quit IRC | 16:28 | |
| *** cdent has joined #openstack-oslo | 16:39 | |
| dhellmann | harlowja : it would be good to get https://review.openstack.org/#/c/353029/ into next week's releases if we can | 16:40 |
| patchbot | dhellmann: patch 353029 - stevedore - do not emit warnings for missing hooks | 16:40 |
| dhellmann | jd__ : ^^ | 16:40 |
| *** ihrachys has joined #openstack-oslo | 16:41 | |
| *** spzala has quit IRC | 16:41 | |
| *** spzala has joined #openstack-oslo | 16:42 | |
| *** ihrachys has quit IRC | 16:44 | |
| *** ihrachys has joined #openstack-oslo | 16:46 | |
| *** spzala has quit IRC | 16:46 | |
| *** ihrachys has quit IRC | 16:49 | |
| *** ihrachys has joined #openstack-oslo | 16:49 | |
| *** mpotdar_ has quit IRC | 16:51 | |
| *** lpetrut has quit IRC | 16:52 | |
| *** sambetts is now known as sambetts|afk | 16:56 | |
| sdake_ | dhellmann ya its using the setuptools metadata - but the problem is this leads to a version inconsistency when we work from within git | 16:56 |
| sdake_ | if master is installed for example, and i am working on the stable branch in git, our code does the wrong thing wrt tagging images | 16:57 |
| sdake_ | it uses the pbr version from teh system rather then git | 16:57 |
| sdake_ | what I was curious about is if there is a mechanism to override that, or if that is just how it is and need to learn to love it ;) | 16:57 |
| *** yamahata has quit IRC | 16:58 | |
| sdake_ | my expectation would be that if pbr finds git, it uses git tags, if it doesn't find git, it doesn't use git tags | 16:58 |
| sdake_ | but I don't know how it works under the hood | 16:58 |
| sdake_ | jroll ^^ | 16:58 |
| jroll | don't have two things installed side-by-side, would be my recommendation :) | 16:59 |
| sdake_ | jroll well we only hve one thing installed and another from git | 16:59 |
| sdake_ | jroll but ya i get the intent of the message - dont have anything installed | 16:59 |
| sdake_ | is that the only solution? | 17:00 |
| dhellmann | sdake_ : is this a case of using the same working dir with multiple branches? | 17:00 |
| jroll | sdake_: is the thing from git installed? | 17:00 |
| sdake_ | jroll it is not installed | 17:00 |
| *** lucasagomes is now known as lucas-dinner | 17:00 | |
| jroll | sdake_: can I see your pbr code? | 17:00 |
| *** tesseract- has quit IRC | 17:00 | |
| sdake_ | dhellmann - the reproducer for us is pip instlal kolla=2.0.2 - git clone kolla - git checkout stable/mitaka | 17:01 |
| sdake_ | dhellmann then from within git dir, pbr produces the answer 3.0.0 intead of 2.0.2 | 17:01 |
| dhellmann | if the git directory isn't the installed version, then yeah the package metadata is going to be used | 17:01 |
| sdake_ | sorry pip install master | 17:01 |
| sdake_ | messed that reproducer up | 17:01 |
| sdake_ | jroll moment let me hunt it down | 17:02 |
| jroll | right, if you import pbr and use it to find a version, I believe it only uses the installed version | 17:02 |
| sdake_ | https://github.com/openstack/kolla/blob/master/kolla/version.py#L15 | 17:02 |
| sdake_ | if there is no installed version it uses the git tags to make a version determination | 17:02 |
| jroll | right | 17:03 |
| sdake_ | i guess what I want is the order of precdence reversed | 17:03 |
| sdake_ | current order: if any version installed -> that version is used next ignore git | 17:04 |
| sdake_ | desired order: if any version installed -> if git present use git, otherwise use system version | 17:05 |
| sdake_ | no such capacity in pbr? | 17:05 |
| *** e0ne has quit IRC | 17:05 | |
| jroll | not that I know of, but I'm nowhere near an expert | 17:05 |
| sdake_ | jroll cool thanks for helping brainstorm atleast | 17:05 |
| dhellmann | it's hard-coded to use git as a fallback | 17:05 |
| sdake_ | jroll i hardly understand this stuff | 17:05 |
| sdake_ | dhellmann got it hard coded | 17:06 |
| sdake_ | so answer is no way to override precedence | 17:06 |
| jroll | sdake_: magic! | 17:06 |
| dhellmann | and to use pkg_resources to look at the metadata if it's found | 17:06 |
| dhellmann | is there any reason the git version can't just be installed? | 17:06 |
| sdake_ | dhellmann they change - meaning have to git install each time - pip install -e works here, but not for everyone's environment | 17:07 |
| dhellmann | yeah | 17:07 |
| jroll | so one suggestion, without knowing what the task is (building images?) | 17:07 |
| sdake_ | for some distros pip -e install is busted completely - dont know why | 17:07 |
| jroll | maybe have a tox job for building those when running from git checkouts | 17:07 |
| dhellmann | you're not using pip from system packages, I hope | 17:07 |
| jroll | then it'll always be the current git tree version, because virtualenvs | 17:07 |
| sdake_ | dhellmann no pip upgrade is our first step | 17:08 |
| dhellmann | k | 17:08 |
| sdake_ | or one of them anyway | 17:08 |
| sdake_ | dhellmann - sysstem packages of pip - that is something i never debug with people | 17:09 |
| sdake_ | dhellmann that may make pip install -e work on all distros | 17:10 |
| dhellmann | sdake_ : one way your suggestion would break things is that pbr assumes any git repo is the package for which it's being asked a version, because it assumes the only reason the metadata isn't present is that you're running in a test configuration of some sort. so if we change the precedence order, and you're in a random git directory, you'll get the version for that for every package you ask about (unless we make the | 17:10 |
| dhellmann | git logic smarter somehow) | 17:10 |
| dhellmann | IOW, the git logic makes a lot of assumptions | 17:10 |
| sdake_ | dhellmann roger - I think I can work something out that involves a workflow for devs vs a workflow for eval/deploy | 17:11 |
| sdake_ | which is "dont install the system packages" | 17:11 |
| dhellmann | yeah, I think that's going to be your most reliable way forward | 17:11 |
| sdake_ | rather don't install the kolla package ia pip | 17:11 |
| jroll | sdake_: I think tox targets would work really well here | 17:11 |
| sdake_ | jroll interestingly tox I think worked for someone (the virtualenv part) | 17:12 |
| jroll | right, it's intended to :) | 17:12 |
| dhellmann | tox may be another good option, if you set up a tox environment to run the services then you'll always get what's in the current repo | 17:12 |
| sdake_ | but that is relaly hard to work with as an eval type person | 17:12 |
| sdake_ | our cmds take options | 17:12 |
| dhellmann | I'm not sure what an "eval type person" is | 17:12 |
| sdake_ | don't want to confuse peopel with multliple commands and usign tox | 17:12 |
| sdake_ | dhellmann - someone consuming the software | 17:12 |
| sdake_ | vs developing or it | 17:13 |
| sdake_ | for it | 17:13 |
| dhellmann | ah | 17:13 |
| jroll | right, tox is for devs | 17:13 |
| sdake_ | ya - devs can handle whatever we throw at em :) | 17:13 |
| sdake_ | this is about making it easy on the operators | 17:13 |
| dhellmann | yeah, I mean you can do stuff like "tox -e venv -- command -w ith args" but that's not great as an eval tool | 17:13 |
| jroll | do operators work out of git repos, generally? | 17:13 |
| sdake_ | dhellmann right - diffeerent set of cmds then what they would typically use | 17:13 |
| sdake_ | jroll our instructions recommend it - there is a review up to change that to pip install kolla (from pypi) | 17:14 |
| sdake_ | but it has a version pin ==2.0.2 to get mitaka for eg | 17:14 |
| sdake_ | and people don't like that in the review | 17:14 |
| jroll | ah | 17:15 |
| sdake_ | (because we have to maintain a list of pins to use.. in the docs) | 17:15 |
| sdake_ | cool well folks thanks for the brainstorm session | 17:15 |
| sdake_ | it was fantasitc to have | 17:15 |
| openstackgerrit | Gevorg Davoian proposed openstack/oslo.messaging: [zmq] Implement retries for unacknowledged CALLs https://review.openstack.org/353529 | 17:15 |
| *** lpetrut has joined #openstack-oslo | 17:15 | |
| sdake_ | i have to do some thinking - but think I have enough to work with | 17:15 |
| jroll | awesome :) | 17:16 |
| sdake_ | jroll dhellmann ++! | 17:16 |
| dhellmann | sdake_ : good! | 17:17 |
| openstackgerrit | Gevorg Davoian proposed openstack/oslo.messaging: [zmq] Implement retries for unacknowledged CALLs https://review.openstack.org/353529 | 17:19 |
| *** amotoki has quit IRC | 17:24 | |
| *** ihrachys has quit IRC | 17:25 | |
| *** amotoki has joined #openstack-oslo | 17:25 | |
| *** tonytan4ever has quit IRC | 17:25 | |
| *** rcernin has joined #openstack-oslo | 17:26 | |
| *** lpetrut has quit IRC | 17:27 | |
| *** amotoki has quit IRC | 17:28 | |
| *** Rockyg has quit IRC | 17:32 | |
| *** yamamoto has joined #openstack-oslo | 17:34 | |
| openstackgerrit | Merged openstack/oslo.context: Manually specify from_dict parameters https://review.openstack.org/341250 | 17:36 |
| *** yamamoto has quit IRC | 17:38 | |
| dansmith | harlowja: dims: this should really make it in for newton, is that plausible? https://review.openstack.org/#/c/355981/3 | 17:38 |
| patchbot | dansmith: patch 355981 - oslo.versionedobjects - Fix remotable object change tracking | 17:38 |
| dansmith | wiating on a project config change to fix a broken job before we can land | 17:38 |
| *** yamahata has joined #openstack-oslo | 17:40 | |
| *** stevemar_ has joined #openstack-oslo | 17:43 | |
| *** stevemar_ has quit IRC | 17:43 | |
| *** hockeynut has quit IRC | 17:48 | |
| *** jdandrea has joined #openstack-oslo | 17:52 | |
| *** spzala has joined #openstack-oslo | 17:53 | |
| *** jdandrea_ has quit IRC | 17:55 | |
| *** spzala has quit IRC | 17:58 | |
| *** hockeynut has joined #openstack-oslo | 17:59 | |
| *** code-R has quit IRC | 18:00 | |
| *** spzala has joined #openstack-oslo | 18:00 | |
| harlowja | dansmith sure, let's try to get it | 18:02 |
| dansmith | okay | 18:03 |
| openstackgerrit | Merged openstack/oslo.messaging: Add docstring for get_notification_transport https://review.openstack.org/355053 | 18:07 |
| *** dkehn has quit IRC | 18:07 | |
| *** dkehn_ has quit IRC | 18:08 | |
| openstackgerrit | Merged openstack/stevedore: do not emit warnings for missing hooks https://review.openstack.org/353029 | 18:12 |
| *** mtanino has quit IRC | 18:13 | |
| *** ozamiatin has joined #openstack-oslo | 18:15 | |
| *** dkehn has joined #openstack-oslo | 18:15 | |
| *** dkehn_ has joined #openstack-oslo | 18:20 | |
| *** ociuhandu has quit IRC | 18:24 | |
| *** tonytan4ever has joined #openstack-oslo | 18:26 | |
| amrith | harlowja, quick question. what's your take on https://review.openstack.org/#/c/339787/? do you think it's going to make it in time for newton? I'm guessing no but want to make sure. thx | 18:28 |
| patchbot | amrith: patch 339787 - oslo.config - Add Range type | 18:28 |
| *** mpotdar_ has joined #openstack-oslo | 18:29 | |
| *** mpotdar_ has quit IRC | 18:29 | |
| harlowja | amrith seems like it could | 18:30 |
| harlowja | if it merges soon :-P | 18:30 |
| *** mpotdar_ has joined #openstack-oslo | 18:30 | |
| amrith | thx harlowja | 18:31 |
| *** tonytan4ever has quit IRC | 18:31 | |
| *** rlrossit_ has joined #openstack-oslo | 18:38 | |
| *** rlrossit has quit IRC | 18:41 | |
| *** hoonetorg has quit IRC | 18:53 | |
| *** lpetrut has joined #openstack-oslo | 18:57 | |
| harlowja | amrith ok, some comments added, small things that i'd like before that one merges | 19:03 |
| *** thumpba has joined #openstack-oslo | 19:05 | |
| *** jamielennox has quit IRC | 19:06 | |
| *** hoonetorg has joined #openstack-oslo | 19:06 | |
| *** jamielennox has joined #openstack-oslo | 19:09 | |
| *** jeckersb is now known as jeckersb_gone | 19:11 | |
| *** snizar_ has quit IRC | 19:13 | |
| *** snizar has quit IRC | 19:13 | |
| *** spzala has quit IRC | 19:18 | |
| *** spzala has joined #openstack-oslo | 19:19 | |
| *** spzala_ has joined #openstack-oslo | 19:22 | |
| *** spzala has quit IRC | 19:23 | |
| *** ociuhandu has joined #openstack-oslo | 19:25 | |
| bknudson | got an issue with oslo.db live db testing -- looks like the database doesn't get cleaned up after the TestCase finishes and therefor the next TestCase fails. | 19:26 |
| *** spzala_ has quit IRC | 19:26 | |
| bknudson | so things were working fine when we had a single TestCase using "FIXTURE = test_base.MySQLOpportunisticFixture", but now we've got 2 and the tests are failing. | 19:26 |
| bknudson | https://review.openstack.org/#/c/349939/22/keystone/tests/unit/test_sql_upgrade.py | 19:26 |
| patchbot | bknudson: patch 349939 - keystone - Add expand, data migration and contract logic to k... | 19:26 |
| *** edmondsw has quit IRC | 19:28 | |
| *** jeckersb_gone is now known as jeckersb | 19:34 | |
| *** eandersson_ has quit IRC | 19:36 | |
| harlowja | zzzeek andytime to look into that ^ | 19:38 |
| harlowja | *anytime | 19:38 |
| zzzeek | bknudson: did someting change that is suddenly causing this | 19:39 |
| harlowja | k | 19:39 |
| bknudson | zzzeek: it's a new review, so the review is making the change. | 19:39 |
| zzzeek | bknudson: ok... | 19:40 |
| bknudson | the live testing has been working great for several months | 19:40 |
| *** dimtruck is now known as zz_dimtruck | 19:43 | |
| bknudson | interesting fact -- the failures are not consistent. First time I ran it locally saw 2 failures, now not seeing any, and CI only got 1 failure. | 19:44 |
| zzzeek | bknudson: that's not unusual assuming you're working w/ multi-process testing | 19:44 |
| zzzeek | bknudson: it's likely these tests nver cleaned up within the schema | 19:44 |
| bknudson | oh, right, will switch to single-threaded. | 19:44 |
| zzzeek | bknudson: however...the current fixtures normally would set up and tear down the whole DB for each test unless you're using the optimizing function at the package level | 19:45 |
| *** spzala has joined #openstack-oslo | 19:45 | |
| zzzeek | in which case it shoudl do a drop all | 19:45 |
| *** code-R has joined #openstack-oslo | 19:45 | |
| -openstackstatus- NOTICE: The volume for logs.openstack.org filled up rather suddenly, causing a number of jobs to fail with a POST_FAILURE result and no logs; we're manually expiring some logs now to buy breathing room, but any changes which hit that in the past few minutes will need to be rechecked and/or approved again | 19:45 | |
| bknudson | zzzeek: I never did figure out how to use the optimizing loader... on my todo list. | 19:45 |
| bknudson | maybe it's time. | 19:46 |
| zzzeek | bknudson: well it's broke anyway so wait til harlowja and others can get aroud to reviewing my patches I mention everyday on IRC here :) | 19:46 |
| bknudson | :) | 19:46 |
| bknudson | maybe that's why I didn't do it from the beginning | 19:46 |
| harlowja | yes boss | 19:46 |
| zzzeek | bknudson: it would work at the package level though ... | 19:46 |
| *** code-R_ has joined #openstack-oslo | 19:49 | |
| zzzeek | bknudson: um. anyway you can make sure the error reproduces in pure single-threaded and perhaps you can use pdb to see what the DB state is doing ? it should be in a differently-named database for every test. e.g. DROP DATABASE when a test ends, brand new CREATE DATABASE for new test | 19:49 |
| *** spzala has quit IRC | 19:49 | |
| bknudson | will give it a shot. | 19:49 |
| zzzeek | bknudson: unless! your test is not actaully hitting mysql / Posgresql and is accidentally using sqlite | 19:49 |
| zzzeek | bknudson: ah. this might be. | 19:49 |
| zzzeek | because neutron had the same problem | 19:49 |
| bknudson | the live tests are pretty slow | 19:50 |
| *** spzala has joined #openstack-oslo | 19:50 | |
| zzzeek | bknudson: hm | 19:50 |
| bknudson | much slower than the sqlite live tests | 19:50 |
| zzzeek | bknudson: when I see: with sql.session_for_write() as session: , how is session_for_write() getting a hold of "self.engine" from the test case? | 19:51 |
| bknudson | zzzeek: you're right to be worried about this... I had to do some trickery... let me see if I can find it. | 19:52 |
| zzzeek | bknudson: seems like using the enginefacade is new in this patch? | 19:52 |
| *** code-R has quit IRC | 19:52 | |
| zzzeek | bknudson: so, the current whole test_base thing doesn't have much of a system for that, except it writes to the *global* enginefacade, which nobody seems to use | 19:52 |
| zzzeek | hence all the patches I have in review | 19:52 |
| bknudson | there were uses of sql.session_for_write() before, so not a new thing: http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_sql_upgrade.py#n296 | 19:56 |
| zzzeek | bknudson: any chance that's using sqlite and not the mysql / PG db | 19:58 |
| zzzeek | bknudson: neutron had this problem | 19:58 |
| bknudson | I'm going to try the breakpoint thing and see if I can catch the db create / teardown. | 19:59 |
| zzzeek | bknudson: easy test, "with sql.session_for_write() as session: session.execute("select * from mysql.user")" | 19:59 |
| zzzeek | or seothuing that will crash on sqlie | 19:59 |
| bknudson | zzzeek: ok, added a test to verify it's mysql and that worked as expected. | 20:02 |
| zzzeek | bknudson: weird. where is sql.session_for_Write() getting plugged in ? | 20:02 |
| zzzeek | bknudson: oh. you;re using the global enginefacade :) | 20:03 |
| zzzeek | in tests, at least. | 20:03 |
| bknudson | I'm trying to find the test code that sets up the global facade... | 20:04 |
| bknudson | because I'm pretty sure I needed to do some odd stuff. | 20:04 |
| redrobot | ohai oslo friends | 20:07 |
| bknudson | zzzeek: https://review.openstack.org/#/c/294246/9/keystone/common/sql/core.py | 20:07 |
| patchbot | bknudson: patch 294246 - keystone - Switch migration tests to oslo.db DbTestCase (MERGED) | 20:07 |
| redrobot | I have a question about oslo.config | 20:07 |
| zzzeek | bknudson: yeah found that | 20:08 |
| redrobot | does oslo.config support adding config sections in the way that tox does its ini config | 20:08 |
| redrobot | for example [testenv] and then [testenv:foo] [testenv:bar] etc | 20:09 |
| bknudson | zzzeek: I *think* that was to support sqlite in-memory db (since the same engine must always be used) | 20:09 |
| zzzeek | bknudson: well needed for the opportunistic DBs also. anyway, i dont see anthing that woudl interfere w/ the DROP DB on every test | 20:10 |
| *** mpotdar_ has quit IRC | 20:14 | |
| bknudson | ok, working on pdb | 20:17 |
| *** mpotdar_ has joined #openstack-oslo | 20:18 | |
| *** e0ne has joined #openstack-oslo | 20:20 | |
| *** Jokke_ has quit IRC | 20:35 | |
| *** Jokke_ has joined #openstack-oslo | 20:37 | |
| bknudson | according to pdb the database is dropped on every test (not TestCase) | 20:37 |
| bknudson | might explain why it takes so long to run the tests. | 20:39 |
| bknudson | doesn't explain why it looks like not cleaning up. | 20:40 |
| *** rcernin has quit IRC | 20:40 | |
| *** snizar has joined #openstack-oslo | 20:42 | |
| *** snizar_ has joined #openstack-oslo | 20:42 | |
| *** ansmith has quit IRC | 20:47 | |
| bknudson | zzzeek: put a print statement in the drop db -- http://paste.openstack.org/show/559095/ happens on every test... | 20:50 |
| zzzeek | bknudson: looking | 20:54 |
| zzzeek | bknudson: k so the next thing is, go in your tests, pdb, do a "select database()", make sure its the name you expect. do "show tables", make sure there are none of them. | 20:56 |
| zzzeek | bknudson: you can do it in process w/ session.execute("show tables").fetchall() | 20:56 |
| bknudson | here's a bit of info -- it's actually the sqlite "live" test that's failing. | 20:56 |
| zzzeek | bknudson: that changes everything :) | 20:57 |
| bknudson | y, I'm wondering if there isn't an assumption about that db cleaning itself. | 20:57 |
| zzzeek | bknudson: so...if you take all the Mysql/PG opportunistic stuff out of the module entirely, or similar, just test at the sqlite level | 20:57 |
| bknudson | run in this order, the sqlite one fails: keystone.tests.unit.test_sql_upgrade.PostgreSQLOpportunisticContractSchemaUpgradeTestCase.test_001_created_at_made_non_nullable keystone.tests.unit.test_sql_upgrade.SqlContractSchemaUpgradeTests.test_001_created_at_made_non_nullable | 20:57 |
| zzzeek | bknudson: confirmed it's dependent on the PG and/or MySQL test being run ? | 20:58 |
| *** edmondsw has joined #openstack-oslo | 20:58 | |
| *** e0ne has quit IRC | 21:00 | |
| bknudson | commented out the mysql & postgresql TetsCases, no failure. | 21:00 |
| boden | harlowja jd__ : tenacity is very cool. I have 1 question: https://github.com/jd/tenacity/issues/11 best I can tell it suits all my use cases already | 21:05 |
| harlowja | boden cools | 21:05 |
| *** sdague has quit IRC | 21:06 | |
| harlowja | jd__ might be able to answer that i think | 21:06 |
| boden | why isn’t it in global requirements yet!! ; ) | 21:06 |
| harlowja | lol | 21:06 |
| *** snizar_ has quit IRC | 21:06 | |
| *** snizar has quit IRC | 21:06 | |
| openstackgerrit | Merged openstack/oslo.db: Link enginefacade to test database provisioning https://review.openstack.org/352520 | 21:07 |
| *** zz_dimtruck is now known as dimtruck | 21:07 | |
| jd__ | boden: I didn't release it but I will soon :) | 21:07 |
| jd__ | boden: I replied on GitHub :) | 21:08 |
| *** gordc has quit IRC | 21:08 | |
| boden | nice work… minimal but very functional | 21:08 |
| harlowja | jd__ aims to please | 21:10 |
| harlowja | lol | 21:10 |
| boden | jd__ do we still think “marked wait” (https://github.com/rholder/retrying/pull/51) should live outside of the repo? | 21:11 |
| zzzeek | bknudson: k, next pdb is, get into the sqlite test there and see whats in the DB | 21:12 |
| jd__ | boden: you can send a PR with that function implemented :) | 21:12 |
| jd__ | it's modular now so we don't need to refactor __init__ or whatever | 21:13 |
| boden | jd__ will do | 21:13 |
| *** snizar has joined #openstack-oslo | 21:13 | |
| *** snizar_ has joined #openstack-oslo | 21:13 | |
| jd__ | if you think it's worth it – I'm more into wait_functions that computes times automatically than tuples, but whatever :) | 21:13 |
| boden | I’m trying to solve some actual uses in openstack, I’m not into tuples or anything ; ) | 21:14 |
| bknudson | zzzeek: see this : http://paste.openstack.org/show/559100/ -- everything looks fine except the column value is off. | 21:14 |
| bknudson | this is wackier than I imagined it would be. | 21:14 |
| harlowja | whats this about tuples | 21:16 |
| *** jeckersb is now known as jeckersb_gone | 21:16 | |
| *** spzala has quit IRC | 21:17 | |
| *** spzala has joined #openstack-oslo | 21:18 | |
| bknudson | zzzeek: if I change the order of the tests now it's postgres that fails: http://paste.openstack.org/show/559101/ | 21:18 |
| zzzeek | bknudson: any cahnce you can config and/or hardcode connection_debug=50 for your oslo.db create_engine and start looking in the SQL output ? | 21:19 |
| *** kgiusti has left #openstack-oslo | 21:20 | |
| bknudson | I'm going to make a quick check with another pdb when the col is created / updated and then try connection_debug=50. | 21:20 |
| bknudson | might have something to do with the migrate version tables... | 21:22 |
| *** spzala has quit IRC | 21:22 | |
| *** rlrossit_ has quit IRC | 21:22 | |
| *** cdent has quit IRC | 21:23 | |
| *** pradk has quit IRC | 21:29 | |
| *** boden has quit IRC | 21:29 | |
| bknudson | we've got 3 repos now, and for some reason it's running the contract migration when a function was called to run the expand migrations | 21:30 |
| openstackgerrit | Merged openstack/oslo.messaging: Clean outdated docstring and comment https://review.openstack.org/353253 | 21:31 |
| *** lpetrut has quit IRC | 21:44 | |
| *** mpotdar_ has quit IRC | 21:44 | |
| *** dtardivel has quit IRC | 21:47 | |
| *** sdake_ has quit IRC | 22:06 | |
| *** amotoki has joined #openstack-oslo | 22:09 | |
| *** amotoki_ has joined #openstack-oslo | 22:10 | |
| *** amotoki has quit IRC | 22:14 | |
| *** spzala has joined #openstack-oslo | 22:17 | |
| *** spzala has quit IRC | 22:17 | |
| *** spzala has joined #openstack-oslo | 22:17 | |
| *** sdake has joined #openstack-oslo | 22:21 | |
| *** ozamiatin has quit IRC | 22:22 | |
| *** hockeynut has quit IRC | 22:22 | |
| *** lucas-dinner has quit IRC | 22:23 | |
| *** eeiden has quit IRC | 22:25 | |
| *** syjulian has quit IRC | 22:25 | |
| *** jeckersb_gone has quit IRC | 22:25 | |
| *** yamamoto has joined #openstack-oslo | 22:30 | |
| *** jeckersb_gone has joined #openstack-oslo | 22:30 | |
| *** lucasagomes has joined #openstack-oslo | 22:30 | |
| *** abashmak has joined #openstack-oslo | 22:34 | |
| abashmak | Hello, anybody on this channel? | 22:35 |
| *** sdake has quit IRC | 22:38 | |
| harlowja | abashmak hello | 22:42 |
| abashmak | Hi harlowja | 22:42 |
| abashmak | I have a question | 22:42 |
| *** syjulian has joined #openstack-oslo | 22:43 | |
| abashmak | I'm trying to override a config option using set_override, the option has a defined list of choices specified, but there is a test that provides a different value that's not in the list | 22:43 |
| abashmak | I thought I could use set_override to change the choices list temporarily, then clear_override afterwards | 22:44 |
| abashmak | Perhaps I misunderstand the purpose of set_override | 22:45 |
| harlowja | ya, i don't think set_override means change the option all together | 22:46 |
| harlowja | i think it just means override the values (from set of valid values) | 22:46 |
| harlowja | where that set is still valid | 22:47 |
| harlowja | seems like if | 22:47 |
| harlowja | 'different value that's not in the list' is a valid value, then it should be in the 'ok' set? | 22:47 |
| abashmak | it's a test value | 22:47 |
| abashmak | so we don't want it in the ok list | 22:48 |
| harlowja | so u are testing oslo.config? | 22:49 |
| *** eeiden has joined #openstack-oslo | 22:49 | |
| *** amotoki_ has quit IRC | 22:50 | |
| abashmak | No | 22:50 |
| abashmak | at least I don't believe so :) | 22:50 |
| harlowja | ya, i think i get what u are trying to do | 22:51 |
| harlowja | have u tried turning off the `enforce_type` to set_override | 22:52 |
| harlowja | at least https://github.com/openstack/oslo.config/blob/master/oslo_config/fixture.py#L51 seems like that may help | 22:52 |
| abashmak | Yeah, tried that, the set_override call goes through, but later the test option is still rejected | 22:54 |
| harlowja | hmmmm | 22:55 |
| abashmak | in fact I had it set to False from the start, when I changed to enforce=True, it actually barfed on the set_override call | 22:55 |
| harlowja | part that i can't seem to get out of my mind is that it seems like it should be getting rejected :-P | 22:55 |
| harlowja | because it really isn't valid, even if u just want to bypass that during a test | 22:55 |
| abashmak | I actually had the same question as you, but being new thought the test devs knew better. Looking back, I should've clarified it from the start | 22:56 |
| abashmak | Let me pose that as an existential question to the experts on my project | 22:57 |
| harlowja | dhellmann do u know the best way to do that (is it even possible) ^ | 22:57 |
| harlowja | dhellmann knows oslo.config better than mysel | 22:57 |
| harlowja | *myself | 22:57 |
| *** spzala has quit IRC | 22:58 | |
| *** spzala has joined #openstack-oslo | 22:58 | |
| dhellmann | abashmak : what is your test trying to do? | 22:58 |
| dhellmann | that is, are you testing the option class or are you trying to set an option to a "bad" value and test your code somehow? | 22:59 |
| abashmak | This is in Glance, it's testing the images API. We have several options for communicating with the DAL (database abstraction layer): including SQLAlchemy or an optional Registry service. These are specified as choices in the 'data_api' config. The test is testing using the Registry. But instead of providing a valid value, it's setting a mock value. The only reason that I can think of, being that the Registry is a separate service and perhaps | 23:01 |
| abashmak | not running in the functional test environment | 23:01 |
| *** spzala has quit IRC | 23:03 | |
| dhellmann | abashmak: do you have a link to the code? I'm still not understanding the issue. | 23:04 |
| abashmak | one sec | 23:05 |
| abashmak | dhellmann: https://github.com/openstack/glance/blob/master/glance/tests/functional/v2/test_images.py#L2906 | 23:06 |
| abashmak | and here's where the valid choices are being set: https://review.openstack.org/#/c/354332/7/glance/common/config.py | 23:07 |
| patchbot | abashmak: patch 354332 - glance - Improving help text for data access API option | 23:07 |
| *** amotoki has joined #openstack-oslo | 23:08 | |
| dhellmann | abashmak : ok, so the test is setting the data api module name but the patch you linked there is trying to restrict the valid set of names that can be used in that option, right? | 23:08 |
| abashmak | that's right | 23:09 |
| dhellmann | how does the api_server object load the registry API, where is that code? | 23:09 |
| abashmak | Looking... | 23:10 |
| *** amotoki has quit IRC | 23:13 | |
| abashmak | It's not obvious to me | 23:13 |
| dhellmann | normally the way I'd do this is mock out the function that instantiates the data_api thing so it returns a value without looking at the config at all, but I don't know if that's abstracted enough | 23:13 |
| dhellmann | the other way to do it is to use stevedore to load the plugins, and to use a test manager instead of a real manager, since you can inject whatever plugins you want that way | 23:13 |
| dhellmann | I don't know if glance is using stevedore, though, because if it was I wouldn't expect to see these long module names | 23:14 |
| abashmak | Sounds like I need to dig into and learn more about the code and change the test in one of the ways you suggest | 23:16 |
| dhellmann | yeah | 23:16 |
| abashmak | OK, thank you for your help dhellmann, harlowja | 23:17 |
| dhellmann | abashmak : have a look at glance/db/__init__.py -- I think the get_api() function in that module is the one you would need to mock | 23:18 |
| dhellmann | abashmak : http://paste.openstack.org/show/559757/ | 23:18 |
| abashmak | I see it. | 23:19 |
| abashmak | thanks! | 23:20 |
| dhellmann | abashmak : sure, I hope that helps! | 23:20 |
| *** amotoki has joined #openstack-oslo | 23:35 | |
| *** amotoki has quit IRC | 23:41 | |
| bknudson | zzzeek: sqlalchemy-migrate keeps a cache of objects & looks up when you create something, so here's from pdb: http://paste.openstack.org/show/560282/ | 23:44 |
| bknudson | somehow the PythonScript object for key '/opt/stack/keystone/keystone/common/sql/expand_repo/versions/001_make_password_create_at_non_nullable.py' is '/opt/stack/keystone/keystone/common/sql/contract_repo/versions/001_make_password_create_at_non_nullable.pyc' | 23:45 |
| bknudson | when it should be '/opt/stack/keystone/keystone/common/sql/expand_repo/versions/001_make_password_create_at_non_nullable.pyc' | 23:45 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!