| *** Eva-i has left #openstack-oslo | 00:16 | |
| *** zz_dimtruck is now known as dimtruck | 00:23 | |
| *** dims_ has joined #openstack-oslo | 00:27 | |
| *** dims has quit IRC | 00:28 | |
| *** vilobhmm11 has joined #openstack-oslo | 00:28 | |
| *** dims has joined #openstack-oslo | 00:33 | |
| *** harlowja_at_home has joined #openstack-oslo | 00:33 | |
| *** dims_ has quit IRC | 00:33 | |
| *** dimtruck is now known as zz_dimtruck | 00:39 | |
| *** EinstCrazy has quit IRC | 00:40 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.messaging: Bump rabbit_transient_queues_ttl to 30 mins https://review.openstack.org/292181 | 00:44 |
|---|---|---|
| *** david-lyle has quit IRC | 00:54 | |
| *** david-lyle has joined #openstack-oslo | 00:54 | |
| *** _amrith_ is now known as amrith | 01:35 | |
| *** suro-patz has quit IRC | 01:46 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.messaging: [WIP] Bump rabbit_transient_queues_ttl to 30 mins https://review.openstack.org/292181 | 01:49 |
| *** zz_dimtruck is now known as dimtruck | 02:05 | |
| *** dims has quit IRC | 02:07 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.messaging: [WIP] Bump rabbit_transient_queues_ttl to 30 mins https://review.openstack.org/292181 | 02:11 |
| *** dims has joined #openstack-oslo | 02:21 | |
| *** amrith is now known as _amrith_ | 02:23 | |
| *** vilobhmm11 has quit IRC | 02:27 | |
| *** dims has quit IRC | 02:41 | |
| *** gcb has quit IRC | 02:43 | |
| *** gcb has joined #openstack-oslo | 02:46 | |
| *** yamamoto has quit IRC | 02:55 | |
| *** dimtruck is now known as zz_dimtruck | 03:02 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/osprofiler: doc: Log warning when can't get informaiton from git https://review.openstack.org/292191 | 03:08 |
| *** zz_dimtruck is now known as dimtruck | 03:12 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/osprofiler: Dont claim copyright for future years https://review.openstack.org/292192 | 03:17 |
| *** gcb has quit IRC | 03:18 | |
| *** gcb has joined #openstack-oslo | 03:32 | |
| *** gcb has quit IRC | 03:36 | |
| *** dkehn has quit IRC | 03:45 | |
| openstackgerrit | Merged openstack/oslo.config: Fix typo error in docstrings of oslo_config.cfg.py https://review.openstack.org/291722 | 03:48 |
| *** links has joined #openstack-oslo | 03:50 | |
| *** dkehn has joined #openstack-oslo | 03:51 | |
| *** dimtruck is now known as zz_dimtruck | 03:52 | |
| openstackgerrit | Merged openstack/oslo.config: Explicitly exclude tests from bandit scan https://review.openstack.org/292053 | 03:56 |
| *** yamamoto has joined #openstack-oslo | 03:56 | |
| openstackgerrit | Merged openstack/oslo.utils: Explicitly exclude tests from bandit scan https://review.openstack.org/292055 | 03:57 |
| *** ekarlso- has quit IRC | 04:00 | |
| *** mtanino has quit IRC | 04:02 | |
| *** harlowja_at_home has quit IRC | 04:04 | |
| *** andymaier has joined #openstack-oslo | 04:07 | |
| *** ekarlso- has joined #openstack-oslo | 04:14 | |
| *** gcb has joined #openstack-oslo | 04:21 | |
| *** andymaier has quit IRC | 04:23 | |
| *** zz_dimtruck is now known as dimtruck | 04:27 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/osprofiler: Remove outdated version https://review.openstack.org/292203 | 04:44 |
| *** otherwiseguy has quit IRC | 05:02 | |
| *** boris-42 has joined #openstack-oslo | 05:06 | |
| *** vilobhmm11 has joined #openstack-oslo | 05:08 | |
| *** otherwiseguy has joined #openstack-oslo | 05:12 | |
| *** _amrith_ is now known as amrith | 05:29 | |
| *** EinstCrazy has joined #openstack-oslo | 05:31 | |
| *** EinstCrazy has quit IRC | 05:38 | |
| *** amrith is now known as _amrith_ | 05:39 | |
| *** ihrachys has joined #openstack-oslo | 06:02 | |
| openstackgerrit | Feng Xi Yan proposed openstack/oslo.vmware: Add a locale param to allow vSphere SDK locale specification. https://review.openstack.org/218755 | 06:06 |
| *** dimtruck is now known as zz_dimtruck | 06:19 | |
| *** ihrachys_ has joined #openstack-oslo | 06:19 | |
| *** ihrachys has quit IRC | 06:21 | |
| *** zz_dimtruck is now known as dimtruck | 06:23 | |
| *** dimtruck is now known as zz_dimtruck | 06:36 | |
| *** pcaruana has quit IRC | 06:39 | |
| *** vilobhmm111 has joined #openstack-oslo | 06:42 | |
| *** vilobhmm11 has quit IRC | 06:44 | |
| *** ihrachys has joined #openstack-oslo | 06:46 | |
| *** ihrachys_ has quit IRC | 06:47 | |
| *** ihrachys_ has joined #openstack-oslo | 06:53 | |
| *** ihrachys has quit IRC | 06:55 | |
| *** nkrinner has joined #openstack-oslo | 06:58 | |
| *** ihrachys_ has quit IRC | 07:04 | |
| *** patchbot has quit IRC | 07:04 | |
| *** patchbot` has joined #openstack-oslo | 07:04 | |
| *** patchbot` is now known as patchbot | 07:05 | |
| *** dmitryme2 has joined #openstack-oslo | 07:14 | |
| *** rpodolyaka_ has joined #openstack-oslo | 07:14 | |
| *** isq_ has quit IRC | 07:15 | |
| *** pkholkin__ has quit IRC | 07:15 | |
| *** eezhova has quit IRC | 07:15 | |
| *** sileht has quit IRC | 07:15 | |
| *** szaher has quit IRC | 07:15 | |
| *** rpodolyaka has quit IRC | 07:15 | |
| *** dmitryme has quit IRC | 07:15 | |
| *** mkoderer has quit IRC | 07:15 | |
| *** dstanek has quit IRC | 07:15 | |
| *** dmitryme2 is now known as dmitryme | 07:15 | |
| *** rpodolyaka_ is now known as rpodolyaka | 07:15 | |
| *** isq_ has joined #openstack-oslo | 07:16 | |
| *** pkholkin__ has joined #openstack-oslo | 07:16 | |
| *** eezhova has joined #openstack-oslo | 07:16 | |
| *** szaher has joined #openstack-oslo | 07:16 | |
| *** mkoderer has joined #openstack-oslo | 07:16 | |
| *** dstanek has joined #openstack-oslo | 07:16 | |
| *** sileht has joined #openstack-oslo | 07:16 | |
| *** ihrachys has joined #openstack-oslo | 07:30 | |
| *** tesseract has joined #openstack-oslo | 07:52 | |
| *** tesseract is now known as Guest59187 | 07:52 | |
| *** vilobhmm111 has quit IRC | 08:03 | |
| *** vilobhmm11 has joined #openstack-oslo | 08:04 | |
| *** rcernin has joined #openstack-oslo | 08:25 | |
| *** dguitarbite has joined #openstack-oslo | 08:27 | |
| *** dtardivel has joined #openstack-oslo | 08:27 | |
| *** pcaruana has joined #openstack-oslo | 08:43 | |
| haypo | morning | 08:46 |
| *** permalac has joined #openstack-oslo | 08:50 | |
| *** vilobhmm11 has quit IRC | 09:01 | |
| *** gcb has quit IRC | 09:05 | |
| *** mhickey has joined #openstack-oslo | 09:06 | |
| *** ndipanov has joined #openstack-oslo | 09:14 | |
| *** gcb has joined #openstack-oslo | 09:18 | |
| *** lucas-afk is now known as lucasagomes | 09:19 | |
| *** yassine__ has joined #openstack-oslo | 09:27 | |
| *** pblaho has joined #openstack-oslo | 09:33 | |
| *** e0ne has joined #openstack-oslo | 09:52 | |
| *** gcb has quit IRC | 09:53 | |
| *** dims has joined #openstack-oslo | 10:09 | |
| *** sdague has joined #openstack-oslo | 10:13 | |
| *** dims has quit IRC | 10:21 | |
| *** cdent has joined #openstack-oslo | 10:28 | |
| *** dims has joined #openstack-oslo | 10:30 | |
| *** dims has quit IRC | 10:39 | |
| *** dims has joined #openstack-oslo | 10:40 | |
| *** e0ne has quit IRC | 10:45 | |
| *** yassine__ has quit IRC | 10:55 | |
| *** e0ne has joined #openstack-oslo | 10:56 | |
| openstackgerrit | Merged openstack/osprofiler: Remove outdated version https://review.openstack.org/292203 | 11:01 |
| *** takedakn has joined #openstack-oslo | 11:02 | |
| *** _amrith_ is now known as amrith | 11:15 | |
| *** cdent has quit IRC | 11:17 | |
| *** ihrachys has quit IRC | 11:18 | |
| *** e0ne has quit IRC | 11:21 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.messaging: Bump rabbit_transient_queues_ttl to 30 mins https://review.openstack.org/292181 | 11:24 |
| dims | sileht : i was tracking this down all weekend. ^^ | 11:25 |
| dims | dmitryme : please take a look too | 11:25 |
| sileht | dims, well done | 11:26 |
| *** e0ne has joined #openstack-oslo | 11:26 | |
| dims | sileht : hopefully it will help | 11:27 |
| *** cdent has joined #openstack-oslo | 11:27 | |
| dims | thanks sileht | 11:27 |
| *** gordc has joined #openstack-oslo | 11:35 | |
| *** cdent_ has joined #openstack-oslo | 11:37 | |
| dims | gordc : dukhlov : does this look good to you? https://review.openstack.org/#/c/292181/ | 11:37 |
| patchbot | dims: patch 292181 - oslo.messaging - Bump rabbit_transient_queues_ttl to 30 mins | 11:37 |
| *** cdent has quit IRC | 11:37 | |
| *** cdent_ is now known as cdent | 11:37 | |
| *** ujjain- is now known as ujjain | 11:39 | |
| dims | dukhlov : ozamiatin_ : sileht : gordc : dmitryme : Now that we have stable/mitaka, i'd like to see what's the minimum reviews we absolutely need that was not in the last release of oslo.messaging and came up with these two - https://review.openstack.org/#/q/is:open+project:openstack/oslo.messaging+branch:stable/mitaka | 11:40 |
| dims | dukhlov : ozamiatin_ : sileht : gordc : dmitryme : Are there others? (Objective is to make sure rabbitmq+kombu is production ready), the zmq and pika driver we can wait a bit for a release | 11:41 |
| *** cdent has quit IRC | 11:41 | |
| sileht | I have nothing in mind | 11:41 |
| *** cdent has joined #openstack-oslo | 11:42 | |
| *** binarin has joined #openstack-oslo | 11:43 | |
| gordc | dims: those two seem fine to me. not tracking anything else. | 11:44 |
| *** cdent has quit IRC | 11:45 | |
| *** dims_ has joined #openstack-oslo | 11:45 | |
| *** dims has quit IRC | 11:47 | |
| dims_ | binarin : long story, but here's the short version, in https://review.openstack.org/#/c/243845/ we 'x-expires' in a queue and specified a ttl | 11:47 |
| patchbot | dims_: patch 243845 - oslo.messaging - Kombu: make reply and fanout queues expire instead... (MERGED) | 11:47 |
| dims_ | binarin : so when exactly does the the queue expire? is the time counted from the last time something was sent using that queue? | 11:49 |
| dims_ | binarin : what i see is that the ttl clock seems to start when the queue is created, not when it's last used | 11:50 |
| dims_ | binarin : can you please check when you get a chance? | 11:50 |
| binarin | dims_: Yes, I'm already looking into the rabbit source | 11:50 |
| dims_ | thanks binarin : documentation says "Unused means the queue has no consumers, the queue has not been redeclared, and basic.get has not been invoked for a duration of at least the expiration period." https://www.rabbitmq.com/ttl.html | 11:51 |
| *** cdent has joined #openstack-oslo | 11:51 | |
| dims_ | binarin : temporarily, i've bumped ttl in https://review.openstack.org/#/c/292181/ hopefully when it gets into a release it will help | 11:55 |
| patchbot | dims_: patch 292181 - oslo.messaging - Bump rabbit_transient_queues_ttl to 30 mins | 11:55 |
| *** yassine__ has joined #openstack-oslo | 11:57 | |
| *** yamamoto has quit IRC | 12:00 | |
| *** ihrachys has joined #openstack-oslo | 12:01 | |
| *** ihrachys has quit IRC | 12:03 | |
| dmitryme | dims: +1’ed https://review.openstack.org/#/c/292181/ . But frankly I hardly understand, how it happens that a queue lives 10 minutes without consumer. | 12:03 |
| patchbot | dmitryme: patch 292181 - oslo.messaging - Bump rabbit_transient_queues_ttl to 30 mins | 12:03 |
| dmitryme | and nothing to add to stable/liberty from my side | 12:03 |
| dims_ | dmitryme : right | 12:03 |
| *** ihrachys has joined #openstack-oslo | 12:03 | |
| *** kgiusti has joined #openstack-oslo | 12:07 | |
| openstackgerrit | javeme proposed openstack/oslo.messaging: Missing version parameter in can_send_version() https://review.openstack.org/290385 | 12:13 |
| *** amrith is now known as _amrith_ | 12:19 | |
| *** claudiub has joined #openstack-oslo | 12:27 | |
| *** edmondsw has joined #openstack-oslo | 12:31 | |
| *** boris-42 has quit IRC | 12:34 | |
| *** MVenesio has joined #openstack-oslo | 12:40 | |
| openstackgerrit | Merged openstack/oslo.messaging: Pika: fix sending fanout messages https://review.openstack.org/289430 | 12:40 |
| *** links has quit IRC | 12:46 | |
| *** links has joined #openstack-oslo | 12:46 | |
| *** yamamoto has joined #openstack-oslo | 12:48 | |
| *** mtanino has joined #openstack-oslo | 12:52 | |
| *** lucasagomes is now known as lucas-hungry | 12:54 | |
| *** cdent has quit IRC | 13:01 | |
| *** cdent has joined #openstack-oslo | 13:02 | |
| *** EinstCrazy has joined #openstack-oslo | 13:04 | |
| *** rlrossit has joined #openstack-oslo | 13:05 | |
| *** cdent has quit IRC | 13:06 | |
| *** _amrith_ is now known as amrith | 13:16 | |
| *** takedakn has quit IRC | 13:21 | |
| dims_ | folks, please check your local times :) we are meeting at 1600 UTC | 13:29 |
| *** krotscheck has joined #openstack-oslo | 13:32 | |
| *** gcb has joined #openstack-oslo | 13:33 | |
| *** cdent has joined #openstack-oslo | 13:35 | |
| *** jpena is now known as jpena|lunch | 13:48 | |
| krotscheck | dhellmann: Just went through and rechecked the various patches that failed in the cors bug. Other than that, it looks like most of them are landing. | 13:52 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:00 | |
| *** ig0r_ has joined #openstack-oslo | 14:02 | |
| *** lucas-hungry is now known as lucasagomes | 14:03 | |
| *** mriedem has joined #openstack-oslo | 14:04 | |
| *** boris-42 has joined #openstack-oslo | 14:05 | |
| *** amrith is now known as _amrith_ | 14:06 | |
| *** _amrith_ is now known as amrith | 14:06 | |
| *** salv-orlando has joined #openstack-oslo | 14:08 | |
| *** zz_dimtruck is now known as dimtruck | 14:12 | |
| *** david-lyle has quit IRC | 14:16 | |
| *** david-lyle has joined #openstack-oslo | 14:19 | |
| *** yamamoto has quit IRC | 14:25 | |
| *** pcaruana has quit IRC | 14:28 | |
| *** ansiwen has quit IRC | 14:29 | |
| *** mriedem is now known as mriedem_meeting | 14:30 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.db: Remove sqlalchemy < 1.0.0 compatible layer https://review.openstack.org/292413 | 14:31 |
| *** yamahata has quit IRC | 14:31 | |
| *** ansiwen has joined #openstack-oslo | 14:33 | |
| dhellmann | krotscheck : thanks for staying on top of those | 14:33 |
| *** mtanino has quit IRC | 14:34 | |
| *** yamamoto has joined #openstack-oslo | 14:35 | |
| *** yamamoto has quit IRC | 14:36 | |
| krotscheck | dhellmann: Of course. | 14:36 |
| *** binarin has quit IRC | 14:37 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Refactoring consumer side https://review.openstack.org/283049 | 14:41 |
| *** gcb has quit IRC | 14:41 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Refactoring consumer side https://review.openstack.org/283049 | 14:41 |
| *** gcb has joined #openstack-oslo | 14:43 | |
| *** mriedem_meeting is now known as mriedem | 14:49 | |
| dims_ | oslo-cores, please help me fill in https://etherpad.openstack.org/p/mitaka-oslo-achievements | 14:49 |
| haypo | dims_: is the death of openstack/common/ in oslo-incubator an archivement? :) | 14:51 |
| openstackgerrit | Julien Danjou proposed openstack/oslo.middleware: cors: prevent WebOb setting a default Content-Type https://review.openstack.org/290482 | 14:51 |
| dims_ | haypo : goes into the last bucket :) | 14:52 |
| haypo | (in fact, i'm not sure that it's 100% dead) | 14:52 |
| haypo | dims_: is it dead or is it still moving? | 14:52 |
| *** yamamoto has joined #openstack-oslo | 14:52 | |
| dims_ | haypo : we can say needs cleanup from projects | 14:53 |
| dims_ | haypo : we have released all the code as libs. what's left is cleanup | 14:53 |
| krotscheck | dhellmann: Just to confirm - that default config hook name has to be unique, but the actual string doesn't matter, yes? | 14:54 |
| *** jpena|lunch is now known as jpena | 14:56 | |
| dhellmann | krotscheck : the string does matter. it must match the name of one of the plugins registered for the application to return a list of options | 14:58 |
| krotscheck | dhellmann: ....RIGHT | 14:58 |
| * krotscheck goes back to confirm his patches. | 14:58 | |
| dhellmann | krotscheck : if a project has 2 apps, and they want different defaults, it can register 2 functions and have 2 input files to the config generator using the different names | 14:58 |
| dhellmann | s/using/containing | 14:58 |
| rbradfor | haypo, has all code become libraries? from looking at what's still in use some does not seem to have an obvious migration path | 15:06 |
| haypo | rbradfor: no. cliutils.py simply died :) | 15:06 |
| haypo | rbradfor: it was decided to keep unsynchronized copies in each projects, and these projects must be updated to cliff | 15:07 |
| rbradfor | haypo, do you see all the client libs moving to cliff. there is probably 15+ using cliutils | 15:10 |
| dims_ | krotscheck : wanna add CORS stuff to https://etherpad.openstack.org/p/mitaka-oslo-achievements ? | 15:11 |
| krotscheck | dims_: Sure! | 15:11 |
| dims_ | haypo : something about python3 readiness? | 15:11 |
| haypo | rbradfor: sorry, is it a question? | 15:11 |
| krotscheck | dims_: New features? | 15:12 |
| haypo | dims_: oslo was already ready in the previous cycle. but i added some helpers to make porting easier | 15:12 |
| dims_ | krotscheck : sure | 15:12 |
| dims_ | haypo : ok, counts as new features? :) | 15:12 |
| dims_ | rbradfor : what else can we add? https://etherpad.openstack.org/p/mitaka-oslo-achievements ? | 15:12 |
| *** amotoki has quit IRC | 15:14 | |
| *** ozamiatin has joined #openstack-oslo | 15:16 | |
| rbradfor | dims_, cannot say a lot of my small contributions getting started this cycle are noteworthy | 15:23 |
| *** links has quit IRC | 15:23 | |
| *** ig0r__ has joined #openstack-oslo | 15:23 | |
| *** regXboi has joined #openstack-oslo | 15:25 | |
| *** ig0r_ has quit IRC | 15:26 | |
| rbradfor | little consolation, however it does help me with what's important for a full newton cycle. | 15:29 |
| *** mriedem is now known as mriedem_meeting | 15:30 | |
| *** yamamoto has quit IRC | 15:30 | |
| *** salv-orlando has quit IRC | 15:31 | |
| *** salv-orlando has joined #openstack-oslo | 15:33 | |
| *** harlowja_at_home has joined #openstack-oslo | 15:38 | |
| *** AJaeger has joined #openstack-oslo | 15:53 | |
| *** gcb has quit IRC | 15:54 | |
| dims_ | 4 mins to oslo meeting | 15:56 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.serialization: msgpack: fix datetime serialization https://review.openstack.org/291715 | 15:56 |
| harlowja_at_home | meeting! | 15:57 |
| *** jecarey has joined #openstack-oslo | 15:57 | |
| harlowja_at_home | sileht, thanks for fixing those, guess bugs i never saw :-P | 16:00 |
| harlowja_at_home | (when making said code) | 16:00 |
| *** pcaruana has joined #openstack-oslo | 16:09 | |
| *** jaypipes has joined #openstack-oslo | 16:10 | |
| *** cdent has quit IRC | 16:12 | |
| *** nkrinner has quit IRC | 16:13 | |
| *** EinstCrazy has quit IRC | 16:14 | |
| *** mriedem_meeting is now known as mriedem | 16:21 | |
| *** yamamoto has joined #openstack-oslo | 16:31 | |
| *** Guest59187 has quit IRC | 16:31 | |
| haypo | sileht: i'm not sure that https://review.openstack.org/#/c/291715/ is 100% safe, but i'm ok to come back to the code later if someone complains ;) | 16:34 |
| patchbot | haypo: patch 291715 - oslo.serialization - msgpack: fix datetime serialization | 16:34 |
| haypo | sileht: (i just approved your change) | 16:34 |
| sileht | haypo: thx | 16:35 |
| *** cdent has joined #openstack-oslo | 16:39 | |
| *** yamamoto has quit IRC | 16:41 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 16:46 | |
| haypo | sileht: it looks like timezone (used in OpenStack) are always encoded to ASCII | 16:48 |
| haypo | sileht: hopefully, the issue goes away with python 3 :-D | 16:49 |
| *** rcernin has quit IRC | 16:56 | |
| *** sputnik13 has joined #openstack-oslo | 17:04 | |
| *** vilobhmm11 has joined #openstack-oslo | 17:07 | |
| *** suro-patz has joined #openstack-oslo | 17:07 | |
| *** yamahata has joined #openstack-oslo | 17:07 | |
| *** dims has joined #openstack-oslo | 17:08 | |
| *** vilobhmm111 has joined #openstack-oslo | 17:09 | |
| *** dims_ has quit IRC | 17:09 | |
| *** vilobhmm11 has quit IRC | 17:11 | |
| *** yassine__ has quit IRC | 17:15 | |
| *** e0ne has quit IRC | 17:17 | |
| openstackgerrit | Merged openstack/oslo.messaging: Bump rabbit_transient_queues_ttl to 30 mins https://review.openstack.org/292181 | 17:40 |
| *** jecarey_ has joined #openstack-oslo | 17:50 | |
| *** alejandrito has joined #openstack-oslo | 17:51 | |
| harlowja_at_home | dims, http://lists.openstack.org/pipermail/openstack-dev/2016-March/089276.html (and others) | 17:51 |
| harlowja_at_home | included a few newton references, lol | 17:52 |
| dims | lol | 17:52 |
| *** jecarey has quit IRC | 17:53 | |
| *** salv-orl_ has joined #openstack-oslo | 17:54 | |
| *** salv-orlando has quit IRC | 17:57 | |
| harlowja_at_home | ya, whoever names these summits needs to know all the secret meanings of these words, ha | 17:57 |
| openstackgerrit | Merged openstack/oslo.serialization: msgpack: fix datetime serialization https://review.openstack.org/291715 | 17:57 |
| harlowja_at_home | fig newton, iaasac newton, apple newton | 17:58 |
| harlowja_at_home | probably other newtons, lol | 17:58 |
| harlowja_at_home | although 'apple newton' shouldn't be used, unless u want to talk about a failure | 17:58 |
| * harlowja_at_home had one of those once | 17:58 | |
| *** browne has joined #openstack-oslo | 17:58 | |
| harlowja_at_home | apples first attempt at a pda that failed (iphone finally worked out) | 17:58 |
| harlowja_at_home | https://en.wikipedia.org/wiki/Apple_Newton | 17:58 |
| *** salv-orl_ has quit IRC | 17:59 | |
| harlowja_at_home | wayne newton | 18:00 |
| harlowja_at_home | lol | 18:00 |
| dims | haha | 18:02 |
| *** jpena is now known as jpena|off | 18:05 | |
| AJaeger | fungi, rpodolyaka, zzzeek, dims: We (=OpenStack Infra) currently always set up for python27, python34 etc jobs MySQL and PostgreSQL adatabases. We want to only setup databases when needed and I need some help from you. | 18:05 |
| AJaeger | Is a project using oslo.db always needing a setup database? Or how to determine if one needs it? | 18:06 |
| AJaeger | I mean: During the test runs... | 18:06 |
| *** vilobhmm111 has quit IRC | 18:07 | |
| fungi | ideal would be some pattern we can look for to tell whether projects are likely relying on opportunistic database testing with fixtures/framework support provided by oslo.db | 18:07 |
| *** ihrachys has quit IRC | 18:10 | |
| *** e0ne has joined #openstack-oslo | 18:11 | |
| *** ozamiatin has quit IRC | 18:12 | |
| *** harlowja_at_home has quit IRC | 18:14 | |
| *** mhickey has quit IRC | 18:15 | |
| kevinbenton | zzzeek: what do you suggest I do for https://review.openstack.org/#/c/291795/ ? The session has already been rolled back by the transaction context manager | 18:16 |
| patchbot | kevinbenton: patch 291795 - neutron - Call expunge_all on session before retrying requests | 18:16 |
| *** permalac has quit IRC | 18:19 | |
| kevinbenton | zzzeek: for context, this is the leftover object in the session that i was chatting with you about the other day that would go away when using pdb | 18:20 |
| *** lucasagomes is now known as lucas-dinner | 18:25 | |
| *** cdent has quit IRC | 18:37 | |
| *** yamahata has quit IRC | 18:40 | |
| *** yamahata has joined #openstack-oslo | 18:40 | |
| *** ozamiatin has joined #openstack-oslo | 18:44 | |
| zzzeek | kevinbenton: added a comment to the launchpad issue illustrating weakrefs should not matter | 18:47 |
| *** vilobhmm11 has joined #openstack-oslo | 18:51 | |
| *** cdent has joined #openstack-oslo | 18:55 | |
| kevinbenton | zzzeek: managed to repro with your example and a relationship. left code on the bug report | 19:18 |
| zzzeek | ok | 19:18 |
| zzzeek | kevinbenton: passive_deletes = 'all' == probably wrong, why is that there? | 19:19 |
| kevinbenton | zzzeek: is that the cause. it's something that has been in the neutron model for quite some time. i will need to git blame to see why it's there | 19:20 |
| kevinbenton | zzzeek: but if that's it, that's great news because we can probably just dump it | 19:20 |
| zzzeek | kevinbenton: yes, because SQLA has no idea you're deleting those | 19:20 |
| zzzeek | kevinbenton: it should be: cascade='all, delete-orphan' and passive_deletes=True | 19:20 |
| kevinbenton | zzzeek: ok. cool. let me figure out why that was done | 19:21 |
| kevinbenton | zzzeek: thx | 19:21 |
| zzzeek | kevinbenton: whoever put passive_deletes='all' there was basically doing the "try everything until it works" solution which is what I'm trying to interrupt here :) | 19:22 |
| *** claudiub|2 has joined #openstack-oslo | 19:22 | |
| *** ozamiatin has quit IRC | 19:23 | |
| *** claudiub has quit IRC | 19:25 | |
| kevinbenton | zzzeek: it looks like it was done as some of the quota work https://review.openstack.org/#/c/188883/ | 19:25 |
| patchbot | kevinbenton: patch 188883 - neutron - Introduce usage data tracking for Neutron (MERGED) | 19:25 |
| kevinbenton | zzzeek: unfortunately nobody asked about it in the code review so i will need to get in touch with Salvatore to figure out why he added it | 19:26 |
| kevinbenton | zzzeek: (fixed_ips is the relationship that is causing the issue) | 19:27 |
| *** suro-patz1 has joined #openstack-oslo | 19:28 | |
| *** suro-patz has quit IRC | 19:29 | |
| kevinbenton | zzzeek: this still fails. http://paste.openstack.org/show/490430/ | 19:36 |
| *** fhermeni has joined #openstack-oslo | 19:37 | |
| kevinbenton | zzzeek: it even fails with no passive_deletes option | 19:37 |
| kevinbenton | zzzeek: and also with no cascade option | 19:37 |
| *** yamahata has quit IRC | 19:38 | |
| fhermeni | hi | 19:38 |
| zzzeek | kevinbenton: looking | 19:44 |
| zzzeek | kevinbenton: ok that is a much more subtle variant on the same thing | 19:46 |
| zzzeek | and ive actually never seen that one before | 19:46 |
| kevinbenton | zzzeek: the wonky thing is the lookup using the relationship | 19:47 |
| zzzeek | kevinbenton: well sort of | 19:47 |
| zzzeek | kevinbenton: i know what this is, and it is very easy to resolve, but im not sure what i can tell you about this never happening b.c. i think openstack's patterns lend themselves to this kind of thing | 19:48 |
| zzzeek | kevinbenton: if you do: | 19:48 |
| zzzeek | s.flush() | 19:48 |
| zzzeek | b1 = B(a_id=1, address='1234') | 19:48 |
| zzzeek | s.add(b1) | 19:48 |
| zzzeek | then its fine | 19:48 |
| zzzeek | because | 19:48 |
| zzzeek | it knows that b1 is part of that transaction. | 19:48 |
| zzzeek | if you do: s.add(B(...)) | 19:48 |
| zzzeek | then it gets gc'ed | 19:48 |
| zzzeek | then you load it again | 19:48 |
| zzzeek | the Session has no idea it was the one you just added in that transaction | 19:48 |
| zzzeek | kevinbenton: there's something to be improved here in SQLA possibly, not usre | 19:50 |
| kevinbenton | zzzeek: ack. unfortunately holding that reference will have to carry through lots of code because the record creation for an allocation is isolated to a function https://github.com/openstack/neutron/blob/master/neutron/db/db_base_plugin_common.py#L94-L109 | 19:52 |
| zzzeek | kevinbenton: yes i figured that's whats happening | 19:52 |
| zzzeek | kevinbenton: for your decorator I'd just put session.close() at the end | 19:53 |
| zzzeek | as long as its run at the toplevel transaction | 19:53 |
| zzzeek | if its doing savepoints then you'd not want that | 19:53 |
| kevinbenton | zzzeek: ack. it's not doing savepoints. we have been good about decorating at the very top-level | 19:53 |
| zzzeek | yeah id just close() it | 19:54 |
| zzzeek | or even just make a new Session | 19:54 |
| kevinbenton | zzzeek: yeah, at one point i was just discarding the session on the context so a new one would get created | 19:54 |
| kevinbenton | zzzeek: but it felt a little heavy-handed | 19:54 |
| kevinbenton | zzzeek: at least now i understand why it's happening | 19:55 |
| kevinbenton | zzzeek: and have a better justification for it | 19:55 |
| zzzeek | kevinbenton: on the SQLA side we don't hold onto thigns that arent referenced brcause if someone is inserting 100K recs it would be foolish for us to make strong refs | 19:55 |
| zzzeek | kevinbenton: you *can* just throw it in session.info also | 19:56 |
| *** mtanino has joined #openstack-oslo | 19:56 | |
| kevinbenton | zzzeek: ah, that might be less invasive | 19:56 |
| zzzeek | kevinbenton: e.g. this could be even on SQLA side "make sure this rolls back" type of thing | 19:56 |
| zzzeek | kevinbenton: but it will look strange to people | 19:56 |
| kevinbenton | zzzeek: well it will be isolated to one particular part of the code with a nice TODO to clean this up later | 19:56 |
| zzzeek | kevinbenton: OK, im jsut not sure what the long term solution to this one wuld be | 19:57 |
| kevinbenton | zzzeek: the decorator modification is messier because the decorator doesn't know which argument is the context | 19:57 |
| kevinbenton | zzzeek: but i think it makes sense to have a clean session in the long run just as a way to completely ensure everything is started from scratch | 19:57 |
| *** dtardivel has quit IRC | 19:58 | |
| *** claudiub|2 has quit IRC | 20:01 | |
| zzzeek | kevinbenton: so this works | 20:03 |
| zzzeek | s.flush() | 20:03 |
| zzzeek | b1 = B(a_id=1, address='1234') | 20:03 |
| zzzeek | s.add(b1) | 20:03 |
| zzzeek | s.info['my_b1'] = b1 | 20:03 |
| zzzeek | del b1 | 20:03 |
| *** ig0r_ has joined #openstack-oslo | 20:03 | |
| kevinbenton | zzzeek: yep, just modified the logic in ipam to add it and it works as well | 20:05 |
| kevinbenton | zzzeek: adding a note now | 20:05 |
| zzzeek | kevinbenton: OK. i just find it to bea shortcoming in SQLA that one even has to worry about a ref here | 20:05 |
| *** ig0r__ has quit IRC | 20:07 | |
| kevinbenton | zzzeek: can you leave a comment on the bug explaining that the root cause is because this is GC'ed? | 20:07 |
| *** salv-orlando has joined #openstack-oslo | 20:08 | |
| *** dimtruck is now known as zz_dimtruck | 20:08 | |
| AJaeger | zzzeek: did you see my question in the backscroll and can help infra, please? | 20:09 |
| zzzeek | kevinbenton: i cna but im doing 8 things at onec here, if you want to give it a shot | 20:10 |
| kevinbenton | zzzeek: no prob | 20:10 |
| zzzeek | AJaeger: my backscroll is unreliable let me check | 20:10 |
| *** amrith is now known as _amrith_ | 20:10 | |
| zzzeek | AJaeger: negatory, if you want to repaste for me otherwise i have to go look ona server at the logs | 20:11 |
| zzzeek | AJaeger: oh here, i was looking on openstack-infra | 20:11 |
| zzzeek | AJaeger: yes | 20:11 |
| *** zz_dimtruck is now known as dimtruck | 20:12 | |
| zzzeek | AJaeger: oslo.db has a front-to-back version of this but you have to buy into it a little bit | 20:12 |
| *** MVenesio has quit IRC | 20:12 | |
| zzzeek | AJaeger: well i guess you could use provisioning externally too... would need to know what you're trying to do | 20:13 |
| *** kgiusti has quit IRC | 20:13 | |
| AJaeger | zzzeek: here ;) Shall I repaste? | 20:17 |
| *** AJaeger has quit IRC | 20:17 | |
| *** dguitarbite has quit IRC | 20:17 | |
| *** AJaeger has joined #openstack-oslo | 20:17 | |
| AJaeger | zzzeek: sorry, lost connectivty but read your comment on eavesdrop | 20:17 |
| AJaeger | zzzeek, the problem that fungi and myself try to solve is: Currently python27 jobs setup databases for each repository. | 20:18 |
| AJaeger | But most repos don't need a database for their python27 jobs. | 20:18 |
| zzzeek | AJaeger: yes are these tempest jobs | 20:18 |
| AJaeger | not tempest, we talk about unittests | 20:18 |
| zzzeek | AJaeger: OK | 20:18 |
| AJaeger | So, gate-nova-python27 | 20:18 |
| AJaeger | how can we figure out which repos need database access - so that only these get database setup? | 20:19 |
| zzzeek | AJaeger: so how do we define what "setup" means? | 20:19 |
| zzzeek | AJaeger: just "CREATE DATABASE" "DROP DATABASE", or actualyl running a DB in some kind of container or something? | 20:20 |
| AJaeger | setup means: We create a gate-nova-python27-db job that sets up the database access as it's done right now and remove it from gate-nova-python27 | 20:20 |
| zzzeek | AJaeger: OK is there a DB server already running, and you create a username, or theyre using openstack-citest ? | 20:21 |
| AJaeger | setup is http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/macros.yaml#n643 | 20:21 |
| fungi | yeah, we install mysql and postgresql servers, precreate openstack_citest accounts with basically full database control, and also precreate openstack_citest databases which some jobs use | 20:21 |
| AJaeger | and http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/macros.yaml#n681 | 20:21 |
| fungi | wondering whether there are some projects relying on testing features of oslo.db which we need to spot separately from just grepping for "openstack_citest" in their repos directly | 20:22 |
| *** thingee has joined #openstack-oslo | 20:22 | |
| zzzeek | AJaeger / fungi OK if the DB is already running, and I assume the test suite gets exclusive control of that DB, what is the downside to just having a persistent openstack_citest user? concern that the DB messes up the user or some kind of security concern ? | 20:23 |
| thingee | fhermeni: hey are you interested in talking about this with a wider audience tomorrow at the cross-project meeting? https://wiki.openstack.org/wiki/Meetings/CrossProjectMeeting | 20:23 |
| AJaeger | zzzeek: we want to allow projects to install their own package lists. Right now since we always setup the database, they would need to install database packages even if they don't need them. | 20:24 |
| fungi | zzzeek: we're whittling down what we preconfigure on our base images. we don't install any database services on them at all now, but we've temporarily worked around not knowing which jobs need database support to run their tests by having all unit test and coverage jobs install mysql and portgres and set them up similarly to how our old-style images did | 20:24 |
| *** dguitarbite has joined #openstack-oslo | 20:25 | |
| AJaeger | So, we like to make it possible that a project that doesn't need a database, has the option to install a package set iwthout database | 20:25 |
| fungi | er, s/portgres/postgres/ | 20:25 |
| kevinbenton | zzzeek: https://review.openstack.org/291795 thanks for all of your help as always! | 20:26 |
| fungi | zzzeek: so the reason we're asking is that we're trying to figure out which jobs really do make use of these preconfigured databases. one way is to look for "openstack_citest" in all their repos and try to investigate from that angle, but we're worried there may be indirect use through oslo.db too | 20:27 |
| fungi | so we want to make sure we find those as well if so | 20:27 |
| fhermeni | thingee: Hi thingee. I think I can, but Joshua is a bit more used to such procedures first | 20:27 |
| zzzeek | fungi: OK well if you have scripts that run outside of a test, and you want to be handed a repo and generically determine what kinds of databases that suite will want to see, that repo would have to publish that metadata somehow | 20:28 |
| zzzeek | fungi: i wouldn't want to do it by guessing | 20:28 |
| zzzeek | fungi: id say we'd be looking in either pbr / setup.cfg / tox.ini something like that for a known indicator | 20:28 |
| thingee | fhermeni: ok, either of you can lead it. Just need someone that understands the subject to brief introduce it and answer questions to the group. | 20:28 |
| thingee | fhermeni: it would also be good to identify the immediate projects this involves. | 20:29 |
| zzzeek | fungi: otherwise if you're looking for heuristics that would tell you from looking at the code itself if it needs a DB then I can figure that out for you but that seems a little more brittle | 20:29 |
| fhermeni | thingee: It will be 9:00 UTC, so 10 am in France. My kid will be asleep, so I will be able to attend anyway | 20:29 |
| fungi | zzzeek: i'll try to rephrase again | 20:29 |
| *** jecarey_ has quit IRC | 20:29 | |
| fungi | zzzeek: does oslo.db provide opportunistic database testing fixtures/framework that may assume the presence of a database? | 20:30 |
| zzzeek | fungi: there is probably a function in there called "has_database" or somethign like that, i can check | 20:30 |
| fhermeni | thingee: ok. I can discuss about the objective and what we had in mind. Some questions might however be too technical for me wrt. my knowlegde about OS architecture and principles. Joshua matters then for that second point | 20:30 |
| zzzeek | fungi: there's also the whole test harness that will run tests only if such a database is present | 20:31 |
| thingee | fhermeni: actually 21:00 utc so it'll be late in france | 20:31 |
| zzzeek | fungi: but that sounds like the opposite of what you're asking | 20:31 |
| dims | fungi : keys off this environment variable http://codesearch.openstack.org/?q=OS_TEST_DBAPI_ADMIN_CONNECTION&i=nope&files=&repos= | 20:31 |
| thingee | 22:00 I believe | 20:31 |
| zzzeek | fungi: like, if you run oslo.db tests on a machien that doesnt have mysql or pg installed, it detects that and skips all those tests | 20:31 |
| fungi | zzzeek: thanks. and yeah that's what i was mostly asking about. because if we try to run these jobs without a database to discover whether or not they use one, then opportunistic testing will cause us to silently strip database tests upstream | 20:31 |
| zzzeek | fungi: all of the openstack projects that test against mysql/pg to my knowledge have such a system | 20:32 |
| *** vilobhmm111 has joined #openstack-oslo | 20:32 | |
| fhermeni | thingee: indeed but sooner would then have been difficult as it is still partly a “family time”. So 22:00 is ok to me | 20:32 |
| zzzeek | tehy all opportunistically check for the DB being present and should skip tests if not | 20:32 |
| fungi | zzzeek: oslo.db tests but also tests of other projects using oslo.db which may use test features from oslo.db to the same end | 20:32 |
| thingee | fhermeni: excellent, I'll add it to the agenda. I will ping both you and josh | 20:32 |
| zzzeek | fungi: other openstack projets which use MySQLOpportunisticTestBase and PosgresqlOpportunisticTestBAse, yes | 20:32 |
| fungi | zzzeek: eventually, i want to start a broader discussion across the project about figuroug out how to disable opportunism in tests, since it's terribly fragile | 20:33 |
| zzzeek | fungi: most of them use that, though not consistently | 20:33 |
| AJaeger | so, should we then say: If oslo.db is in requirements, let's assume it uses databases? | 20:33 |
| dims | fungi : zzzeek : we look for urls in OS_TEST_DBAPI_ADMIN_CONNECTION env variable and check if any one of them works and then use that | 20:33 |
| zzzeek | AJaeger: sure | 20:33 |
| dims | to run opportunistic tests | 20:33 |
| fungi | AJaeger: that seems like a safe addition to our initial list | 20:33 |
| fungi | thanks zzzeek and dims! | 20:34 |
| *** vilobhmm11 has quit IRC | 20:34 | |
| zzzeek | AJaeger: there's the tiniest chance they only use SQLite but that is not currently the case for any projects | 20:34 |
| AJaeger | do we need to check whether a project is pulling in oslo.db indirectly? | 20:34 |
| fhermeni | thingee: about the people involved by the spec, it matters to exchange will people that are providing schedulers to ensure the language allows to express their feature and people talking to the scheduler to check for the language practical aspects | 20:35 |
| *** edmondsw has quit IRC | 20:35 | |
| fungi | i'd love to come up with a way to make it possible for devs to skip database tests easily but still force them to run in our ci so that absence of database setup doesn't just cause them to be silently skipped | 20:35 |
| fungi | AJaeger: probably any project which declares a requirement on oslo.db, and any project qhich declares a requirement on any of those requirements, recursively until we find teh closed set | 20:36 |
| *** dims_ has joined #openstack-oslo | 20:36 | |
| * AJaeger hates these closure calculations ;( | 20:36 | |
| lifeless | fungi: env variables ftw | 20:37 |
| thingee | fhermeni: in the cross-project specs you want to get consensus on a concept. This can include technical details, but some implementation details can be handled within each project on their own. That doesn't fall on you to do each spec, but someone on the team that understands the project well to draft. You just want agreement at a higher level from the OpenStack projects. | 20:37 |
| thingee | so each project this would matter to on agree. Just the immediate ones that come to mind. We can figure out the full list in meeting since each project should have a rep available. | 20:37 |
| *** dims has quit IRC | 20:38 | |
| fungi | lifeless: right, we likely need a cross-project spec and/or pti update to provide an appropriate model for how to default database tests (and any other sort of currently opportunistic testing) to on and still be able to disable it with an envvar or whatever | 20:38 |
| thingee | fhermeni: in addition, the hope with cross-project specs is to surface existing solutions to solve to implement the concept. To make sure everyone knows and doesn't go off duplicating effort in their own project. | 20:39 |
| AJaeger | fungi, only 52 repos in openstack namespace import oslo.db directly | 20:39 |
| dims_ | haha "only" :) | 20:39 |
| fungi | that's still quite a few | 20:39 |
| fungi | dims_: well, that's out of 1010 repos we have in gerrit now | 20:39 |
| fhermeni | thingee: looks ok to me | 20:39 |
| fungi | so ~0.5% | 20:40 |
| dims_ | fungi : wow | 20:40 |
| fungi | sorry, 5% | 20:40 |
| AJaeger | fungi: 5 percent! | 20:40 |
| fungi | indeed. i math terribly | 20:40 |
| AJaeger | but you noticed quickly ;) | 20:40 |
| fungi | i tend to recheck my calculations, i just often hit enter before i do so :/ | 20:41 |
| lifeless | fungi: I was more thinking | 20:42 |
| lifeless | fungi: we define a variable meaning 'skipping db tests are errors' | 20:43 |
| AJaeger | fungi, dims_ , zzzeek http://paste.openstack.org/show/490434/ is result of grep | 20:43 |
| AJaeger | lifeless: yeah! | 20:43 |
| lifeless | fungi: and then look in the subunit output for matching skips | 20:43 |
| lifeless | fungi: so there's no per-project changes to do | 20:43 |
| lifeless | fungi: (potentially; we can of course do it locally to - agree that a spec probably needed) | 20:44 |
| lifeless | though, I'd like to see someone JFDI it somewhere and use a spec to drive wider adoption | 20:44 |
| AJaeger | fungi, all these repos are "server" typo, so shouldn't show up in requirements... | 20:47 |
| fungi | lifeless: right, we can implement on a limited scale first as a trial before trying to convince everyone else it's beneficial | 20:49 |
| fungi | AJaeger: i hope that's true, though it didn't used to be (for example we had things declaring glance as a requirement once upon a time) | 20:50 |
| fungi | but that was in the days before middlewares and such | 20:50 |
| lifeless | stadium... | 20:50 |
| AJaeger | fungi, yeah, we need to double check - I checked a few but should automate... | 20:51 |
| fungi | oh, right, tons of neutron-stadium repos depend on neutron | 20:52 |
| fungi | or did | 20:52 |
| fungi | no idea whether they still do | 20:52 |
| AJaeger | they use neutron-lib now -and that's in the list ;( | 20:54 |
| *** bapalm has quit IRC | 20:55 | |
| *** bapalm has joined #openstack-oslo | 20:58 | |
| AJaeger | fungi, so how to continue? I can change these repos to use the db jobs - and then remove the db from the default. Do you want to write an email or a spec? | 20:59 |
| *** e0ne has quit IRC | 21:00 | |
| *** haypo has left #openstack-oslo | 21:04 | |
| zzzeek | kevinbenton: I likely have a fix for this but it will be in 1.1 | 21:19 |
| *** sdague has quit IRC | 21:19 | |
| *** ryansb has left #openstack-oslo | 21:22 | |
| kevinbenton | zzzeek: cool. What's the fix? Keep a ref if a transaction is active? | 21:25 |
| *** jdandrea_ is now known as jdandrea | 21:25 | |
| zzzeek | kevinbenton: no, when i raise that error, check the row | 21:25 |
| zzzeek | kevinbenton: im working up a fix w/ an issue we can reference in the gerrit | 21:25 |
| kevinbenton | zzzeek: sgtm | 21:25 |
| *** dimtruck is now known as zz_dimtruck | 21:27 | |
| *** zz_dimtruck is now known as dimtruck | 21:29 | |
| *** yamahata has joined #openstack-oslo | 21:36 | |
| *** vilobhmm111 has quit IRC | 21:37 | |
| *** vilobhmm11 has joined #openstack-oslo | 21:37 | |
| *** regXboi has quit IRC | 21:40 | |
| zzzeek | kevinbenton: this is a huge fix so, this is why i always want to understand :) | 21:45 |
| kevinbenton | zzzeek: well now you have a test case at least :) | 21:46 |
| kevinbenton | zzzeek: I actually dislike our allocation pattern anyway because now it emits a select statement for something we just created :( | 21:51 |
| kevinbenton | zzzeek: regardless of this bug | 21:52 |
| *** dims_ has quit IRC | 21:53 | |
| fungi | AJaeger: with the number being that high, i'm unconvinced all 50+ repos are actually testing database interactions in their unit tests | 21:55 |
| fungi | AJaeger: so we probably should look for the other more direct markers mentioned (i.e., OS_TEST_DBAPI_ADMIN_CONNECTION and OpportunisticTestBase substrings as well as openstack_citest) | 21:57 |
| *** jdandrea_ has joined #openstack-oslo | 21:57 | |
| *** jdandrea has quit IRC | 21:57 | |
| *** mriedem is now known as mriedem_away | 21:57 | |
| *** rlrossit has quit IRC | 22:03 | |
| kevinbenton | zzzeek: the only thing I can't quite reconcile is how messing around in pdb was fixing it | 22:13 |
| *** MVenesio has joined #openstack-oslo | 22:13 | |
| zzzeek | kevinbenton: pdb changes gc stuff highly | 22:13 |
| zzzeek | object was not getting gc'ed in that case | 22:14 |
| kevinbenton | zzzeek: I see. So by having the set_trace in that allocation code it was preventing gc from collecting when that function exited or something? | 22:15 |
| zzzeek | kevinbenton: likely | 22:15 |
| *** MVenesio has quit IRC | 22:18 | |
| *** yamahata has quit IRC | 22:26 | |
| *** dims has joined #openstack-oslo | 22:26 | |
| *** dimtruck is now known as zz_dimtruck | 22:30 | |
| *** e0ne has joined #openstack-oslo | 22:34 | |
| *** cdent has quit IRC | 22:36 | |
| *** alejandrito has quit IRC | 22:42 | |
| *** e0ne has quit IRC | 22:49 | |
| *** dims has quit IRC | 22:50 | |
| *** yamahata has joined #openstack-oslo | 22:56 | |
| *** yamamoto_ has joined #openstack-oslo | 22:56 | |
| *** yamamoto_ has quit IRC | 23:00 | |
| *** pradk has quit IRC | 23:02 | |
| *** ozamiatin has joined #openstack-oslo | 23:06 | |
| *** harlowja_at_home has joined #openstack-oslo | 23:10 | |
| *** gordc has quit IRC | 23:11 | |
| fhermeni | harlowja_at_home: hi ! | 23:15 |
| harlowja_at_home | fhermeni, hey hey | 23:15 |
| harlowja_at_home | isn't it late for u ;) | 23:15 |
| fhermeni | harlowja_at_home: indeed | 23:15 |
| harlowja_at_home | soooooo scheduler stuff | 23:16 |
| harlowja_at_home | ill comment on that spec soon | 23:16 |
| harlowja_at_home | fhermeni, though if u want to attend a TC meeting | 23:18 |
| harlowja_at_home | u can bug the TC about it, may or may not help ,lol | 23:18 |
| fhermeni | harlowja_at_home: that is my point. Your opinion ? | 23:19 |
| fhermeni | harlowja_at_home: as I still have a limited perspective on OS principles, it might not be easy to me | 23:19 |
| harlowja_at_home | that's ok imho, someone who rocks the boat might not be bad | 23:20 |
| harlowja_at_home | sylvian is coming at the view point of nova | 23:20 |
| harlowja_at_home | fhermeni, maybe i'll showup, this boat imho needs rocking | 23:20 |
| harlowja_at_home | there isn't time enough to wait for nova to do something | 23:20 |
| harlowja_at_home | and wait for others to do something | 23:21 |
| harlowja_at_home | and wait and wait | 23:21 |
| harlowja_at_home | openstack imho be dead by then | 23:21 |
| harlowja_at_home | lol | 23:21 |
| fhermeni | harlowja_at_home: :D | 23:21 |
| harlowja_at_home | basically it will probably end up in the area of 'what is nova' | 23:21 |
| harlowja_at_home | because if nova wants to do cross-project scheduling, fine | 23:21 |
| harlowja_at_home | but it doesn't | 23:21 |
| *** dims has joined #openstack-oslo | 23:21 | |
| harlowja_at_home | dims, howdy | 23:22 |
| harlowja_at_home | dims, jaypipes might have thoughts if they are around | 23:22 |
| harlowja_at_home | fhermeni, jaypipes do u know when the 'nova scheduler' weekly meetings are? | 23:22 |
| fhermeni | harlowja_at_home: nova is also looking for moving our the driver parts as I remember ? | 23:23 |
| harlowja_at_home | fhermeni, i'd like to see the documentation on that split-out also, idk what the up to date docs even are, lol | 23:23 |
| harlowja_at_home | fhermeni, there are many ambitious plans :-P | 23:23 |
| harlowja_at_home | that ones been an ambition for years | 23:23 |
| harlowja_at_home | reality == (not so much afaik) | 23:23 |
| harlowja_at_home | jaypipes, dims btw, this is in regard to commentary at https://review.openstack.org/#/c/210549/ (if u get time to read it) | 23:24 |
| patchbot | harlowja_at_home: patch 210549 - openstack-specs - Super scheduling (WIP) | 23:24 |
| harlowja_at_home | super-duper-scheduling, lol | 23:24 |
| fhermeni | damn spam bot, there are everywhere :D | 23:25 |
| dims | harlowja_at_home : i believe it's https://wiki.openstack.org/wiki/Meetings/Scheduler | 23:26 |
| dims | fhermeni : i requested it for our channel :) | 23:26 |
| *** dims has quit IRC | 23:27 | |
| harlowja_at_home | Mondays at 14:00 UTC (#openstack-meeting-alt) | 23:28 |
| harlowja_at_home | oops guess i missed it | 23:28 |
| harlowja_at_home | lol | 23:28 |
| harlowja_at_home | fhermeni, that might be better in your tz | 23:29 |
| harlowja_at_home | its 6am mine, lol | 23:29 |
| harlowja_at_home | i haven't seen 6am in a while, lol | 23:29 |
| fhermeni | harlowja_at_home: Very bad timing for you indeed. How is the time choosed ? Depending on the majority TZ ? | 23:30 |
| harlowja_at_home | fhermeni, usually ya | 23:30 |
| fhermeni | so bed time now. If you have some idea/points/.. that I should mention. Mail them to me | 23:33 |
| *** fhermeni has quit IRC | 23:34 | |
| *** zz_dimtruck is now known as dimtruck | 23:41 | |
| *** sputnik13 has quit IRC | 23:46 | |
| kevinbenton | zzzeek: ouch, now holding that reference is causing a test to fail that creates ports, deletes them, and re-creates them all using the same session | 23:47 |
| openstackgerrit | Jamie Lennox proposed openstack/oslo.config: Handle some native python types in config generation https://review.openstack.org/292647 | 23:48 |
| *** sputnik13 has joined #openstack-oslo | 23:51 | |
| *** yamahata has quit IRC | 23:53 | |
| *** sputnik13 has quit IRC | 23:53 | |
| *** salv-orl_ has joined #openstack-oslo | 23:54 | |
| *** _amrith_ is now known as amrith | 23:55 | |
| *** salv-orlando has quit IRC | 23:57 | |
| *** yamamoto_ has joined #openstack-oslo | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!