| *** ashleighfarnham has quit IRC | 00:02 | |
| *** Rockyg has quit IRC | 00:05 | |
| *** sdake has quit IRC | 00:07 | |
| *** dims has joined #openstack-oslo | 00:12 | |
| *** sputnik13 has quit IRC | 00:12 | |
| dims | harlowja: around? can you please bless the pip-missing-requirements ones? https://review.openstack.org/#/q/status:open+topic:pip-missing-reqs,n,z | 00:13 |
|---|---|---|
| harlowja | yuppers | 00:13 |
| * harlowja getting his holy water ready | 00:13 | |
| openstackgerrit | Komei Shimamura proposed openstack/oslo-specs: Blueprint Spec for Adding Kafka Driver https://review.openstack.org/189003 | 00:13 |
| *** tsekiyam_ has joined #openstack-oslo | 00:13 | |
| dims | thanks harlowja | 00:14 |
| harlowja | holy water gun ready | 00:14 |
| harlowja | starts to shoot holywatergu | 00:14 |
| harlowja | *gun | 00:14 |
| harlowja | *super soaker like* | 00:14 |
| *** mtanino has quit IRC | 00:16 | |
| dims | haha | 00:16 |
| *** tsekiyama has quit IRC | 00:17 | |
| openstackgerrit | Komei Shimamura proposed openstack/oslo-specs: Specification for Adding Kafka Driver https://review.openstack.org/189003 | 00:18 |
| *** tsekiyam_ has quit IRC | 00:18 | |
| harlowja | dims spotted dups | 00:19 |
| harlowja | https://review.openstack.org/#/c/195837/ | 00:19 |
| harlowja | https://review.openstack.org/#/c/195826/ | 00:19 |
| harlowja | u pick | 00:19 |
| harlowja | ha | 00:19 |
| dims | nuked one! | 00:20 |
| harlowja | woah, oslo.messaging requires webob | 00:23 |
| dims | middleware thingy | 00:23 |
| harlowja | ah | 00:23 |
| harlowja | k | 00:23 |
| *** amotoki has joined #openstack-oslo | 00:27 | |
| harlowja | ok, think i exhaused my hold watergun | 00:28 |
| harlowja | *exhausted | 00:28 |
| harlowja | *ye olde holy watergun | 00:28 |
| *** boris-42 has joined #openstack-oslo | 00:31 | |
| *** ozamiatin has quit IRC | 00:31 | |
| *** manishg has quit IRC | 00:42 | |
| *** jamespd has joined #openstack-oslo | 00:47 | |
| jamespd | Is there a good place to learn about the semantics of unique_id and msg_id? I'm reading the amqpdriver code, but don't see any good docs regarding what these two are used for. | 00:49 |
| harlowja | lifeless dims if u interested, i'm trying to form https://github.com/harlowja/taskflow_paper/blob/master/paper.tex (maybe can get that published somewhere); just an outline right now (but hopefully grows into something useful), ha | 00:53 |
| *** abitha has quit IRC | 00:54 | |
| * harlowja seeking any volunterrs that want to be involved :_P | 00:54 | |
| harlowja | :-P | 00:54 |
| *** ViswaV has quit IRC | 00:55 | |
| *** ViswaV has joined #openstack-oslo | 00:56 | |
| lifeless | harlowja: would I have to write tex? | 00:57 |
| lifeless | :) | 00:57 |
| harlowja | :-P | 00:57 |
| harlowja | maybe a little? | 00:57 |
| harlowja | lol | 00:57 |
| harlowja | its not that hard :-P | 00:57 |
| dims | harlowja: very interesting, will let you know :) | 00:58 |
| harlowja | k | 00:58 |
| * dims still has not read the papers you pointed out over the weekend | 00:58 | |
| harlowja | hahaha | 00:58 |
| harlowja | there's to many | 00:58 |
| harlowja | but some interesting ones there | 00:59 |
| harlowja | * http://tempsend.com/4FD06CA829/7826/paper.pdf (what that paper.tex looks like rendered) | 00:59 |
| harlowja | anyway, something i was thinking of doing | 01:00 |
| harlowja | cause u know, why the heck not, ha | 01:01 |
| dims | ++ harlowja | 01:01 |
| *** ashleighfarnham has joined #openstack-oslo | 01:03 | |
| *** ashleighfarnham has quit IRC | 01:07 | |
| lifeless | harlowja: so maybe. I dunno ? :) | 01:10 |
| harlowja | ha, something to consider, idk either :-P | 01:11 |
| lifeless | like, I don't think I've contributed much to taskflow | 01:12 |
| lifeless | I've kibbitzed here and there | 01:12 |
| lifeless | if you wanted to put my name in the fine print, that would be cool | 01:12 |
| harlowja | well professor, its ok | 01:12 |
| harlowja | lol | 01:12 |
| harlowja | :-P | 01:12 |
| lifeless | I'm certainly happy to read and critique drafts | 01:12 |
| harlowja | cool | 01:12 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 01:13 | |
| * harlowja we need more openstack people to produce this stuff IMHO :-P | 01:13 | |
| lifeless | [btw I've used beamer for presentations in the past | 01:13 |
| lifeless | the tex thing was a tease | 01:13 |
| lifeless | I'm certainly stale on it, but not scared | 01:13 |
| harlowja | just a weird-funky syntax, ha | 01:14 |
| openstackgerrit | Merged openstack/oslo.rootwrap: Add tox target to find missing requirements https://review.openstack.org/195839 | 01:14 |
| harlowja | *yet another weird-funky syntax | 01:14 |
| openstackgerrit | Merged openstack/oslo.middleware: Add tox target to find missing requirements https://review.openstack.org/195830 | 01:15 |
| lifeless | YAWFS | 01:17 |
| harlowja | :) | 01:17 |
| lifeless | must make project for that now. | 01:17 |
| harlowja | sounds like a new filesystem, ha | 01:17 |
| harlowja | YAW FS | 01:18 |
| lifeless | You forgot the YEE | 01:20 |
| lifeless | YEE YAW FS | 01:21 |
| harlowja | ha | 01:33 |
| dims | harlowja: cinder managed to get their test working with newer o.vo? | 01:36 |
| *** timsim has joined #openstack-oslo | 01:38 | |
| *** tobe has joined #openstack-oslo | 01:38 | |
| *** timsim has left #openstack-oslo | 01:38 | |
| *** tobe has quit IRC | 01:39 | |
| *** yamahata has quit IRC | 01:43 | |
| *** kzaitsev_mb has joined #openstack-oslo | 02:01 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 02:05 | |
| *** tobe has joined #openstack-oslo | 02:16 | |
| *** mriedem_away has quit IRC | 02:16 | |
| dims | ativelkov: ping | 02:20 |
| *** manishg has joined #openstack-oslo | 02:24 | |
| *** kzaitsev_mb has quit IRC | 02:24 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.versionedobjects: Revert "Compound fields now coerce their elements' values" https://review.openstack.org/196926 | 02:26 |
| *** ashleighfarnham has joined #openstack-oslo | 02:27 | |
| *** manishg has quit IRC | 02:28 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.versionedobjects: Revert "Compound fields now coerce their elements' values" https://review.openstack.org/196926 | 02:28 |
| *** stevemar has joined #openstack-oslo | 02:35 | |
| harlowja | dims unsure | 02:37 |
| dims | harlowja: unsure about? | 02:38 |
| harlowja | *about cinder fixing there test | 02:38 |
| harlowja | for o.vo | 02:38 |
| dims | ack. looks like they were able to merge stuff and i just ran their py34 by hand and that works too | 02:38 |
| harlowja | k | 02:38 |
| harlowja | guess something got tweaked somewhere then | 02:38 |
| dims | so for Nova i figured out that i need a revert ^^^ | 02:39 |
| dims | y | 02:39 |
| harlowja | revert all the things | 02:40 |
| harlowja | back to commit #0 | 02:40 |
| harlowja | oh, u probably don't mean that kind of revert | 02:40 |
| harlowja | ha | 02:40 |
| dims | no this one was a single review | 02:40 |
| harlowja | hmmm, looks like a revert of things to me :-P | 02:41 |
| harlowja | yup yup, verified it changed some code back to something else | 02:41 |
| *** mriedem has joined #openstack-oslo | 02:45 | |
| *** manishg has joined #openstack-oslo | 02:50 | |
| harlowja | dims do u know if anyone uses that eventlet backdoor stuff, have u heard of people actually using it? | 02:58 |
| harlowja | that one probably can be made into a real thread (with some kind of additions/changes when eventlet is active) but meh, or just delete that code, lol | 02:59 |
| *** kevinbenton has joined #openstack-oslo | 03:08 | |
| *** manishg has quit IRC | 03:12 | |
| openstackgerrit | Joshua Harlow proposed openstack/oslo.service: Make the loopingcall code less dependent on eventlet https://review.openstack.org/194479 | 03:17 |
| harlowja | eezhova let me know what u think of ^ | 03:17 |
| harlowja | it does work :-P | 03:17 |
| *** yamahata has joined #openstack-oslo | 03:41 | |
| *** rjaiswal has quit IRC | 03:44 | |
| *** ViswaV has quit IRC | 03:44 | |
| *** ViswaV has joined #openstack-oslo | 03:45 | |
| *** dims has quit IRC | 03:48 | |
| *** mriedem has quit IRC | 03:53 | |
| *** manishg has joined #openstack-oslo | 03:56 | |
| *** sdake has joined #openstack-oslo | 04:09 | |
| *** sdake_ has joined #openstack-oslo | 04:10 | |
| *** david-lyle has quit IRC | 04:13 | |
| *** achanda has joined #openstack-oslo | 04:13 | |
| *** sdake has quit IRC | 04:13 | |
| *** david-lyle has joined #openstack-oslo | 04:14 | |
| *** zz_jgrimm has quit IRC | 04:15 | |
| *** ihrachyshka has joined #openstack-oslo | 04:24 | |
| *** ihrachyshka has quit IRC | 04:25 | |
| *** ihrachyshka_ has joined #openstack-oslo | 04:25 | |
| *** sdake has joined #openstack-oslo | 04:25 | |
| *** zz_jgrimm has joined #openstack-oslo | 04:26 | |
| *** sdake_ has quit IRC | 04:29 | |
| *** achanda has quit IRC | 04:30 | |
| *** tobe has quit IRC | 04:31 | |
| openstackgerrit | Merged openstack/oslo.versionedobjects: Revert "Compound fields now coerce their elements' values" https://review.openstack.org/196926 | 04:32 |
| *** hogepodge has quit IRC | 04:32 | |
| *** tobe has joined #openstack-oslo | 04:32 | |
| *** hogepodge has joined #openstack-oslo | 04:32 | |
| *** manishg has quit IRC | 04:33 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 04:39 | |
| *** zz_jgrimm has quit IRC | 04:39 | |
| *** zz_jgrimm has joined #openstack-oslo | 04:42 | |
| openstackgerrit | Steve Martinelli proposed openstack/oslo.concurrency: Remove redundant fileutils https://review.openstack.org/196945 | 04:42 |
| *** amotoki_ has joined #openstack-oslo | 04:49 | |
| *** arnaud____ has joined #openstack-oslo | 04:50 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Add ability to reset an engine via a `reset` method https://review.openstack.org/186239 | 04:54 |
| *** arnaud____ has quit IRC | 04:54 | |
| *** sdake_ has joined #openstack-oslo | 04:59 | |
| *** arnaud____ has joined #openstack-oslo | 05:00 | |
| *** sdake has quit IRC | 05:03 | |
| *** sputnik13 has joined #openstack-oslo | 05:04 | |
| *** stevemar has quit IRC | 05:05 | |
| *** stevemar has joined #openstack-oslo | 05:06 | |
| *** rpodolyaka1 has joined #openstack-oslo | 05:07 | |
| *** ihrachyshka_ has quit IRC | 05:13 | |
| *** amotoki__ has joined #openstack-oslo | 05:24 | |
| *** amotoki_ has quit IRC | 05:24 | |
| *** arnaud____ has quit IRC | 05:31 | |
| *** jaosorior has joined #openstack-oslo | 05:38 | |
| *** hogepodge has quit IRC | 05:42 | |
| *** hogepodge has joined #openstack-oslo | 05:42 | |
| *** nkrinner has joined #openstack-oslo | 05:44 | |
| *** sputnik1_ has joined #openstack-oslo | 05:45 | |
| *** sputnik13 has quit IRC | 05:48 | |
| *** achanda has joined #openstack-oslo | 05:48 | |
| openstackgerrit | Steve Martinelli proposed openstack/oslo-specs: Graduate fileutils to oslo.utils and oslo.policy https://review.openstack.org/170685 | 05:49 |
| *** rjaiswal has joined #openstack-oslo | 05:50 | |
| *** browne has quit IRC | 05:53 | |
| openstackgerrit | Eric Brown proposed openstack/oslo.config: Expose min and max to IntOpt https://review.openstack.org/196955 | 05:59 |
| *** browne has joined #openstack-oslo | 06:00 | |
| *** achanda has quit IRC | 06:04 | |
| *** stevemar has quit IRC | 06:05 | |
| *** stevemar has joined #openstack-oslo | 06:06 | |
| *** arnaud____ has joined #openstack-oslo | 06:06 | |
| *** ihrachyshka_ has joined #openstack-oslo | 06:07 | |
| *** stevemar has quit IRC | 06:09 | |
| *** SridharGaddam has joined #openstack-oslo | 06:16 | |
| *** rpodolyaka1 has quit IRC | 06:19 | |
| *** achanda has joined #openstack-oslo | 06:26 | |
| *** sdake_ has quit IRC | 06:29 | |
| openstackgerrit | Eric Brown proposed openstack/oslo.config: Expose min and max to IntOpt https://review.openstack.org/196955 | 06:32 |
| *** browne has quit IRC | 06:52 | |
| *** rpodolyaka1 has joined #openstack-oslo | 06:58 | |
| *** pblaho_ has joined #openstack-oslo | 07:00 | |
| *** arnaud____ has quit IRC | 07:01 | |
| *** boris-42 has quit IRC | 07:02 | |
| *** achanda has quit IRC | 07:03 | |
| *** pblaho has quit IRC | 07:04 | |
| *** sputnik1_ has quit IRC | 07:05 | |
| *** sputnik13 has joined #openstack-oslo | 07:05 | |
| *** pblaho_ has quit IRC | 07:06 | |
| *** pblaho has joined #openstack-oslo | 07:06 | |
| *** sputnik13 has quit IRC | 07:09 | |
| openstackgerrit | Victor Sergeyev proposed openstack/oslo.messaging: Remove qpid-config call https://review.openstack.org/194723 | 07:28 |
| *** subscope has quit IRC | 07:31 | |
| *** e0ne has joined #openstack-oslo | 07:33 | |
| *** e0ne has quit IRC | 07:37 | |
| *** rpodolyaka1 has quit IRC | 07:40 | |
| *** ndipanov has quit IRC | 07:42 | |
| *** ndipanov has joined #openstack-oslo | 07:44 | |
| *** subscope has joined #openstack-oslo | 07:47 | |
| *** amotoki__ has quit IRC | 07:50 | |
| *** yamahata has quit IRC | 07:52 | |
| *** sdake_ has joined #openstack-oslo | 07:55 | |
| *** subscope has quit IRC | 07:55 | |
| *** shardy_ has joined #openstack-oslo | 07:58 | |
| *** shardy has quit IRC | 07:59 | |
| *** pblaho_ has joined #openstack-oslo | 08:02 | |
| *** shardy_ has quit IRC | 08:03 | |
| *** shardy has joined #openstack-oslo | 08:04 | |
| *** pblaho has quit IRC | 08:05 | |
| *** jaosorior has quit IRC | 08:06 | |
| *** pblaho_ has quit IRC | 08:07 | |
| *** stevemar has joined #openstack-oslo | 08:07 | |
| *** pblaho has joined #openstack-oslo | 08:08 | |
| *** stevemar has quit IRC | 08:10 | |
| *** rpodolyaka1 has joined #openstack-oslo | 08:11 | |
| *** subscope has joined #openstack-oslo | 08:11 | |
| *** droyal has quit IRC | 08:16 | |
| *** tobe has quit IRC | 08:17 | |
| *** rpodolyaka1 has quit IRC | 08:21 | |
| *** rpodolyaka1 has joined #openstack-oslo | 08:23 | |
| *** subscope has quit IRC | 08:24 | |
| *** rpodolyaka1 has quit IRC | 08:25 | |
| *** rpodolyaka1 has joined #openstack-oslo | 08:28 | |
| *** yassine_ has joined #openstack-oslo | 08:34 | |
| *** Kennan has quit IRC | 08:38 | |
| *** subscope has joined #openstack-oslo | 08:38 | |
| *** Kennan has joined #openstack-oslo | 08:38 | |
| *** dguitarbite has joined #openstack-oslo | 08:42 | |
| *** sdake_ has quit IRC | 08:45 | |
| -openstackstatus- NOTICE: OpenStack CI is down due to hard drive failures | 08:49 | |
| *** ChanServ changes topic to "OpenStack CI is down due to hard drive failures" | 08:49 | |
| *** rjaiswal has quit IRC | 08:54 | |
| *** e0ne has joined #openstack-oslo | 09:02 | |
| *** e0ne is now known as e0ne_ | 09:15 | |
| *** ozamiatin has joined #openstack-oslo | 09:17 | |
| *** kzaitsev_mb has joined #openstack-oslo | 09:24 | |
| *** e0ne_ has quit IRC | 09:26 | |
| *** e0ne has joined #openstack-oslo | 09:29 | |
| *** kzaitsev_mb has quit IRC | 09:31 | |
| *** rpodolyaka1 has quit IRC | 09:39 | |
| *** rpodolyaka1 has joined #openstack-oslo | 09:42 | |
| *** subscope has quit IRC | 09:44 | |
| *** rpodolyaka1 has quit IRC | 09:46 | |
| *** stevemar has joined #openstack-oslo | 09:56 | |
| *** stevemar has quit IRC | 10:00 | |
| *** subscope has joined #openstack-oslo | 10:01 | |
| *** arnaud____ has joined #openstack-oslo | 10:02 | |
| *** arnaud____ has quit IRC | 10:07 | |
| *** e0ne is now known as e0ne_ | 10:23 | |
| *** e0ne_ is now known as e0ne | 10:25 | |
| *** shardy_ has joined #openstack-oslo | 10:43 | |
| *** shardy has quit IRC | 10:45 | |
| *** shardy_ has quit IRC | 10:48 | |
| *** shardy has joined #openstack-oslo | 10:49 | |
| *** rpodolyaka1 has joined #openstack-oslo | 10:50 | |
| *** dims has joined #openstack-oslo | 11:07 | |
| *** exploreshaifali has joined #openstack-oslo | 11:13 | |
| haypo | dims: hi. we need to something to enhance tests before oslo releases | 11:26 |
| dims | haypo: yep | 11:27 |
| haypo | dims: to catch more bugs. it's annoying to notify bugs only after a release. the cost is fix them is much higher, and bugs are annoying - as you probably already know :-D | 11:27 |
| haypo | dims: are you using a script? | 11:27 |
| haypo | dims: did you have time to investigate why oslo.db was installed using the version 0.12.0 and not the master branch? | 11:28 |
| dims | haypo: figured out what was wrong - am using https://review.openstack.org/#/c/186413/ | 11:28 |
| dims | haypo: see the install_command - https://review.openstack.org/#/c/186413/9/tox.ini,cm | 11:28 |
| dims | that was causing older versions to be installed | 11:28 |
| haypo | oh, i didn't notice this line | 11:28 |
| *** rpodolyaka1 has quit IRC | 11:28 | |
| dims | if you see latest runs, you can see that we pick up right version of oslo from master | 11:29 |
| dims | haypo: see example http://logs.openstack.org/13/186413/9/check/gate-nova-python27/9a85e81/testr_results.html.gz | 11:30 |
| dims | it shows the problem we saw in nova yesterday | 11:30 |
| haypo | dims: cool | 11:30 |
| *** rpodolyaka1 has joined #openstack-oslo | 11:30 | |
| haypo | dims: how did you write the patch https://review.openstack.org/#/c/186413/ ? | 11:30 |
| dims | haypo: see requirements.txt | 11:30 |
| *** subscope has quit IRC | 11:31 | |
| haypo | dims: i saw it. but i would like if you wrote the patch manually, or if you have a tool for that | 11:31 |
| haypo | dims: what do you think of having a CI (where C stands for continuous :-)) for running tempest with the master branch of oslo.* ? | 11:32 |
| haypo | dims: it's also annoying to only detect bugs while preparing a release | 11:32 |
| dhellmann | haypo: we run like tempest on every patch, the problem is the unit tests in the other projects | 11:32 |
| dims | haypo: i have one of those too - https://review.openstack.org/#/c/186418/ | 11:33 |
| dims | ++ dhellmann | 11:33 |
| haypo | dhellmann: i don't understand how the oslo.versionedobject regression was not detected before | 11:33 |
| dhellmann | I don't know what happened there, did it break something? | 11:34 |
| haypo | dhellmann: yes, cinder & nova... | 11:34 |
| haypo | dhellmann: https://review.openstack.org/#/c/196926/ i don't see tempest here | 11:34 |
| dims | haypo: 186413 was working well before, something changed in the harness which reinstalled old versions over the one specified in requirements and i did not notice that | 11:34 |
| haypo | dhellmann: maybe, tempest is only run for some oslo projects, but not all of them? | 11:34 |
| dhellmann | haypo: gate-tempest-dsvm-neutron-src-oslo.versionedobjects | 11:34 |
| dims | right | 11:34 |
| *** manishg has joined #openstack-oslo | 11:35 | |
| dims | haypo: see pip freeze of that job | 11:35 |
| haypo | dims: ok, i understood for 186413 | 11:36 |
| dims | dhellmann: the jobs the release script generates do not log anything do they? (thinking if i can release o.vo 0.5.1) | 11:36 |
| haypo | dhellmann: oh ok. hum. does it mean that the test failure in nova was only in nova unit tests, not in tempest? | 11:36 |
| dims | haypo: yep | 11:36 |
| haypo | ok. so i have to rephrase my question: should we have a CI running nova unit tests with all oslo.* master branches? or add a check job running nova unit tests to oslo.* projects? | 11:37 |
| haypo | maybe more than just nova. also cinder & neutron? | 11:38 |
| haypo | cinder, neutron, nova are the 3 most important projects in openstack, right? | 11:38 |
| dims | haypo: not sure what you are asking, i just pointed you to 2 reviews, one runs oslo.* master against nova itself, the other one runs oslo.* against devstack for a bunch more projects | 11:38 |
| dhellmann | dims: they do log, but they're not visible through gerrit. you can use "git os-job" (pip install git-os-job) and the version number to find the URL to the logs | 11:39 |
| haypo | dims: i would like to automate this to detect bugs earlier | 11:39 |
| haypo | dims: i don't want to wait until we prepare a release to detect a regression | 11:39 |
| haypo | (it's too late) | 11:39 |
| dims | haypo: sure, ideas beyond those 2 reviews are welcome | 11:39 |
| dhellmann | haypo: it is not practical to run all of the unit tests for all projects against all of the oslo libraries that use them every time we send in a patch. Submitting patches to oslo.config or oslo.log would take up all of our available test nodes at this point | 11:40 |
| haypo | dhellmann: you mean that it would cost too much resources to openstack infra? | 11:40 |
| dhellmann | haypo: yes. We would use our entire quota just to run a single patch through the check queue for some of the more popular libraries | 11:41 |
| haypo | dhellmann: well, i understand this practical issue :-) in this case, we have to double check manual checks done before a release | 11:41 |
| dims | haypo: ++ please kick tires on thu/fri | 11:41 |
| dhellmann | haypo: there's a script in oslotest that does what you want, serially | 11:41 |
| haypo | and maybe run them once per week | 11:41 |
| haypo | dhellmann: can this script be used on a desktop PC? | 11:42 |
| dims | haypo: and we do already do a go-no-go on monday meeting | 11:42 |
| dhellmann | it takes something like 12 hours to run, which is longer than jobs are allowed to live in CI, so I use it by hand (I'm using it to pre-test the namespace removals) | 11:42 |
| haypo | dhellmann: (how long does it take) | 11:42 |
| haypo | 12 hours, wow :) | 11:42 |
| dhellmann | haypo: I use it on a cloud vm, but it would work anywhere you could run the tests. It might not work on OS X | 11:42 |
| dhellmann | the length actually depends on how many projects use the libs, of course, so some are less | 11:43 |
| haypo | dhellmann: maybe we can ask to get dedicated resources from the openstack infra to run such test? | 11:43 |
| haypo | my idea is to run this script and then send a report | 11:44 |
| haypo | would it be insane to run it every day? | 11:44 |
| haypo | in an ideal world without any bug, it would not send any email :) | 11:44 |
| *** stevemar has joined #openstack-oslo | 11:45 | |
| *** manishg has quit IRC | 11:45 | |
| *** jaosorior has joined #openstack-oslo | 11:45 | |
| *** manishg has joined #openstack-oslo | 11:45 | |
| *** subscope has joined #openstack-oslo | 11:47 | |
| haypo | such email would not be send to openstack-dev, maybe to users who explicitly asked (ex: me!) | 11:47 |
| *** manishg has quit IRC | 11:47 | |
| *** stevemar has quit IRC | 11:48 | |
| haypo | (we may start with the oslo liaison guys) | 11:48 |
| *** rpodolyaka1 has quit IRC | 11:51 | |
| *** rpodolyaka1 has joined #openstack-oslo | 11:52 | |
| dims | haypo: get familiar with that script, see what it takes and we can propose something | 11:55 |
| dims | try it out a few times :) | 11:55 |
| haypo | dims: "what it takes" what do you mean? | 11:56 |
| dims | machine cpu/ram, scripts around it | 11:56 |
| dims | to pull everything from git, scripts to collect error | 11:57 |
| dims | s | 11:57 |
| haypo | tools/oslo_run_cross_tests doesn't look to install the development version of oslo.* projects | 11:58 |
| haypo | is it the right script? | 11:58 |
| *** prad has joined #openstack-oslo | 12:02 | |
| *** marcusvrn has joined #openstack-oslo | 12:07 | |
| dhellmann | haypo: if you want to test oslo.config, you go to the oslo.config source directory and run "tox -e venv -- oslo_run_pre_release_tests" | 12:09 |
| dhellmann | haypo: you need to have already checked out all of the sources for openstack using the script for that in oslo-incubator/tools | 12:10 |
| dhellmann | oslo_run_pre_release_tests uses a config file (~/.oslo.conf) where you can set the repo_root to point to the directory where you have the source checked out | 12:11 |
| * dhellmann needs to document these things | 12:11 | |
| haypo | "dhellmann needs to document these things" yeah, that would be nice :) | 12:11 |
| haypo | dhellmann, dims : i sent an email to openstack-dev to discuss how we can enhance the oslo release process | 12:13 |
| haypo | (i tried to summurize our discussion here) | 12:14 |
| *** stpierre has joined #openstack-oslo | 12:14 | |
| *** subscope has quit IRC | 12:17 | |
| *** amrith is now known as _amrith_ | 12:25 | |
| haypo | dhellmann: it would be great if you can reply to my email to give the commands to run tests | 12:29 |
| haypo | dhellmann: sorry, i'm unable to rebuild the puzzle | 12:29 |
| dhellmann | haypo: I'm working on that right now | 12:29 |
| haypo | dhellmann: coolness | 12:29 |
| haypo | dims: serisouly, i want to blame anyone. i don't care if you missed a bug. i only care of making the process better ;) | 12:30 |
| haypo | woops | 12:30 |
| dhellmann | dims: I wonder if we should ask the nova folks to remove that install customization with the --force-reinstall option from their tox file? | 12:30 |
| haypo | dims: serisouly, i *don't* want to blame anyone. i don't care if you missed a bug. i only care of making the process better ;) | 12:30 |
| dims | haypo: no worries. | 12:31 |
| haypo | dims: i know that it's an hard task to cook a release | 12:32 |
| *** ozamiatin has quit IRC | 12:32 | |
| *** gordc_afk is now known as gordc | 12:38 | |
| *** kgiusti has joined #openstack-oslo | 12:42 | |
| *** shardy has quit IRC | 12:43 | |
| *** shardy has joined #openstack-oslo | 12:43 | |
| *** manishg has joined #openstack-oslo | 12:47 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslotest: Add documentation for cross-project unit testing https://review.openstack.org/197050 | 12:47 |
| dhellmann | haypo, dims : ^^ | 12:47 |
| *** bknudson has joined #openstack-oslo | 12:47 | |
| *** manishg has quit IRC | 12:51 | |
| *** rpodolyaka1 has quit IRC | 12:53 | |
| *** manishg has joined #openstack-oslo | 12:54 | |
| *** bknudson has quit IRC | 12:54 | |
| *** zzzeek has joined #openstack-oslo | 13:06 | |
| ativelkov | hi dims | 13:06 |
| ativelkov | seems like that was my patch which caused the trouble, or at least part of it. Now trying to understand how to detect such issues before they hit the gate | 13:07 |
| dims | ativelkov: we were just talking about it. one way was doc'ed by doug just now - https://review.openstack.org/197050 | 13:09 |
| dims | ativelkov: here's another way i was using https://review.openstack.org/#/c/186413/ | 13:10 |
| *** bknudson has joined #openstack-oslo | 13:10 | |
| *** rpodolyaka1 has joined #openstack-oslo | 13:12 | |
| *** manishg has quit IRC | 13:14 | |
| ativelkov | Got it. And so it seems like it should be done for cinder as well | 13:14 |
| dims | ativelkov: problem is someone has to run it thu/fri everyweek and follow up when errors show up | 13:23 |
| ativelkov | dims: May we add some more tempest jobs to oslo so they are run on each commit? | 13:25 |
| openstackgerrit | Brant Knudson proposed openstack/oslo.serialization: Stop using deprecated oslo_utils.timeutils.strtime https://review.openstack.org/197069 | 13:26 |
| dims | ativelkov: counter productive, too much load on info and not scalable big-tentwise (how many can we add) | 13:28 |
| dims | problem is the unit tests are tripping up | 13:28 |
| dims | we have a dsvm job already for each oslo project that runs tempes | 13:28 |
| *** e0ne is now known as e0ne_ | 13:29 | |
| ativelkov | So, the problem is only unit-tests? I thought that as least some of the failures reported at https://bugs.launchpad.net/oslo.versionedobjects/+bug/1469865 were tempest jobs | 13:30 |
| openstack | Launchpad bug 1469865 in OpenStack Compute (nova) "oslo.versionedobjects breaks nova/cinder tests" [Critical,In progress] - Assigned to Victor Stinner (victor-stinner) | 13:30 |
| bknudson | if the unit tests don't catch it then fix the unit tests | 13:30 |
| dims | bknudson: ++ | 13:31 |
| *** boris-42 has joined #openstack-oslo | 13:31 | |
| *** _amrith_ is now known as amrith | 13:33 | |
| *** stevemar has joined #openstack-oslo | 13:34 | |
| *** manishg has joined #openstack-oslo | 13:35 | |
| dansmith | ativelkov: are you working on the fix for that? | 13:37 |
| *** stevemar has quit IRC | 13:37 | |
| *** manishg has quit IRC | 13:38 | |
| *** mriedem has joined #openstack-oslo | 13:39 | |
| *** e0ne_ has quit IRC | 13:39 | |
| ativelkov | dansmith: yes. Seems like the reason is the way how deepcopy reinstantiates CoercedCollections, such as CoercesList, CoercedDict and CoercedSet. For set it is unable to properly infer the constuctor agrs | 13:39 |
| dansmith | ativelkov: okay | 13:39 |
| ativelkov | Trying to find the best way to do it without having custom constructors | 13:40 |
| dansmith | okay, thanks for working on it | 13:40 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:40 | |
| *** zz_jgrimm is now known as jgrimm | 13:42 | |
| *** stevemar has joined #openstack-oslo | 13:42 | |
| *** e0ne has joined #openstack-oslo | 13:46 | |
| openstackgerrit | Steve Martinelli proposed openstack/oslo.concurrency: Remove redundant fileutils https://review.openstack.org/196945 | 13:47 |
| *** manishg has joined #openstack-oslo | 13:52 | |
| dims | ativelkov: thanks! | 13:53 |
| *** pblaho has quit IRC | 13:53 | |
| *** manishg_ has joined #openstack-oslo | 13:55 | |
| *** manishg has quit IRC | 13:55 | |
| *** yamahata has joined #openstack-oslo | 13:58 | |
| *** jecarey has joined #openstack-oslo | 14:00 | |
| *** dims has quit IRC | 14:01 | |
| *** dims has joined #openstack-oslo | 14:01 | |
| *** browne has joined #openstack-oslo | 14:05 | |
| *** rjaiswal has joined #openstack-oslo | 14:08 | |
| *** mtanino has joined #openstack-oslo | 14:11 | |
| *** pblaho has joined #openstack-oslo | 14:11 | |
| *** ozamiatin has joined #openstack-oslo | 14:12 | |
| *** ihrachyshka_ has quit IRC | 14:13 | |
| *** ihrachyshka has joined #openstack-oslo | 14:14 | |
| *** dbite has joined #openstack-oslo | 14:19 | |
| *** manishg_ has quit IRC | 14:23 | |
| *** shardy_ has joined #openstack-oslo | 14:24 | |
| *** shardy has quit IRC | 14:24 | |
| *** shardy_ has quit IRC | 14:29 | |
| *** shardy has joined #openstack-oslo | 14:30 | |
| *** pblaho has quit IRC | 14:39 | |
| *** pblaho has joined #openstack-oslo | 14:39 | |
| *** rushiagr_away is now known as rushiagr | 14:45 | |
| *** rpodolyaka1 has quit IRC | 14:51 | |
| *** subscope has joined #openstack-oslo | 14:52 | |
| krotscheck | dims, dhellmann: It's been a week since I've addressed gordc's questions and comments on https://review.openstack.org/#/c/191238/ and https://review.openstack.org/#/c/194268/ . In order to not delay this any further, could you two take a look? | 14:53 |
| *** rpodolyaka1 has joined #openstack-oslo | 14:53 | |
| *** ChanServ changes topic to "#openstack-oslo" | 14:54 | |
| -openstackstatus- NOTICE: The log volume was repaired and brought back online at 14:00 UTC. Log links today from before that time may be missing, and changes should be rechecked if fresh job logs are desired for them. | 14:54 | |
| *** e0ne is now known as e0ne_ | 14:55 | |
| *** shardy has quit IRC | 14:56 | |
| *** shardy has joined #openstack-oslo | 14:56 | |
| dhellmann | krotscheck: I wonder if he was offline last week? I'll take a look | 14:58 |
| *** jaypipes has quit IRC | 14:58 | |
| *** e0ne_ is now known as e0ne | 14:58 | |
| *** e0ne is now known as e0ne_ | 14:59 | |
| *** e0ne_ is now known as e0ne | 14:59 | |
| dhellmann | krotscheck: https://review.openstack.org/#/c/191238/5 is going to be backwards incompatible, right? are we concerned about that? | 14:59 |
| *** nkrinner has quit IRC | 15:00 | |
| krotscheck | dhellmann: Well, I don't know to what extent people are using the middleware. I know that I myself haven't landed anything in openstack, but I can't speak for shadow-downstream. | 15:00 |
| krotscheck | dhellmann: So I'm personally not concerned. | 15:00 |
| krotscheck | But my level of not-being-concerned may vary with others'. | 15:01 |
| *** exploreshaifali has quit IRC | 15:01 | |
| dhellmann | krotscheck: yeah, if we've released it we can't break it because it's next to impossible to discover users | 15:02 |
| dhellmann | this one is particularly challenging because it requires a change to the paste.ini file for a project, but if we haven't landed those ourselves I think we're probably ok to deprecate the old class in favor of the new API, but we have to retain the old API | 15:03 |
| *** shardy has quit IRC | 15:12 | |
| *** ihrachyshka has quit IRC | 15:12 | |
| krotscheck | dhellmann: Alright, I'll see what I can do. | 15:15 |
| dhellmann | krotscheck: the "simple" fix is to literally make a new class and mark the old one deprecated, but if you want to get fancier that's fine | 15:15 |
| dhellmann | this is a case where duplicating code feels fine, because you don't want that old behavior to change at all | 15:16 |
| krotscheck | dhellmann: Honestly, I think I'm going to make the config non-oslo-specific. | 15:16 |
| dhellmann | krotscheck: I'm not sure what you mean? | 15:16 |
| krotscheck | Well, the constructor takes config as a property, however at the moment it makes the assumption that it's an instance of oslo.config. I'm wondering if there's a decent way to have it sanely detect that and adjust itself accordingly. | 15:17 |
| *** shardy has joined #openstack-oslo | 15:18 | |
| dhellmann | krotscheck: ah, I see, yeah, probably | 15:18 |
| dhellmann | I mean, you could do an explicit check and convert it to something more generic in that case | 15:18 |
| *** tsekiyama has joined #openstack-oslo | 15:20 | |
| krotscheck | dhellmann: Yeah, something like that. I'll play around with it, see what I can make the tests accept. | 15:23 |
| dhellmann | krotscheck: ok. like I said, it's perfectly ok to be expedient in this case, too, so don't spin your wheels on it | 15:24 |
| *** krotscheck has left #openstack-oslo | 15:24 | |
| *** haypo has left #openstack-oslo | 15:24 | |
| *** krotscheck has joined #openstack-oslo | 15:24 | |
| krotscheck | Aight | 15:24 |
| *** yamahata has quit IRC | 15:26 | |
| *** ozamiatin has quit IRC | 15:27 | |
| dhellmann | dims: is this the test that was failing in nova with the versioned objects release? http://paste.openstack.org/show/328826/ | 15:29 |
| dhellmann | that actually looks like an oslo.db issue, so probably not? | 15:30 |
| dims | dhellmann: that was the oslo.db one | 15:30 |
| dims | dhellmann: o.vo failures were about __init__ having wrong number of parameters | 15:30 |
| dhellmann | ok, I got that from the cross-tests in oslo.versionedobjects | 15:30 |
| dhellmann | hmm, I only saw the one failure | 15:31 |
| dhellmann | oh, heh, the change was reverted | 15:31 |
| dhellmann | duh | 15:31 |
| dims | dhellmann: y, we reverted one change and issued 0.5.1 | 15:31 |
| dhellmann | k | 15:31 |
| dims | dhellmann: i am looking at travis-ci.org to see if we can setup something there | 15:32 |
| dhellmann | dims: we should probably just talk to the infra team about running some of these longer jobs -- it's fine to run all of the unit tests in one node in this case if we can prevent the job from timing out | 15:33 |
| dims | dhellmann: y, will look at a few things including that | 15:33 |
| openstackgerrit | Joshua Harlow proposed openstack/futurist: Add future waiting helper module https://review.openstack.org/196269 | 15:34 |
| *** ViswaV_ has joined #openstack-oslo | 15:40 | |
| *** ViswaV has quit IRC | 15:41 | |
| openstackgerrit | Michael Krotscheck proposed openstack/oslo.middleware: Support PasteDeploy https://review.openstack.org/191238 | 15:44 |
| krotscheck | dhellmann: ^^ That should do it. Thoughts on the approach? | 15:44 |
| *** manishg has joined #openstack-oslo | 15:44 | |
| krotscheck | dhellmann: I still have to update the documentation. | 15:45 |
| *** amrith is now known as _amrith_ | 15:48 | |
| *** manishg has quit IRC | 15:49 | |
| *** abitha has joined #openstack-oslo | 15:50 | |
| openstackgerrit | Thomas Goirand proposed openstack-dev/cookiecutter: Add insentive to write a long description https://review.openstack.org/197146 | 15:51 |
| stevemar | dims: did the oslo releases not happen yesterday? | 15:51 |
| dims | stevemar: we did a bunch. which one did you need? | 15:51 |
| stevemar | dims: oslo.cache | 15:51 |
| dims | stevemar: let me check my notes | 15:52 |
| dims | stevemar: there was only one change in tests. | 15:53 |
| dims | stevemar: at the time when i ran my report | 15:54 |
| dims | looks like a couple more went in later in the day | 15:54 |
| *** sdake has joined #openstack-oslo | 15:54 | |
| dims | stevemar: lets get the 4 in the queue in (https://review.openstack.org/#/q/project:openstack%2Foslo.cache+is:open,n,z) and i'll cut another release. ok? | 15:56 |
| *** achanda has joined #openstack-oslo | 15:56 | |
| dims | bknudson: ^^ | 15:56 |
| *** sdake_ has joined #openstack-oslo | 15:56 | |
| stevemar | dims: oh sure | 15:57 |
| ativelkov | dims: I think I've got the fix for versioned_object but I am stuck trying to verify it: attempt to do a clean checkout of nova to run its unittest leads to a number of tests failing with "No module named request" exceptions. I am definetely doing something wrong, but don't know what exactly. | 15:58 |
| openstackgerrit | Michael Krotscheck proposed openstack/oslo.middleware: Support PasteDeploy https://review.openstack.org/191238 | 15:58 |
| stevemar | bknudson: whats the deal with reworking the exceptions for oslo.cache | 15:59 |
| stevemar | they are currently overkill? | 15:59 |
| bknudson | stevemar: mixing http response codes into exceptions doesn't make sense in a library | 15:59 |
| stevemar | that too | 16:00 |
| *** achanda has quit IRC | 16:00 | |
| *** sdake has quit IRC | 16:00 | |
| openstackgerrit | Michael Krotscheck proposed openstack/oslo.middleware: Support PasteDeploy https://review.openstack.org/191238 | 16:01 |
| openstackgerrit | Michael Krotscheck proposed openstack/oslo.middleware: Support PasteDeploy https://review.openstack.org/191238 | 16:01 |
| openstackgerrit | Merged openstack/oslo.cache: Remove unnecessary file https://review.openstack.org/196636 | 16:02 |
| stevemar | bknudson: no auto docs for the backends? | 16:02 |
| krotscheck | ok, dhellmann. This patch addresses your last set of comments. ^^ | 16:03 |
| dhellmann | krotscheck: +2 | 16:04 |
| openstackgerrit | Merged openstack/oslo.service: Add tox target to find missing requirements https://review.openstack.org/195829 | 16:05 |
| openstackgerrit | Merged openstack/oslo.service: Add greenlet to requirements https://review.openstack.org/195835 | 16:05 |
| *** shardy has quit IRC | 16:06 | |
| bknudson | stevemar: are they part of the public API? | 16:06 |
| *** shardy has joined #openstack-oslo | 16:07 | |
| *** ViswaV_ has quit IRC | 16:08 | |
| *** ViswaV has joined #openstack-oslo | 16:09 | |
| openstackgerrit | Merged openstack/oslo.cache: Cleanup exceptions https://review.openstack.org/196478 | 16:09 |
| *** ViswaV has quit IRC | 16:09 | |
| *** arnaud____ has joined #openstack-oslo | 16:10 | |
| *** shardy_ has joined #openstack-oslo | 16:17 | |
| *** shardy has quit IRC | 16:18 | |
| stevemar | bknudson: commented on more stuff | 16:19 |
| *** beekneemech is now known as bnemec | 16:19 | |
| dims | ativelkov: fix Routes==1.13 in requirements.txt | 16:22 |
| ativelkov | dims: thanks | 16:22 |
| dims | ativelkov: https://bugs.launchpad.net/nova/+bug/1469883 | 16:22 |
| openstack | Launchpad bug 1469883 in OpenStack Compute (nova) "Unit tests fail with Routes==2.1" [Undecided,New] | 16:22 |
| *** shardy_ has quit IRC | 16:23 | |
| *** e0ne is now known as e0ne_ | 16:23 | |
| *** shardy has joined #openstack-oslo | 16:24 | |
| dims | ativelkov: that reminds me to ping dhellmann about https://review.openstack.org/#/c/196867/ :) | 16:25 |
| ativelkov | dims: now it complains on "cannot import name apirequest". Missing something else? | 16:26 |
| *** jaypipes has joined #openstack-oslo | 16:26 | |
| *** _amrith_ is now known as amrith | 16:26 | |
| dims | ativelkov: did you recreate the .tox ? | 16:27 |
| openstackgerrit | Brant Knudson proposed openstack/oslo.cache: Make symbols private https://review.openstack.org/196463 | 16:27 |
| *** dontalton has joined #openstack-oslo | 16:29 | |
| openstackgerrit | Brant Knudson proposed openstack/oslo.cache: Make symbols private https://review.openstack.org/196463 | 16:29 |
| ativelkov | dims: nope. Thanks, looks like it helps. | 16:31 |
| *** e0ne_ has quit IRC | 16:34 | |
| *** achanda has joined #openstack-oslo | 16:35 | |
| *** e0ne has joined #openstack-oslo | 16:36 | |
| *** manishg has joined #openstack-oslo | 16:38 | |
| *** rpodolyaka1 has quit IRC | 16:41 | |
| *** manishg has quit IRC | 16:43 | |
| *** rpodolyaka1 has joined #openstack-oslo | 16:44 | |
| *** yassine_ has quit IRC | 16:47 | |
| *** openstackgerrit has quit IRC | 16:50 | |
| *** openstackgerrit has joined #openstack-oslo | 16:51 | |
| *** manishg has joined #openstack-oslo | 16:58 | |
| *** manishg has quit IRC | 16:59 | |
| *** manishg has joined #openstack-oslo | 16:59 | |
| *** manishg has quit IRC | 17:00 | |
| *** rpodolyaka1 has quit IRC | 17:02 | |
| *** e0ne has quit IRC | 17:05 | |
| *** ViswaV has joined #openstack-oslo | 17:07 | |
| *** manishg has joined #openstack-oslo | 17:07 | |
| *** samueldmq has quit IRC | 17:08 | |
| openstackgerrit | Merged openstack/oslo.cache: Module docs https://review.openstack.org/196464 | 17:09 |
| *** sdake has joined #openstack-oslo | 17:18 | |
| *** rpodolyaka1 has joined #openstack-oslo | 17:18 | |
| *** rpodolyaka3 has joined #openstack-oslo | 17:19 | |
| *** rpodolyaka1 has quit IRC | 17:19 | |
| *** rpodolyaka3 has quit IRC | 17:21 | |
| *** sdake_ has quit IRC | 17:21 | |
| * harlowja thinks everyone in US has left for vacation | 17:27 | |
| harlowja | oh wells | 17:27 |
| *** zzzeek_ has joined #openstack-oslo | 17:28 | |
| *** zzzeek_ has quit IRC | 17:31 | |
| *** achanda_ has joined #openstack-oslo | 17:38 | |
| *** zzzeek_ has joined #openstack-oslo | 17:40 | |
| *** abitha has quit IRC | 17:40 | |
| *** achanda has quit IRC | 17:40 | |
| *** achanda has joined #openstack-oslo | 17:42 | |
| *** zzzeek_ has quit IRC | 17:42 | |
| *** kzaitsev_mb has joined #openstack-oslo | 17:43 | |
| *** jecarey has quit IRC | 17:43 | |
| *** pradk has joined #openstack-oslo | 17:44 | |
| *** achanda_ has quit IRC | 17:44 | |
| openstackgerrit | Merged openstack/automaton: Add a bigger CD-player state-machine doctest(ed) example https://review.openstack.org/191477 | 17:45 |
| openstackgerrit | Merged openstack/futurist: Add badge support to README.rst https://review.openstack.org/196279 | 17:45 |
| *** mgarza_ has joined #openstack-oslo | 17:46 | |
| *** sputnik13 has joined #openstack-oslo | 17:46 | |
| *** sdake_ has joined #openstack-oslo | 17:47 | |
| openstackgerrit | Merged openstack/oslo.concurrency: Remove redundant fileutils https://review.openstack.org/196945 | 17:47 |
| openstackgerrit | Merged openstack/oslo.reports: Add tox target to find missing requirements https://review.openstack.org/195840 | 17:48 |
| openstackgerrit | Merged openstack/oslo.serialization: Stop using deprecated oslo_utils.timeutils.strtime https://review.openstack.org/197069 | 17:48 |
| openstackgerrit | Merged openstack/oslo.reports: Add oslo.utils dependency https://review.openstack.org/195841 | 17:49 |
| *** sdake has quit IRC | 17:50 | |
| *** sdake_ is now known as sdake | 17:50 | |
| *** ihrachyshka has joined #openstack-oslo | 17:52 | |
| *** kzaitsev_mb has quit IRC | 17:52 | |
| *** jecarey has joined #openstack-oslo | 17:56 | |
| *** manishg has quit IRC | 17:58 | |
| *** yamahata has joined #openstack-oslo | 17:58 | |
| *** e0ne has joined #openstack-oslo | 17:58 | |
| *** ViswaV_ has joined #openstack-oslo | 18:01 | |
| *** ViswaV has quit IRC | 18:01 | |
| openstackgerrit | Joshua Harlow proposed openstack/oslo.service: Make it easier to use the eventlet backdoor locally https://review.openstack.org/196324 | 18:03 |
| *** e0ne is now known as e0ne_ | 18:04 | |
| *** ViswaV_ has quit IRC | 18:05 | |
| *** ViswaV has joined #openstack-oslo | 18:06 | |
| *** e0ne_ is now known as e0ne | 18:07 | |
| *** ericksonsantos has joined #openstack-oslo | 18:14 | |
| *** manishg has joined #openstack-oslo | 18:17 | |
| *** e0ne is now known as e0ne_ | 18:18 | |
| *** rushiagr is now known as rushiagr_away | 18:18 | |
| *** rpodolyaka1 has joined #openstack-oslo | 18:19 | |
| *** achanda has quit IRC | 18:21 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslo.versionedobjects: Sort child versions before generating fingerprint https://review.openstack.org/197212 | 18:22 |
| *** e0ne_ is now known as e0ne | 18:23 | |
| dims | dansmith: ^^ need one more thing before we can drop fingerprint code from nova and use the one in o.vo | 18:25 |
| dansmith | dims: hah, we're actually just ripping that out of nova right now | 18:26 |
| *** rpodolyaka1 has quit IRC | 18:26 | |
| dims | dansmith: so we won't need this? | 18:27 |
| *** shardy has quit IRC | 18:27 | |
| dansmith | well, | 18:27 |
| *** tsekiyama has quit IRC | 18:27 | |
| dansmith | it's in the fixture, which means we can't remove it I guess | 18:27 |
| *** shardy has joined #openstack-oslo | 18:27 | |
| dansmith | but soonish nova won't use it so none of our fingerprints will include it | 18:27 |
| dansmith | so probably good to fix it in o.vo I guess | 18:27 |
| dims | right | 18:28 |
| *** manishg has quit IRC | 18:37 | |
| *** pradk has quit IRC | 18:37 | |
| *** tsekiyama has joined #openstack-oslo | 18:38 | |
| *** arnaud____ has quit IRC | 18:41 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Add ability to reset an engine via a `reset` method https://review.openstack.org/186239 | 18:42 |
| *** rpodolyaka1 has joined #openstack-oslo | 18:43 | |
| *** sdake has quit IRC | 18:46 | |
| *** manishg has joined #openstack-oslo | 18:47 | |
| mtanino | dims: Thank you for pushing https://review.openstack.org/194357 | 18:47 |
| dims | mtanino: you have been waiting too long. thanks for your patience | 18:47 |
| mtanino | dims: I will try to backport that fix to other components. thanks, | 18:47 |
| *** sdake has joined #openstack-oslo | 18:48 | |
| openstackgerrit | Merged openstack-dev/cookiecutter: Add insentive to write a long description https://review.openstack.org/197146 | 18:56 |
| *** ozamiatin has joined #openstack-oslo | 18:58 | |
| *** browne has quit IRC | 19:00 | |
| *** kgiusti has left #openstack-oslo | 19:01 | |
| dims | stevemar: looks like 196463 is still in merging | 19:01 |
| openstackgerrit | Merged openstack/oslo.concurrency: Add tox target to find missing requirements https://review.openstack.org/196000 | 19:05 |
| dstanek | dhellmann: what's the difference between using oslo_policy.opts.set_defaults() vs. oslo_policy.opts.register() - other than register being more explicit | 19:08 |
| *** achanda has joined #openstack-oslo | 19:10 | |
| dhellmann | dstanek: set_defaults() does not then assume you will go cfg.CONF.oslo_policy.some_option = value | 19:11 |
| *** ozamiatin has quit IRC | 19:11 | |
| dhellmann | so if we change "some_option" to "new_name_for_option" your code won't break, because we'll make set_defaults() handle the transition | 19:11 |
| openstackgerrit | Merged openstack/oslo.cache: Make symbols private https://review.openstack.org/196463 | 19:11 |
| dhellmann | dstanek: the minute you allow something outside of the library to use a configuration option directly through the config object, that option is frozen and cannot be change | 19:12 |
| dstanek | dhellmann: do you're saying we shouldn't use the config fixture anymore? | 19:12 |
| *** manishg has quit IRC | 19:12 | |
| dhellmann | dstanek: the config fixture is appropriate for tests within a repo to set the options within that repo, but it's not a good idea for setting options in code from another repo | 19:12 |
| dhellmann | so keystone can use it to set their own options, but should not use it to set options for a library -- that's why we provide the set_defaults() API | 19:13 |
| dhellmann | API(s) | 19:13 |
| *** shardy has quit IRC | 19:13 | |
| *** e0ne is now known as e0ne_ | 19:14 | |
| dstanek | dhellmann: that's interesting. i've not seen that used anywhere yet. do we document that we should be using that? | 19:14 |
| *** e0ne_ is now known as e0ne | 19:15 | |
| dhellmann | dstanek: I'm not sure we do have that written down, I should write up a policy document | 19:15 |
| dhellmann | it mostly comes up when folks try to add APIs like this to the libraries and we say no, so education is happening that way, but it's not optimal | 19:15 |
| *** manishg has joined #openstack-oslo | 19:16 | |
| dstanek | so right now oslo_policy's set_defaults only allows one of the options to be set :-( | 19:16 |
| *** manishg_ has joined #openstack-oslo | 19:17 | |
| dhellmann | dstanek: that's ok, we can add some more. Which ones do you need to set? | 19:17 |
| *** mgarza_ has quit IRC | 19:18 | |
| openstackgerrit | Merged openstack/automaton: Add runners to features.rst & add a runner base & update docstrings https://review.openstack.org/196394 | 19:19 |
| *** rpodolyaka1 has quit IRC | 19:20 | |
| dstanek | dhellmann: i think keystone only needs the one defined. we should add a reset method that tests can call in a teardown | 19:20 |
| *** manishg has quit IRC | 19:20 | |
| *** rpodolyaka1 has joined #openstack-oslo | 19:21 | |
| dhellmann | dstanek: I think set_defaults() expects a config object as argument, and if you past a ConfigFixture it will register the options in a way that unregisters them later | 19:22 |
| dstanek | dhellmann: does that mean oslo_log.register_options() should not longer be used? | 19:22 |
| dstanek | dhellmann: the config fixture won't let me set an option until it is defined | 19:23 |
| dstanek | that's why i wanted to call register | 19:23 |
| dhellmann | dstanek: that's a special case. We have to register some of the log options before we can parse the config file. I forget why. | 19:23 |
| dhellmann | dstanek: does the config fixture not have a register_opt method? I thought it did | 19:23 |
| dhellmann | yeah, it does | 19:24 |
| dstanek | dhellmann: doesn't that defeat the purpose of why you want to do with set_defaults? | 19:24 |
| dhellmann | you're keeping the level of indirection intact | 19:24 |
| dhellmann | your code is not ever accessing the config option as an attribute directly, so it's safe for us to change the name of the option | 19:25 |
| dstanek | either way my tests will depend on knowing the config file option right? | 19:25 |
| dhellmann | hang on, let me put together an example | 19:25 |
| *** kzaitsev_mb has joined #openstack-oslo | 19:26 | |
| openstackgerrit | Merged openstack/oslo.messaging: Add tox target to find missing requirements https://review.openstack.org/195819 | 19:28 |
| dims | stevemar: 0.1.2 oslo.cache is out | 19:28 |
| openstackgerrit | Merged openstack/automaton: Add badge support to README.rst https://review.openstack.org/196281 | 19:30 |
| openstackgerrit | Merged openstack/automaton: Disallow adding transitions from terminal states https://review.openstack.org/195323 | 19:30 |
| dhellmann | dstanek: oh, bother, what I told you doesn't actually work. We can make it work, it just doesn't right now. | 19:30 |
| *** crc32 has joined #openstack-oslo | 19:33 | |
| dhellmann | dstanek: patch coming shortly | 19:35 |
| dstanek | dhellmann: patch for register_opt? | 19:35 |
| dhellmann | dstanek: to make the fixture work the way you need it to, yeah | 19:35 |
| dhellmann | bah, now I can't make the oslo.config tests run | 19:39 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use encodeutils for exception -> string function https://review.openstack.org/192458 | 19:39 |
| dims | dhellmann: a couple of pending reviews? | 19:40 |
| dhellmann | dims, lifeless : have you seen http://paste.openstack.org/show/329443/ before? | 19:40 |
| *** rpodolyaka1 has quit IRC | 19:40 | |
| dhellmann | dims: I don't see anything that looks related | 19:40 |
| dhellmann | I suspect something is wrong with my local setup, but I can't imagine what | 19:41 |
| dims | dhellmann: i saw it here in CI just a half hour ago, haven't looked yet http://logs.openstack.org/49/195849/4/check/gate-oslo.config-python26/48adfd1/testr_results.html.gz | 19:41 |
| dhellmann | that version of fixtures was released today, I wonder if it has a bug | 19:42 |
| dhellmann | dims: downgrading fixtures to 1.2.0 fixed it | 19:44 |
| dhellmann | it looks like a problem with calling cleanup more than once on a fixture | 19:46 |
| *** Rockyg has joined #openstack-oslo | 19:51 | |
| stevemar | dims: ty! | 19:52 |
| *** pblaho has quit IRC | 19:52 | |
| *** yamahata has quit IRC | 19:54 | |
| *** yamahata has joined #openstack-oslo | 19:54 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack-dev/pbr: Updated from global requirements https://review.openstack.org/197249 | 19:54 |
| *** achanda has quit IRC | 19:54 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/futurist: Updated from global requirements https://review.openstack.org/196262 | 19:55 |
| dhellmann | dims: ok, it looks like the right fix is for us to rewrite these fixture tests, so I'll work on that | 19:55 |
| *** achanda has joined #openstack-oslo | 19:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/mox3: Updated from global requirements https://review.openstack.org/197257 | 19:57 |
| *** browne has joined #openstack-oslo | 19:58 | |
| *** rpodolyaka1 has joined #openstack-oslo | 19:58 | |
| *** amrith is now known as _amrith_ | 19:59 | |
| *** achanda has quit IRC | 19:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.concurrency: Updated from global requirements https://review.openstack.org/194835 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.config: Updated from global requirements https://review.openstack.org/197262 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.db: Updated from global requirements https://review.openstack.org/197263 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.log: Updated from global requirements https://review.openstack.org/197264 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.messaging: Updated from global requirements https://review.openstack.org/192382 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.middleware: Updated from global requirements https://review.openstack.org/197265 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.rootwrap: Updated from global requirements https://review.openstack.org/197266 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.utils: Updated from global requirements https://review.openstack.org/197267 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.versionedobjects: Updated from global requirements https://review.openstack.org/196263 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.vmware: Updated from global requirements https://review.openstack.org/197268 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslotest: Updated from global requirements https://review.openstack.org/197269 | 20:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tooz: Updated from global requirements https://review.openstack.org/194837 | 20:04 |
| lifeless | dhellmann: hi, thats fallout from 1.3.0 | 20:05 |
| lifeless | dhellmann: what it means is that somewhere this is happening: | 20:14 |
| lifeless | f = Fixture() | 20:14 |
| lifeless | f.setUp() | 20:14 |
| lifeless | f.cleanUp() | 20:14 |
| lifeless | f.cleanUp() | 20:14 |
| lifeless | dhellmann: which was never legal | 20:14 |
| dhellmann | yeah, I have a patch to fix it up | 20:14 |
| lifeless | dhellmann: linky linky? | 20:14 |
| dhellmann | lifeless: not submitted yet, I'm still working on the one I actually wanted to merge :-) | 20:15 |
| dhellmann | I'll submit both together | 20:15 |
| lifeless | k | 20:15 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: fix fixture tests for fixtures 1.3.x https://review.openstack.org/197299 | 20:26 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Add set_default() method to fixture https://review.openstack.org/197300 | 20:26 |
| dhellmann | lifeless: ^^ | 20:26 |
| dims | dhellmann: so will work with both old and new fixtures right? | 20:27 |
| dhellmann | dstanek: the thing you need is in ^^ | 20:28 |
| dhellmann | dims: yes | 20:28 |
| dims | thanks | 20:28 |
| dhellmann | dims: it changes the way we were using the fixture, but doesn't change its implementation | 20:28 |
| dims | dhellmann: add one more test with a group? https://review.openstack.org/#/c/197300/1/oslo_config/tests/test_fixture.py,cm | 20:29 |
| dhellmann | dims: sure | 20:30 |
| *** mgarza has joined #openstack-oslo | 20:31 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Add set_default() method to fixture https://review.openstack.org/197300 | 20:31 |
| *** e0ne has quit IRC | 20:35 | |
| lifeless | dhellmann: some thoughts posted on 299 | 20:38 |
| *** rpodolyaka1 has quit IRC | 20:41 | |
| dhellmann | lifeless: replied | 20:44 |
| dstanek | dhellmann: i'll give that a try, but it still ties the consumer project to the library via the config opt name, which is what i thought you were arguing against when you said to use set_default | 20:45 |
| *** jaosorior has quit IRC | 20:46 | |
| dhellmann | dstanek: no, it ties you to the library via the name of the argument to set_defaults(). That happens to be the same as the config name, but it does not have to be. If we ever change a config option, or move options from the DEFAULT group to a private group, that can all happen transparently. | 20:46 |
| dhellmann | dstanek: I never posted my example, hang on | 20:46 |
| dhellmann | dstanek: http://paste.openstack.org/show/329594 | 20:46 |
| *** ozamiatin has joined #openstack-oslo | 20:47 | |
| dhellmann | dstanek: I admit the distinction is subtle, but the technicality of it is important | 20:47 |
| dstanek | dhellmann: i get the sentiment, but it feels wrong that they would ever be different. | 20:47 |
| lifeless | dhellmann: reviewed 300 too now | 20:47 |
| dhellmann | dstanek: that's how we'll address renames, though, so it may happen during a deprecation period | 20:47 |
| dhellmann | dstanek: I would expect the old name to trigger a deprecation warning, and then apply the value to the new name | 20:48 |
| dstanek | dhellmann: i thought we already had a way to deprecate config options in oslo.config | 20:48 |
| dhellmann | dstanek: we do | 20:49 |
| dstanek | dhellmann: isn't that sufficient? | 20:49 |
| dhellmann | dstanek: I'm just saying that when that happens, the API for setting the default programmatically would also need to change | 20:49 |
| dhellmann | dstanek: I'm in a meeting, i can't give this my full attention. Can we chat tomorrow? | 20:49 |
| dstanek | dhellmann: sure | 20:49 |
| lifeless | krotscheck: dhellmann: I don't want the wsgi stuff to be wedged between two reviewers. Can we make a timeslice to jointly discuss today ? | 20:50 |
| dhellmann | dstanek: thanks, and sorry | 20:50 |
| dstanek | dhellmann: no reason to be sorry! we can char tomorrow | 20:50 |
| dstanek | chat | 20:50 |
| dhellmann | lifeless: 66 minutes from now-ish? | 20:55 |
| lifeless | krotscheck: ^ work for you ? | 20:56 |
| lifeless | dhellmann: it certainly works for me | 20:56 |
| *** stevemar has quit IRC | 20:58 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Add set_default() method to fixture https://review.openstack.org/197300 | 21:07 |
| dhellmann | lifeless: I rolled your suggestions into 300 since it was going to fail pep8 checks anyway | 21:07 |
| krotscheck | lifeless: Eh? Wsgi stuff? | 21:22 |
| krotscheck | Are we talking about the TypeError thing? | 21:22 |
| lifeless | krotscheck: yeah | 21:25 |
| lifeless | krotscheck: I just want to make sure you're unblocked | 21:25 |
| krotscheck | lifeless: I'm not really blocked. That add was your idea ;) | 21:25 |
| krotscheck | lifeless: I can live without it :) | 21:25 |
| lifeless | oh, I see, the getargspec one landed | 21:26 |
| lifeless | I presume its not in a release yet. | 21:26 |
| lifeless | ok, so I'll arm wrestle with doug on it and we'll get back to you :) | 21:26 |
| lifeless | krotscheck: thanks for following up | 21:27 |
| krotscheck | lifeless: Aight, no problem | 21:29 |
| krotscheck | lifeless: You could just git review -d it and make any modifications yourself :) | 21:29 |
| lifeless | so there are three routes; as-is, double-dispatch (the current patch in question), or new-method (which avoids all the failure modes) | 21:30 |
| lifeless | I'll talk with dhellmann in 34m+- and we'll see :) | 21:30 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.i18n: clarify translation policy https://review.openstack.org/197339 | 21:30 |
| openstackgerrit | Merged openstack/mox3: Updated from global requirements https://review.openstack.org/197257 | 21:32 |
| lifeless | dhellmann: you missed the changes I separately suggested on 300 itself in patchset 2 | 21:33 |
| *** mriedem has quit IRC | 21:41 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Add set_default() method to fixture https://review.openstack.org/197300 | 21:42 |
| dhellmann | lifeless: done ^^ | 21:42 |
| *** kzaitsev_mb has quit IRC | 21:46 | |
| openstackgerrit | Joshua Harlow proposed openstack/automaton: Document `process_event` return and return type https://review.openstack.org/197345 | 21:48 |
| *** eezhova has quit IRC | 21:48 | |
| *** sreshetnyak has quit IRC | 21:49 | |
| dhellmann | lifeless, krotscheck : ok | 21:50 |
| krotscheck | yeees? | 21:52 |
| dhellmann | krotscheck: I think lifeless wanted to talk about https://review.openstack.org/#/c/194268/1/oslo_middleware/base.py,cm | 21:53 |
| dhellmann | is there some background behind that change? there's not much detail in the commit message | 21:53 |
| krotscheck | dhellmann, lifeless: Ok, so the issue is that lifeless doesn't like using getargspec, because it's heavy and could have a noticable impact on performance across 1000's of requests. His proposed alternative, using except TypeError:, is a bit too aggressive in its exception handling. | 21:54 |
| krotscheck | I'm just the poor schmuck who decided to propose lifeless's suggested change. I should have just let him do it. | 21:54 |
| dhellmann | krotscheck: ok, if we have a performance concern then it would be good to have that in the commit message explaining why the change is desired | 21:55 |
| dhellmann | I'm still not sure about the exception handling thing, though. I dislike that it's such a common exception | 21:55 |
| dhellmann | someone could pass an int where we expect a string and we could get that and run the request again | 21:55 |
| krotscheck | dhellmann: I'm going to tag out to lifeless. I'm ok with the getargspec bit. | 21:55 |
| dhellmann | krotscheck: ok | 21:55 |
| *** ozamiatin has quit IRC | 21:57 | |
| dhellmann | lifeless: I'm going to step away for a couple of minutes, but will be back | 21:57 |
| *** jecarey has quit IRC | 21:57 | |
| *** stpierre has quit IRC | 21:57 | |
| *** ihrachyshka has quit IRC | 21:58 | |
| lifeless | ok back | 21:59 |
| dhellmann | lifeless: krotscheck explained that the change is coming from a performance concern? | 22:00 |
| lifeless | dhellmann: well | 22:00 |
| lifeless | dhellmann: I raised some issues on the original change | 22:01 |
| lifeless | dhellmann: and proposed this approach | 22:01 |
| dhellmann | I'm trying to understand what those are | 22:01 |
| lifeless | dhellmann: the original change has several issues - a) its heavyweight and per-request, b) it doesn't permit decorators or *args or *kwargs function definitions in subclasses | 22:01 |
| lifeless | which is surprising | 22:01 |
| dhellmann | lifeless: ok. I am probably less familiar with the wsgi spec than you. Is process_response() even a part of that spec? do we need to be worried about someone using this middleware to wrap something we didn't write? | 22:04 |
| lifeless | its not | 22:04 |
| lifeless | but | 22:04 |
| *** yamahata has quit IRC | 22:04 | |
| dhellmann | ok, that's fine | 22:04 |
| lifeless | nothing in webob is :) | 22:04 |
| dhellmann | so that means we can be more flexible | 22:04 |
| dhellmann | or not | 22:04 |
| lifeless | right | 22:04 |
| lifeless | so in a lot of ways i'd like to rip webob out entirely | 22:04 |
| lifeless | but thats an entirely different discussion | 22:05 |
| lifeless | if all our middleware is in the one repo | 22:05 |
| lifeless | then we don't need the backwards compat shim | 22:05 |
| lifeless | if its not in the one repo, then we're essentially dealing with the whole openstack ecosystem | 22:05 |
| lifeless | which means IMO that we should think about this as a public API evolution | 22:05 |
| dhellmann | well, no, I'm trying to understand how this stuff is intended to be used because it seems like adding this argument is a hugely dangerous change if we're going to say sometimes it's there and sometimes its not and the way we tell if it's not is if we try to use it an get TypeError, because all sorts of things throw TypeError | 22:06 |
| dhellmann | right, exactly | 22:06 |
| dhellmann | so, I get the point about looking for the special name | 22:06 |
| lifeless | so lets step back a second | 22:06 |
| dhellmann | and we should at least change it to allow *args | 22:06 |
| openstackgerrit | Joshua Harlow proposed openstack/automaton: Allow providing and using a 'nested_start_state_fetcher' https://review.openstack.org/197351 | 22:06 |
| dhellmann | it's less clear how we could handle **kwds | 22:06 |
| lifeless | we've three ways to deal with 'add an argument to an override point' | 22:06 |
| lifeless | - introspection [which landed] | 22:06 |
| lifeless | - new override point- e.g. process_response2 | 22:07 |
| lifeless | - try-and-handle-typeerror | 22:07 |
| harlowja | also fyi, https://docs.python.org/dev/library/inspect.html#inspect.getargspec | 22:07 |
| harlowja | 'This function will be removed in Python 3.6.' | 22:07 |
| harlowja | just fyi :-/ | 22:07 |
| lifeless | rather than ratholling on the typeerror aspect | 22:07 |
| lifeless | maybe we should just do the graceful thing and use a new function name | 22:07 |
| lifeless | if hasattr(self, 'process_response2'): call-that else call-process_response() | 22:08 |
| dhellmann | sure, that works. I don't know what that's going to break now that we have this -- is it released? | 22:08 |
| lifeless | I don't think so | 22:08 |
| lifeless | it landed in the last week I think ? | 22:08 |
| *** gordc is now known as gordc_afk | 22:09 | |
| dhellmann | yeah, it was released | 22:09 |
| lifeless | argh | 22:09 |
| lifeless | so the graceful way is to add parse_response2 | 22:09 |
| dhellmann | looks like it was in version 2.2.0 | 22:09 |
| lifeless | with the CORS middleware (the only in-tree user) updated to use tha | 22:10 |
| dhellmann | I'm ok with doing that | 22:10 |
| lifeless | and add a deprecation warning to the argspec path | 22:10 |
| lifeless | krotscheck: are you willing to do that? I realise its adding more work to your plate. perhaps if you have a few idle minutes? | 22:11 |
| dhellmann | it would be interesting to see if we even need to do the deprecation, but safer to just do it | 22:11 |
| krotscheck | Sec, reading backscroll. | 22:12 |
| lifeless | dhellmann: someone might pick up 2.2.0 in 3 months then use the thing, then be surprised in e.g. 2.4. | 22:12 |
| *** Rockyg has quit IRC | 22:12 | |
| dhellmann | lifeless: we can make the next release 3.0 | 22:12 |
| dhellmann | I'd be pretty surprised if someone outside of openstack was using this lib in this way | 22:13 |
| lifeless | dhellmann: I'd rather not do API breaks rapidly. I know we can... | 22:13 |
| dhellmann | yeah | 22:13 |
| lifeless | dhellmann: but I really like being pleasant for consumers | 22:13 |
| dhellmann | yeah, I'm ok with the deprecations, just trying to cut down on supporting dead weight | 22:14 |
| * krotscheck is amused that we're already talking about deprecating things when we don't even know that anyone's using the middleware yet. | 22:15 | |
| dhellmann | krotscheck: yeah, that's the problem, we have no way of telling, so we have to be conservative | 22:15 |
| morganfainberg | dims, dhellmann: added you both to https://review.openstack.org/#/c/196329/ when you have a few | 22:16 |
| krotscheck | dhellmann: See, that's where we disagree. I'm more of a forgiveness > permission type. | 22:16 |
| dhellmann | krotscheck: I'll have to send you some of the hate mail I get | 22:16 |
| krotscheck | Mind you, breaking changes in anything but major versions are bad juju | 22:16 |
| krotscheck | dhellmann: I'll warm up the popcorn. | 22:17 |
| krotscheck | (And this is why I love Java. Because if I update a library, it won't compile.) | 22:17 |
| dhellmann | krotscheck: it's just a different type of break | 22:17 |
| dhellmann | morganfainberg: I don't know anything about the background there | 22:18 |
| *** manishg_ has quit IRC | 22:18 | |
| *** dims_ has joined #openstack-oslo | 22:19 | |
| morganfainberg | dhellmann: so the sanity check doesn't really catch anything in the gate | 22:19 |
| morganfainberg | what it does do is catch when our deployers have already got a broken - non-utf-8 table | 22:19 |
| morganfainberg | and then they are wedged | 22:19 |
| dhellmann | why are they wedged? | 22:19 |
| morganfainberg | because the sanity check runs before and after migrations | 22:20 |
| morganfainberg | you can't unwedge w/o a try/except or specific disable of the sanity check in the db_sync cli took | 22:20 |
| morganfainberg | tool* | 22:20 |
| krotscheck | lifeless: I don't think I quite get what you're suggesting. I think it might be easier on all of us, and reduce churn, if you take over this patch. | 22:20 |
| morganfainberg | the migration can't disable the sanity check | 22:20 |
| dhellmann | morganfainberg: ok | 22:20 |
| morganfainberg | so you're wedged when any table is not utf-8. ive never seen this catch things in gate - i've seen it wedge multiple deployers -- i think this is the 4th time? | 22:20 |
| lifeless | krotscheck: I have negative available cycles. Remember your complaint about TC bandwidth? | 22:21 |
| dhellmann | morganfainberg: should we just turn this off in oslo.db? | 22:21 |
| lifeless | krotscheck: we're dependent on being able to convince other people to do things, or we end up just busy and not doing what we're meant to be doing. | 22:21 |
| morganfainberg | dhellmann: thats my view. but i'm also happy to just disable it in keystone (i think we're the only ones that have ended up wedged) | 22:21 |
| *** manishg has joined #openstack-oslo | 22:21 | |
| krotscheck | lifeless: Ok, so we've got a problem where a change needs to be made, but the people who are most qualified to make it are tied up with other things. | 22:21 |
| morganfainberg | dhellmann: the real fix is to make sure we specify utf-8 on the migrations | 22:21 |
| lifeless | krotscheck: I think you're entirely qualified. I'll write a paragraph up on the review and see if it helps clarify things. | 22:21 |
| dhellmann | morganfainberg: ok. Maybe submit a second patch to oslo.db, then we can discuss them separately with rpodolyaka and viktors | 22:21 |
| krotscheck | lifeless: I'm saying that I'm also busy | 22:22 |
| *** dims has quit IRC | 22:22 | |
| morganfainberg | yeah wanted to run it buy you and dims before the oslo.db change | 22:22 |
| dhellmann | morganfainberg: ok, now I'm caught up :-) | 22:22 |
| morganfainberg | dhellmann: cool i'll also propose [as promised, now that we have proper stable branches] the little code to elegantly say "no no no no downgrades" to oslo.db | 22:23 |
| *** pleia2 has quit IRC | 22:23 | |
| dhellmann | morganfainberg: ack | 22:24 |
| *** pleia2 has joined #openstack-oslo | 22:25 | |
| ViswaV | Hi Need some help with olso.messaging. Our RabbitMQ servers are filling up with reply_* queues which are created when you make a sync call(). The calls are succeeding and returning output from remote server. But the queues are not getting cleaned up. | 22:25 |
| ViswaV | I even invoke transport.cleanup() after the sync call completes. Is there some config param I need to set or is this a known issue? | 22:26 |
| *** bknudson has quit IRC | 22:26 | |
| dhellmann | ViswaV: I think you're going to want sileht, but he's in Europe so it's late there right now | 22:27 |
| ViswaV | Thx dhellmann. Will try reaching out to him via email or check back again later in the day | 22:27 |
| lifeless | krotscheck: so, everyone is busy. We landed something that has added technical debt to the project - its added use of a deprecated function (getargspec) - and we've got a fairly small better path. | 22:28 |
| krotscheck | lifeless: Would love to keep discussing this, however I've got a doctor's appointment I have to be at. Back in 1 hour + commute. | 22:28 |
| lifeless | krotscheck: ok | 22:29 |
| lifeless | krotscheck: I have added a comment that hopefully clarifies on the review | 22:29 |
| dstanek | dhellmann: was https://review.openstack.org/#/c/197300 to fix my issue? | 22:30 |
| *** jgrimm is now known as zz_jgrimm | 22:31 | |
| *** mriedem has joined #openstack-oslo | 22:32 | |
| dhellmann | dstanek: yes | 22:33 |
| *** manishg has quit IRC | 22:36 | |
| openstackgerrit | Thomas Goirand proposed openstack-dev/oslo-cookiecutter: Add insentive to write a long description https://review.openstack.org/197357 | 22:36 |
| *** dontalton has quit IRC | 22:37 | |
| dstanek | dhellmann: it doesn't actually fix it because it still want me to register the options before i call set_default | 22:38 |
| dstanek | dhellmann: i'm trying the latest patch now | 22:39 |
| dhellmann | dstanek: the set_defaults() function inside the library should register them for you -- the docstring illustrates how you should use it | 22:40 |
| *** zz_jgrimm is now known as jgrimm | 22:40 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Add FAQ entry for why we do not treat config options as API https://review.openstack.org/197359 | 22:41 |
| dhellmann | dstanek: ^^ | 22:41 |
| dstanek | dhellmann: http://paste.openstack.org/show/329910/ | 22:41 |
| dhellmann | dstanek: where's your app code? | 22:41 |
| dhellmann | or that test or whatever | 22:42 |
| dstanek | it doesn't look like it gets to the point where oslo_config's set_default is called | 22:42 |
| dstanek | dhellmann: the diff is here http://paste.openstack.org/show/329911/ | 22:42 |
| dstanek | then i run the test with 'tox -e py27 -- keystone.tests.unit.test_v2_keystoneclient_sql.ClientDrivenSqlTestCase.test_authenticate_disabled_tenant' | 22:42 |
| dhellmann | dstanek: change self.config_fixture.set_default(...) to oslo_policy.opts.set_defaults(self.config_fixture, ...) | 22:43 |
| dhellmann | dstanek: you're calling the low-level API directly, but you should be passing the fixture to the high-level API | 22:44 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/oslo.db: Updated from global requirements https://review.openstack.org/197263 | 22:44 |
| dhellmann | dstanek: make it oslo_policy.opts.set_defaults(self.config_fixture, policy_file=dirs.etc('policy.json')) | 22:45 |
| dhellmann | the group name should be hidden from you outside of the lib | 22:45 |
| dstanek | dhellmann: ah, ok. let me give that a try | 22:46 |
| dhellmann | dstanek: dinner time here, I'll catch up with you tomorrow and make sure you've got it working | 22:48 |
| *** ozamiatin has joined #openstack-oslo | 22:50 | |
| krotscheck | lifeless: I'm afraid I don't really see how this patch should be high priority for me, sorry. | 23:01 |
| *** dims_ has quit IRC | 23:01 | |
| krotscheck | I'll put it on my list, but I can't guarantee that I'll get around to it by the time liberty ships, what with paternity and all. | 23:01 |
| *** manishg has joined #openstack-oslo | 23:03 | |
| lifeless | krotscheck: I can understand that. I need to figure out how to reconcile everyone's focus-on-just-narrow-slice with managing techdebt, and not hairing off and just doing-myself. E.g. meeting TC needs etc. | 23:04 |
| lifeless | krotscheck: because, the knee jerk thing is to seek lots of control, start -1ing and -2ing everything thats not perfect etc | 23:04 |
| lifeless | krotscheck: which has lots of negative side effects | 23:04 |
| lifeless | krotscheck: I appreciate your frankness however | 23:05 |
| openstackgerrit | Adam Gandelman proposed openstack/tooz: Ensure run_watchers called from mixin, not base class https://review.openstack.org/197367 | 23:05 |
| *** ashleighfarnham has quit IRC | 23:06 | |
| harlowja | lifeless i thought me and u were gonna fix all of openstack man | 23:07 |
| harlowja | lol | 23:07 |
| harlowja | by tommorow noon | 23:07 |
| *** tsbot_ has quit IRC | 23:07 | |
| lifeless | harlowja: if we can discard compat, sure :) | 23:08 |
| harlowja | append -redux on all the things, and be done with it | 23:08 |
| lifeless | harlowja: nova2.0 | 23:09 |
| harlowja | :-P | 23:09 |
| harlowja | nova X | 23:12 |
| harlowja | *has nothing to do with os X* | 23:12 |
| harlowja | lol | 23:12 |
| openstackgerrit | Merged openstack/tooz: Updated from global requirements https://review.openstack.org/194837 | 23:13 |
| openstackgerrit | Merged openstack/oslo.middleware: Updated from global requirements https://review.openstack.org/197265 | 23:13 |
| *** mgarza has quit IRC | 23:15 | |
| *** mgarza_ has joined #openstack-oslo | 23:15 | |
| *** mgarza_ has quit IRC | 23:16 | |
| *** ozamiatin has quit IRC | 23:16 | |
| *** zzzeek has quit IRC | 23:18 | |
| *** ozamiatin has joined #openstack-oslo | 23:18 | |
| lifeless | iNova | 23:18 |
| openstackgerrit | Merged openstack/oslo.utils: Updated from global requirements https://review.openstack.org/197267 | 23:21 |
| openstackgerrit | Merged openstack/automaton: Add `is_actionable_event` checking function https://review.openstack.org/191955 | 23:27 |
| *** yamahata has joined #openstack-oslo | 23:28 | |
| *** ozamiatin has quit IRC | 23:28 | |
| openstackgerrit | Merged openstack/oslo.log: Updated from global requirements https://review.openstack.org/197264 | 23:30 |
| openstackgerrit | Merged openstack/oslo.concurrency: Updated from global requirements https://review.openstack.org/194835 | 23:33 |
| openstackgerrit | Merged openstack/oslo.versionedobjects: Updated from global requirements https://review.openstack.org/196263 | 23:38 |
| *** dims has joined #openstack-oslo | 23:45 | |
| *** dims has quit IRC | 23:46 | |
| *** _amrith_ is now known as amrith | 23:47 | |
| *** manishg has quit IRC | 23:53 | |
| openstackgerrit | Joshua Harlow proposed openstack/oslo.service: Ensure the stop watch is always stopped https://review.openstack.org/197375 | 23:53 |
| *** sputnik1_ has joined #openstack-oslo | 23:54 | |
| *** zzzeek has joined #openstack-oslo | 23:55 | |
| *** sputnik13 has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!