| *** dimtruck is now known as zz_dimtruck | 00:02 | |
| dave-mccowan | rm_work i noticed with the previous patch that a lot of the quotas test cases failed. i might have jumped to the conclusion that it was caused by removing the project_id code. i'll dig into it more, depending if this patch works. :-) | 00:05 |
|---|---|---|
| rm_work | hmm | 00:05 |
| rm_work | i wonder if it was being re-set for a reason? | 00:05 |
| rm_work | but i don't see how it'd fail | 00:05 |
| rm_work | i will also look again after this | 00:05 |
| rm_work | once the gates run | 00:05 |
| dave-mccowan | rm_work :-( stuck at queued? | 00:23 |
| *** zz_dimtruck is now known as dimtruck | 00:23 | |
| *** openstackstatus has quit IRC | 00:24 | |
| openstackgerrit | Priti Desai proposed openstack/barbican: Authorized API Requests https://review.openstack.org/251593 | 00:24 |
| *** openstackstatus has joined #openstack-barbican | 00:24 | |
| *** ChanServ sets mode: +v openstackstatus | 00:24 | |
| *** pdesai has quit IRC | 00:26 | |
| *** Stanzi has joined #openstack-barbican | 00:57 | |
| *** stanzi_ has joined #openstack-barbican | 00:57 | |
| *** stanzi_ has quit IRC | 01:01 | |
| *** Stanzi has quit IRC | 01:01 | |
| *** dimtruck is now known as zz_dimtruck | 01:07 | |
| *** openstackgerrit has quit IRC | 01:22 | |
| *** openstackgerrit has joined #openstack-barbican | 01:22 | |
| *** jhfeng has joined #openstack-barbican | 01:24 | |
| *** jhfeng has quit IRC | 02:08 | |
| *** stanzi has joined #openstack-barbican | 02:31 | |
| *** stanzi has quit IRC | 02:36 | |
| *** Stanzi_ has joined #openstack-barbican | 02:47 | |
| *** stanzi has joined #openstack-barbican | 02:47 | |
| *** stanzi has quit IRC | 02:51 | |
| *** Stanzi_ has quit IRC | 02:52 | |
| *** Stanzi has joined #openstack-barbican | 03:03 | |
| *** Stanzi has quit IRC | 03:07 | |
| *** Stanzi has joined #openstack-barbican | 03:18 | |
| *** stanzi_ has joined #openstack-barbican | 03:18 | |
| *** yfujioka has quit IRC | 03:19 | |
| *** yuanying has quit IRC | 03:23 | |
| *** stanzi_ has quit IRC | 03:25 | |
| *** stanzi_ has joined #openstack-barbican | 03:25 | |
| *** dave-mccowan has quit IRC | 03:44 | |
| *** stanzi_ has quit IRC | 03:54 | |
| *** Stanzi has quit IRC | 03:54 | |
| *** stanzi has joined #openstack-barbican | 03:55 | |
| *** Stanzi_ has joined #openstack-barbican | 03:55 | |
| *** Stanzi_ has quit IRC | 04:06 | |
| *** stanzi has quit IRC | 04:06 | |
| *** Stanzi_ has joined #openstack-barbican | 04:06 | |
| *** stanzi has joined #openstack-barbican | 04:06 | |
| *** yuanying has joined #openstack-barbican | 04:07 | |
| *** kebray has joined #openstack-barbican | 04:07 | |
| *** kebray has quit IRC | 04:09 | |
| *** kebray has joined #openstack-barbican | 04:15 | |
| *** stanzi has quit IRC | 04:16 | |
| *** Stanzi_ has quit IRC | 04:16 | |
| *** kebray has quit IRC | 04:18 | |
| *** kebray has joined #openstack-barbican | 04:20 | |
| *** gyee has quit IRC | 04:52 | |
| *** kebray_ has joined #openstack-barbican | 05:03 | |
| *** kebray has quit IRC | 05:05 | |
| *** kebray_ has quit IRC | 05:11 | |
| *** kebray has joined #openstack-barbican | 05:13 | |
| openstackgerrit | Fernando Diaz proposed openstack/castellan: Add some documentation on enabling logging https://review.openstack.org/251649 | 06:03 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-barbicanclient: Updated from global requirements https://review.openstack.org/251661 | 06:09 |
| *** Kevin_Zheng has joined #openstack-barbican | 06:10 | |
| *** jaosorior has joined #openstack-barbican | 06:21 | |
| *** Nirupama has joined #openstack-barbican | 07:07 | |
| *** kebray has quit IRC | 07:15 | |
| *** jaosorior has quit IRC | 07:25 | |
| *** jaosorior has joined #openstack-barbican | 07:26 | |
| *** shohel has joined #openstack-barbican | 07:44 | |
| *** NazcaLines has joined #openstack-barbican | 07:48 | |
| *** NazcaLines has quit IRC | 08:02 | |
| *** everjeje has joined #openstack-barbican | 08:40 | |
| *** xek has joined #openstack-barbican | 09:07 | |
| *** yuanying has quit IRC | 09:30 | |
| *** dave-mccowan has joined #openstack-barbican | 11:02 | |
| *** chlong has quit IRC | 11:35 | |
| *** chlong has joined #openstack-barbican | 11:35 | |
| *** rhagarty has quit IRC | 11:37 | |
| *** rhagarty has joined #openstack-barbican | 11:40 | |
| *** Nirupama has quit IRC | 11:41 | |
| *** jaosorior has quit IRC | 12:42 | |
| *** jaosorior has joined #openstack-barbican | 12:43 | |
| *** jaosorior has quit IRC | 12:47 | |
| *** jaosorior has joined #openstack-barbican | 12:47 | |
| *** alee has quit IRC | 13:27 | |
| *** kfarr has joined #openstack-barbican | 13:42 | |
| *** kfarr has quit IRC | 13:43 | |
| *** kfarr has joined #openstack-barbican | 13:43 | |
| *** kfarr has quit IRC | 13:56 | |
| *** jmckind has joined #openstack-barbican | 14:23 | |
| *** alee has joined #openstack-barbican | 14:30 | |
| *** lisaclark1 has joined #openstack-barbican | 14:54 | |
| *** cbits has joined #openstack-barbican | 14:57 | |
| *** lisaclark1 has quit IRC | 15:01 | |
| *** spotz_zzz is now known as spotz | 15:06 | |
| *** kfarr has joined #openstack-barbican | 15:09 | |
| *** rellerreller has joined #openstack-barbican | 15:09 | |
| *** lisaclark1 has joined #openstack-barbican | 15:10 | |
| *** jhfeng has joined #openstack-barbican | 15:17 | |
| *** lisaclark1 has quit IRC | 15:22 | |
| *** zz_dimtruck is now known as dimtruck | 15:28 | |
| *** edtubill has joined #openstack-barbican | 15:32 | |
| openstackgerrit | Kaitlin Farr proposed openstack/barbican: Add reno for release notes management https://review.openstack.org/248918 | 15:33 |
| *** lisaclark1 has joined #openstack-barbican | 15:36 | |
| *** cbits has left #openstack-barbican | 15:40 | |
| *** silos has joined #openstack-barbican | 15:43 | |
| *** stanzi_ has joined #openstack-barbican | 15:45 | |
| *** Stanzi has joined #openstack-barbican | 15:45 | |
| *** darrenmoffat has quit IRC | 15:47 | |
| *** darrenmoffat has joined #openstack-barbican | 15:48 | |
| *** kfarr has quit IRC | 16:01 | |
| *** kebray has joined #openstack-barbican | 16:05 | |
| *** kebray has quit IRC | 16:16 | |
| *** diazjf has joined #openstack-barbican | 16:19 | |
| *** xaeth_afk is now known as xaeth | 16:22 | |
| *** fnaval has joined #openstack-barbican | 16:26 | |
| *** kebray has joined #openstack-barbican | 16:28 | |
| *** kebray has quit IRC | 16:28 | |
| *** kebray has joined #openstack-barbican | 16:28 | |
| *** jhfeng has quit IRC | 16:30 | |
| *** Stanzi has quit IRC | 16:37 | |
| *** stanzi_ has quit IRC | 16:37 | |
| openstackgerrit | Thomas Dinkjian proposed openstack/barbican: Added ACL tests https://review.openstack.org/239521 | 16:47 |
| *** woodster_ has joined #openstack-barbican | 16:51 | |
| *** jhfeng has joined #openstack-barbican | 16:54 | |
| *** shohel has quit IRC | 16:55 | |
| *** stanzi has joined #openstack-barbican | 16:58 | |
| *** Stanzi_ has joined #openstack-barbican | 16:58 | |
| *** kfarr has joined #openstack-barbican | 16:58 | |
| kfarr | elmiko diazjf I'm looking at https://review.openstack.org/#/c/251558/ | 16:59 |
| openstackgerrit | Merged openstack/castellan: Add gate hooks for Castellan functional tests https://review.openstack.org/229115 | 16:59 |
| kfarr | elmiko, will merging it and a new release fix the downstream problems? | 16:59 |
| *** kebray has quit IRC | 17:00 | |
| elmiko | kfarr: looking | 17:01 |
| diazjf | kfarr, elmiko, sounds good. lmk if you guys need anything | 17:02 |
| elmiko | kfarr: yea, i think if we get that change in a release it will alleviate the issues i'm seeing | 17:02 |
| *** stanzi has quit IRC | 17:02 | |
| *** Stanzi_ has quit IRC | 17:03 | |
| kfarr | diazjf what documentation are you thinking of adding? | 17:04 |
| *** kebray has joined #openstack-barbican | 17:05 | |
| diazjf | kfarr, just how to enable default logging should be fine. elmiko, if you can help me out on how to actually pass config options for logging, but I guess that should be a future patch. | 17:05 |
| kfarr | oh wait, I just noticed this one: https://review.openstack.org/#/c/251649/ | 17:06 |
| elmiko | diazjf: i did a first pass over the doc patch. i need to add some comments | 17:06 |
| diazjf | kfarr yeah thats the one :) elmiko thanks! | 17:06 |
| elmiko | imo, the enable_logging call is mainly for apps that have not configured oslo_log usage | 17:07 |
| elmiko | diazjf: i'll revisit the doc patch this afternoon | 17:08 |
| diazjf | elmiko, thanks. Currently is just a small how to on enabling logging if it wasn't setup before | 17:08 |
| elmiko | diazjf: yea, it might be nice to start a new section on logging, but i want to think about it a little more | 17:09 |
| diazjf | elmiko sounds good | 17:12 |
| kfarr | elmiko, diazjf, is there some minor documentation we could add before we cut a new release though? | 17:13 |
| kfarr | maybe reworking the whole logging section could come later? | 17:13 |
| elmiko | kfarr: yea, i think we should add something in the docs before a release | 17:14 |
| *** gyee has joined #openstack-barbican | 17:14 | |
| *** Stanzi has joined #openstack-barbican | 17:15 | |
| openstackgerrit | Merged openstack/castellan: Allow for default logging configuration to be user enabled https://review.openstack.org/251558 | 17:15 |
| *** stanzi_ has joined #openstack-barbican | 17:15 | |
| *** xaeth has quit IRC | 17:17 | |
| elmiko | diazjf: added my suggestions onto 251649 | 17:19 |
| *** Stanzi has quit IRC | 17:19 | |
| *** Stanzi has joined #openstack-barbican | 17:19 | |
| *** pdesai has joined #openstack-barbican | 17:20 | |
| *** nelsnelson has joined #openstack-barbican | 17:20 | |
| diazjf | elmiko, I'll take a look now | 17:20 |
| *** xaeth has joined #openstack-barbican | 17:27 | |
| *** stanzi_ has quit IRC | 17:27 | |
| *** stanzi_ has joined #openstack-barbican | 17:28 | |
| diazjf | kfarr, elmiko, updated https://review.openstack.org/#/c/251649/ be good for now and in the future we will add a how to on setting up oslo_log | 17:30 |
| *** lisaclark1 has quit IRC | 17:32 | |
| diazjf | elmiko would you suggest something like http://paste.openstack.org/show/480544/ in the future but describing what context format, and log levels can be set? | 17:42 |
| *** pdesai has quit IRC | 17:42 | |
| *** silos has left #openstack-barbican | 17:45 | |
| *** Guest42 has joined #openstack-barbican | 17:45 | |
| *** Guest42 has quit IRC | 17:49 | |
| *** rellerreller has quit IRC | 17:59 | |
| *** kebray has quit IRC | 18:02 | |
| *** jaosorior has quit IRC | 18:02 | |
| *** lisaclark1 has joined #openstack-barbican | 18:09 | |
| *** kebray has joined #openstack-barbican | 18:13 | |
| *** stanzi_ has quit IRC | 18:15 | |
| *** Stanzi has quit IRC | 18:15 | |
| *** kebray has quit IRC | 18:17 | |
| *** lisaclark1 has quit IRC | 18:20 | |
| elmiko | diazjf: i don't think we really need to get too deep into how to configure oslo_log, but mentioning the log levels is probably a good idea for a future patch | 18:21 |
| *** lisaclark1 has joined #openstack-barbican | 18:23 | |
| *** fnaval has quit IRC | 18:24 | |
| *** lisaclark1 has quit IRC | 18:24 | |
| *** diazjf has quit IRC | 18:24 | |
| *** edtubill has quit IRC | 18:27 | |
| *** lisaclark1 has joined #openstack-barbican | 18:28 | |
| *** jhfeng has quit IRC | 18:29 | |
| *** Stanzi has joined #openstack-barbican | 18:35 | |
| *** stanzi_ has joined #openstack-barbican | 18:35 | |
| *** jhfeng has joined #openstack-barbican | 18:40 | |
| *** stanzi_ has quit IRC | 18:40 | |
| *** Stanzi has quit IRC | 18:40 | |
| *** silos has joined #openstack-barbican | 18:41 | |
| *** stanzi has joined #openstack-barbican | 18:43 | |
| *** Stanzi_ has joined #openstack-barbican | 18:43 | |
| *** kebray has joined #openstack-barbican | 18:49 | |
| *** kebray has quit IRC | 18:52 | |
| *** kebray has joined #openstack-barbican | 18:54 | |
| *** lisaclark1 has quit IRC | 18:56 | |
| *** kebray has quit IRC | 18:58 | |
| *** gyee has quit IRC | 19:02 | |
| openstackgerrit | Steve Heyman proposed openstack/barbican: Added container consumer end to end flow testcase https://review.openstack.org/241830 | 19:10 |
| *** lisaclark1 has joined #openstack-barbican | 19:12 | |
| *** silos has quit IRC | 19:14 | |
| *** lisaclark1 has quit IRC | 19:15 | |
| kfarr | Anyone (especially cores!) available to review diazjf's documentation patch? It's a quick review :) https://review.openstack.org/#/c/251649/ | 19:15 |
| *** silos has joined #openstack-barbican | 19:17 | |
| *** edtubill has joined #openstack-barbican | 19:24 | |
| *** lisaclark1 has joined #openstack-barbican | 19:25 | |
| openstackgerrit | Merged openstack/castellan: Add some documentation on enabling logging https://review.openstack.org/251649 | 19:28 |
| *** lisaclark1 has quit IRC | 19:28 | |
| *** lisaclark1 has joined #openstack-barbican | 19:33 | |
| *** lisaclark1 has quit IRC | 19:34 | |
| *** lisaclark1 has joined #openstack-barbican | 19:34 | |
| *** kebray has joined #openstack-barbican | 19:38 | |
| *** stanzi has quit IRC | 19:39 | |
| *** Stanzi_ has quit IRC | 19:39 | |
| *** kebray has quit IRC | 19:39 | |
| *** stanzi_ has joined #openstack-barbican | 19:40 | |
| *** Stanzi has joined #openstack-barbican | 19:40 | |
| kfarr | Asking for another quick Castellan review: https://review.openstack.org/#/c/251199/ | 19:45 |
| spotz | kfarr there's a +1 | 19:46 |
| *** kebray has joined #openstack-barbican | 19:46 | |
| *** diazjf has joined #openstack-barbican | 19:48 | |
| *** kebray has quit IRC | 19:48 | |
| kfarr | Thanks spotz! | 19:48 |
| spotz | kfarr you're welcome:) | 19:48 |
| elmiko | kfarr: question about that review, does pbr provide the [wheel] entry? (i'm only seeing [bdist_wheel] in the setuptools docs) | 19:56 |
| kfarr | elmiko, I am not so sure, I had a similar question, but the other openstack clients seem to use [wheel] | 19:58 |
| kfarr | maybe someone on infra knows, I'll try to ask | 19:58 |
| elmiko | kfarr: ok, no worries, i just wasn't very familiar with the wheel stuff | 20:01 |
| arunkant_ | Can someone please review this change...https://review.openstack.org/#/c/248324/ ? | 20:05 |
| *** pdesai has joined #openstack-barbican | 20:06 | |
| *** diazjf has quit IRC | 20:12 | |
| *** jhfeng has quit IRC | 20:15 | |
| *** rellerreller has joined #openstack-barbican | 20:20 | |
| rellerreller | kfarr ping | 20:20 |
| kfarr | rellerreller pong | 20:21 |
| rellerreller | kfarr what is this https://review.openstack.org/#/c/251199/1/setup.cfg ? | 20:21 |
| spotz | arunkant_ there's a valid comment already in there. I can -1 or do you want to patch? | 20:21 |
| kfarr | rellerreller I think redrobot just recently set up the auto-publish jobs for castellan releases | 20:22 |
| kfarr | and when that happened, it started releasing a version that was only compatible with py2 | 20:23 |
| arunkant_ | spotz: Which one..typo one ? | 20:23 |
| spotz | yeah | 20:23 |
| kfarr | rellerreller and so I think it's causing problems when I tried to update the global requirements version | 20:23 |
| kfarr | rellerreller the fix is to add that the changes in that patch. the other openstack python-client libraries have that line | 20:24 |
| rellerreller | kfarr ok, I will take a look | 20:24 |
| arunkant_ | I was not planning to address that, being just typo. If you prefer, I can fix that. | 20:25 |
| arunkant_ | spotz ^^ | 20:27 |
| elmiko | kfarr, rellerreller, i *think* this is the relevant documentation about the wheel/universal stuff https://wheel.readthedocs.org/en/latest/index.html?highlight=universal#defining-the-python-version | 20:28 |
| spotz | Well if you don't I'll go back and do it:) just cleaner but yes it's minor | 20:28 |
| elmiko | kfarr, rellerreller, although from what i can tell, the wheel tool provides the [wheel] section support in setup.cfg | 20:29 |
| *** jmckind has quit IRC | 20:32 | |
| spotz | Ok I +1'd | 20:32 |
| openstackgerrit | Merged openstack/castellan: Mark castellan as being a universal wheel https://review.openstack.org/251199 | 20:34 |
| *** pdesai has quit IRC | 20:34 | |
| kfarr | elmiko, it was a success! There's also a new version of castellan out | 20:44 |
| kfarr | 0.3.1 | 20:44 |
| *** diazjf has joined #openstack-barbican | 20:45 | |
| *** jhfeng has joined #openstack-barbican | 20:46 | |
| elmiko | kfarr: awesome, thanks! | 20:46 |
| elmiko | kfarr: should we adjust https://review.openstack.org/#/c/248911/ to refect 0.3.1 as the global req version? | 20:47 |
| openstackgerrit | Fernando Diaz proposed openstack/barbican-specs: Blueprint defining user defined metadata for Barbican Secrets https://review.openstack.org/229995 | 20:48 |
| kfarr | elmiko, yes I am planning on doing that. Last time I updated global requirements on the same day as a release though, it failed jenkins, but passed the next day when I rechecked. | 20:49 |
| elmiko | kfarr: ack, thanks =) | 20:49 |
| kfarr | I'm guessing there's a cache of the libraries or something that gets updated at midnight, but I'm really not sure | 20:49 |
| elmiko | right, makes sense | 20:49 |
| *** lisaclark1 has quit IRC | 20:54 | |
| rm_work | dave-mccowan: yeah i am not sure what's going on with my patch, haven't had a chance to really debug in devstack yet. the last week has been a bit awkward, getting pulled in a bunch of directions | 20:58 |
| rm_work | hope to be able to look at it today though | 20:58 |
| rm_work | I kinda thought this would be fairly straightforward | 20:58 |
| *** rellerreller has quit IRC | 21:00 | |
| dave-mccowan | rm_work i found what was causing the 500 return. hopefully that's it. (see code review comment) | 21:19 |
| rm_work | Ah, kk yeah just saw the email come in | 21:19 |
| *** Stanzi has quit IRC | 21:19 | |
| *** stanzi_ has quit IRC | 21:19 | |
| rm_work | whoops | 21:19 |
| openstackgerrit | Steve Heyman proposed openstack/barbican: Added container consumer end to end flow testcase https://review.openstack.org/241830 | 21:20 |
| openstackgerrit | Adam Harwell proposed openstack/barbican: Remove consumer check for project_id to match containers https://review.openstack.org/251168 | 21:21 |
| rm_work | dave-mccowan: good catch, thanks for looking through that | 21:21 |
| rm_work | dave-mccowan: will see if that works, meanwhile i am still building my new devstack env | 21:21 |
| *** silos has quit IRC | 21:22 | |
| *** edtubill has quit IRC | 21:23 | |
| *** jmckind has joined #openstack-barbican | 21:25 | |
| openstackgerrit | Fernando Diaz proposed openstack/barbican-specs: Blueprint defining user defined metadata for Barbican Secrets https://review.openstack.org/229995 | 21:26 |
| *** silos has joined #openstack-barbican | 21:28 | |
| arunkant_ | kfarr: there? | 21:32 |
| *** lisaclark1 has joined #openstack-barbican | 21:35 | |
| kfarr | arunkant_, yes! | 21:35 |
| kfarr | arunkant_ I saw your comment | 21:36 |
| arunkant_ | kfarr: Okay..does it clarify the concern you have ? | 21:37 |
| *** lisaclark1 has quit IRC | 21:38 | |
| kfarr | arunkant_ yes mostly. Still thinking about it | 21:38 |
| arunkant_ | kfarr: We are using audit middleware for auditing and this used to work with barbican Kilo but it got broken when we switched to liberty. And issue identified is related to two different version of CONF handle | 21:40 |
| *** jkf has joined #openstack-barbican | 21:40 | |
| arunkant_ | kfarr: So now we are making those CONF handle to be same as it should not matter whether barbican is reading its configuration or other library component is trying to use 'service' configuration | 21:43 |
| rm_work | dave-mccowan: whoops missed updating that in the test code too T_T one more patchset | 21:43 |
| arunkant_ | kfarr: Does it make sense? | 21:43 |
| openstackgerrit | Adam Harwell proposed openstack/barbican: Remove consumer check for project_id to match containers https://review.openstack.org/251168 | 21:44 |
| kfarr | arunkant_, yes I think I understand the bug you are trying to fix, but I am just concerned about reassigning a global variable | 21:47 |
| arunkant_ | kfarr: barbican.common.config.CONF object is always updated whenever its referenced..so we need to keep cfg.CONF in sync with that. This was a change made in liberty. | 21:49 |
| arunkant_ | kfarr: this was the change in liberty (https://review.openstack.org/#/c/187297/3/barbican/common/config.py,cm_ where it was refactored from using CONF = cfg.CONF | 21:52 |
| kfarr | arunkant_ but what about, in new_config(), if you told it to update and re-register cfg.CONF instead of creating a new ConfigOpts object | 21:57 |
| arunkant_ | kfarr: In this review, the change seems to be done around that..https://review.openstack.org/#/c/187297/3/barbican/common/config.py,cm (line 28) | 21:58 |
| elmiko | arunkant_: i was kind of curious why barbican doesn't just start by using cfg.CONF as base? (instead of the other way around) | 21:59 |
| *** kebray has joined #openstack-barbican | 21:59 | |
| arunkant_ | kfarr: From that review, it seems they do want to create new ConfigOpts from scratch. | 21:59 |
| kfarr | arunkant_ from that review, from the way I understand it, it seemed like the goal was to just register everything at the same time to avoid a race condition, and creating a new ConfigOpts was just a side effect | 22:01 |
| kfarr | and it seemed like the race condition would still be fixed if it just registered everything to cfg.CONF instead | 22:02 |
| kfarr | unless I am wrong? | 22:02 |
| arunkant_ | Yes..it will work. But the way I see current code creates a new ConfigOpts whenever new_config() is invoked. I am not sure what race condition it was as server is just starting up. | 22:05 |
| *** pdesai has joined #openstack-barbican | 22:05 | |
| elmiko | i was trying to understand that part, is it required to create a new ConfgOpts object for each usage? | 22:06 |
| elmiko | i've usually seen the design pattern where the cfg.CONF object is updated and then used throughout the application. but i thought maybe barbican had a specific usage for the new_config pattern. | 22:07 |
| arunkant_ | elmiko: Earlier code used to have similar logic and it was updating existing object. But that logic was refactored as https://review.openstack.org/#/c/187297/3/barbican/common/config.py,cm | 22:07 |
| elmiko | arunkant_: was there a reason to not use the global CONF object? | 22:08 |
| elmiko | i mean, it seems like barbican.common.config.CONF is duplicating that functionality | 22:09 |
| arunkant_ | elmiko: I am not fully aware about that change. Its creating a different CONF object instance which is different from cfg.CONF handle. Ideally they should be same. | 22:10 |
| arunkant_ | elmiko: Also agree, we don't need to recreate ConfigOpts again and again as some module do invoke new_config() method | 22:11 |
| elmiko | arunkant_: right, seems like it would be easier to just update cfg.CONF and let all the modules use that. | 22:12 |
| elmiko | i thought maybe there was a historical reason for barbican to do it this way | 22:12 |
| arunkant_ | Not historical..it was changed in liberty | 22:13 |
| elmiko | ah ok | 22:15 |
| *** diazjf has quit IRC | 22:16 | |
| arunkant_ | kfarr: What do you recommend ? I can make that change in new_config() method, cannot tell if it will bring race condition or not? | 22:16 |
| kfarr | arunkant_ I'm still not convinced that patch needed to create a new ConfigOpts every time, I think your bug could be solved if you swapped out line 167 | 22:20 |
| kfarr | to say 'conf = cfg.CONF' instead of creating a new ConfigOpts | 22:21 |
| kfarr | I think it still solves the race condition issue | 22:21 |
| kfarr | The only potential reasoning for creating a new ConfigOpts is maybe you get a DuplicateOptError if you try to register the same thing twice | 22:21 |
| *** gyee has joined #openstack-barbican | 22:22 | |
| arunkant_ | kfarr: I have not made any change around creation of ConfigOpts() and new instances of ConfigOpts is existing functionality | 22:22 |
| elmiko | agreed with kfarr, moving towards using the cfg.CONF object is a safer option | 22:23 |
| elmiko | but if you change `conf=cfg.CONF`, then there wouldn't be much reason to re-register all the options | 22:24 |
| *** pdesai has quit IRC | 22:24 | |
| *** lisaclark1 has joined #openstack-barbican | 22:24 | |
| *** silos has left #openstack-barbican | 22:25 | |
| arunkant_ | elmiko: Agreed, there are number of place where new_config() method is invoked . https://github.com/openstack/barbican/blob/master/barbican/plugin/store_crypto.py#L24 | 22:27 |
| elmiko | arunkant_: yea, it's all over the place. but all those references could be replaced by cfg.CONF if it was used. | 22:27 |
| *** kebray has quit IRC | 22:27 | |
| arunkant_ | elmiko, kfarr: If you look into the related change..it was removed from cfg.CONF to use this new_config object. https://review.openstack.org/#/c/187297/3/barbican/plugin/store_crypto.py,cm | 22:29 |
| elmiko | interesting... | 22:29 |
| *** lisaclark2 has joined #openstack-barbican | 22:30 | |
| elmiko | this is where my engineering history for barbican is weak, i'm not sure why that change was made. | 22:30 |
| elmiko | the main thing that looked weird to me on the latest review was writing back to cfg.CONF, that seems like an anti-pattern | 22:31 |
| *** kebray has joined #openstack-barbican | 22:31 | |
| *** lisaclark1 has quit IRC | 22:32 | |
| arunkant_ | elmiko: yes..the reason, in current code cfg.CONF and barbican.common.config.CONF handles are two totally different instances. Now somehow need to keep them in sync | 22:33 |
| *** lisaclark2 has quit IRC | 22:33 | |
| *** lisaclark1 has joined #openstack-barbican | 22:33 | |
| elmiko | arunkant_: ack, thank you for the explanation | 22:33 |
| elmiko | my question is, why not make barbican.common.config.CONF=cfg.CONF? | 22:34 |
| *** stanzi has joined #openstack-barbican | 22:35 | |
| *** Stanzi_ has joined #openstack-barbican | 22:35 | |
| *** everjeje has quit IRC | 22:35 | |
| elmiko | (kinda like what kfarr just proposed) | 22:35 |
| arunkant_ | yes..that can be done provided all barbican specific options are registered to cfg.CONF but that's not the case. | 22:36 |
| elmiko | ah, ok | 22:37 |
| elmiko | is there something preventing them from being registered to cfg.CONF, for example are different ConfigOpts needed by different parts of the code? | 22:38 |
| arunkant_ | elmiko: I don't think so..may be the concern was that different modules load config opts at different time as they were defined in related module. | 22:39 |
| arunkant_ | It is no longer the case as now all of the config opts are defined at one place. | 22:40 |
| arunkant_ | So will try to do what kfarr suggested and will see then what are the comments then | 22:41 |
| arunkant_ | kfarr , elmiko: thanks for your inputs. | 22:41 |
| elmiko | arunkant_: cool, i'll make sure to take a look at the new one | 22:42 |
| kfarr | arunkant_ thanks for being patient with your explanations! | 22:42 |
| elmiko | +1 | 22:42 |
| *** pdesai has joined #openstack-barbican | 22:43 | |
| *** jhfeng has quit IRC | 22:43 | |
| *** pdesai has quit IRC | 22:51 | |
| *** kebray has quit IRC | 22:53 | |
| *** lisaclark1 has quit IRC | 22:59 | |
| *** Stanzi_ has quit IRC | 22:59 | |
| *** stanzi has quit IRC | 22:59 | |
| *** jmckind has quit IRC | 23:01 | |
| *** lisaclark1 has joined #openstack-barbican | 23:01 | |
| *** xaeth is now known as xaeth_afk | 23:02 | |
| *** dimtruck is now known as zz_dimtruck | 23:10 | |
| *** kebray has joined #openstack-barbican | 23:16 | |
| *** stanzi has joined #openstack-barbican | 23:20 | |
| *** Stanzi_ has joined #openstack-barbican | 23:20 | |
| *** pdesai has joined #openstack-barbican | 23:22 | |
| *** fnaval has joined #openstack-barbican | 23:23 | |
| *** kebray has quit IRC | 23:24 | |
| *** stanzi has quit IRC | 23:24 | |
| *** fnaval has quit IRC | 23:24 | |
| *** Stanzi_ has quit IRC | 23:25 | |
| *** david-lyle has quit IRC | 23:25 | |
| *** fnaval has joined #openstack-barbican | 23:25 | |
| *** alee has quit IRC | 23:30 | |
| *** kfarr has quit IRC | 23:31 | |
| *** spotz is now known as spotz_zzz | 23:34 | |
| *** kebray has joined #openstack-barbican | 23:34 | |
| *** kebray has quit IRC | 23:35 | |
| *** kebray has joined #openstack-barbican | 23:36 | |
| *** kebray has quit IRC | 23:36 | |
| *** Stanzi has joined #openstack-barbican | 23:41 | |
| *** Stanzi has quit IRC | 23:45 | |
| *** jorgem is now known as jorgem[away] | 23:53 | |
| *** lisaclark1 has quit IRC | 23:53 | |
| *** lisaclark1 has joined #openstack-barbican | 23:56 | |
| *** alee has joined #openstack-barbican | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!