| *** packet has quit IRC | 00:00 | |
| *** bradjones has quit IRC | 00:01 | |
| *** bradjones has joined #openstack-horizon | 00:03 | |
| *** sbfox has joined #openstack-horizon | 00:03 | |
| *** jgravel has joined #openstack-horizon | 00:13 | |
| *** DaveJ__ has quit IRC | 00:28 | |
| *** yingjun has joined #openstack-horizon | 00:32 | |
| *** nikunj2512_ has quit IRC | 00:32 | |
| *** nikunj2512_ has joined #openstack-horizon | 00:34 | |
| openstackgerrit | Merged openstack/horizon: Integration tests - 3 new form fields https://review.openstack.org/133095 | 00:51 |
|---|---|---|
| *** ongk has quit IRC | 00:52 | |
| *** ongk has joined #openstack-horizon | 00:52 | |
| *** ongk has quit IRC | 00:52 | |
| *** radez is now known as radez_g0n3 | 00:54 | |
| *** hinnant has quit IRC | 00:55 | |
| *** hurgleburgler has quit IRC | 00:55 | |
| *** sbfox has quit IRC | 00:57 | |
| *** lhcheng_ has joined #openstack-horizon | 00:59 | |
| *** lhcheng has quit IRC | 01:02 | |
| *** ZZelle_ has quit IRC | 01:08 | |
| *** Sukhdev has quit IRC | 01:14 | |
| *** rm_work is now known as rm_work|away | 01:17 | |
| *** liuyulong has joined #openstack-horizon | 01:19 | |
| *** jgravel_ has joined #openstack-horizon | 01:20 | |
| *** jgravel has quit IRC | 01:23 | |
| *** aliceR has quit IRC | 01:30 | |
| *** bradjones has quit IRC | 01:32 | |
| *** kumartin has quit IRC | 01:32 | |
| *** gyee has quit IRC | 01:37 | |
| *** dboik has joined #openstack-horizon | 01:37 | |
| *** zhenguo has joined #openstack-horizon | 01:39 | |
| *** wuhg has joined #openstack-horizon | 01:47 | |
| *** dboik_ has joined #openstack-horizon | 01:53 | |
| *** mitz has joined #openstack-horizon | 01:54 | |
| *** liuyulong has quit IRC | 01:56 | |
| *** dboik has quit IRC | 01:56 | |
| *** sbfox has joined #openstack-horizon | 01:59 | |
| *** lhcheng_ has quit IRC | 02:00 | |
| openstackgerrit | Thai Tran proposed openstack/horizon: WIP Consolidating conf and scripts https://review.openstack.org/141457 | 02:02 |
| openstackgerrit | Thai Tran proposed openstack/horizon: WIP Consolidating conf and scripts https://review.openstack.org/141457 | 02:03 |
| *** mitz_ has joined #openstack-horizon | 02:04 | |
| *** nlahouti has joined #openstack-horizon | 02:09 | |
| *** david-lyle is now known as david-lyle_afk | 02:10 | |
| *** jasondotstar has joined #openstack-horizon | 02:11 | |
| *** Longgeek has quit IRC | 02:11 | |
| *** sbfox has quit IRC | 02:12 | |
| *** nlahouti has quit IRC | 02:13 | |
| *** nlahouti has joined #openstack-horizon | 02:13 | |
| *** hurgleburgler has joined #openstack-horizon | 02:14 | |
| *** jgravel_ has quit IRC | 02:18 | |
| *** jgravel has joined #openstack-horizon | 02:19 | |
| *** amotoki has joined #openstack-horizon | 02:38 | |
| *** yingjun has quit IRC | 02:45 | |
| *** nikunj2512_ has quit IRC | 02:49 | |
| *** liuyulong has joined #openstack-horizon | 02:59 | |
| *** lblanchard has joined #openstack-horizon | 03:01 | |
| *** Poornima has joined #openstack-horizon | 03:01 | |
| *** lhcheng has joined #openstack-horizon | 03:01 | |
| *** lblanchard has quit IRC | 03:03 | |
| *** nlahouti has quit IRC | 03:03 | |
| *** liuyulong has quit IRC | 03:04 | |
| *** lhcheng has quit IRC | 03:06 | |
| *** lblanchard has joined #openstack-horizon | 03:06 | |
| *** lblanchard has quit IRC | 03:10 | |
| *** dboik_ has quit IRC | 03:13 | |
| *** lblanchard has joined #openstack-horizon | 03:13 | |
| *** dboik has joined #openstack-horizon | 03:16 | |
| *** cptsalty has quit IRC | 03:27 | |
| *** julim has joined #openstack-horizon | 03:29 | |
| *** nlahouti has joined #openstack-horizon | 03:41 | |
| *** masco has joined #openstack-horizon | 03:45 | |
| *** teju has joined #openstack-horizon | 03:46 | |
| *** dboik has quit IRC | 03:56 | |
| *** sbfox has joined #openstack-horizon | 03:59 | |
| *** hurgleburgler has quit IRC | 04:03 | |
| *** sbfox has quit IRC | 04:12 | |
| openstackgerrit | Kanchan Gupta proposed openstack/horizon: Proper error message is shown. https://review.openstack.org/140614 | 04:14 |
| *** echevemaster has quit IRC | 04:16 | |
| *** echevemaster has joined #openstack-horizon | 04:18 | |
| *** harlowja is now known as harlowja_away | 04:19 | |
| openstackgerrit | Thai Tran proposed openstack/horizon: WIP Consolidating conf and scripts https://review.openstack.org/141457 | 04:19 |
| *** lblanchard has quit IRC | 04:19 | |
| openstackgerrit | Merged openstack/horizon: Test's random name gen now uses uuid4 https://review.openstack.org/141906 | 04:20 |
| *** sayan has joined #openstack-horizon | 04:21 | |
| openstackgerrit | Thai Tran proposed openstack/horizon: WIP Consolidating conf and scripts https://review.openstack.org/141457 | 04:23 |
| *** lblanchard has joined #openstack-horizon | 04:31 | |
| *** jgravel_ has joined #openstack-horizon | 04:36 | |
| *** hurgleburgler has joined #openstack-horizon | 04:36 | |
| *** jgravel has quit IRC | 04:40 | |
| *** yingjun has joined #openstack-horizon | 04:40 | |
| *** lblanchard has quit IRC | 04:45 | |
| *** lhcheng has joined #openstack-horizon | 04:50 | |
| *** hurgleburgler has quit IRC | 04:51 | |
| *** esp has left #openstack-horizon | 04:52 | |
| *** lhcheng has quit IRC | 04:55 | |
| *** jgravel__ has joined #openstack-horizon | 04:58 | |
| *** jgravel_ has quit IRC | 05:01 | |
| *** Poornima has quit IRC | 05:01 | |
| *** ongk has joined #openstack-horizon | 05:01 | |
| *** nikunj2512 has joined #openstack-horizon | 05:09 | |
| *** flwang has quit IRC | 05:11 | |
| *** tzumainn has quit IRC | 05:14 | |
| *** jgravel_ has joined #openstack-horizon | 05:14 | |
| *** lhcheng has joined #openstack-horizon | 05:16 | |
| openstackgerrit | Nikunj Aggarwal proposed openstack/horizon: Fixes resource name problem in "Resources Usage" tab https://review.openstack.org/125307 | 05:17 |
| *** jgravel__ has quit IRC | 05:18 | |
| openstackgerrit | Nikunj Aggarwal proposed openstack/horizon: Adds the security group id besides the name in Manage Rules https://review.openstack.org/130956 | 05:19 |
| *** jgravel__ has joined #openstack-horizon | 05:21 | |
| *** nlahouti has quit IRC | 05:24 | |
| *** jgravel_ has quit IRC | 05:25 | |
| *** Veera_ has joined #openstack-horizon | 05:26 | |
| *** Poornima has joined #openstack-horizon | 05:26 | |
| *** nlahouti has joined #openstack-horizon | 05:26 | |
| *** rushiagr_away is now known as rushiagr | 05:39 | |
| openstackgerrit | Nikunj Aggarwal proposed openstack/horizon: Adds support to accept IPv6 in VIP Address for LB Pool https://review.openstack.org/142670 | 05:48 |
| *** jasondotstar has quit IRC | 05:48 | |
| *** jasondotstar has joined #openstack-horizon | 05:49 | |
| openstackgerrit | Arvinder Singh proposed openstack/horizon: Expose Image owner info in the image table https://review.openstack.org/142671 | 05:54 |
| openstackgerrit | Arvinder Singh proposed openstack/horizon: Expose Image owner info in the image table https://review.openstack.org/142672 | 05:54 |
| *** ongk has quit IRC | 05:58 | |
| openstackgerrit | Swati Shukla proposed openstack/horizon: Download Reports of Horizon Detailed Views, in dfferent formats https://review.openstack.org/142676 | 06:01 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/horizon: Imported Translations from Transifex https://review.openstack.org/142677 | 06:04 |
| *** k4n0 has joined #openstack-horizon | 06:10 | |
| *** echevemaster has quit IRC | 06:13 | |
| *** yingjun has quit IRC | 06:19 | |
| openstackgerrit | Lin Hua Cheng proposed openstack/horizon: Fix Firewalls panel to override the right method https://review.openstack.org/142687 | 06:33 |
| *** cody-somerville has quit IRC | 06:34 | |
| *** jgravel has joined #openstack-horizon | 06:37 | |
| *** jgravel__ has quit IRC | 06:37 | |
| *** jtomasek has joined #openstack-horizon | 06:38 | |
| *** jgravel_ has joined #openstack-horizon | 06:50 | |
| *** jgravel has quit IRC | 06:52 | |
| *** mrunge has joined #openstack-horizon | 06:58 | |
| *** rushiagr is now known as rushiagr_away | 07:02 | |
| openstackgerrit | Arvinder Singh proposed openstack/horizon: Use bdmv2 format when instance is boot from volume https://review.openstack.org/135253 | 07:05 |
| *** sayan has quit IRC | 07:07 | |
| *** yingjun has joined #openstack-horizon | 07:15 | |
| *** rdopiera has joined #openstack-horizon | 07:15 | |
| *** rdopiera has quit IRC | 07:16 | |
| *** rdopiera has joined #openstack-horizon | 07:16 | |
| *** jcoufal has joined #openstack-horizon | 07:18 | |
| *** jgravel has joined #openstack-horizon | 07:18 | |
| *** jgravel_ has quit IRC | 07:21 | |
| openstackgerrit | Arvinder Singh proposed openstack/horizon: Expose Image owner info in the image table https://review.openstack.org/142671 | 07:23 |
| openstackgerrit | Nikunj Aggarwal proposed openstack/horizon: Allows to specify ip during port create https://review.openstack.org/142695 | 07:26 |
| rdopiera | hello everyone! | 07:29 |
| nikunj2512 | hello | 07:30 |
| *** jgravel_ has joined #openstack-horizon | 07:32 | |
| *** my_openstack_use has joined #openstack-horizon | 07:32 | |
| my_openstack_use | hello | 07:32 |
| openstackgerrit | Vincent Untz proposed openstack/horizon: Fix display of ports when associating floating IP https://review.openstack.org/142605 | 07:32 |
| mrunge | radomir, good morning! | 07:33 |
| rdopiera | mrunge: sorry, I wasn't clear about my pto, I'm back already, until Christmas | 07:34 |
| *** jgravel has quit IRC | 07:34 | |
| mrunge | rdopiera, no worries! I hope you could enjoy your time | 07:34 |
| my_openstack_use | Horizon documentation states here that you can add new panels using pluggable setting files and if panel group is specified as 'default', the panel will show up on its own: http://docs.openstack.org/developer/horizon/topics/settings.html#pluggable-settings-for-panels | 07:35 |
| *** alexpilotti has joined #openstack-horizon | 07:35 | |
| my_openstack_use | We used to have Icehouse and used this feature | 07:35 |
| mrunge | rdopiera, sad, you weren't available for a good wrestle match with r1chardj0n3s_afk yesterday at the meeting | 07:35 |
| rdopiera | mrunge: yeah, I just read the logs | 07:36 |
| my_openstack_use | now we have upgraded to juno (version 2014.2) and this geature doesn't seem to work anymore | 07:36 |
| my_openstack_use | all panels added through pluggable settings now show up in 'Others', even if the panel group specified already exists | 07:38 |
| my_openstack_use | is this a bug or has this feature been removed on purpose? | 07:39 |
| rdopiera | my_openstack_use: sounds like a bug | 07:39 |
| mrunge | my_openstack_use, if I remember correctly, this is currently a known issue and a fix is under way | 07:39 |
| mrunge | that should be this one here: https://review.openstack.org/#/c/128133/ | 07:40 |
| my_openstack_use | mrunge: Ok, thanks for quick notice, I'll be looking forward to the fix | 07:41 |
| *** lhcheng has quit IRC | 07:41 | |
| *** nlahouti has quit IRC | 07:42 | |
| *** my_openstack_use has left #openstack-horizon | 07:43 | |
| *** nlahouti has joined #openstack-horizon | 07:43 | |
| *** jgravel has joined #openstack-horizon | 07:45 | |
| *** jgravel_ has quit IRC | 07:47 | |
| *** jgravel_ has joined #openstack-horizon | 07:49 | |
| *** jgravel has quit IRC | 07:52 | |
| *** e0ne has joined #openstack-horizon | 07:53 | |
| *** e0ne has quit IRC | 07:54 | |
| *** pcaruana has joined #openstack-horizon | 07:56 | |
| *** jtomasek has quit IRC | 07:57 | |
| *** jgravel__ has joined #openstack-horizon | 07:58 | |
| *** nlahouti has quit IRC | 08:00 | |
| *** jgravel_ has quit IRC | 08:01 | |
| openstackgerrit | Vincent Untz proposed openstack/horizon: Fix display of ports when associating floating IP https://review.openstack.org/142605 | 08:01 |
| *** sayan has joined #openstack-horizon | 08:10 | |
| *** Shohei has joined #openstack-horizon | 08:13 | |
| *** amotoki has quit IRC | 08:15 | |
| *** jgravel_ has joined #openstack-horizon | 08:15 | |
| *** jgravel__ has quit IRC | 08:17 | |
| *** jgravel__ has joined #openstack-horizon | 08:18 | |
| *** jgravel_ has quit IRC | 08:21 | |
| *** zz_ttrifonov is now known as ttrifonov | 08:28 | |
| *** romainh has joined #openstack-horizon | 08:33 | |
| nikunj2512 | mrunge: Hi, need your help with the testcase for this patch - https://review.openstack.org/#/c/142695/1 | 08:35 |
| nikunj2512 | mrunge: I am not able to figure out what is the issue | 08:35 |
| *** JeanBriceCombebi has joined #openstack-horizon | 08:40 | |
| *** exploreshaifali has joined #openstack-horizon | 08:44 | |
| *** jgravel__ has quit IRC | 08:44 | |
| *** JeanBriceCombebi has quit IRC | 08:44 | |
| *** jgravel has joined #openstack-horizon | 08:45 | |
| *** Ala has joined #openstack-horizon | 08:46 | |
| *** sayan_ has joined #openstack-horizon | 08:47 | |
| *** jgravel_ has joined #openstack-horizon | 08:53 | |
| *** sayan_ has quit IRC | 08:54 | |
| *** alexpilotti has quit IRC | 08:56 | |
| *** jgravel has quit IRC | 08:57 | |
| *** akrivoka has joined #openstack-horizon | 08:57 | |
| *** JeanBriceCombebi has joined #openstack-horizon | 08:58 | |
| *** jgravel__ has joined #openstack-horizon | 09:02 | |
| *** jgravel_ has quit IRC | 09:05 | |
| *** jpich has joined #openstack-horizon | 09:07 | |
| *** jgravel_ has joined #openstack-horizon | 09:08 | |
| *** athomas has joined #openstack-horizon | 09:10 | |
| *** jgravel__ has quit IRC | 09:11 | |
| *** MaxV has joined #openstack-horizon | 09:17 | |
| *** ygbo has joined #openstack-horizon | 09:17 | |
| *** jgravel__ has joined #openstack-horizon | 09:17 | |
| mrunge | nikunj2512, please add a test to verify, the form accepts ipv6 addresses | 09:19 |
| openstackgerrit | Nikita Konovalov proposed openstack/horizon: [Data_processing] fix job_executions table https://review.openstack.org/125927 | 09:20 |
| *** jgravel has joined #openstack-horizon | 09:20 | |
| *** jgravel_ has quit IRC | 09:21 | |
| *** gerchardon has quit IRC | 09:22 | |
| nikunj2512 | mrunge: i am not taking about the patch which you reviewed. please open the URI in browser. It is a different review | 09:22 |
| *** jgravel__ has quit IRC | 09:22 | |
| mrunge | nikunj2512, ok, I see 4 unaddressed comments | 09:23 |
| mrunge | nikunj2512, that is failing, because of a test failed. did you try to run ./run_tests.sh at all? | 09:24 |
| nikunj2512 | mrunge: that i can fix.. please see the zuul log... http://logs.openstack.org/95/142695/1/check/gate-horizon-python27/5d953b7/ | 09:24 |
| nikunj2512 | mrunge: it gave -1 because test failed.. and from last night i am trying to fix them and still not able to :-( | 09:25 |
| nikunj2512 | mrunge: Yes.. it is failing in my dev system.. i uploaded the patch because i thought i will ask for help once patch was up because zuul will tell what is failed | 09:26 |
| mrunge | nikunj2512, you mean something like this: UnexpectedMethodCallError: Unexpected method call network_get.__call__(<class 'django.http.HttpRequest'>, '82288d84-e0a5-42ac-95be-e6af08727e42') -> None | 09:26 |
| mrunge | I see | 09:26 |
| nikunj2512 | mrunge: Yes.. it is failing in my dev system.. i uploaded the patch because i thought i will ask for help once patch was up because zuul will tell what is failed | 09:26 |
| nikunj2512 | mrunge: Yes.. | 09:26 |
| nikunj2512 | mrunge: i am not figureout why | 09:26 |
| mrunge | nikunj2512, I'm in the middle of something else. maybe I'll have time to come back to that later; sorry | 09:28 |
| nikunj2512 | mrunge: ok.. can you ping me when you have time | 09:28 |
| nikunj2512 | ? | 09:28 |
| *** yingjun has quit IRC | 09:29 | |
| mrunge | nikunj2512, I can try that. but this shouldn't stop you from trying to solve this yourself | 09:29 |
| openstackgerrit | Swati Shukla proposed openstack/horizon: LoadBalancers' Add Member form has missing asterisk for 'Members' list https://review.openstack.org/139546 | 09:29 |
| nikunj2512 | mrunge: ok... i am still trying to solve it | 09:30 |
| *** zhenguo has quit IRC | 09:30 | |
| *** jgravel_ has joined #openstack-horizon | 09:31 | |
| *** Poornima has quit IRC | 09:32 | |
| *** jamielennox is now known as jamielennox|away | 09:32 | |
| *** JeanBriceCombebi has quit IRC | 09:33 | |
| *** jgravel has quit IRC | 09:34 | |
| *** JeanBriceCombebi has joined #openstack-horizon | 09:36 | |
| *** Poornima has joined #openstack-horizon | 09:36 | |
| *** e0ne has joined #openstack-horizon | 09:39 | |
| openstackgerrit | Swati Shukla proposed openstack/horizon: Add Member form's "Member Address" field - Required asterisk missing https://review.openstack.org/137331 | 09:39 |
| openstackgerrit | Merged openstack/horizon: Check input for Log length https://review.openstack.org/102712 | 09:52 |
| *** sayan has quit IRC | 10:03 | |
| *** e0ne is now known as e0ne_ | 10:05 | |
| *** rushiagr_away is now known as rushiagr | 10:06 | |
| *** jgravel__ has joined #openstack-horizon | 10:08 | |
| *** e0ne_ is now known as e0ne | 10:09 | |
| *** jgravel_ has quit IRC | 10:12 | |
| *** ZZelle has quit IRC | 10:14 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/horizon: Updated from global requirements https://review.openstack.org/142713 | 10:14 |
| *** romainh1 has joined #openstack-horizon | 10:19 | |
| *** romainh has quit IRC | 10:19 | |
| *** cedrics has joined #openstack-horizon | 10:20 | |
| *** jgravel_ has joined #openstack-horizon | 10:20 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tuskar-ui: Updated from global requirements https://review.openstack.org/142716 | 10:21 |
| *** jtomasek has joined #openstack-horizon | 10:24 | |
| *** jgravel__ has quit IRC | 10:24 | |
| *** jgravel has joined #openstack-horizon | 10:24 | |
| *** jgravel_ has quit IRC | 10:26 | |
| *** alexpilotti has joined #openstack-horizon | 10:35 | |
| *** athomas has quit IRC | 10:43 | |
| openstackgerrit | Merged openstack/horizon: Base Glance Metadata Definitions Admin UI https://review.openstack.org/104063 | 10:48 |
| openstackgerrit | Nikunj Aggarwal proposed openstack/horizon: Allows to specify ip during port create https://review.openstack.org/142695 | 10:49 |
| *** athomas has joined #openstack-horizon | 10:51 | |
| *** JeanBriceCombebi has quit IRC | 10:51 | |
| openstackgerrit | Merged openstack/tuskar-ui: Updated from global requirements https://review.openstack.org/142716 | 10:56 |
| *** jgravel has quit IRC | 11:09 | |
| *** nikunj2512 has quit IRC | 11:10 | |
| *** jgravel has joined #openstack-horizon | 11:15 | |
| *** cedrics has quit IRC | 11:19 | |
| *** aix has quit IRC | 11:23 | |
| *** masco has quit IRC | 11:30 | |
| *** Poornima has quit IRC | 11:37 | |
| *** sayan has joined #openstack-horizon | 11:39 | |
| *** exploreshaifali has quit IRC | 11:47 | |
| *** jgravel_ has joined #openstack-horizon | 11:48 | |
| *** jgravel has quit IRC | 11:48 | |
| *** Poornima has joined #openstack-horizon | 11:51 | |
| *** athomas has quit IRC | 11:53 | |
| *** nikunj2512_ has joined #openstack-horizon | 11:54 | |
| openstackgerrit | Matthias Runge proposed openstack/django_openstack_auth: add last_activity to session https://review.openstack.org/142737 | 11:56 |
| *** dsneddon has quit IRC | 11:59 | |
| *** athomas has joined #openstack-horizon | 11:59 | |
| *** Poornima has quit IRC | 11:59 | |
| *** dsneddon has joined #openstack-horizon | 12:00 | |
| *** Longgeek has joined #openstack-horizon | 12:06 | |
| *** cedrics has joined #openstack-horizon | 12:09 | |
| *** jgravel__ has joined #openstack-horizon | 12:11 | |
| *** e0ne is now known as e0ne_ | 12:13 | |
| *** jgravel_ has quit IRC | 12:15 | |
| *** ZZelle has joined #openstack-horizon | 12:15 | |
| *** vkmc has joined #openstack-horizon | 12:22 | |
| *** e0ne_ is now known as e0ne | 12:22 | |
| *** masco has joined #openstack-horizon | 12:25 | |
| *** alexpilotti has quit IRC | 12:26 | |
| openstackgerrit | Merged openstack/horizon: Imported Translations from Transifex https://review.openstack.org/142677 | 12:28 |
| *** aix has joined #openstack-horizon | 12:32 | |
| *** Longgeek has quit IRC | 12:33 | |
| *** Longgeek has joined #openstack-horizon | 12:33 | |
| *** teju has quit IRC | 12:36 | |
| *** JeanBriceCombebi has joined #openstack-horizon | 12:37 | |
| *** JeanBriceCombebi has quit IRC | 12:42 | |
| openstackgerrit | Vincent Untz proposed openstack/horizon: Fix display of ports when associating floating IP https://review.openstack.org/142605 | 12:43 |
| zigo | r1chardj0n3s_afk: FYI, now I know bower, and I even less want to see any occurrence of it on my system. | 12:43 |
| zigo | And I think it's a *very* bad idea to use such a thing. | 12:43 |
| zigo | Sure, the XStatic stuff is more work, but at least we have a little bit control of things that we use. | 12:43 |
| zigo | What I'm very nervous is also to pull from any random git repo, this seems really not appropriate at all, and prone to all sorts of issues (security or otherwise). | 12:44 |
| zigo | Last, bower creates a directory structure which will be really annoying to deal with. | 12:45 |
| zigo | I know about bower since I had to work with it for Fuel (and its web interface). | 12:45 |
| zigo | I don't like at all the forest of symlinks I had to create. | 12:45 |
| rdopiera | zigo: for Horizon you won't need symlinks | 12:47 |
| zigo | rdopiera: How come? Horizon wont use the directory structure of bower then? | 12:47 |
| rdopiera | zigo: we can use the STATICFILES_DIRS setting | 12:47 |
| rdopiera | zigo: we would have that setting different for when using bower, and different when packaged | 12:48 |
| rdopiera | zigo: it's used to "mount" the static directories to urls | 12:49 |
| rdopiera | zigo: right now it takes the paths from xstatic | 12:49 |
| rdopiera | zigo: but it can instead have them hardcoded | 12:49 |
| rdopiera | zigo: then you have all paths in one place | 12:49 |
| zigo | rdopiera: This doesn't always work. | 12:50 |
| rdopiera | zigo: how come? | 12:50 |
| rdopiera | zigo: see https://github.com/openstack/horizon/blob/master/openstack_dashboard/settings.py#L153-L188 | 12:50 |
| zigo | Because what bower will download may have a different directory structure than the JS lib has. | 12:50 |
| zigo | I already pointed this out... | 12:50 |
| zigo | So pointing to a directory may just not be enough. | 12:51 |
| rdopiera | zigo: how would that happen? | 12:51 |
| zigo | For a single file, it does, for multiple, sometimes it doesn't. | 12:51 |
| rdopiera | zigo: bower downloads and builds the library the same way you would for the package | 12:51 |
| rdopiera | zigo: so how would it come up with different paths? | 12:52 |
| zigo | rdopiera: Very simple. Upstream releases a version of the lib. Then decides for whatever reason that the directory structure should change. Then in Debian, in order to not break 239489234 packages, we just keep the old directory structure... | 12:52 |
| rdopiera | zigo: aah, but that's specifically a Debian problem, and I suppose you have a solution for it already, otherwise you wouldn't do it | 12:52 |
| zigo | Sure, we *COULD* fix the 239489234 packages of the Debian archive instead ... | 12:52 |
| zigo | rdopiera: The solution is a symlink forest ... | 12:53 |
| rdopiera | zigo: sure, but xstatic doesn't help there | 12:53 |
| zigo | Having a symlink for each js file. | 12:53 |
| zigo | rdopiera: It does, because that's where I do the symlinks. | 12:53 |
| rdopiera | zigo: if you effectively fork a library, you have to maintain that fork, that's normal | 12:53 |
| openstackgerrit | Merged openstack/horizon: Updated description of update user form https://review.openstack.org/142349 | 12:53 |
| zigo | rdopiera: That's not up to me to decide for all of the Debian archive. | 12:53 |
| rdopiera | zigo: so the number of symlinks is the same, it's just the fact that they are all in one place instead of multiple places? | 12:53 |
| openstackgerrit | Merged openstack/horizon: Updated from global requirements https://review.openstack.org/142713 | 12:53 |
| zigo | I only maintain a subset... | 12:54 |
| rdopiera | zigo: also, you can use that STATICFILES_DIRS even for individual files, if you need to have that fine-grained control | 12:54 |
| zigo | rdopiera: Also, are we going to ensure versions that we use with bower? I'm scared it becomes hell with versions. | 12:54 |
| zigo | Like, it's going to get the very latest, always ... | 12:55 |
| zigo | Which would become not practical for me. | 12:55 |
| rdopiera | zigo: from what I was told, bower lets you pin versions just like pip does | 12:55 |
| rdopiera | zigo: how is a 100 symlinks practical, but a piece of settings file with 100 paths not? | 12:55 |
| *** zeus has joined #openstack-horizon | 12:58 | |
| *** zeus has quit IRC | 12:58 | |
| *** zeus has joined #openstack-horizon | 12:58 | |
| zigo | rdopiera: Constantly re-writing a patch for settings.py on each beta release of Horizon (or even: on each modification of settings.py if we consider that I would like to get in the gate to do package validation on each commits) isn't an idea I'm vouching for. | 12:59 |
| zigo | rdopiera: IMO, at least, the static file config should go in a specific file. | 12:59 |
| rdopiera | zigo: sure, we can have it in a separate file, and included | 13:00 |
| zigo | Please ! :) | 13:00 |
| rdopiera | I was thinking to do it that way from the beginning, even | 13:00 |
| rdopiera | but wanted to simplify my explanations | 13:00 |
| *** sayan has quit IRC | 13:01 | |
| *** tzumainn has joined #openstack-horizon | 13:01 | |
| rdopiera | zigo: so, basically you would have one such file for bower, pointing to wherever bower puts stuff | 13:01 |
| rdopiera | zigo: and each distro would have its own | 13:01 |
| rdopiera | zigo: pointing to wherever they install that stuff | 13:02 |
| rdopiera | zigo: if Debian has some kind of a python api for finding stuff, you could even have that in there, instead of a static list! | 13:03 |
| rdopiera | just like we use xstatic right now | 13:03 |
| zigo | We don't. | 13:03 |
| mrunge | rdopiera, zigo something like python-bindings to the packaging system? | 13:04 |
| zigo | Though I could write something which would somehow make things automated. | 13:04 |
| zigo | Like reading /var/lib/dpkg/info/libjs<something>.list, and compare what bower downloaded... | 13:04 |
| rdopiera | mrunge: something like 'dpkg -L lib-js-jquery | grep jquery.js' ;) | 13:05 |
| mrunge | yeah, or python-bindings to rpm? | 13:06 |
| mrunge | we have them ;-) | 13:06 |
| *** lblanchard has joined #openstack-horizon | 13:06 | |
| mrunge | but I'd somehow expect some resistance against using that | 13:07 |
| rdopiera | mrunge: it would slow down the startup, | 13:07 |
| rdopiera | mrunge: so in the end you probably want static paths generated, even if automatically | 13:07 |
| mrunge | rdopiera, but you could even cache static paths generated at startup | 13:08 |
| rdopiera | what I'm trying to explain is that STATICFILES_DIRS gives you the ability to map any directory, subdirectory or individual file to any url unser /static, which gives you all the power of symlinks and more | 13:08 |
| *** athomas has quit IRC | 13:08 | |
| mrunge | not saying, this is a really good idea, as e.g zigo might disagree in using RPM as package manager | 13:08 |
| *** tmazur_ has joined #openstack-horizon | 13:09 | |
| rdopiera | mrunge: somehow I believe in debuggability of settings, and nothing is easier to inspect and debug than a statically generated file :) | 13:09 |
| zigo | Oh, I'm not planning on doing this at run time! :) | 13:09 |
| rdopiera | caches are magical | 13:09 |
| zigo | This would be just a packaging trick ... | 13:10 |
| mrunge | :D | 13:10 |
| rdopiera | zigo: and if you do it, it will work with any Django app | 13:10 |
| mrunge | those mappings would most probably solve our biggest headaches | 13:10 |
| mrunge | and it's a different story, how they're getting generated | 13:10 |
| rdopiera | mrunge: we had them in there since the very beginning | 13:11 |
| rdopiera | mrunge: we just have to start using them | 13:11 |
| mrunge | yes! | 13:12 |
| rdopiera | so then we would have different ways of getting the files onto the computer (packages/bower) and different ways of generating the paths file (static/from bower log/from package metadata) | 13:12 |
| *** JeanBriceCombebi has joined #openstack-horizon | 13:13 | |
| rdopiera | r1chardj0n3s_afk: does that sound doable? | 13:13 |
| * rdopiera has no idea if what bower generates is somehow repeatable and if not, if it's possible to inspect it to get the right paths | 13:14 | |
| openstackgerrit | Ana Krivokapic proposed openstack/tuskar-ui: Don't show spinner when deployment failed https://review.openstack.org/142771 | 13:14 |
| *** athomas has joined #openstack-horizon | 13:15 | |
| mrunge | rdopiera, what I got from bower is: it just downloads required files to your directory | 13:15 |
| mrunge | so paths are somehow known and relative to your actual path | 13:16 |
| rdopiera | awesome | 13:16 |
| rdopiera | so my proposition is this: 1. bower configuration file, listing all the libraries and their versions (instead of using automatic dependency resolution of Bower), 2. a static_settings.py file with the STATICFILES_DIRS setting it in, that gets included in the settings.py file and replaced by packagers for particular distributions | 13:19 |
| rdopiera | what do you think? | 13:19 |
| rdopiera | s/included/imported | 13:20 |
| mrunge | does 1. really work? | 13:22 |
| mrunge | if you have a dep of a dep...? | 13:22 |
| rdopiera | well, I assume bower has an equivalent of the 'pip freeze' command | 13:23 |
| rdopiera | that you would use to generate that file | 13:23 |
| mrunge | I remember trying to build up a dependency tree by hand | 13:23 |
| mrunge | just to know, which packages I need to build | 13:23 |
| rdopiera | it doesn't have to be a tree, just a flat list | 13:24 |
| mrunge | anything | 13:24 |
| rdopiera | it has to have it, and if not, we will write a tool that does it | 13:24 |
| rdopiera | and a check on the gate that makes sure no extra packages are pulled in | 13:24 |
| mrunge | but I thought a package a just requires package b. and package b come with it's own dependency file | 13:25 |
| mrunge | that would make a tree | 13:25 |
| rdopiera | mrunge: yes, so what you do is you run bower on just the packages you need, and then list all packages installed | 13:25 |
| mrunge | rdopiera, yes, I got that. but: your package can not know deps of depending packages | 13:26 |
| rdopiera | mrunge: yours doesn't, bower does | 13:26 |
| mrunge | and we all know packages are changing fast in js world | 13:26 |
| rdopiera | hopefully not witchout bumping the version numbers | 13:26 |
| rdopiera | without | 13:26 |
| mrunge | so, the question is: how does bower know this? | 13:26 |
| mrunge | how to flatten the dep tree | 13:27 |
| rdopiera | mrunge: it downloads the package and looks into its metadata | 13:27 |
| rdopiera | mrunge: I'm not sure how it resolves version conflicts, though, but that's not important for us -- we will have the versions pinned | 13:27 |
| mrunge | ... installs packages, reads deps and keeps on downloading, until deps are satisfied | 13:28 |
| rdopiera | mrunge: yup, but that's not a problem | 13:28 |
| mrunge | ok, fine then | 13:28 |
| rdopiera | I will write an e-mail to the dev list then | 13:29 |
| *** raildo has joined #openstack-horizon | 13:30 | |
| openstackgerrit | Merged openstack/tuskar-ui: Don't show spinner when deployment failed https://review.openstack.org/142771 | 13:30 |
| rdopiera | I have one more concern | 13:30 |
| *** JeanBriceCombebi has quit IRC | 13:30 | |
| rdopiera | what about static file libraries that are not JavaScript? | 13:31 |
| rdopiera | such as font-awesome or bootstrap? | 13:31 |
| rdopiera | we will still need to package them for Bower | 13:31 |
| *** ZZelle has quit IRC | 13:31 | |
| mrunge | isn't xstatic an answer? | 13:32 |
| *** JeanBriceCombebi has joined #openstack-horizon | 13:32 | |
| mrunge | javascript world lives from copying stuff around | 13:33 |
| *** Ala has quit IRC | 13:34 | |
| mrunge | rdopiera, fontawesome is available via bower | 13:34 |
| mrunge | the same is true for bootstrap | 13:35 |
| rdopiera | hmm, ok | 13:35 |
| rdopiera | and I suppose if we find something that isn't, we can make a bower package for it | 13:35 |
| *** Ala has joined #openstack-horizon | 13:38 | |
| *** jcoufal_ has joined #openstack-horizon | 13:40 | |
| *** JeanBriceCombebi has quit IRC | 13:41 | |
| *** jcoufal has quit IRC | 13:43 | |
| *** JeanBriceCombebi has joined #openstack-horizon | 13:43 | |
| *** _crobertsrh is now known as crobertsrh | 13:44 | |
| *** Longgeek has quit IRC | 13:44 | |
| *** Longgeek has joined #openstack-horizon | 13:45 | |
| *** krykowski has joined #openstack-horizon | 14:01 | |
| *** e0ne_ has joined #openstack-horizon | 14:07 | |
| *** e0ne has quit IRC | 14:08 | |
| *** tosky has joined #openstack-horizon | 14:14 | |
| *** radez_g0n3 is now known as radez | 14:18 | |
| *** peristeri has joined #openstack-horizon | 14:27 | |
| *** k4n0 has quit IRC | 14:29 | |
| *** dboik has joined #openstack-horizon | 14:31 | |
| *** ZZelle has joined #openstack-horizon | 14:33 | |
| openstackgerrit | Radomir Dopieralski proposed openstack/tuskar-ui: Fix TemplateSyntaxError in the server-side filetring of node list views https://review.openstack.org/142800 | 14:37 |
| *** ZZelle has quit IRC | 14:38 | |
| openstackgerrit | Sam Betts proposed openstack/horizon: Reduced code duplication in creating page headers https://review.openstack.org/142802 | 14:42 |
| *** packet has joined #openstack-horizon | 14:43 | |
| *** neelashah has joined #openstack-horizon | 14:51 | |
| openstackgerrit | Matt Borland proposed openstack/horizon: Add collectstatic commands to install doc https://review.openstack.org/141885 | 14:52 |
| *** hurgleburgler has joined #openstack-horizon | 14:54 | |
| *** raildo has quit IRC | 14:54 | |
| *** wuhg has quit IRC | 15:03 | |
| *** krykowski has quit IRC | 15:04 | |
| *** Ala has quit IRC | 15:05 | |
| *** krykowski_ has joined #openstack-horizon | 15:05 | |
| *** Ala has joined #openstack-horizon | 15:05 | |
| *** david-lyle_afk has quit IRC | 15:07 | |
| *** Ala has quit IRC | 15:08 | |
| *** esp has joined #openstack-horizon | 15:09 | |
| *** vijendar has quit IRC | 15:10 | |
| *** mrunge has quit IRC | 15:12 | |
| *** JeanBriceCombebi has quit IRC | 15:13 | |
| *** bradjones has joined #openstack-horizon | 15:16 | |
| *** woodm1979 has joined #openstack-horizon | 15:17 | |
| *** Ala has joined #openstack-horizon | 15:19 | |
| *** JeanBriceCombebi has joined #openstack-horizon | 15:20 | |
| *** Ala has quit IRC | 15:21 | |
| *** Ala has joined #openstack-horizon | 15:24 | |
| *** EmilyW has joined #openstack-horizon | 15:28 | |
| *** mattfarina has joined #openstack-horizon | 15:28 | |
| *** tmazur_ has quit IRC | 15:29 | |
| *** rdopiera has quit IRC | 15:33 | |
| *** vijendar has joined #openstack-horizon | 15:42 | |
| *** absubram has joined #openstack-horizon | 15:45 | |
| *** woodm19791 has joined #openstack-horizon | 15:50 | |
| *** woodm1979 has quit IRC | 15:50 | |
| *** rushiagr is now known as rushiagr_away | 15:50 | |
| *** krykowski_ has quit IRC | 15:52 | |
| *** ttrifonov is now known as zz_ttrifonov | 16:00 | |
| *** krykowski has joined #openstack-horizon | 16:02 | |
| *** david-lyle has joined #openstack-horizon | 16:02 | |
| *** EmilyW has quit IRC | 16:04 | |
| *** nlahouti has joined #openstack-horizon | 16:05 | |
| *** ongk has joined #openstack-horizon | 16:09 | |
| *** nlahouti has quit IRC | 16:09 | |
| *** rushiagr_away is now known as rushiagr | 16:14 | |
| *** krykowski has quit IRC | 16:15 | |
| *** nlahouti has joined #openstack-horizon | 16:16 | |
| *** raildo has joined #openstack-horizon | 16:17 | |
| *** hurgleburgler has quit IRC | 16:18 | |
| *** ongk has quit IRC | 16:19 | |
| *** hurgleburgler has joined #openstack-horizon | 16:19 | |
| *** ongk has joined #openstack-horizon | 16:20 | |
| *** hurgleburgler has quit IRC | 16:21 | |
| openstackgerrit | Abishek Subramanian proposed openstack/horizon: Fix multi-nic issue with nexus plugin https://review.openstack.org/65793 | 16:23 |
| *** nlahouti has quit IRC | 16:24 | |
| *** nlahouti has joined #openstack-horizon | 16:25 | |
| *** MaxV has quit IRC | 16:28 | |
| *** ygbo has quit IRC | 16:28 | |
| *** bradjones has quit IRC | 16:28 | |
| *** kumartin has joined #openstack-horizon | 16:31 | |
| *** bradjones has joined #openstack-horizon | 16:32 | |
| *** romainh1 has left #openstack-horizon | 16:32 | |
| *** pcaruana has quit IRC | 16:33 | |
| *** Ala has quit IRC | 16:37 | |
| *** Ala has joined #openstack-horizon | 16:37 | |
| *** pawels has joined #openstack-horizon | 16:44 | |
| *** neelashah has quit IRC | 16:44 | |
| *** exploreshaifali has joined #openstack-horizon | 16:45 | |
| *** exploreshaifali has quit IRC | 16:45 | |
| openstackgerrit | Abishek Subramanian proposed openstack/horizon: Refactor project instance test https://review.openstack.org/142839 | 16:46 |
| *** exploreshaifali has joined #openstack-horizon | 16:46 | |
| *** exploreshaifali has quit IRC | 16:46 | |
| *** exploreshaifali has joined #openstack-horizon | 16:46 | |
| openstackgerrit | Paul Karikh proposed openstack/horizon: Disable password autocomplete in the login form https://review.openstack.org/142844 | 16:56 |
| *** masco has quit IRC | 16:59 | |
| *** sayan has joined #openstack-horizon | 17:03 | |
| *** gyee has joined #openstack-horizon | 17:05 | |
| *** tqtran has joined #openstack-horizon | 17:06 | |
| *** EmilienM has quit IRC | 17:08 | |
| *** EmilienM has joined #openstack-horizon | 17:09 | |
| *** rm_work|away is now known as rm_work | 17:09 | |
| *** bpokorny has joined #openstack-horizon | 17:15 | |
| *** jpich has quit IRC | 17:15 | |
| *** pawels1 has joined #openstack-horizon | 17:15 | |
| *** EmilyW has joined #openstack-horizon | 17:16 | |
| *** ayoung is now known as ayoung-afk | 17:16 | |
| *** pawels has quit IRC | 17:17 | |
| *** bpokorny has quit IRC | 17:19 | |
| *** kumartin has quit IRC | 17:20 | |
| *** regebro has quit IRC | 17:21 | |
| thiagop | david-lyle: ping | 17:23 |
| david-lyle | thiagop: pong | 17:23 |
| thiagop | hey, I was trying to understand why you suggested the use of can_access on that patch to remove the hard-coded admin | 17:23 |
| david-lyle | thiagop: the check is incomplete and much more complicated | 17:24 |
| david-lyle | not a simple role check | 17:24 |
| david-lyle | I posted comments before I simply -2'd it | 17:25 |
| thiagop | we were working on a patch that takes care of correcting the RBAC check on the panels, but using the "policy_rules" property | 17:25 |
| david-lyle | it's missing a lot of the nuance of policy | 17:25 |
| david-lyle | thiagop: IMO the right solution is not RBAC the panel itself in almost all cases, but simply merge the admin and project panels and use RBAC to expose actions | 17:26 |
| *** dboik has quit IRC | 17:27 | |
| david-lyle | things like host-aggregates and hypervisors are simple enough to policy checks to block with a role check, but keystone's version of admin is not enough | 17:27 |
| thiagop | https://review.openstack.org/#/c/126988/ | 17:28 |
| thiagop | I was seeing pkarikh's patch as a first move towards that, not a definitive solution | 17:29 |
| david-lyle | thiagop: that's closer. Ideally, we just need to check the read action policy | 17:29 |
| thiagop | but I understand your concerns | 17:29 |
| thiagop | if I can't read, I can't modify either, right?! | 17:30 |
| david-lyle | not necessarily | 17:30 |
| david-lyle | that's policy driven | 17:30 |
| thiagop | but on Horizon's context (web context), I need to show something before editing it. | 17:31 |
| david-lyle | In fact the separation is very desirable | 17:31 |
| david-lyle | I may want read only access to say hypervisors | 17:31 |
| david-lyle | just not edit | 17:31 |
| david-lyle | so read is enough to get into the panel | 17:32 |
| david-lyle | edit is filtered on the panel | 17:32 |
| david-lyle | see the identity/projects panel | 17:32 |
| david-lyle | that's the pattern | 17:32 |
| david-lyle | the policy checks https://github.com/openstack/horizon/blob/master/openstack_dashboard/dashboards/identity/projects/panel.py#L29 are for read only | 17:33 |
| thiagop | right | 17:33 |
| david-lyle | but munging admin of identity and nova or cinder is in no way accurate | 17:33 |
| david-lyle | they all can specify that in their policy implementation | 17:33 |
| openstackgerrit | Matt Borland proposed openstack/horizon: Tests pass run in languages other than English https://review.openstack.org/142851 | 17:34 |
| david-lyle | so panels that pull in the kitchen sink, like instances become very complicated | 17:34 |
| openstackgerrit | Matt Borland proposed openstack/horizon: Tests pass in languages other than English https://review.openstack.org/142851 | 17:35 |
| thiagop | right, right. :) | 17:35 |
| thiagop | I'll take a look on our patch to consider only the listing rules | 17:36 |
| *** dboik has joined #openstack-horizon | 17:37 | |
| david-lyle | ok, ping me if you have questions | 17:37 |
| thiagop | ty! | 17:38 |
| *** lhcheng has joined #openstack-horizon | 17:39 | |
| *** sambetts has quit IRC | 17:40 | |
| openstackgerrit | Matt Borland proposed openstack/horizon: Tests pass in languages other than English https://review.openstack.org/142851 | 17:41 |
| *** lhcheng has quit IRC | 17:42 | |
| openstackgerrit | Matt Borland proposed openstack/horizon: Tests pass in languages other than English https://review.openstack.org/142851 | 17:42 |
| *** nikunj2512_ has quit IRC | 17:43 | |
| *** cody-somerville has joined #openstack-horizon | 17:44 | |
| mattfarina | david-lyle ping | 17:45 |
| mattfarina | david-lyle i was hoping to get a few minutes of your time to talk about https://blueprints.launchpad.net/horizon/+spec/domain-based-rbac | 17:45 |
| david-lyle | mattfarina: sure | 17:46 |
| *** gyee has quit IRC | 17:48 | |
| mattfarina | david-lyle so, I'm really interesting in helping this happen. | 17:48 |
| *** sbfox has joined #openstack-horizon | 17:48 | |
| mattfarina | david-lyle did you have a few minutes to get on hangouts to chat about it? | 17:48 |
| david-lyle | sure | 17:49 |
| *** lhcheng has joined #openstack-horizon | 17:49 | |
| *** woodm1979 has joined #openstack-horizon | 18:00 | |
| *** woodm19791 has quit IRC | 18:00 | |
| *** sbfox has quit IRC | 18:06 | |
| *** cedrics has quit IRC | 18:08 | |
| *** e0ne_ has quit IRC | 18:09 | |
| *** sbfox has joined #openstack-horizon | 18:17 | |
| *** jtomasek has quit IRC | 18:21 | |
| *** pawels1 has quit IRC | 18:22 | |
| *** robcresswell has joined #openstack-horizon | 18:22 | |
| robcresswell | Evening, does anybody know how to parse ':' in the Django templating language? | 18:23 |
| robcresswell | My Google-fu is failing me | 18:23 |
| *** harlowja_away is now known as harlowja | 18:23 | |
| *** lhcheng_ has joined #openstack-horizon | 18:28 | |
| *** lhcheng has quit IRC | 18:31 | |
| *** aix has quit IRC | 18:33 | |
| *** aliceR has joined #openstack-horizon | 18:37 | |
| *** rushiagr is now known as rushiagr_away | 18:37 | |
| *** ericpeterson has joined #openstack-horizon | 18:37 | |
| *** Nate has joined #openstack-horizon | 18:39 | |
| openstackgerrit | Matt Borland proposed openstack/horizon: Add collectstatic commands to install doc https://review.openstack.org/141885 | 18:40 |
| *** jcoufal_ has quit IRC | 18:41 | |
| *** e0ne has joined #openstack-horizon | 18:46 | |
| *** neelashah has joined #openstack-horizon | 18:46 | |
| *** e0ne is now known as e0ne_ | 18:48 | |
| *** athomas has quit IRC | 18:49 | |
| *** bradjones has quit IRC | 18:49 | |
| *** robcresswell has quit IRC | 18:50 | |
| *** bradjones has joined #openstack-horizon | 18:51 | |
| openstackgerrit | Matt Borland proposed openstack/horizon: Tests pass in languages other than English https://review.openstack.org/142851 | 18:51 |
| *** EmilyW has quit IRC | 18:52 | |
| *** Nate has quit IRC | 18:56 | |
| openstackgerrit | Abishek Subramanian proposed openstack/horizon: Fix multi-nic issue with nexus plugin https://review.openstack.org/65793 | 18:58 |
| *** e0ne_ has quit IRC | 19:00 | |
| *** lhcheng_ is now known as lhcheng | 19:01 | |
| *** colinmcnamara has joined #openstack-horizon | 19:02 | |
| *** nlahouti has quit IRC | 19:02 | |
| *** nlahouti has joined #openstack-horizon | 19:03 | |
| *** e0ne has joined #openstack-horizon | 19:07 | |
| *** rushiagr_away is now known as rushiagr | 19:14 | |
| openstackgerrit | Abishek Subramanian proposed openstack/horizon: Display profile information in network details https://review.openstack.org/76787 | 19:15 |
| openstackgerrit | Szymon Wróblewski proposed openstack/horizon: Create a reusable angular based wizard https://review.openstack.org/142880 | 19:15 |
| *** EmilyW has joined #openstack-horizon | 19:20 | |
| *** ayoung-afk is now known as ayoung | 19:21 | |
| *** colinmcnamara has quit IRC | 19:21 | |
| openstackgerrit | Thai Tran proposed openstack/horizon: WIP Consolidating conf and scripts https://review.openstack.org/141457 | 19:23 |
| *** rushiagr is now known as rushiagr_away | 19:26 | |
| *** aliceR has quit IRC | 19:26 | |
| openstackgerrit | Kelly Domico proposed openstack/horizon: Fix popover positions in network topology https://review.openstack.org/142886 | 19:28 |
| *** sayan has quit IRC | 19:28 | |
| tsufiev | david-lyle, hello! Speaking of RBAC policies and different dashboard parts... | 19:32 |
| *** colinmcnamara has joined #openstack-horizon | 19:33 | |
| *** hurgleburgler has joined #openstack-horizon | 19:34 | |
| *** salty has joined #openstack-horizon | 19:34 | |
| tsufiev | david-lyle, what do you think of giving each dashboard element (Dashboard/Panel/Table/Tab/Form/Workflow step) a natural address based on its place in the UI elements hierarchy? Then the RBAC policy could be attached directly to each element without actually adding some specific policy checks into the dashboard code | 19:34 |
| openstackgerrit | Thai Tran proposed openstack/horizon: Consolidating conf and scripts https://review.openstack.org/141457 | 19:35 |
| *** salty is now known as Guest13542 | 19:35 | |
| openstackgerrit | Thai Tran proposed openstack/horizon: Consolidating conf and scripts https://review.openstack.org/141457 | 19:35 |
| *** Guest13542 is now known as MrSalty | 19:35 | |
| tsufiev | the whole point here is freeing operators/deployers of editing python code, they could just add another rule into policy.json file, tied to some UI address on the left side, and a combination of existing rules on the right side | 19:36 |
| *** bradjones has quit IRC | 19:36 | |
| *** bradjones has joined #openstack-horizon | 19:40 | |
| *** EmilyW has quit IRC | 19:41 | |
| david-lyle | tsufiev: adding the policy checks are attempting to make the policy.json files the only place that need editing | 19:42 |
| david-lyle | horizon could really care less about what the user can or can't do :) it's just proxying that for the services so users aren't encountered by endless errors | 19:44 |
| tsufiev | david-lyle, agree. But isn't it tedious to specify some policy rule for every UI element possible :)? | 19:44 |
| *** woodm1979 has quit IRC | 19:44 | |
| david-lyle | it's not for the UI element, it's for specific actions in the end services | 19:44 |
| david-lyle | can I add a volume | 19:44 |
| david-lyle | can I edit a network | 19:44 |
| david-lyle | can I list all networks | 19:45 |
| david-lyle | hiding a panel is done as a convenience based on the actions of that panel | 19:45 |
| david-lyle | if I can't do anything but encounter a blank page and an error on a panel, why are you letting me see it? | 19:46 |
| david-lyle | and that's it really | 19:46 |
| tsufiev | david-lyle, e.g. tabs that are lazy-loaded - I cannot know whether they do have some data until I send ajax request | 19:46 |
| tqtran | tsufiev: i dont envision the policy on an element basis, that would be insane. I'm thinking more on a page basis. | 19:47 |
| david-lyle | tsufiev: not that they don't have data, that you don't even have the possibility of seeing data because the service doesn't feel like you have sufficient authority | 19:47 |
| david-lyle | tqtran: policy is tied to API calls | 19:48 |
| david-lyle | actions | 19:48 |
| tqtran | david-lyle, tsufiev: and since all actions trigger a page, you can say that its bound to action. its almost synonymous | 19:48 |
| david-lyle | list, get, update, delete | 19:48 |
| tsufiev | tqtran, what's about hiding separate steps of a workflow? they are less than page | 19:48 |
| david-lyle | but if you can list the data, the page should be available | 19:48 |
| tsufiev | we had at least one customer who wanted this feature | 19:48 |
| david-lyle | tsufiev: if that step is necessary to complete the workflow, you shouldn't see the workflow at all | 19:49 |
| david-lyle | if it's an optional step, that's not really policy as implemented today, unless you say users with role X can't complete task Y | 19:49 |
| *** Nate has joined #openstack-horizon | 19:50 | |
| tsufiev | david-lyle, okay, I get your point | 19:50 |
| tsufiev | tqtran, but why the element-wise policy is insane? well, I suppose it may have some perfomance issues, but just wanted to hear your reasoning :) | 19:51 |
| *** nlahouti has quit IRC | 19:51 | |
| *** ZZelle has joined #openstack-horizon | 19:51 | |
| david-lyle | tsufiev: I agree the policy checks are tedious | 19:52 |
| tqtran | tsufiev: i think is insane the way we have it now. there will be checks all over the place. but using angular, that might be feasible | 19:52 |
| *** colinmcnamara has quit IRC | 19:52 | |
| *** ericpeterson has quit IRC | 19:53 | |
| david-lyle | tqtran: the only way this improves is when something like https://review.openstack.org/#/c/136980/ gets merged all over | 19:53 |
| david-lyle | then the policy checks change direction | 19:54 |
| david-lyle | you could get the list of approved actions and filter out the non-allowed from the page | 19:54 |
| david-lyle | but policy is still at the per CRUD operation level | 19:54 |
| david-lyle | that is RBAC | 19:55 |
| *** ericpeterson has joined #openstack-horizon | 19:55 | |
| david-lyle | it was indeed simpler in a two role world | 19:55 |
| david-lyle | but it's not very useful outside of token installations | 19:55 |
| tqtran | david-lyle: what if angular have access to the policy file client-side? | 19:56 |
| tqtran | david-lyle: only downside is that we'll have to manually update the policy file, which is a pain | 19:57 |
| david-lyle | that's fine, but it's still the same logic moved to the client | 19:57 |
| tsufiev | david-lyle, well, this still makes us to decide whether to show or not some page based on a set of approved actions this page contains - if someone wants to restrict this page explicitly and unconditionally for some user roles in his cloud setup, most probably he will have to edit the dashboard code | 19:57 |
| david-lyle | I think ideally, you would make a call back to the horizon server to get the list of approved actions | 19:57 |
| david-lyle | rather than sending down all the policy files | 19:57 |
| david-lyle | tsufiev: overrides.py and add override of panel method, def can_access(): return False | 19:58 |
| david-lyle | that is code though | 19:59 |
| tqtran | david-lyle: so you would make this call for approved actions once? or for each action you want to take? | 19:59 |
| david-lyle | tqtran: to prematurely optimize, I would batch that call | 19:59 |
| david-lyle | you know the list at load time | 19:59 |
| tqtran | david-lyle: ok, lets say you batch that call when you login | 19:59 |
| david-lyle | not then, per view at this point | 20:00 |
| tqtran | david-lyle: =/ why not for all panels? | 20:00 |
| david-lyle | that's not right for a single page model | 20:00 |
| david-lyle | I suppose you could, but that's a big list | 20:00 |
| tqtran | david-lyle: i would rather see the policy file fetched once, then angular will handle showing/hiding of elements | 20:00 |
| david-lyle | but then you get the can access list and cache it | 20:01 |
| tqtran | david-lyle: which brings me to my next point, why not make the policy file static? | 20:01 |
| david-lyle | by policy file, you mean policy file for each service | 20:01 |
| tqtran | david-lyle: yes | 20:01 |
| david-lyle | and reimplement the policy engine on the client side | 20:02 |
| david-lyle | ? | 20:02 |
| david-lyle | not excited about that | 20:02 |
| tsufiev | david-lyle, yep, still has to edit some code. But perhaps time has not yet come to move every user setting into .json file :). I don't want to solve the problem that nobody considers as a problem (yet) | 20:02 |
| openstackgerrit | ayoung proposed openstack/django_openstack_auth: Use keystone auth plugins https://review.openstack.org/140308 | 20:02 |
| tqtran | david-lyle: well, we're not reimplementing the policy engine, we are merely letting angular show/hide base on what it reads in the json | 20:03 |
| tsufiev | tqtran, you're reading my thoughts, I also wanted to ask you today how are you going to deal with policies in angular )) | 20:03 |
| david-lyle | tqtran: which is evaluating policy | 20:03 |
| david-lyle | hence the policy engine | 20:03 |
| david-lyle | read a policy.json | 20:03 |
| david-lyle | they are nested rules | 20:03 |
| david-lyle | that ultimately check role | 20:04 |
| david-lyle | it's not a simple as it seems | 20:04 |
| tqtran | david-lyle: i'll have to take a closer look at policy files, you might be right | 20:04 |
| tqtran | tsufiev: fast answer: i dont know yet lol | 20:04 |
| tsufiev | lol )) | 20:05 |
| tqtran | tsufiev: as of now, we are not handling showing/hiding of stuff. we let user take action and API throw error if action is not permitted | 20:05 |
| tqtran | tsufiev: but the focus is on identity users, so its not that complicated | 20:06 |
| tqtran | tsufiev: taking it step by step, im sure we will get to that discussion at some point | 20:06 |
| tsufiev | tqtran, yeah. I guess ng-show/ng-hide could do this work, but how are they going to get info for deciding that - that's the question... | 20:07 |
| *** Nate has quit IRC | 20:07 | |
| tsufiev | sending one request for each element seems quite inefficient | 20:07 |
| tqtran | tsufiev: ok, this is what i imagine. you have a policy service. and in your controller, you require this policy service | 20:07 |
| tqtran | tsufiev: and you use directive ng-show="function" where function will do the policy check for you. | 20:08 |
| tqtran | tsufiev: and policy file would exist client-side, so no calls is required. its all done client-side | 20:09 |
| *** JeanBriceCombebi has quit IRC | 20:09 | |
| tqtran | tsufiev: but im treating it like a blackbox, obviously, it more complicated im sure | 20:09 |
| tsufiev | tqtran, I guess this is 'next' phase of angular migration? | 20:09 |
| tsufiev | (sending policies to client-side) | 20:09 |
| david-lyle | tqtran: indeed | 20:09 |
| tqtran | tsufiev: yeah probably, when we get to it | 20:09 |
| *** Longgeek has quit IRC | 20:10 | |
| david-lyle | whipper snappers, back in my day | 20:10 |
| tqtran | lol whipper snappers? | 20:10 |
| david-lyle | https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#q=whippersnapper | 20:10 |
| tqtran | whippersnapper - A term generally used by old persons, to insult the younger generation | 20:10 |
| tqtran | lol | 20:10 |
| david-lyle | cheers | 20:11 |
| tqtran | hahaha, blame it on tsufiev | 20:12 |
| david-lyle | get off my lawn | 20:12 |
| david-lyle | tqtran, tsufiev, we'll figure it out | 20:12 |
| * tsufiev adds a new word to his vocabulary | 20:12 | |
| tsufiev | though I'm not old enough to use it )) | 20:13 |
| tsufiev | david-lyle, sure! | 20:14 |
| david-lyle | I think you'll want to leave the policy engine on the server (remain in sync with the rest of openstack) but batch call to get an access list | 20:14 |
| david-lyle | the subsequent checks on the angular side check that secondary construct | 20:14 |
| tqtran | david-lyle: are we manually updating the policy files today? | 20:15 |
| david-lyle | unfortunately | 20:15 |
| *** raildo is now known as raildo_away | 20:15 | |
| tsufiev | btw, we may want to use something like olso.lib updater - but for policies | 20:15 |
| david-lyle | thats a different issue | 20:15 |
| david-lyle | inflight | 20:16 |
| tsufiev | I suspect that they're currently a bit outdated in horizon | 20:16 |
| tsufiev | (for neutron at least) | 20:16 |
| *** Nate has joined #openstack-horizon | 20:16 | |
| *** Nate has quit IRC | 20:17 | |
| *** NateTheGreat has joined #openstack-horizon | 20:18 | |
| tsufiev | david-lyle, do you think that pkarikh's patch for un-hardcoding admin role could be changed to fit in overall policy architecture? | 20:18 |
| *** bradjones has quit IRC | 20:18 | |
| NateTheGreat | Anybody know what might through a PV_MISSING_DRIVER error only on automatic disk paritioning? | 20:19 |
| tsufiev | NateTheGreat, is it really happening in horizon? sounds like cinder issue | 20:20 |
| *** vijendar has quit IRC | 20:20 | |
| NateTheGreat | We dont' have cinder installed, just glance | 20:21 |
| NateTheGreat | we get a different error for not having cinder ;P | 20:21 |
| david-lyle | tsufiev: It could, needs a lot of rework. the -2 was because it's mis-interpreting admin and that _has_ to be fixed | 20:21 |
| NateTheGreat | It seems to work fine when doing manual partitioning of the disk, or through the CLI | 20:22 |
| david-lyle | NateTheGreat: do you have access to the stack trace? | 20:31 |
| *** jgravel_ has joined #openstack-horizon | 20:31 | |
| *** NateTheGreat_ has joined #openstack-horizon | 20:32 | |
| NateTheGreat_ | david-lyle: if you can point me to it, I can get it. | 20:32 |
| NateTheGreat_ | david-lyle: this may also be of some help https://ask.openstack.org/en/question/56596/horizion-automatic-disk-partitioning-fails-but-manual-works-fine/ | 20:34 |
| david-lyle | look at /var/log/apache2/horizon_error.log | 20:34 |
| *** NateTheGreat has quit IRC | 20:34 | |
| david-lyle | and it worked from the CLI? | 20:35 |
| *** jgravel__ has quit IRC | 20:35 | |
| NateTheGreat_ | Yes, it has to do with the automatic paritioning, cause both the CLI and horizon manual partition works | 20:35 |
| NateTheGreat_ | I guess that specific error doesn't appear in the error.log under /apache | 20:39 |
| david-lyle | NateTheGreat_: if you pass --disk-config=AUTO to the nova boot command line, does it fail as well? | 20:40 |
| *** sbfox has quit IRC | 20:40 | |
| NateTheGreat_ | it gives me unrecognized arguments: --disk-config=AUTO | 20:41 |
| NateTheGreat_ | that may present a problem... | 20:41 |
| david-lyle | auto-partitioning is an extension | 20:41 |
| david-lyle | you may not have it enabled | 20:41 |
| NateTheGreat_ | that would be my guess. I didn't actually configure most of it. | 20:41 |
| NateTheGreat_ | Where is it in the docs? | 20:41 |
| *** flwang has joined #openstack-horizon | 20:43 | |
| david-lyle | http://developer.openstack.org/api-ref-compute-v2-ext.html#ext-os-disk-config-servers is the API | 20:43 |
| david-lyle | not well versed in enabling extension | 20:44 |
| david-lyle | s | 20:44 |
| david-lyle | but I assume in the config file | 20:44 |
| *** harlowja has quit IRC | 20:45 | |
| *** harlowja_ has joined #openstack-horizon | 20:45 | |
| NateTheGreat_ | alright, I'll go see if I can find it. Could this possibly be pushed into a dependency for horizon? as its the default method | 20:45 |
| david-lyle | extensions really shouldn't be the default, that's a problem in Horizon | 20:46 |
| *** jtomasek has joined #openstack-horizon | 20:46 | |
| NateTheGreat_ | Right, I think that is our problem. or at least put it as part of the docs in the horizon part as maybe a warning? or a how to on installing it? | 20:46 |
| david-lyle | NateTheGreat_: this might be helpful https://github.com/rackerlabs/os_diskconfig_python_novaclient_ext | 20:47 |
| *** NateTheGreat has joined #openstack-horizon | 20:47 | |
| NateTheGreat | Sorry my work connection keeps dumping me. | 20:48 |
| david-lyle | this might be helpful https://github.com/rackerlabs/os_diskconfig_python_novaclient_ext | 20:48 |
| david-lyle | but that seems odd | 20:49 |
| david-lyle | I think that may be out of date | 20:49 |
| david-lyle | might be better to ask in nova or dev | 20:49 |
| NateTheGreat | okay, I'll give it a shot over there. Thanks for the help :) | 20:50 |
| david-lyle | I'm not sure how it's installed | 20:50 |
| *** NateTheGreat_ has quit IRC | 20:50 | |
| david-lyle | let me know if you find out we're doing something really dumb in Horizon | 20:51 |
| david-lyle | and we'll patch it up | 20:51 |
| NateTheGreat | Okay. I'd say having automatic be the default if its not part of the base might need a flip. Or throw teh dependency in. I'll get back to you on the fix I have | 20:52 |
| *** vijendar has joined #openstack-horizon | 20:52 | |
| david-lyle | I agree, just not sure what happens when not auto | 20:52 |
| NateTheGreat | It doesn't resize the disk to the flavor for you. if you are in manual you have to do it yourself | 20:53 |
| david-lyle | right, but not from horizon | 20:54 |
| NateTheGreat | Correct, that is at the OS level | 20:55 |
| david-lyle | We have an extension check around that option I believe, rechecking code | 20:56 |
| *** romainh has joined #openstack-horizon | 20:56 | |
| tsufiev | david-lyle, ok, I'll tell pkarikh not to lose heart and keep going ) | 20:57 |
| david-lyle | NateTheGreat: we do have an extension check for DiskConfig, so the nova server is reporting that extension is enabled | 20:57 |
| david-lyle | tsufiev: good, just couldn't go in as is, and someone else +2'd it | 20:58 |
| david-lyle | the goal is good | 20:58 |
| NateTheGreat | Hmmm.. thats quite odd..wondering what happened then. | 20:59 |
| *** kumartin has joined #openstack-horizon | 21:03 | |
| *** NateTheGreat has quit IRC | 21:06 | |
| *** akrivoka has quit IRC | 21:07 | |
| *** jasondotstar has quit IRC | 21:10 | |
| *** jasondotstar has joined #openstack-horizon | 21:12 | |
| *** neelashah has quit IRC | 21:14 | |
| *** exploreshaifali has quit IRC | 21:17 | |
| *** zeus has quit IRC | 21:20 | |
| *** dboik has quit IRC | 21:20 | |
| *** sbfox has joined #openstack-horizon | 21:20 | |
| *** romainh has left #openstack-horizon | 21:29 | |
| *** ayoung has quit IRC | 21:29 | |
| *** woodm1979 has joined #openstack-horizon | 21:32 | |
| *** jtomasek has quit IRC | 21:35 | |
| *** ayoung has joined #openstack-horizon | 21:37 | |
| *** flwang has quit IRC | 21:41 | |
| *** flwang has joined #openstack-horizon | 21:41 | |
| ericpeterson | I have an easy bug that needs some fixin', if anyone wants low hanging fruit. filing it now | 21:46 |
| *** julim has quit IRC | 21:48 | |
| *** dboik_ has joined #openstack-horizon | 21:52 | |
| *** jgravel__ has joined #openstack-horizon | 21:55 | |
| *** jgravel_ has quit IRC | 21:58 | |
| ericpeterson | https://bugs.launchpad.net/horizon/+bug/1404032 | 21:58 |
| woodm1979 | ericpeterson: is it happening on the non-admin version of the page? | 22:01 |
| woodm1979 | I should slightly rephrase: Is it ALSO happening on the non-admin version of the page? | 22:02 |
| ericpeterson | prolly, but I have not accessed that tenant's stuff yet | 22:03 |
| woodm1979 | ok | 22:03 |
| ericpeterson | seems like the error handling has been "improved" somehow | 22:03 |
| ericpeterson | it eventually gets to this line: https://github.com/openstack/horizon/blob/master/horizon/exceptions.py#L359 | 22:05 |
| woodm1979 | That just means that there's 2 level's of try -> except: exceptions.handle() | 22:06 |
| ericpeterson | I'm not sure that re-raise is a good idea, especially when escalate defaults to False | 22:06 |
| woodm1979 | The re-raise is just there to preserve the original. (because we're handling it twice.) | 22:07 |
| *** lblanchard has quit IRC | 22:07 | |
| woodm1979 | I'll see of someone here wants to tackle this. | 22:07 |
| ericpeterson | the outside block https://github.com/openstack/horizon/blob/master/openstack_dashboard/dashboards/project/volumes/volumes/tables.py#L263 | 22:07 |
| ericpeterson | that outside block.... it sees the exception re-raised and it exits out of there.... never going any farther | 22:08 |
| ericpeterson | which is weird. if escalate is true and I am telling a method to handle the exception for me, why can't it just you know.... handle it ? | 22:09 |
| woodm1979 | heh. | 22:09 |
| ericpeterson | I mispoke. escalate is False. which still begs my question | 22:10 |
| woodm1979 | Well, that last line (the re-raise) shoudl only be called on the 2nd run through the code. But a nova NotFound exception should be handled prior to this. | 22:11 |
| woodm1979 | wait, exceptions.handle has changed quite a bit. There used to be 3 different buckets that exceptions coudl be in. | 22:12 |
| woodm1979 | No it looks different. | 22:13 |
| *** tosky has quit IRC | 22:13 | |
| woodm1979 | a lot different. | 22:13 |
| ericpeterson | yeah, it has been "improved" | 22:13 |
| ericpeterson | I think part of it is that the Nova NotFound exception is not considered recoverable. that might be true in most situations, but not for this one | 22:15 |
| woodm1979 | Well, I'm not sure how the code is getting past: https://github.com/openstack/horizon/blob/master/horizon/exceptions.py#L352 | 22:15 |
| ericpeterson | All I can say is the last line gets called or catches my breakpoint | 22:16 |
| ericpeterson | and then poo hits the fan | 22:16 |
| ericpeterson | its fairly easy to recreate in your env, just re-raise that exception in there and enjoy the carnage | 22:17 |
| woodm1979 | Interesting. I'm not sure I like the "improvements" to the exception handler code. | 22:18 |
| *** ongk has quit IRC | 22:19 | |
| *** kdomico has joined #openstack-horizon | 22:19 | |
| *** cbader has quit IRC | 22:21 | |
| ericpeterson | I think I have it | 22:21 |
| * ericpeterson continues to bird dog | 22:22 | |
| ericpeterson | got it. that exception handler is jacked up | 22:22 |
| *** peristeri has quit IRC | 22:23 | |
| ericpeterson | https://github.com/openstack/horizon/blob/master/horizon/exceptions.py#L345 thats the bug | 22:23 |
| ericpeterson | exc_handler['exc'] is of type tuple | 22:23 |
| woodm1979 | lol, is that backwards? | 22:23 |
| ericpeterson | its a tuple of various exception types | 22:23 |
| ericpeterson | so it's close.... just not quite | 22:24 |
| woodm1979 | nah, that can be a tuple. | 22:24 |
| openstackgerrit | Thai Tran proposed openstack/horizon: Single page navigation https://review.openstack.org/114392 | 22:24 |
| woodm1979 | Return whether class C is a subclass (i.e., a derived class) of class B. | 22:24 |
| woodm1979 | When using a tuple as the second argument issubclass(X, (A, B, ...)), | 22:24 |
| *** neelashah has joined #openstack-horizon | 22:25 | |
| woodm1979 | damn it, now I have to actually try to fix it too. :-) | 22:25 |
| ericpeterson | yeah, see? | 22:25 |
| ericpeterson | I take it back, I might have found something on my side | 22:28 |
| woodm1979 | LOL | 22:29 |
| ericpeterson | I see stuff that points to monasca issues | 22:30 |
| ericpeterson | which comes from https://github.com/stackforge/monasca-ui/blob/ab0cd448f272dda4ac4f6bb3361254dc44b721f8/monitoring/enabled/_50_admin_add_monitoring_panel.py | 22:30 |
| ericpeterson | could be the plugin thing is also contributing to this? | 22:30 |
| woodm1979 | Eh? Just to handle that exception? | 22:30 |
| woodm1979 | that's … rough. | 22:30 |
| *** neelashah has quit IRC | 22:30 | |
| ericpeterson | the tuple of not found exceptions is (<class 'monascaclient.exc.HTTPNotFound'>,) | 22:30 |
| ericpeterson | so that is a bit incomplete I would say | 22:31 |
| *** tqtran has quit IRC | 22:31 | |
| woodm1979 | Yeah, I think that's a few short. | 22:33 |
| woodm1979 | :-) | 22:33 |
| *** mattfarina has quit IRC | 22:37 | |
| *** Amogh has joined #openstack-horizon | 22:39 | |
| ericpeterson | woodm1979: still around? I think I got it, if you are interested | 22:48 |
| woodm1979 | Yeah, what was the deal? | 22:48 |
| ericpeterson | lhcheng might care too | 22:48 |
| ericpeterson | https://github.com/openstack/horizon/blob/master/openstack_dashboard/utils/settings.py#L113 | 22:48 |
| ericpeterson | that line.... it replaces the exceptions like NotFound with whatever the plugin wanted to add | 22:49 |
| woodm1979 | Aaaah. | 22:49 |
| woodm1979 | Yup. | 22:49 |
| ericpeterson | so like https://github.com/stackforge/monasca-ui/blob/master/monitoring/enabled/_50_admin_add_monitoring_panel.py will be the new value for NotFound | 22:49 |
| woodm1979 | The update is a good idea, except that A) they're tuples, and B) we're updating the larger dict. | 22:50 |
| ericpeterson | it is acting more like REPLACE_EXCEPTIONS instead of ADD_EXCEPTIONS | 22:50 |
| woodm1979 | yup. | 22:50 |
| lhcheng | ericpeterson, good catch. | 22:51 |
| ericpeterson | ping lhcheng | 22:51 |
| ericpeterson | good, your online ;) | 22:51 |
| lhcheng | ericpeterson: tuskar-ui had a workaround on that issue: https://github.com/openstack/tuskar-ui/blob/master/tuskar_ui/exceptions.py | 22:51 |
| woodm1979 | That's aweful. | 22:52 |
| lhcheng | But probably better to just fix it, then tuskar doesn't won't need the hack. | 22:52 |
| lhcheng | woodm1979: yeah, I am surprised that they didn't file a bug to just fix it in horizon. | 22:52 |
| woodm1979 | Not THAT difficult of a fix. | 22:53 |
| ericpeterson | https://bugs.launchpad.net/horizon/+bug/1404032 updated defect | 22:53 |
| ericpeterson | yeah. it's ugly but whatever. I will ping monasca to update too | 22:54 |
| lhcheng | yeah, the workaround will cause problems if there are more than one plug-in adding exception. | 22:54 |
| ericpeterson | but ideally, horizon needs to fix this | 22:54 |
| ericpeterson | yep. horizon can support all the plugins you want, so long as there is one, and it is tuskar ;) | 22:54 |
| lhcheng | ericpeterson: ++ horizon must fix the issue. | 22:55 |
| ericpeterson | I thought this was going to be an easy issue. kinda tricky | 22:58 |
| *** david-ly_ has joined #openstack-horizon | 22:58 | |
| *** packet has quit IRC | 22:58 | |
| woodm1979 | ericpeterson: I don't have either tuskar nor monasca enabled. Does this two-line-fix solve your issue: | 23:01 |
| woodm1979 | for category, exc_list in config.get('ADD_EXCEPTIONS', {}): | 23:01 |
| woodm1979 | exceptions[category] += exc_list | 23:01 |
| woodm1979 | I'm a little concerned that we may get duplicates, but I'm not sure if that's a big deal, nor if it'll happen. | 23:01 |
| ericpeterson | dups are not a big deal imo | 23:02 |
| *** david-lyle has quit IRC | 23:03 | |
| *** mitz has quit IRC | 23:03 | |
| *** flwang has quit IRC | 23:03 | |
| *** dsneddon has quit IRC | 23:03 | |
| lhcheng | if the exception got re-categorize to another, it would result to inconsistent error handling. not sure what is the correct approach though. | 23:03 |
| woodm1979 | We could validate the list once everything is done. | 23:03 |
| ericpeterson | well, if there was code telling it to do that, then that would be the problem also | 23:03 |
| ericpeterson | woodm1979 - your stuff is not quite right | 23:04 |
| ericpeterson | fine tuning | 23:04 |
| ericpeterson | close tho | 23:04 |
| *** jamielennox|away is now known as jamielennox | 23:05 | |
| woodm1979 | derp: forgot .items() | 23:05 |
| woodm1979 | feeling a bit stupid. | 23:05 |
| *** mitz has joined #openstack-horizon | 23:06 | |
| woodm1979 | for category, exc_list in config.get('ADD_EXCEPTIONS', {}).items(): | 23:06 |
| woodm1979 | exceptions[category] += exc_list | 23:06 |
| ericpeterson | closer, still not it ;) | 23:06 |
| ericpeterson | your += needs to work differently | 23:06 |
| *** dsneddon has joined #openstack-horizon | 23:06 | |
| ericpeterson | need to consider when the value is not there yet | 23:06 |
| *** e0ne has quit IRC | 23:07 | |
| ericpeterson | for category, exc_list in config.get('ADD_EXCEPTIONS', {}).iteritems(): | 23:07 |
| ericpeterson | exceptions[category] = exceptions.get(category, ()) + exc_list | 23:07 |
| ericpeterson | something like that | 23:07 |
| ericpeterson | but ultimately, that doesn't work either.... I have it | 23:08 |
| ericpeterson | the problem is exceptions is an empty dict at the top. it should read the default horizon value already there | 23:09 |
| *** colinmcnamara has joined #openstack-horizon | 23:09 | |
| *** flwang has joined #openstack-horizon | 23:10 | |
| woodm1979 | oh yeah, should have used dict.setdefault. | 23:11 |
| woodm1979 | that's what I get for coding things that I'm not actually using. | 23:11 |
| ericpeterson | I give up. not sure what it is, but we don't have the fix yet... still no worky worky | 23:11 |
| *** kumartin has quit IRC | 23:11 | |
| woodm1979 | Is there a reasonable way to enable tuskar or monasca in dev-stack? | 23:12 |
| ericpeterson | I think monasca is not too bad. its an egg, you might even be able to add it via a git requirements line | 23:12 |
| *** e0ne has joined #openstack-horizon | 23:13 | |
| woodm1979 | OK: here's what should happen: exceptions should be a defaultdict | 23:13 |
| woodm1979 | then the defaults get set normally. | 23:13 |
| woodm1979 | >>> from collections import defaultdict | 23:14 |
| woodm1979 | >>> a = defaultdict(tuple) | 23:14 |
| woodm1979 | >>> a['yo'] += ('yo', 'mamma') | 23:14 |
| woodm1979 | >>> a | 23:14 |
| woodm1979 | defaultdict(<type 'tuple'>, {'yo': ('yo', 'mamma')}) | 23:14 |
| woodm1979 | easy peasy | 23:14 |
| * ericpeterson wonders whose momma woodm1979 is talking about | 23:15 | |
| woodm1979 | Clearly the key is "yo" so ... | 23:16 |
| ericpeterson | so is that spanish? would it be your mother? | 23:16 |
| woodm1979 | lol | 23:16 |
| *** e0ne has quit IRC | 23:17 | |
| *** EmilienM is now known as EmilienM|pto | 23:21 | |
| ericpeterson | http://paste.openstack.org/show/152913/ woodm1979 I take it back, I think your fix does work | 23:25 |
| ericpeterson | free bug with solution attached.... https://bugs.launchpad.net/horizon/+bug/1404032 | 23:26 |
| *** lblanchard has joined #openstack-horizon | 23:30 | |
| *** peristeri has joined #openstack-horizon | 23:31 | |
| openstackgerrit | Lin Hua Cheng proposed openstack/django_openstack_auth: Add call to KS V3 revoke_token on logout https://review.openstack.org/142934 | 23:31 |
| *** jgravel_ has joined #openstack-horizon | 23:35 | |
| *** jgravel__ has quit IRC | 23:37 | |
| *** jgravel has joined #openstack-horizon | 23:37 | |
| *** ericpeterson has quit IRC | 23:38 | |
| *** jgravel_ has quit IRC | 23:39 | |
| *** absubram has quit IRC | 23:44 | |
| openstackgerrit | Jamie Lennox proposed openstack/django_openstack_auth: Use keystone auth plugins https://review.openstack.org/121281 | 23:50 |
| *** gyee has joined #openstack-horizon | 23:51 | |
| *** lblanchard has quit IRC | 23:53 | |
| *** tosky has joined #openstack-horizon | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!