| *** lecalcot has joined #murano | 01:18 | |
| *** sergmelikyan has quit IRC | 02:44 | |
| *** rakhmerov has quit IRC | 02:45 | |
| *** ativelkov has quit IRC | 02:46 | |
| *** ativelkov has joined #murano | 02:48 | |
| *** rakhmerov has joined #murano | 02:49 | |
| lecalcot | Hey guys. Is Georgy online? | 03:01 |
|---|---|---|
| lecalcot | Or Maksym? I'm not sure what handles they go by on IRC | 03:03 |
| asalkeld | lecalcot, it's "gokrokve" | 03:05 |
| lecalcot | Thanks, asalkeld. You seem to be the only one active on solum and murano channels tonight. :) | 03:06 |
| asalkeld | lecalcot, it's 'cos I am in Australia:-) | 03:06 |
| asalkeld | and it's Monday here | 03:06 |
| asalkeld | Sunday elsewhere | 03:06 |
| lecalcot | Ha! I'm the only one silly enough to be working on Sunday night. :) | 03:07 |
| *** lazy_prince2 has joined #murano | 05:09 | |
| *** k4n0 has joined #murano | 05:13 | |
| *** lazy_prince2 has quit IRC | 05:19 | |
| *** smurashov has joined #murano | 06:28 | |
| tnurlygayanov | bharath please, use pip instead of bzr | 07:27 |
| tnurlygayanov | bharath we have only first version of Murano in bzr, it was just experiment ;) now we use pip to install Murano, or you can install murano from the git repo (for dashboard it is https://github.com/stackforge/murano-dashboard) | 07:29 |
| *** lazy_prince has quit IRC | 09:23 | |
| *** lazy_prince has joined #murano | 09:41 | |
| *** k4n0 has quit IRC | 09:57 | |
| *** k4n0 has joined #murano | 10:10 | |
| *** k4n0 has quit IRC | 10:35 | |
| *** killer_prince has joined #murano | 11:04 | |
| *** sergmelikyan has joined #murano | 11:06 | |
| *** lazy_prince has quit IRC | 11:07 | |
| *** killer_prince is now known as lazy_prince | 11:07 | |
| *** sergmelikyan has quit IRC | 11:10 | |
| *** lazy_prince2 has joined #murano | 11:11 | |
| *** lazy_prince2 is now known as killer_prince | 11:11 | |
| *** sergmelikyan has joined #murano | 11:12 | |
| *** tnurlygayanov has quit IRC | 11:15 | |
| *** killer_prince has quit IRC | 11:20 | |
| *** asalkeld has quit IRC | 11:59 | |
| *** lazy_prince has quit IRC | 12:05 | |
| sergmelikyan | ruhe, regarding this patch https://review.openstack.org/103990 | 12:09 |
| sergmelikyan | Guys in Horizon released fix that adapts master and IceHouse to .18 release of python-novalclient | 12:10 |
| sergmelikyan | we are going to wait till new version of python-novaclient is released or upgrade tox to install latest master? | 12:11 |
| *** lecalcot has quit IRC | 12:18 | |
| *** killer_prince has joined #murano | 12:21 | |
| *** killer_prince is now known as lazy_prince | 12:21 | |
| sergmelikyan | Guys, I would like to remind that we have 28 bugs that are crave for attention - j.mp/murano-j-bugs :) | 12:34 |
| sergmelikyan | Same with reviews :) | 12:34 |
| sergmelikyan | j2 release are really close, so we need to focus on bugs & review | 12:35 |
| openstackgerrit | A change was merged to stackforge/murano: Unit tests for engine's YAQL functions https://review.openstack.org/104959 | 12:35 |
| sergmelikyan | I think we decided to move murano-app-incubator to murano-apps. Does someone aware about status of this activity? | 12:44 |
| sergmelikyan | ruhe? | 12:44 |
| ruhe | sergmelikyan: sorry, i was afk. reading scrollback now | 12:47 |
| ruhe | sergmelikyan: re python-novaclient. i expect a new hot-fix release of novaclient really soon. it should resolve our problems | 12:48 |
| *** chandankumar has joined #murano | 12:48 | |
| sergmelikyan | I have asked in #o-nova but they didn't answer me :( | 12:49 |
| ruhe | sergmelikyan: re app incubator. this one is one me. i didn't do that yet. can you add an action item for our weekly irc meeting? i'll try to submit a request to infra before the meeting | 12:49 |
| sergmelikyan | ruhe, patch that fixes issue is still not merged to python-novaclient, received -1 from Jenkins on gate | 12:49 |
| sergmelikyan | ruhe, sure | 12:49 |
| ruhe | sergmelikyan: see this discussion. it has details about expected release of python-novaclient - http://lists.openstack.org/pipermail/openstack-dev/2014-July/039566.html | 12:50 |
| sergmelikyan | ruhe, yep, already seen this thread | 12:52 |
| sergmelikyan | on weekend we had very interesting conversation about how Murano works, maybe someone is interested: http://eavesdrop.openstack.org/irclogs/%23murano/%23murano.2014-07-12.log | 13:02 |
| sergmelikyan | starting: 2014-07-12T15:00:43 | 13:02 |
| *** chandankumar has quit IRC | 13:18 | |
| *** chandankumar has joined #murano | 13:18 | |
| sergmelikyan | stan_lagun, I am working on https://bugs.launchpad.net/murano/+bug/1339762 | 13:20 |
| stan_lagun | i know | 13:20 |
| sergmelikyan | I thought that we can introduce HeatResource base class that can handle two things for us: | 13:20 |
| sergmelikyan | 1)resource name generation | 13:20 |
| sergmelikyan | 2)attribute publishing to outputs section (and later clean-up) | 13:21 |
| sergmelikyan | This is generic solution for bug above | 13:21 |
| sergmelikyan | other way is simply add generated name for underlying resource used in Instance | 13:21 |
| stan_lagun | Though not a bad idea this doesn't solve the problem in general. We still need completely different interface to Heat. And once we have it the problem will be solved automatically | 13:23 |
| stan_lagun | So this solution only save you from copy-pasting of the same code (that is also good) | 13:24 |
| sergmelikyan | Yeah, I remember discussion about new interface to interact with Heat... I have only concerns about when it is going to be available, and do this idea usefull in J scope | 13:24 |
| stan_lagun | I thing we must implement it for J3 | 13:24 |
| sergmelikyan | I think so too... | 13:25 |
| stan_lagun | even if you use base class in core library there is no way to enforce this for user apps that can also access Heat | 13:26 |
| sergmelikyan | But, reality can be harsh - j3 cycle is really short and mostly covered by different freezes. | 13:26 |
| sergmelikyan | Can you at least prepare short etherpad + BP describing idea? I think sjmc7 had some thoughts regarding interaction with Heat. Something related to Taskflow (not sure how Taskflow may be fitted here, but anyway) | 13:27 |
| sergmelikyan | We can discuss this today on the meeting | 13:28 |
| stan_lagun | may be. Anyway this will be the next big change. I thing this is priority #1 from our plans. So even if its not J3 then its K1 | 13:28 |
| sergmelikyan | So can you spend 10-15m on etherpad? | 13:29 |
| stan_lagun | sure | 13:29 |
| sergmelikyan | And what about my idea? Should we introduce it now (and refactor out once we have new interface) or should I just make fix for now? | 13:29 |
| sergmelikyan | Introducing it now may have it's own consequences - it's hard to refactor out something like this from Core Library. It is definetly going to be used by app published to create they own packages. | 13:30 |
| sergmelikyan | *publishers | 13:31 |
| stan_lagun | I'd say just generate random name (like in Environment). Although the class may be useful it introduces another abstraction with not so well defined scope and which is likely to be removed in next version. But if you just generate random name in place this will be implementation detail that could be changed without introducing breaking changes | 13:31 |
| sergmelikyan | Yep, I am saying same thing. | 13:32 |
| stan_lagun | btw, I didn't took that bug because you marked it as low-hanging-fruit. A thought you plan to give it away | 13:33 |
| stan_lagun | Also I'm not sure this need to be fixed at all | 13:34 |
| sergmelikyan | Why? | 13:34 |
| sergmelikyan | I have found this bug during real deployment when Object Model had two apps with two instances, but each instance had same name (but different ID through) | 13:35 |
| stan_lagun | Because you are not supposed to have 2 instances with the same name. Instance name should be part of user input. Even if for now dashboard hides it from the user name generation should happen on dashboard side and be removed as soon as entire object model becomes editable there | 13:37 |
| stan_lagun | on engine (MPL) side we need to think how to check values for being unique | 13:38 |
| stan_lagun | but that can be done later | 13:38 |
| sergmelikyan | Yep, but this requirement is not validated and not defined anywhere. I was really confused at first, but found issue very fast. | 13:38 |
| sergmelikyan | sure | 13:38 |
| stan_lagun | we have many things that are not validated. API cannot validate most of the things | 13:39 |
| sergmelikyan | But anyway I think we should fix this issue - it is quite important, and fact that we can't validate it now raises importance even morew | 13:39 |
| sergmelikyan | So as bottom line I suggest to fix this issue like you suggested (I will place comment on this issue), since fix is quite trivial, it will not take too much time to make a patch. | 13:42 |
| stan_lagun | The error need to be fixed. But conceptually generating random names for instances is not a good solution. UI forms have a unit naming pattern (with # chars). It will become useless if you generate random name instead | 13:42 |
| sergmelikyan | hmm... didn't thought about this... | 13:43 |
| sergmelikyan | I am more and more inclined to thought that we shouldn't fix this issue... | 13:44 |
| sergmelikyan | ruhe, guys, what do you think? | 13:44 |
| stan_lagun | can we have the same instance name in 2 different HOT templates within 1 tenant? | 13:46 |
| sergmelikyan | sure | 13:47 |
| sergmelikyan | I think... | 13:47 |
| stan_lagun | Need th check this. This will create 2 identical record in Horizon | 13:48 |
| sergmelikyan | Instance name is appended with stack-id | 13:48 |
| *** sjmc7 has joined #murano | 13:48 | |
| sergmelikyan | sjmc7, morning :) | 13:48 |
| stan_lagun | and what about hostname? | 13:48 |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Refactor functional tests https://review.openstack.org/106767 | 13:50 |
| sjmc7 | hi serg | 13:53 |
| sergmelikyan | stan_lagun, looks like used either name property (we do not set this property) or physical_resource_name with limit of 53 symbol but unique: https://github.com/openstack/heat/blob/master/heat/engine/resource.py#L597 | 13:54 |
| sergmelikyan | sjmc7, we were talking about how to fix this issue and do we need to fix it all: https://bugs.launchpad.net/murano/+bug/1339762 | 13:55 |
| sergmelikyan | And in scope of that issue were discussing with stan_lagun new interface for interaction with Heat | 13:56 |
| *** drupalmonkey has joined #murano | 13:56 | |
| sjmc7 | it doesn't seem like a bug so much as an input problem | 13:56 |
| stan_lagun | serg_melikyan: I remember we set it to short 14-chars length name in Windows | 13:57 |
| sjmc7 | how did you end up finding it? | 13:57 |
| sergmelikyan | Created manually object model while were writing integration tests | 13:57 |
| sergmelikyan | *was | 13:57 |
| sergmelikyan | sjmc7, yep, we come-up with same solution - leave it as it is and mark issue as invalid. Plus document requirement to have unique name for Instance in env scope | 13:58 |
| sergmelikyan | do you agree? | 13:59 |
| sjmc7 | as long as it's possible to create multiple identical instances, yes | 13:59 |
| sjmc7 | e.g. if i have something that requires a master node and several slave nodes | 14:00 |
| sjmc7 | and there's some way to take care of naming | 14:00 |
| stan_lagun | https://blueprints.launchpad.net/murano/+spec/heat-patch-list | 14:02 |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Refactor functional tests https://review.openstack.org/106767 | 14:15 |
| *** lazy_prince is now known as killer_prince | 14:17 | |
| sjmc7 | sergmelikyan, ruhe - https://bugs.launchpad.net/python-novaclient/+bug/1340596 looks like it will fix the problem causing horizon not to function properly with last week's novaclient release | 14:58 |
| sjmc7 | ah... in fact, horizon got fixed on friday | 14:58 |
| sjmc7 | https://review.openstack.org/#/c/106311/ | 14:59 |
| sjmc7 | ruhe - for these kind of failures (e.g. for https://review.openstack.org/#/c/103990/) do we just leave a comment 'recheck bug 1340596'? or is that for CI tests only | 15:02 |
| ruhe | sjmc7: 'recheck bug 1340596' should help; let's try it | 15:05 |
| ruhe | sjmc7: will you add this comment to gerrit? | 15:06 |
| sjmc7 | yeah, will do | 15:15 |
| sjmc7 | was just checking i was following etiquette | 15:15 |
| ruhe | maybe we should document usage somewhere. basically, if you need to re-trigger all the jobs (inluding unit tests, pep8, dsvm jobs and murano-ci) you can comment with "recheck no bug" or "recheck bug xxx"; "recheck murano-ci" will re-trigger only murano-ci | 15:19 |
| sjmc7 | it's documented on the gerrit workflow instructions | 15:33 |
| sjmc7 | it's just not clear whether that's only for gate tests | 15:33 |
| ruhe | sjmc7: murano-ci uses the same infrastructure tools (zuul, nodepool, etc). so it responds on the same commands | 15:39 |
| sjmc7 | yep | 15:39 |
| sjmc7 | ah.. the recheck will fail until this is back-ported to the icehouse release | 15:43 |
| ruhe | sjmc7: i guess we can update horizon dependency (in tox.ini) to horizon-master | 15:46 |
| ruhe | i'm not sure what effect it will have on overall stability though | 15:46 |
| sjmc7 | who knows what else will break :) yeah, i suppose we should | 15:46 |
| sjmc7 | i think they should merge it quite soon | 15:47 |
| sjmc7 | it's not a big change | 15:48 |
| openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-dashboard: [WIP] Update horizon dependency to master https://review.openstack.org/106796 | 15:48 |
| ruhe | let's see how it works :) | 15:48 |
| *** tnurlygayanov has joined #murano | 16:06 | |
| sjmc7 | :D | 16:06 |
| stan_lagun | sjmc7: May I ask you why haven't you fixed https://review.openstack.org/#/c/105545/ ? | 16:11 |
| sjmc7 | oversight. will do it now | 16:11 |
| stan_lagun | It seems like we have broken dashboard in master for a long time already | 16:11 |
| sjmc7 | yeah | 16:11 |
| stan_lagun | I'm wondering why integration tests do not fail | 16:12 |
| sjmc7 | different version perhaps? there is a problem that horizon doesn't put upper bounds on versions | 16:15 |
| sjmc7 | so 105545 may actually break against icehouse, thinking about it | 16:15 |
| sjmc7 | i'll make it deal with both | 16:16 |
| stan_lagun | maybe it would be better check returned tupple length instead of just unpucking it | 16:16 |
| sjmc7 | yep | 16:17 |
| sjmc7 | changing it now | 16:17 |
| stan_lagun | just add [:2] to the end of statement | 16:29 |
| sjmc7 | yeah | 16:32 |
| openstackgerrit | Timur Sufiev proposed a change to stackforge/murano-dashboard: Enhance versionutils.deprecated to work with classes https://review.openstack.org/103502 | 16:38 |
| *** lecalcot has joined #murano | 16:40 | |
| ruhe | seems like upgrade to horizon-master fixes failing unit tests | 16:40 |
| sjmc7 | woo! ship it! | 16:42 |
| openstackgerrit | Timur Sufiev proposed a change to stackforge/python-muranoclient: Provide method for calling action in an environment https://review.openstack.org/99502 | 16:46 |
| ruhe | sjmc7: just need to fix failing murano-ci selenium job. there is a minor issue on jenkins slaves | 16:46 |
| sjmc7 | ok | 16:46 |
| openstackgerrit | Timur Sufiev proposed a change to stackforge/murano-dashboard: Initial implementation of Actions UI https://review.openstack.org/99360 | 16:47 |
| openstackgerrit | Steve McLellan proposed a change to stackforge/murano-dashboard: Expect _prev from glance.image_list_detailed https://review.openstack.org/105545 | 16:53 |
| *** chandankumar has quit IRC | 17:05 | |
| sjmc7 | ruhe - are we deciding to go with horizon master? | 17:17 |
| ruhe | sjmc7: i think nothing prevents us from doing so | 17:18 |
| sjmc7 | except that it'll affect stability | 17:18 |
| sjmc7 | i think i'd prefer to keep it tagged. maybe use juno.1 or something | 17:19 |
| ruhe | there are two options - 1) test against stable release (icehouse) during the development cycle and switch to master (or RC) right before the next stable release 2) test against master | 17:19 |
| ruhe | option 1) will produce a lot of trouble right before the release | 17:20 |
| sjmc7 | yeah | 17:20 |
| sjmc7 | we can go with master and see if it causes problems | 17:20 |
| sjmc7 | can always go back | 17:20 |
| ruhe | sjmc7: your idea seems like a nice 3rd option :) | 17:20 |
| sjmc7 | if we go with the .{1,2,3} releases at least we restrict the problems before releases to the last 6 weeks or so | 17:21 |
| sjmc7 | let's go with master for now | 17:22 |
| sjmc7 | and see if it causes trouble | 17:22 |
| ruhe | sjmc7: all right. murano-ci should be fixed in a couple of minutes. will re-trigger it on my patch then | 17:23 |
| ruhe | i'm experimenting with running selenium tests inside dsvm job. that should help to move selenium jobs to common openstack infrastructure (ci.openstack.org instead of murano-ci) | 17:25 |
| sjmc7 | ok | 17:36 |
| *** lecalcot has quit IRC | 17:48 | |
| *** lecalcot has joined #murano | 17:58 | |
| *** slagun has joined #murano | 18:02 | |
| *** stan_lagun has quit IRC | 18:05 | |
| openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-dashboard: Update horizon dependency to master https://review.openstack.org/106796 | 18:28 |
| ruhe | https://review.openstack.org/#/c/106796/ should be ready to go | 18:28 |
| ruhe | and that's a good time for me to run away and catch the last bus to my home :) | 18:29 |
| *** lecalcot has quit IRC | 19:17 | |
| *** lecalcot has joined #murano | 19:18 | |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Refactor functional tests https://review.openstack.org/106767 | 19:36 |
| sjmc7 | hi katyafervent. for https://review.openstack.org/#/c/106796/, what is it needs updating in settings.py ? | 19:56 |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Refactor functional tests https://review.openstack.org/106767 | 20:19 |
| openstackgerrit | Ekaterina Chernova proposed a change to stackforge/murano-dashboard: Refactor functional tests https://review.openstack.org/106767 | 20:40 |
| *** asalkeld has joined #murano | 20:51 | |
| openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano-dashboard: Update horizon dependency to master https://review.openstack.org/106796 | 21:03 |
| openstackgerrit | Steve McLellan proposed a change to stackforge/murano-dashboard: Expect _prev from glance.image_list_detailed https://review.openstack.org/105545 | 21:05 |
| slagun | sjmc7: maybe reduce images, _more = glance.image_list_detailed(request)[:2] to images = glance.image_list_detailed(request)[0] ? We don't use _more anyway | 21:07 |
| sjmc7 | can do, don't mind | 21:08 |
| openstackgerrit | Steve McLellan proposed a change to stackforge/murano-dashboard: Expect _prev from glance.image_list_detailed https://review.openstack.org/105545 | 21:10 |
| openstackgerrit | A change was merged to stackforge/murano: Improve dsl testing framework foundation https://review.openstack.org/106077 | 21:18 |
| ruhe | sjmc7: re https://review.openstack.org/#/c/106563/ ; it doesn't need an alembic migration; migration already declares this field as sa.Text() - http://git.openstack.org/cgit/stackforge/murano/tree/murano/db/migration/alembic_migrations/versions/001_inital_version.py#n208 | 21:26 |
| sjmc7 | ok! | 21:27 |
| sjmc7 | i retract my -1 | 21:27 |
| ruhe | thanks :) | 21:28 |
| *** lecalcot has quit IRC | 21:28 | |
| sjmc7 | why are you still working? go to sleep | 21:28 |
| *** lecalcot has joined #murano | 21:29 | |
| ruhe | i need to adopt my schedule after world football cup. play-offs started at midnight for us :) | 21:30 |
| sjmc7 | :D | 21:30 |
| sjmc7 | russia didn't get beyond the group stages... | 21:30 |
| *** openstackgerrit has quit IRC | 21:31 | |
| *** openstackgerrit has joined #murano | 21:32 | |
| ruhe | they're not in the list of my favourite teams :) | 21:32 |
| sjmc7 | :) the final was a good game | 21:32 |
| *** killer_prince has quit IRC | 21:34 | |
| ruhe | yes it was, germans deserve gold medals this time (even though they're not in the list of my favourite teams) | 21:34 |
| *** killer_prince has joined #murano | 21:41 | |
| *** killer_prince is now known as lazy_prince | 21:41 | |
| openstackgerrit | A change was merged to stackforge/murano: Fix issue with incorrect model on MySQL https://review.openstack.org/106563 | 22:02 |
| *** lazy_prince has quit IRC | 22:19 | |
| *** killer_prince has joined #murano | 22:23 | |
| *** killer_prince is now known as lazy_prince | 22:24 | |
| *** lecalcot has quit IRC | 22:52 | |
| *** lecalcot has joined #murano | 22:53 | |
| *** lecalcot has quit IRC | 22:57 | |
| *** drupalmonkey has quit IRC | 23:11 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!