| *** chlong has quit IRC | 00:20 | |
| *** chlong has joined #openstack-sahara | 00:20 | |
| *** chlong has quit IRC | 00:41 | |
| *** chlong has joined #openstack-sahara | 00:41 | |
| *** david-lyle has joined #openstack-sahara | 00:42 | |
| *** chlong has quit IRC | 00:48 | |
| *** chlong has joined #openstack-sahara | 00:49 | |
| *** chlong has quit IRC | 00:53 | |
| *** chlong has joined #openstack-sahara | 00:53 | |
| *** chlong has quit IRC | 01:55 | |
| *** chlong has joined #openstack-sahara | 01:55 | |
| *** macjack has quit IRC | 02:05 | |
| *** david-lyle has quit IRC | 02:05 | |
| *** macjack has joined #openstack-sahara | 02:09 | |
| openstackgerrit | Jaxon Wang proposed openstack/sahara: CDH plugin versionhandler refactoring https://review.openstack.org/261192 | 02:24 |
|---|---|---|
| *** chlong has quit IRC | 02:49 | |
| *** chlong has joined #openstack-sahara | 02:49 | |
| *** superflyy has joined #openstack-sahara | 02:50 | |
| *** chlong has quit IRC | 03:17 | |
| *** chlong has joined #openstack-sahara | 03:18 | |
| *** chlong has quit IRC | 03:19 | |
| *** chlong has joined #openstack-sahara | 03:19 | |
| *** david-lyle has joined #openstack-sahara | 03:25 | |
| *** superflyy has quit IRC | 04:01 | |
| *** tingwang has joined #openstack-sahara | 04:51 | |
| *** tellesnobrega is now known as tellesnobrega_af | 05:12 | |
| *** tellesnobrega_af is now known as tellesnobrega | 05:13 | |
| *** hdd has quit IRC | 05:46 | |
| *** rcernin has joined #openstack-sahara | 06:53 | |
| *** tingwang has quit IRC | 07:13 | |
| *** vgridnev has joined #openstack-sahara | 07:17 | |
| *** nkrinner has joined #openstack-sahara | 07:24 | |
| *** tingwang has joined #openstack-sahara | 07:27 | |
| *** tellesnobrega is now known as tellesnobrega_af | 07:34 | |
| *** tingwang has quit IRC | 07:39 | |
| *** esikachev has joined #openstack-sahara | 07:47 | |
| *** tingwang has joined #openstack-sahara | 08:07 | |
| *** witlessb has joined #openstack-sahara | 08:11 | |
| openstackgerrit | lu huichun proposed openstack/sahara: [EDP] Add suspend_job() for sahara edp engine(oozie implementation) https://review.openstack.org/201448 | 08:12 |
| *** AndreyPavlov has joined #openstack-sahara | 08:13 | |
| *** witlessb has quit IRC | 08:15 | |
| *** witlessb has joined #openstack-sahara | 08:19 | |
| openstackgerrit | Jinxing Fang proposed openstack/sahara: Add Sahara-dashboard plugin in devstack local.conf https://review.openstack.org/271688 | 08:31 |
| *** witlessb has quit IRC | 08:33 | |
| *** tmckay has quit IRC | 08:33 | |
| *** witlessb has joined #openstack-sahara | 08:34 | |
| openstackgerrit | lu huichun proposed openstack/sahara: Add release note for deprecate-plugin-vanilla2.6.0 https://review.openstack.org/272428 | 08:40 |
| *** witlessb has quit IRC | 08:43 | |
| *** witlessb has joined #openstack-sahara | 08:47 | |
| *** tmckay has joined #openstack-sahara | 08:50 | |
| *** esikachev has quit IRC | 08:51 | |
| openstackgerrit | lu huichun proposed openstack/sahara: Remove vanilla 2.6.0 code https://review.openstack.org/255111 | 08:59 |
| *** esikachev has joined #openstack-sahara | 09:07 | |
| *** AndreyPavlov has quit IRC | 09:07 | |
| *** witlessb has quit IRC | 09:19 | |
| *** witlessb has joined #openstack-sahara | 09:24 | |
| *** _degorenko|afk is now known as degorenko | 09:56 | |
| *** AndreyPavlov has joined #openstack-sahara | 09:56 | |
| *** witlessb has quit IRC | 09:58 | |
| *** witlessb has joined #openstack-sahara | 10:01 | |
| *** jamielennox is now known as jamielennox|away | 10:15 | |
| *** jamielennox|away is now known as jamielennox | 10:25 | |
| *** witlessb_ has joined #openstack-sahara | 10:31 | |
| *** witlessb has quit IRC | 10:32 | |
| *** witlessb_ is now known as witlessb | 10:32 | |
| *** tosky has joined #openstack-sahara | 10:43 | |
| *** tellesnobrega_af is now known as tellesnobrega | 10:46 | |
| openstackgerrit | Merged openstack/sahara-image-elements: Pass all current env to tox venv https://review.openstack.org/272179 | 10:47 |
| *** tingwang has quit IRC | 10:51 | |
| *** tingwang has joined #openstack-sahara | 10:51 | |
| openstackgerrit | Merged openstack/sahara: Use internal auth url to communicate with swift https://review.openstack.org/269001 | 10:57 |
| *** chlong has quit IRC | 11:05 | |
| *** chlong has joined #openstack-sahara | 11:10 | |
| *** witlessb has quit IRC | 11:29 | |
| *** witlessb has joined #openstack-sahara | 11:30 | |
| *** vgridnev has quit IRC | 11:45 | |
| openstackgerrit | Merged openstack/sahara: Enable sahara-dashboard devstack plugin in sahara plugin https://review.openstack.org/252906 | 12:17 |
| openstackgerrit | Andrey Pavlov proposed openstack/python-saharaclient: Adding ability to unset fields with update calls https://review.openstack.org/272503 | 12:21 |
| *** vgridnev has joined #openstack-sahara | 12:22 | |
| openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config: Add sahara-ci to sahara-scenario https://review.openstack.org/268150 | 12:30 |
| *** witlessb has quit IRC | 12:33 | |
| *** witlessb has joined #openstack-sahara | 12:38 | |
| *** vgridnev has quit IRC | 12:41 | |
| *** vgridnev has joined #openstack-sahara | 12:42 | |
| openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config: Fix ci in saharaclient https://review.openstack.org/272509 | 12:42 |
| *** tmckay has left #openstack-sahara | 12:47 | |
| *** tmckay has joined #openstack-sahara | 12:48 | |
| openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config: Add sahara-ci to sahara-scenario https://review.openstack.org/268150 | 12:52 |
| *** raildo-afk is now known as raildo | 12:59 | |
| *** witlessb has quit IRC | 13:02 | |
| openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config: Add sahara-ci to sahara-scenario https://review.openstack.org/268150 | 13:02 |
| *** witlessb has joined #openstack-sahara | 13:05 | |
| openstackgerrit | Merged openstack/sahara: Fix using regions in all OS clients https://review.openstack.org/269024 | 13:06 |
| *** tingwang has quit IRC | 13:07 | |
| openstackgerrit | Merged openstack/sahara-ci-config: Fix ci in saharaclient https://review.openstack.org/272509 | 13:11 |
| *** raildo is now known as raildo-afk | 13:13 | |
| *** raildo-afk is now known as raildo | 13:14 | |
| *** witlessb has quit IRC | 13:17 | |
| *** witlessb has joined #openstack-sahara | 13:20 | |
| openstackgerrit | Akihiro Motoki proposed openstack/sahara-dashboard: Update devstack plugin https://review.openstack.org/272525 | 13:25 |
| openstackgerrit | Evgeny Sikachev proposed openstack/sahara-scenario: testcommit for ci https://review.openstack.org/272537 | 13:35 |
| *** witlessb has quit IRC | 13:47 | |
| *** witlessb has joined #openstack-sahara | 13:47 | |
| *** witlessb has quit IRC | 14:01 | |
| *** witlessb has joined #openstack-sahara | 14:02 | |
| *** crobertsrh has joined #openstack-sahara | 14:04 | |
| *** tmckay has quit IRC | 14:04 | |
| *** tmckay has joined #openstack-sahara | 14:09 | |
| crobertsrh | vgridnev: Have you noticed that our tests are failing looking for "urlresolvers"? | 14:16 |
| *** witlessb has quit IRC | 14:16 | |
| vgridnev | crobertsrh, yep | 14:16 |
| crobertsrh | I noticed it yesterday and tmckay was just trying to get up and running. | 14:16 |
| vgridnev | And also know reasons | 14:17 |
| crobertsrh | Have you already figured out the fix? | 14:17 |
| crobertsrh | ah, great | 14:17 |
| crobertsrh | I was just about to go looking | 14:17 |
| vgridnev | https://github.com/openstack/horizon/commit/8b2a3417e380a95b3a3b9a20cfbb5b03b9354c15 | 14:17 |
| tmckay | looks like some horizon utils module went missing | 14:17 |
| vgridnev | yes, I was removed in commit above | 14:17 |
| crobertsrh | ok, so it was an intentional change | 14:18 |
| *** witlessb has joined #openstack-sahara | 14:19 | |
| tmckay | that might fix the gate stuff, I guess. But I'm getting this when I try to connect to a running horizon with sahara-dashboard installed: | 14:20 |
| tmckay | blah blah blah | 14:20 |
| tmckay | File "/home/fedora/horizon/.venv/lib/python2.7/site-packages/sahara_dashboard/content/data_processing/job_executions/views.py", line 20, in <module> | 14:20 |
| tmckay | from horizon.utils.urlresolvers import reverse # noqa | 14:20 |
| tmckay | ImportError: No module named urlresolvers | 14:20 |
| tmckay | vgridnev, crobertsrh, seems like a different issue to me ... not a test | 14:21 |
| tmckay | but maybe I misunderstand | 14:22 |
| openstackgerrit | Georgy Dyuldin proposed openstack/sahara-dashboard: Add test 'cluster create-delete with fake plugin' https://review.openstack.org/272558 | 14:28 |
| crobertsrh | vgridnev: I'm guessing we need to change our horizon.utils.urlresolvers to django.core.urlresolvers? Are you already working on a patch, or shall I? | 14:31 |
| vgridnev | I haven't working on this | 14:32 |
| crobertsrh | Ok, I'll take a whack at it. thanks | 14:34 |
| crobertsrh | tmckay, vgridnev: ok, the urlresolvers change takes care of a lot. Now just a few AttributeError: 'module' object has no attribute 'ComplexActionRowRegion' to sort out....any ideas on that vgridnev? | 14:46 |
| *** witlessb has quit IRC | 14:47 | |
| vgridnev | hm, about this I have no ideas right now | 14:47 |
| *** witlessb has joined #openstack-sahara | 14:47 | |
| crobertsrh | Ok, thanks | 14:49 |
| tmckay | crobertsrh, ++, thanks for chasing this | 14:51 |
| crobertsrh | Ok, those errors are likely to be fixed by another patch chain https://review.openstack.org/#/c/262747/4 (I hope) | 14:54 |
| openstackgerrit | Andrey Pavlov proposed openstack/sahara-specs: Distributed periodic tasks spec https://review.openstack.org/272569 | 14:56 |
| openstackgerrit | Chad Roberts proposed openstack/sahara-dashboard: Now using django.core.urlresolvers https://review.openstack.org/272570 | 14:56 |
| crobertsrh | tmckay: ^^^ should get you going. The other errors are in the integration tests and shouldn't affect you. | 14:57 |
| crobertsrh | please review the patch as well. It would be good to get that merged soon. | 14:58 |
| *** witlessb has quit IRC | 14:58 | |
| *** witlessb has joined #openstack-sahara | 14:59 | |
| *** witlessb has quit IRC | 14:59 | |
| *** tosky has quit IRC | 14:59 | |
| *** witlessb has joined #openstack-sahara | 15:00 | |
| *** egafford has joined #openstack-sahara | 15:01 | |
| *** tosky has joined #openstack-sahara | 15:01 | |
| *** tmckay has quit IRC | 15:02 | |
| *** tmckay has joined #openstack-sahara | 15:09 | |
| tmckay | crobertsrh, thanks ^^ | 15:09 |
| crobertsrh | no prob. Let me know how it works for you | 15:09 |
| *** AndreyPavlov has quit IRC | 15:30 | |
| openstackgerrit | Jinxing Fang proposed openstack/sahara: Update the devstack.rst document https://review.openstack.org/271688 | 15:39 |
| openstackgerrit | Andrey Pavlov proposed openstack/sahara-specs: Distributed periodic tasks spec https://review.openstack.org/272569 | 16:07 |
| *** vgridnev has quit IRC | 16:08 | |
| *** vgridnev has joined #openstack-sahara | 16:09 | |
| crobertsrh | vgridnev: It looks like the integration test stuff is also a problem for https://review.openstack.org/272570 to be able to pass. I think that tsufiev's patches to the integration tests are also needed, but it looks like those are also failing to pass right now. Any ideas on how we should un-wedge ourselves? | 16:19 |
| crobertsrh | also, should those integration tests be run as part of py27? | 16:19 |
| *** AndreyPavlov has joined #openstack-sahara | 16:22 | |
| crobertsrh | I need to head out for lunch. Will be back soon. | 16:25 |
| *** chlong is now known as chlong_zzz | 16:28 | |
| *** crobertsrh has quit IRC | 16:31 | |
| *** nkrinner has quit IRC | 16:51 | |
| *** vgridnev has quit IRC | 16:51 | |
| *** hdd has joined #openstack-sahara | 17:05 | |
| *** tellesnobrega is now known as tellesnobrega_af | 17:06 | |
| *** esikachev has quit IRC | 17:11 | |
| *** tellesnobrega_af is now known as tellesnobrega | 17:15 | |
| *** openstackgerrit has quit IRC | 17:17 | |
| *** openstackgerrit has joined #openstack-sahara | 17:18 | |
| *** vgridnev has joined #openstack-sahara | 17:24 | |
| *** jriveram has joined #openstack-sahara | 17:26 | |
| *** crobertsrh has joined #openstack-sahara | 17:34 | |
| *** vgridnev has quit IRC | 17:43 | |
| *** vgridnev has joined #openstack-sahara | 17:46 | |
| crobertsrh | SergeyLukjanov, vgridnev: Question about https://review.openstack.org/272570 It's currently failing due to some integration tests, but I thought that the integration tests were to be run separately from the unit tests (there's a separate job for them). The patch is badly needed since horizon won't even start with the sahara-dashboard plugin at the moment. Any thoughts on what a sensible step would be? | 17:46 |
| *** raildo is now known as raildo-afk | 17:47 | |
| *** tellesnobrega is now known as tellesnobrega_af | 17:48 | |
| *** raildo-afk is now known as raildo | 17:49 | |
| vgridnev | crobertsrh, I think that we should run integration tests separately | 17:50 |
| crobertsrh | Yeah, I agree. There is already a separate job for them. Somehow, they are also running as part of gate-sahara-dashboard-python27 | 17:51 |
| crobertsrh | I'm a little unfamiliar with how/where they are configured. | 17:51 |
| *** tellesnobrega_af is now known as tellesnobrega | 17:51 | |
| vgridnev | actually it's a little bit strange, I think | 17:52 |
| vgridnev | maybe we have some kind of unit tests for integration tests? | 17:53 |
| tmckay | http://paste.openstack.org/show/485049/ | 17:53 |
| tmckay | something wrong with the path, or is there some way to configure the test loader? | 17:54 |
| tmckay | looks to me from 10,000 feet that it's just reading stuff out of that dir ... | 17:54 |
| tmckay | crobertsrh, ^^ | 17:54 |
| crobertsrh | yeah, it does appear to be grabbing all of those from that dir. It's a newer behavior though. Tests were fine until about yesterday, I think. | 17:55 |
| crobertsrh | Of course, that was when I noticed it...could have been broken for a bit longer, I guess. | 17:55 |
| tmckay | yeah, there are api_tests and integration_tests in there. Hmm | 17:55 |
| elmiko | crobertsrh: if you are curious about the testing infra, i *think* this is the repo you want to checkout https://github.com/openstack-infra/project-config/ | 17:55 |
| elmiko | look at the jenkins subdir | 17:55 |
| *** degorenko is now known as _degorenko|afk | 17:56 | |
| vgridnev | elmiko, tmckay actually we don't have any dirs in tox.ini : https://github.com/openstack/sahara-dashboard/blob/master/tox.ini#L17 | 17:58 |
| vgridnev | should we specify something? | 17:58 |
| crobertsrh | Yeah, I was just discovering that | 17:58 |
| crobertsrh | Maybe? | 17:58 |
| crobertsrh | ./run_tests.sh yields "nosetests sahara_dashboard --nocapture --nologcapture --cover-package=openstack_dashboard --cover-inclusive --all-modules --verbosity=1" | 17:59 |
| *** hdd has quit IRC | 17:59 | |
| tmckay | crobertsrh, I wonder if the sahara repo is analagous | 18:00 |
| tmckay | analogous, sorry :) | 18:00 |
| crobertsrh | something must have changed somewhere...these tests had to have passed at some point, right? | 18:01 |
| elmiko | i would *hope* so ;) | 18:01 |
| crobertsrh | Oh right...I think the horizon table action bits changed underneath....causing these tests to fail. | 18:01 |
| tmckay | I would think so. do we know the root cause of the ComplexRowAction | 18:01 |
| crobertsrh | sorry....replaying things in my mind | 18:01 |
| crobertsrh | There is a patch that would possibly fix that, but that patch is currently failing tests as well. | 18:02 |
| crobertsrh | But, ideally, I don't think these tests should be run as part of gate*python27 | 18:02 |
| *** Taz_ has joined #openstack-sahara | 18:04 | |
| rickflare | guys allow me introduce you to Taz. He is a awesome dev and cloud engineer who works with me. He is going to be helping with testing and more soon! | 18:05 |
| tmckay | welcome! | 18:05 |
| Taz_ | hey im a newb | 18:05 |
| rickflare | to sahara yes but to Hadoop he is not. | 18:06 |
| crobertsrh | welcome Taz_ | 18:06 |
| Taz_ | thanks! | 18:07 |
| Taz_ | waiting for devstack to install...takes forever | 18:08 |
| crobertsrh | Heh | 18:09 |
| rickflare | so I finished my RDO install using VM's | 18:09 |
| crobertsrh | everything working smoothly? | 18:09 |
| rickflare | what the directions fail to tell you is that once you are done you will not be able to access horizon without serveral layers of ssh tunnels | 18:09 |
| *** raildo is now known as raildo-afk | 18:10 | |
| rickflare | crobertsrh yes everything install as expected | 18:10 |
| rickflare | but accesses it is a major PITA | 18:10 |
| rickflare | and i cant using the bare metal install | 18:10 |
| rickflare | because all my systems dont have ipmi | 18:10 |
| crobertsrh | bare metal pains have been felt by tmckay recently. You guys could swap war stories, I'm sure. | 18:11 |
| rickflare | I was talking to Taz and one of the things we feel holding up community dev is just getting this up and running | 18:12 |
| rickflare | I found some issues with the documentation and I submitted fixes | 18:12 |
| rickflare | but there is still some rough edges | 18:12 |
| *** esikachev has joined #openstack-sahara | 18:12 | |
| rickflare | the biggest area is just detailed explanation of networking | 18:13 |
| rickflare | and how it should be setup in most environments | 18:13 |
| *** raildo-afk is now known as raildo | 18:13 | |
| rickflare | if that is covered I think it would address 90% of the issues I ran into | 18:13 |
| crobertsrh | Yeah, the networking stuff is really complex in the "non-devstack" world | 18:13 |
| crobertsrh | any feedback/patches are certainly welcomed. It's good to hear back from actual users. | 18:13 |
| rickflare | whats worse crobertsrh is I actually understand the openstack networking | 18:13 |
| crobertsrh | Oh dear. my condolences :) | 18:14 |
| rickflare | whats not explained is how you should setup routing | 18:14 |
| rickflare | like when install RDO they are like ok you can use your network settings here blah blah | 18:14 |
| rickflare | with no mention of like well if you do this you can or can not route to your vms or the vms wont be able to route out to your real network etc | 18:15 |
| rickflare | I was telling egafford how maddening it is | 18:15 |
| rickflare | to have this running and can not route to anything | 18:15 |
| crobertsrh | I wonder if there are just too many possibilities to document them all or if the doc writer was just lazy | 18:15 |
| crobertsrh | Yes, running and not routable is pretty much == not running :) | 18:15 |
| rickflare | well I think it should cover how most will test this | 18:15 |
| rickflare | 99% will have a network they dont control | 18:16 |
| rickflare | IE ISP | 18:16 |
| rickflare | and they have a box that is running kvm and runs the RDO manager etc | 18:16 |
| rickflare | base it off that | 18:16 |
| rickflare | that way when folks want to do more advanced stuff they have a solid baseline. | 18:16 |
| rickflare | Does that make sense? | 18:16 |
| crobertsrh | I think so. I've never had a setup like that, but I certainly understand what you're getting at. | 18:17 |
| rickflare | when you deploy RDO how do you do it? | 18:18 |
| rickflare | or are you strickly a devstack users | 18:18 |
| *** tosky has quit IRC | 18:19 | |
| crobertsrh | I have only done upstream devstack-based dev recently. | 18:20 |
| crobertsrh | When I did use RDO, it was a very basic "try it out" kind of install. | 18:20 |
| egafford | rickflare: This room is going to be very devstack-centric, really. | 18:31 |
| *** witlessb has quit IRC | 18:33 | |
| rickflare | ahh | 18:37 |
| rickflare | ok | 18:37 |
| egafford | #rdo and #tripleo are more focused on the RDO codebase and its installer. | 18:38 |
| rickflare | 10-4 | 18:39 |
| *** vgridnev has quit IRC | 18:45 | |
| *** witlessb has joined #openstack-sahara | 18:47 | |
| *** akuznetsov has joined #openstack-sahara | 18:48 | |
| *** vgridnev has joined #openstack-sahara | 18:48 | |
| *** vgridnev has quit IRC | 18:48 | |
| *** hdd has joined #openstack-sahara | 19:09 | |
| *** akuznetsov has quit IRC | 19:10 | |
| openstackgerrit | Chad Roberts proposed openstack/sahara-dashboard: Fixing imports and no longer always running int tests https://review.openstack.org/272700 | 19:17 |
| crobertsrh | tmckay, elmiko, egafford, vgridnev: Ok, please take a look at ^^^. It combines the 2 needed fixes to get things flowing again. The urlresolvers fix from this morning and a fix to stop running the integration tests when they should not be. | 19:19 |
| crobertsrh | I had to combine the 2 fixes because neither one would pass tests without the other one.....fun :) | 19:20 |
| elmiko | crobertsrh: ack, taking a look | 19:21 |
| egafford | crobertsrh: https://review.openstack.org/#/c/272700/1/sahara_dashboard/content/data_processing/job_binaries/views.py just removes the import without adding a new one. Is this awesome? | 19:21 |
| crobertsrh | Yeah....a bit of fun there | 19:22 |
| egafford | So fun. | 19:22 |
| crobertsrh | evidently, we just imported the other stuff on top of the good stuff | 19:22 |
| crobertsrh | bonus bug fix | 19:22 |
| elmiko | lol | 19:22 |
| egafford | crobertsrh: So the contract is that if (either with_selenium or integration is true,) then the integration tests will run? | 19:25 |
| crobertsrh | Yeah...same in horizon run_tests.sh | 19:25 |
| *** jriveram has quit IRC | 19:31 | |
| *** esikachev has quit IRC | 19:50 | |
| *** rcernin has quit IRC | 19:58 | |
| openstackgerrit | Timur Sufiev proposed openstack/sahara-dashboard: Adapt for all horizon i9n refactoring changes https://review.openstack.org/262747 | 20:05 |
| *** vgridnev has joined #openstack-sahara | 20:06 | |
| crobertsrh | elmiko, tmckay, egafford: tests passed....should we merge it (considering it's totally broken atm)? | 20:10 |
| elmiko | sounds good to me | 20:11 |
| elmiko | it's got +5 currently, i think that should be fine | 20:12 |
| elmiko | done and done... | 20:12 |
| crobertsrh | cool | 20:12 |
| crobertsrh | thanks | 20:12 |
| elmiko | and thank you for putting it together so quickly =) | 20:13 |
| crobertsrh | 'twas a learning experience for sure :) | 20:13 |
| *** egafford has quit IRC | 20:14 | |
| openstackgerrit | Merged openstack/sahara-dashboard: Fixing imports and no longer always running int tests https://review.openstack.org/272700 | 20:14 |
| openstackgerrit | Chad Roberts proposed openstack/sahara-dashboard: Adding support for editing active cluster shares https://review.openstack.org/253223 | 20:18 |
| openstackgerrit | Chad Roberts proposed openstack/sahara-dashboard: Update devstack plugin https://review.openstack.org/272525 | 20:20 |
| vgridnev | can we bring more attention for this one: https://review.openstack.org/#/c/270156/ ? | 20:29 |
| elmiko | vgridnev: ack, i'll take another look | 20:30 |
| *** rcernin has joined #openstack-sahara | 20:32 | |
| *** flwang has left #openstack-sahara | 20:33 | |
| *** crobertsrh has quit IRC | 20:34 | |
| tmckay | vgridnev, I'll take a look too | 20:34 |
| tmckay | also, guys, I wonder if we should revisit this https://review.openstack.org/#/c/232067/ | 20:35 |
| *** AndreyPavlov has quit IRC | 20:36 | |
| tmckay | no substring filtering is a real failing, imho. I wonder if it's worth the (I think) low risk of a change in behavior | 20:36 |
| *** crobertsrh has joined #openstack-sahara | 20:37 | |
| tmckay | crobertsrh, while you were disconnected I posted this: | 20:40 |
| tmckay | 03:35:19 PM) tmckay: also, guys, I wonder if we should revisit this https://review.openstack.org/#/c/232067/ | 20:40 |
| tmckay | (03:36:27 PM) tmckay: no substring filtering is a real failing, imho. I wonder if it's worth the (I think) low risk of a change in behavior | 20:40 |
| crobertsrh | ah, yes. I remember it fondly | 20:41 |
| tmckay | if we wait for v2, it may be a really long time | 20:41 |
| crobertsrh | Yeah, I remember originally, I was all for just doing it. | 20:41 |
| crobertsrh | The concern was for people that might have come to rely on the broken behavior | 20:42 |
| tmckay | elmiko, ^^ | 20:42 |
| crobertsrh | punting to v2 seems like it would only encourage more people to adapt to the brokenness | 20:42 |
| tmckay | yeah. but if we say that filtering was broken, ie it was a bug, and this is the correct behavior ... | 20:42 |
| tmckay | ooo, crazy idea. I wonder if we could provide a config to keep it busted | 20:43 |
| tmckay | defaulted to "filter" but you could set it to "broken" | 20:43 |
| elmiko | hmm | 20:43 |
| elmiko | well, we aren't changing the api contract here, we are just "fixing" a broken search impl. i would think we could make a case for just fixing it | 20:44 |
| tmckay | yeah, your result set may change for a particular search | 20:44 |
| elmiko | i agree about documenting the change *thoroughly* | 20:45 |
| tmckay | but the current situation -- search for exact match, and exact match only, has got to be broken by anyone's estimation | 20:45 |
| elmiko | +1 | 20:45 |
| tmckay | after all, if you make the search string long enough, you can still have exact match. | 20:45 |
| elmiko | i'd say start by making a bug, if we don't aleady have one | 20:45 |
| tmckay | Find the thing you're looking for, cut and paste the name, and then search :-D | 20:46 |
| * elmiko doesn't remember if we already have a bug | 20:46 | |
| elmiko | haha | 20:46 |
| *** esikachev has joined #openstack-sahara | 20:46 | |
| elmiko | at they very least, fixing it to behave more like a substring match would be great imo | 20:46 |
| tmckay | I'll check for a bug. Thanks. I think it's worth serious consideration of getting this in Mitaka | 20:46 |
| elmiko | i'm good with that | 20:46 |
| tmckay | k, thanks | 20:47 |
| tmckay | https://bugs.launchpad.net/sahara/+bug/1503345 | 20:48 |
| openstack | Launchpad bug 1503345 in Sahara "Filtering queries only works for exact matches" [High,In progress] - Assigned to Trevor McKay (tmckay) | 20:48 |
| tmckay | elmiko, ^^ | 20:48 |
| elmiko | awesome | 20:50 |
| elmiko | looks like you posted a fix too ;) | 20:50 |
| elmiko | vgridnev: i have one small nit with the health check spec, otherwise lgtm | 20:50 |
| *** esikachev has quit IRC | 20:51 | |
| openstackgerrit | Vitaly Gridnev proposed openstack/sahara-specs: Cluster verifications spec https://review.openstack.org/270156 | 20:52 |
| tmckay | elmiko, does the API working group have guidelines for filtering? I bet they do | 20:53 |
| * tmckay goes to look | 20:53 | |
| *** raildo is now known as raildo-afk | 20:54 | |
| vgridnev | tmckay, I just revisited your change, I think that it's okay to accept that to mitaka, and also we should definitely avoid undocumented places. | 20:54 |
| elmiko | tmckay: http://specs.openstack.org/openstack/api-wg/guidelines/pagination_filter_sort.html | 20:55 |
| tmckay | elmiko, "TODO: Add guidance on a “LIKE” or regex operator to search text." lol | 20:56 |
| tmckay | we can be trailblazers | 20:56 |
| *** hdd has quit IRC | 20:59 | |
| elmiko | yea! | 21:01 |
| openstackgerrit | Merged openstack/sahara-dashboard: Adding support for editing active cluster shares https://review.openstack.org/253223 | 21:01 |
| elmiko | tmckay: if you come up with something good, propose it as a guideline | 21:01 |
| *** crobertsrh has quit IRC | 21:04 | |
| tmckay | elmiko, gonna play around with Nova, I did once but I forget what I found :) | 21:04 |
| elmiko | k | 21:06 |
| elmiko | tmckay: it may also be interesting to see how searchlight does this? | 21:07 |
| openstackgerrit | Vitaly Gridnev proposed openstack/sahara-dashboard: ACL support in Sahara UI https://review.openstack.org/253418 | 21:07 |
| elmiko | assuming they expose an api | 21:07 |
| tmckay | elmiko, ack, good idea | 21:09 |
| tmckay | elmiko, yeah, nova-api.log shows "1f953c83a044bf4b940335321dc6514/servers/detail?limit=21&name=ha&project_id=91f953c83a044bf4b940335321dc6514", that was a filter on name "ha" and it worked as a substring | 21:12 |
| elmiko | cool | 21:12 |
| tmckay | so, no indication in the url that it is anything other than literal equality (but it's not) | 21:12 |
| elmiko | that doesn't seem too out of whack, it's just a matter of how the server interprets that search command | 21:13 |
| tmckay | yeah. that's our case, too. | 21:13 |
| * tmckay goes off to check the nova code again | 21:13 | |
| elmiko | i mean, if we *really* want to preserve exact match, just make the parameter partial_name or something? | 21:13 |
| elmiko | or add an exact_name param | 21:14 |
| elmiko | or even an exact_match boolean | 21:14 |
| *** Erming_ has quit IRC | 21:14 | |
| elmiko | i think that last one is my fav. | 21:14 |
| *** esikachev has joined #openstack-sahara | 21:14 | |
| elmiko | name=<some name>, which results in a substring match | 21:14 |
| elmiko | add exact=True to get an exact match | 21:14 |
| elmiko | so, like. .../some/resource?name=foo&exact=True | 21:15 |
| tmckay | ooo | 21:15 |
| elmiko | or somesuch... | 21:15 |
| tmckay | that might give us a way to sorta preserve compat in the client | 21:17 |
| elmiko | yea, i mean if you want total backward compat, just add a subtring=True param? | 21:17 |
| elmiko | i would question if we need to keep backward compat here though | 21:18 |
| tmckay | me too | 21:18 |
| elmiko | right, so given that, my preference would be to add an "exact=" param (if needed) | 21:19 |
| tmckay | oh man, nova allows \ continuation lines. Barbarians. | 21:19 |
| elmiko | hehe | 21:19 |
| tmckay | it's really jarring now that I've been in Sahara land :) | 21:19 |
| *** esikachev has quit IRC | 21:21 | |
| elmiko | stranger in a strange land? | 21:22 |
| *** hdd has joined #openstack-sahara | 21:23 | |
| tmckay | elmiko, yeah, the nova stuff is much more complicated, with same fields set to exact, and some set to regex, and then the regex op is based on the database connection type. But there is a "like" case that just uses %value%, like our simple patch | 21:26 |
| tmckay | "same fields" => "some fields" | 21:27 |
| elmiko | ah, interesting | 21:27 |
| tmckay | so, we could start with simple substring matches where appropriate and then build out filtering if necessary | 21:27 |
| elmiko | probably best if we just stick to answering the "exact vs like" debate first ;) | 21:27 |
| openstackgerrit | Vitaly Gridnev proposed openstack/sahara-dashboard: ACL support in Sahara UI https://review.openstack.org/253418 | 21:27 |
| tmckay | oh yeah | 21:27 |
| *** rcernin has quit IRC | 21:28 | |
| elmiko | i could see us defaulting to "like" style match | 21:28 |
| elmiko | especially with things like searchlight gaining more traction | 21:28 |
| elmiko | i have a feeling that anyone writing a client for sahara will be more affected by our choice than end-users who are using sahara through horizon | 21:28 |
| *** hdd has quit IRC | 21:29 | |
| tmckay | yes. I think horizon uses would say "hey! they fixed it!" | 21:29 |
| elmiko | haha | 21:29 |
| elmiko | could be ;) | 21:30 |
| elmiko | like, i totally get why nova would need a very robust filtering api | 21:30 |
| elmiko | but, there is more than just horizon consuming nova. many apps are written to interact with nova through the client apis | 21:30 |
| elmiko | but for sahara, we don't have the same level of usage that nova does | 21:31 |
| elmiko | (although we can dream) | 21:31 |
| *** vgridnev has quit IRC | 21:33 | |
| *** jriveram has joined #openstack-sahara | 21:43 | |
| *** egafford has joined #openstack-sahara | 21:56 | |
| tmckay | elmiko, ack, I don't think we need anything fancy | 22:04 |
| *** Erming_ has joined #openstack-sahara | 22:46 | |
| *** egafford has quit IRC | 22:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!