*** takashin has joined #openstack-placement | 00:00 | |
*** mriedem has quit IRC | 00:05 | |
*** tetsuro has joined #openstack-placement | 00:21 | |
*** efried has joined #openstack-placement | 01:15 | |
*** efried has quit IRC | 02:10 | |
*** efried1 has joined #openstack-placement | 02:10 | |
*** efried1 is now known as efried | 02:12 | |
openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: Follow up for placement usage document https://review.openstack.org/607090 | 02:43 |
---|---|---|
*** e0ne has joined #openstack-placement | 05:06 | |
*** e0ne has quit IRC | 05:06 | |
*** zaneb has quit IRC | 05:45 | |
*** zaneb has joined #openstack-placement | 05:50 | |
*** rubasov has quit IRC | 05:56 | |
*** dims has quit IRC | 06:38 | |
*** dims has joined #openstack-placement | 06:44 | |
*** dims has quit IRC | 06:48 | |
*** dims has joined #openstack-placement | 06:51 | |
*** tetsuro has quit IRC | 06:53 | |
*** e0ne has joined #openstack-placement | 07:00 | |
*** helenafm has joined #openstack-placement | 07:12 | |
*** tetsuro has joined #openstack-placement | 07:38 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Initial integration test using gabbits https://review.openstack.org/601614 | 08:02 |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 08:02 |
*** zaneb has quit IRC | 08:12 | |
*** zaneb has joined #openstack-placement | 08:12 | |
openstackgerrit | Merged openstack/placement master: Fix aggregate members in nested alloc candidates https://review.openstack.org/602639 | 08:13 |
openstackgerrit | Merged openstack/placement master: Add a document for allcation candidates constraint https://review.openstack.org/604243 | 08:13 |
*** ttsiouts has joined #openstack-placement | 08:51 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 08:54 |
*** cdent has joined #openstack-placement | 08:56 | |
*** ttsiouts has quit IRC | 09:00 | |
*** takashin has left #openstack-placement | 09:02 | |
*** ttsiouts has joined #openstack-placement | 09:02 | |
*** ttsiouts has quit IRC | 10:01 | |
*** ttsiouts has joined #openstack-placement | 10:03 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 10:38 |
*** cdent has quit IRC | 10:48 | |
openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: Generate sample policy in placement directory https://review.openstack.org/601866 | 11:01 |
*** ttsiouts has quit IRC | 11:03 | |
*** ttsiouts has joined #openstack-placement | 11:03 | |
*** ttsiouts has quit IRC | 11:08 | |
*** e0ne has quit IRC | 11:10 | |
*** ttsiouts has joined #openstack-placement | 11:13 | |
*** tetsuro has quit IRC | 11:18 | |
openstackgerrit | Merged openstack/placement master: max-complexity=>16: refactor GET /a_c qs parsing https://review.openstack.org/602667 | 11:22 |
*** cdent has joined #openstack-placement | 11:23 | |
openstackgerrit | Merged openstack/placement master: De-nova-ify and reformat contributor guide https://review.openstack.org/599864 | 11:24 |
*** ttsiouts has quit IRC | 11:25 | |
openstackgerrit | Merged openstack/placement master: Fix link from root doc to contributor guide https://review.openstack.org/603431 | 11:27 |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 11:37 |
*** e0ne has joined #openstack-placement | 11:50 | |
*** ttsiouts has joined #openstack-placement | 12:10 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 12:35 |
*** mriedem has joined #openstack-placement | 12:46 | |
cdent | the feedback latency from zuul experiments is high enough that I'm starting to consider setting up zuul in my house | 12:48 |
cdent | but it's not like i need yet another thing to do | 12:48 |
cdent | jaypipes: thanks for the reviews on the integration test stuff. I've tried to explain the magic yaml stuff a bit more. Once concern with those changes is they change the length of a check run from 5 minutes or so, to 30 minutes. which is inevitable but a bummer | 12:49 |
jaypipes | cdent: I am in awe of your YAMLness. :) | 12:52 |
cdent | it helps to know the authors | 12:52 |
jaypipes | heh | 12:52 |
cdent | although they extract a high price | 12:52 |
jaypipes | cdent: I wonder why the change to using tempest plugin (vs the bash scripts) expands the runtime from 5 to 30 minutes? is it because of all the extra git cloning? | 12:53 |
cdent | oh, no, that's not what I mean | 12:53 |
jaypipes | oh, you mean just including this in the check pipeline... gotcha | 12:53 |
*** tssurya has joined #openstack-placement | 12:53 | |
cdent | I mean that having an integration test in addition to the existing docs, pep8, unit, functional tests change... yeah | 12:53 |
jaypipes | cdent: totally worth it, though. | 12:53 |
cdent | yeah, I figured fair price | 12:54 |
openstackgerrit | Merged openstack/placement master: Follow up for placement usage document https://review.openstack.org/607090 | 13:01 |
efried | cdent: Have you seen/processed Bence's email in the "intended purpose" thread? | 13:19 |
cdent | efried: I read it, but haven't fully processed | 13:19 |
efried | Of note, in the patch he references, they're inferring "ownership" based on provider UUID being in the binding profile: https://review.openstack.org/#/c/574783/15/neutron/plugins/ml2/managers.py@888 | 13:19 |
efried | which kind of goes with what you were suggesting yesterday about determining ownership by having stored the provider in some kind of database beforehand. | 13:20 |
* cdent nods | 13:21 | |
*** ttsiouts has quit IRC | 13:23 | |
openstackgerrit | Merged openstack/placement master: Add nova database migration script https://review.openstack.org/603234 | 13:23 |
*** e0ne has quit IRC | 13:32 | |
*** ttsiouts has joined #openstack-placement | 13:33 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Remove placement.db.migration https://review.openstack.org/607185 | 13:34 |
openstackgerrit | Chris Dent proposed openstack/placement master: Remove support for multiple database from migration.py https://review.openstack.org/607186 | 13:34 |
openstackgerrit | Chris Dent proposed openstack/placement master: WIP: add a table create CLI https://review.openstack.org/600161 | 13:37 |
*** cdent has quit IRC | 13:37 | |
*** cdent has joined #openstack-placement | 13:47 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Add placeload to integration test https://review.openstack.org/602484 | 13:58 |
*** ttsiouts has quit IRC | 13:58 | |
*** mriedem has quit IRC | 14:00 | |
*** e0ne has joined #openstack-placement | 14:02 | |
*** mriedem has joined #openstack-placement | 14:05 | |
*** ttsiouts has joined #openstack-placement | 14:12 | |
*** ttsiouts has quit IRC | 14:46 | |
*** ttsiouts has joined #openstack-placement | 15:00 | |
*** e0ne has quit IRC | 15:19 | |
*** cdent has quit IRC | 15:27 | |
*** ttsiouts has quit IRC | 15:42 | |
*** cdent has joined #openstack-placement | 16:03 | |
*** helenafm has quit IRC | 16:03 | |
smcginnis | Howdy placers. I'm looking at Forum submissions. One of them is from someone talking about how there is osc-placement for CLI use, but that there is not a placement Python library that can be used to work with placement. Is that correct? | 16:46 |
cdent | smcginnis: that's only sort of true. you can use the sdk to make a connection to placement easily, and then pass json back and forth. | 16:48 |
cdent | there's open discussion on the value of a python library for encapsulating that json handling | 16:48 |
smcginnis | cdent: Ah, OK. So this may be a decent forum topic? | 16:49 |
cdent | I tend to think it is overkill. As in: if you need a library to manage your resources in an API, then you've made a bad api | 16:49 |
cdent | but I'm weird like that | 16:49 |
cdent | smcginnis: maybe? efried mriedem or jaypipes probably have a stronger opinion since they are less "weird like that" | 16:50 |
smcginnis | It did surprise me since I thought it was supposed to all be through a simple REST API. | 16:50 |
efried | smcginnis: I would be interested in talking about it. I personally do see a certain amount of value in a client lib, but also get cdent's point. Depends how far you want to go. | 16:51 |
efried | smcginnis: Like, if all your client is doing is factoring out a few common bits and pieces like headers, then yeah, just use the sdk. | 16:52 |
smcginnis | Seems like there's a sufficient amount to talk about to warrant a Forum session. | 16:52 |
smcginnis | efried, cdent: Thanks! | 16:52 |
efried | smcginnis: If it's wrapping each operation in an appropriately-named method, less so - but if you're still passing json in and out, then what are you gaining? | 16:53 |
cdent | smcginnis: it will likely munge into "what are best practices for talking to placement" which is worth talking about | 16:53 |
smcginnis | Indeed. Might be a good place to hash that out and get it written down. | 16:54 |
efried | smcginnis: But if you're talking about the lib exposing object-y encapsulations of the data and passing those around instead of straight json, that's definitely something to lib-ify. Question is, is there really a want/need for that? | 16:54 |
efried | [end] | 16:54 |
smcginnis | efried: Yeah, I would hope it's just factoring out a few common helpers from my limited understanding of the needs. | 16:54 |
jaypipes | smcginnis: I won't be in Berlin, just FYI. I think there might be some value in moving ProviderTree and *some* bits of reportclient foo into a library, along with the exception classes/parsing. | 16:55 |
smcginnis | And of course, full processing of trait name meanings. | 16:55 |
* smcginnis runs away for cover | 16:55 | |
* efried armbars smcginnis | 16:55 | |
smcginnis | *tap | 16:55 |
* jaypipes lassos smcginnis with placement rope of imprisonment | 16:58 | |
smcginnis | Hah! | 16:58 |
mriedem | cdent: smcginnis: i left comments on that submission last week | 16:58 |
mriedem | saying it's intentional that we don't hae a python-placement library | 16:58 |
cdent | who made the submission? | 16:59 |
smcginnis | Doesn't look like the chair tool captured any of that. | 16:59 |
smcginnis | Tetsuro Nakamura from NTT. | 16:59 |
smcginnis | Er, that's the moderator. Looks like actual submitter was Pierre Riteau | 17:00 |
mriedem | smcginnis: link to the submission? | 17:00 |
smcginnis | Can you access this? https://www.openstack.org/track-chairs/browse/22806?category=262 | 17:00 |
mriedem | oo | 17:00 |
mriedem | no | 17:00 |
smcginnis | http://paste.openstack.org/show/731282/ | 17:01 |
mriedem | https://www.openstack.org/summit/berlin-2018/vote-for-speakers#/22806 | 17:02 |
cdent | mriedem: I'd like it if you got in on the reviews of https://review.openstack.org/#/c/601614/ . It depends on the devstack changes for extracted placement so isn't in danger of merging any time soonish, but your point of view will make it better before then | 17:12 |
cdent | efried: I'm not sure I'm fully parsing the implied vote in your <rant> on the gabbi yaml | 17:19 |
cdent | the last paragraph doesn't seem to correspond with "I disagree" | 17:19 |
cdent | because my assertion was "I obfuscated for personal enjoyment" | 17:19 |
mriedem | oy | 17:20 |
cdent | yo | 17:21 |
efried | cdent: There should be no need to imply a vote, I'm +2 on the gabbit (only reserving +2 on the patch as a whole due to unfamiliarity with zuul concepts). | 17:43 |
cdent | efried: I don't mean vote on the entire patch, I mean on changing the degree of use of anchors | 17:43 |
efried | Regardless of any personal enjoyment you achieved by your obfuscation, I feel like the result is terse and elegant and comprehensible as long as you know the constructs. | 17:43 |
efried | And not knowing the constructs should not be an excuse for downvoting. | 17:44 |
cdent | okay, that clears it up, thanks | 17:44 |
*** zaneb has left #openstack-placement | 17:46 | |
*** openstackgerrit has quit IRC | 17:51 | |
*** tssurya has quit IRC | 17:52 | |
*** openstackgerrit has joined #openstack-placement | 17:57 | |
openstackgerrit | Eric Fried proposed openstack/placement master: Placement: Remove usage of get_legacy_facade() https://review.openstack.org/607335 | 17:57 |
efried | melwitt, cdent: ^ | 17:58 |
cdent | huzzah | 17:58 |
efried | note: two patches | 17:58 |
*** alex_xu has quit IRC | 18:02 | |
cdent | bbl | 18:11 |
*** openstackgerrit has quit IRC | 18:35 | |
*** openstackgerrit has joined #openstack-placement | 19:10 | |
openstackgerrit | melanie witt proposed openstack/nova-specs master: Update blueprint name so spec matches launchpad https://review.openstack.org/607347 | 19:10 |
openstackgerrit | melanie witt proposed openstack/nova-specs master: Dynamically find releases for move-implemented-specs https://review.openstack.org/592628 | 19:15 |
openstackgerrit | melanie witt proposed openstack/nova-specs master: Add a script for counting blueprints https://review.openstack.org/581914 | 19:16 |
*** e0ne has joined #openstack-placement | 19:32 | |
openstackgerrit | Artom Lifshitz proposed openstack/nova-specs master: Fail count in API https://review.openstack.org/607352 | 19:33 |
*** ttsiouts has joined #openstack-placement | 20:55 | |
*** e0ne has quit IRC | 21:02 | |
*** ttsiouts has quit IRC | 21:10 | |
*** ttsiouts has joined #openstack-placement | 21:11 | |
*** ttsiouts has quit IRC | 21:27 | |
*** ttsiouts has joined #openstack-placement | 21:28 | |
*** cdent has quit IRC | 21:32 | |
*** mriedem has quit IRC | 22:03 | |
*** ttsiouts has quit IRC | 22:04 | |
openstackgerrit | melanie witt proposed openstack/nova-specs master: Add a script for counting blueprints https://review.openstack.org/581914 | 22:20 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!