*** irclogbot_2 has quit IRC | 02:23 | |
*** edmondsw has quit IRC | 02:26 | |
*** edleafe has quit IRC | 02:27 | |
*** ttsiouts has quit IRC | 03:52 | |
*** e0ne has joined #openstack-placement | 06:40 | |
*** e0ne has quit IRC | 07:02 | |
*** belmoreira has joined #openstack-placement | 07:24 | |
*** rubasov has quit IRC | 07:59 | |
*** ttsiouts has joined #openstack-placement | 08:06 | |
*** tssurya has joined #openstack-placement | 08:36 | |
*** rubasov has joined #openstack-placement | 08:45 | |
*** belmoreira has quit IRC | 09:15 | |
*** cdent has joined #openstack-placement | 09:55 | |
*** e0ne has joined #openstack-placement | 10:28 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Slightly improve usage documentation https://review.openstack.org/643913 | 10:35 |
---|---|---|
openstackgerrit | Chris Dent proposed openstack/placement master: Update install docs for single database https://review.openstack.org/643915 | 10:46 |
openstackgerrit | Chris Dent proposed openstack/placement master: Rename and restructure install docs https://review.openstack.org/643919 | 11:08 |
openstackgerrit | Chris Dent proposed openstack/placement master: Rename and restructure install docs https://review.openstack.org/643919 | 11:08 |
openstackgerrit | Chris Dent proposed openstack/placement master: Rename and restructure install docs https://review.openstack.org/643919 | 11:13 |
*** ttsiouts has quit IRC | 11:21 | |
openstackgerrit | Merged openstack/placement master: Reuse common get_providers_with_resource() https://review.openstack.org/643729 | 11:30 |
*** ttsiouts has joined #openstack-placement | 11:47 | |
*** belmoreira has joined #openstack-placement | 11:47 | |
*** edmondsw has joined #openstack-placement | 11:50 | |
*** edleafe has joined #openstack-placement | 12:21 | |
*** e0ne has quit IRC | 12:51 | |
*** e0ne has joined #openstack-placement | 12:52 | |
*** mriedem has joined #openstack-placement | 12:58 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Update the from-pypi install instructions https://review.openstack.org/643938 | 13:23 |
*** ttsiouts has quit IRC | 13:36 | |
*** ttsiouts has joined #openstack-placement | 13:41 | |
*** ttsiouts_ has joined #openstack-placement | 13:42 | |
*** ttsiouts has quit IRC | 13:46 | |
efried | finucannot: When you get a chance: | 13:47 |
efried | https://review.openstack.org/#/c/641497/ <== remove backslashes, edleafe suggested a compromise that is hopefully to everyone's liking | 13:47 |
efried | cdent: http://logs.openstack.org/04/641404/5/check/openstack-tox-docs/1d2faee/html/placement-api-microversion-history.html looks fine to me here on my puter. | 13:47 |
cdent | i've got this: https://tank-binaries.s3.amazonaws.com/a3735085710848c0b3560b0ceb1c82e6.png | 13:48 |
cdent | which I dunno | 13:48 |
efried | weird | 13:49 |
efried | do you override fonts in your browser or something? | 13:49 |
cdent | no | 13:49 |
* cdent tries a different browser | 13:49 | |
efried | It looks okay to me in chrome & firefox on bionic. | 13:49 |
* efried tries Windows... | 13:49 | |
efried | yeah, it looks about the same for me in all those places. None of them look like yours. Are you on a mac or something? | 13:51 |
cdent | yup | 13:51 |
efried | well, that must be why it also looked gimpy on my iphone | 13:51 |
cdent | and its the same across three different browsers | 13:51 |
efried | n-sch meeting in 8 minutes in #openstack-meeting-alt | 13:52 |
*** tetsuro has joined #openstack-placement | 13:57 | |
*** e0ne has quit IRC | 13:58 | |
*** takashin has joined #openstack-placement | 13:58 | |
*** e0ne has joined #openstack-placement | 14:05 | |
mriedem | fwiw mine looks like cdent's, firefox on windows | 14:11 |
mriedem | similar for firefox on bionic | 14:12 |
efried | hm, that's odd, why is my firefox/win and firefox/bionic different? | 14:12 |
cdent | missing fancy fonts? | 14:15 |
efried | stock installs of both for me. | 14:19 |
* cdent shrugs | 14:19 | |
*** e0ne has quit IRC | 14:41 | |
openstackgerrit | OpenStack Release Bot proposed openstack/os-resource-classes stable/stein: Update .gitreview for stable/stein https://review.openstack.org/644028 | 14:42 |
openstackgerrit | OpenStack Release Bot proposed openstack/os-resource-classes stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/644029 | 14:42 |
openstackgerrit | OpenStack Release Bot proposed openstack/os-resource-classes master: Update master for stable/stein https://review.openstack.org/644030 | 14:42 |
openstackgerrit | OpenStack Release Bot proposed openstack/os-traits stable/stein: Update .gitreview for stable/stein https://review.openstack.org/644031 | 14:42 |
openstackgerrit | OpenStack Release Bot proposed openstack/os-traits stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/644032 | 14:42 |
openstackgerrit | OpenStack Release Bot proposed openstack/os-traits master: Update master for stable/stein https://review.openstack.org/644033 | 14:42 |
*** e0ne has joined #openstack-placement | 14:42 | |
openstackgerrit | OpenStack Release Bot proposed openstack/osc-placement stable/stein: Update .gitreview for stable/stein https://review.openstack.org/644043 | 14:43 |
openstackgerrit | OpenStack Release Bot proposed openstack/osc-placement stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/644044 | 14:43 |
openstackgerrit | OpenStack Release Bot proposed openstack/osc-placement master: Update master for stable/stein https://review.openstack.org/644045 | 14:43 |
*** e0ne has quit IRC | 14:53 | |
*** takashin has left #openstack-placement | 14:53 | |
*** tetsuro has quit IRC | 14:54 | |
*** e0ne has joined #openstack-placement | 14:55 | |
*** e0ne has quit IRC | 15:05 | |
*** e0ne has joined #openstack-placement | 15:20 | |
mriedem | ew, we should add a way to append/add a trait to a provider because 'openstack resource provider trait set' is a full overwrite | 15:45 |
mriedem | by default, in devstack i've got 11 auto-reported traits on the libvirt compute node | 15:45 |
mriedem | so adding a custom trait would kind of suck | 15:45 |
cdent | mriedem: there was a reason for that, not sure I recall | 15:46 |
mriedem | lack of generations? | 15:46 |
cdent | but it was probalby a combination of "restfulness" and other things | 15:46 |
mriedem | oh there is a generation on traits | 15:46 |
mriedem | this would be client-side sugar | 15:46 |
mriedem | GET /resource_providers/{uuid}/traits | 15:46 |
mriedem | append traits to the list, | 15:47 |
mriedem | PUT /resource_providers/{uuid}/traits with the generation | 15:47 |
* cdent nods | 15:47 | |
* mriedem storyboards | 15:47 | |
cdent | I had misread that you were talking about osc-placement | 15:47 |
efried | wait, he *is* talking about osc-placement, not about the API, right? | 15:48 |
cdent | (correct efried, that's what I mean) I suspect _that_ was simply because of clicking through the microversions, it doesn't feel like there's been that much attention to user experience | 15:48 |
mriedem | https://storyboard.openstack.org/#!/story/2005258 | 15:50 |
mriedem | yes CLI | 15:50 |
mriedem | i'm working on docs for the compute driver capabilities as traits things, | 15:50 |
mriedem | and have a devstack here where i can give some console examples | 15:50 |
mriedem | and noticed that to add a new custom trait via the CLI, it means i have to either (1) copy all of the existing traits + the new one or (2) add the new one and then sighup the compute service to get the standard ones back (i think?) | 15:51 |
mriedem | maybe not sighup, maybe just wait for update_available_resource to run | 15:51 |
* cdent looks at efried | 15:54 | |
cdent | I stayed out of the "how the compute host figures out owning traits" conversation because I had to keep some focus elsewhere and it was tickling my "eh?" button | 15:54 |
cdent | but yeah, I agree osc-placement should have an add or something like | 15:55 |
cdent | ugh, I do wish storyboard would do a spinner or something after I click "hey, hey, did you get that?" | 15:56 |
efried | mriedem: I think you do have to SIGHUP, yes. | 15:57 |
efried | or wait for cache expiry, which may never happen depending on conf | 15:58 |
efried | just having update_available_resource run isn't enough because we don't refresh unless we have some reason to believe the values we already have are stale. | 15:59 |
*** ttsiouts_ has quit IRC | 16:05 | |
*** ttsiouts has joined #openstack-placement | 16:13 | |
mriedem | not entirely true, | 16:13 |
mriedem | i wiped out all of the traits using "openstack resource provider trait delete <rp_uuid>" and after 60 seconds they were all back again | 16:13 |
mriedem | maybe that's b/c i wiped them all out? | 16:14 |
*** e0ne has quit IRC | 16:14 | |
*** e0ne has joined #openstack-placement | 16:19 | |
mriedem | nope, also overwrite so there were just 2, waited 60 seconds and then they were all back again | 16:20 |
*** tssurya has quit IRC | 16:44 | |
cdent | Did we decide that placement-from-stein is 1.0.0? | 16:47 |
cdent | Or if we didn't decide is there any reason it shouldn't be that? | 16:47 |
mriedem | i'd expect it should be 1.0.0 | 16:53 |
*** ttsiouts has quit IRC | 16:57 | |
*** e0ne has quit IRC | 17:14 | |
cdent | blargh. writing docs is hard | 17:18 |
edleafe | We should start Stein Placement at 37.18.4 just to mess with people's heads | 17:21 |
cdent | good choice | 17:22 |
efried | make the first two parts of the semver match the latest microversion | 17:23 |
cdent | I actually considered that and was what led me to say what I did in the meeting earlier today about no longer following the standard release cycle | 17:24 |
edleafe | mriedem: https://review.openstack.org/#/c/641497/ has my +1 | 17:27 |
efried | cdent: It's not the worst idea. Adding a microversion is a feature and warrants a major ver bump. And then the minor version is there for bug fixes. | 17:29 |
* cdent nods | 17:29 | |
cdent | i think you mean minor and patch? | 17:30 |
efried | yeah, I don't know what the semver fields are called. | 17:30 |
cdent | major.minor.patch | 17:30 |
efried | x.y.z => x.y is latest microversion, .z is bugs | 17:30 |
efried | yeah | 17:30 |
cdent | maybe in U? | 17:31 |
cdent | too much change at once we'll honk people | 17:31 |
efried | did we already propose a release? | 17:31 |
efried | I'm just sorta wondering if that's going to blow up on us at some point. | 17:31 |
cdent | not yet | 17:32 |
cdent | but actually if we did want to stick to microversions it would need to be in T | 17:32 |
cdent | as T would normally be 2.0.0 | 17:32 |
edleafe | How often are we tagging releases? Just once per cycle? Or with each new microversion? | 17:33 |
cdent | this cycle we tagged once when the extracted code "worked" and then will tag again this week | 17:34 |
cdent | tagging each new microversion is an option, nothing prevents it | 17:35 |
cdent | I definitely think we should do far more intermediate releases than we used to do in the past | 17:35 |
cdent | but whether it is by microversion or not is unclear as sometimes we have tons, sometimes not | 17:35 |
edleafe | I'm trying to imagine it from the consumer POV: will frequent releases make things clearer or more confusing to them? | 17:36 |
cdent | a new release will be mostly transparent, since microversions are opt-in | 17:37 |
edleafe | Sure, but will they wonder if they are missing out on performance improvements or bug fixes? | 17:39 |
cdent | we're still obliged to make releasenotes | 17:39 |
cdent | and I think the only people who will care are those who chose to install from source or pypi | 17:39 |
cdent | which may be unicorns | 17:39 |
edleafe | Don't disrespect our unicorn users! | 17:40 |
cdent | I have the utmost respect for them, which is why I think they can cope | 17:40 |
*** mriedem has quit IRC | 18:45 | |
*** mriedem has joined #openstack-placement | 18:48 | |
openstackgerrit | Chris Dent proposed openstack/placement master: Upgrade from rocky-nova docs https://review.openstack.org/644396 | 18:54 |
cdent | edleafe, efried, mriedem : that ^ is a first draft. It's messy and brain hurty enough that i think it best that you give it a good read through before I go any further. besides I'm brain dead and need to go | 18:56 |
efried | Have a good evening cdent | 18:57 |
edleafe | cdent: roger that | 18:57 |
mriedem | ack | 18:59 |
openstackgerrit | Merged openstack/placement master: Get rid of backslash continuations in code https://review.openstack.org/641497 | 19:20 |
openstackgerrit | Merged openstack/placement master: Update install docs for single database https://review.openstack.org/643915 | 19:20 |
*** cdent_ has joined #openstack-placement | 19:59 | |
*** cdent has quit IRC | 20:02 | |
*** cdent_ is now known as cdent | 20:02 | |
*** irclogbot_2 has joined #openstack-placement | 20:12 | |
*** dklyle has quit IRC | 20:14 | |
*** dklyle has joined #openstack-placement | 20:18 | |
*** irclogbot_2 has quit IRC | 20:25 | |
*** irclogbot_2 has joined #openstack-placement | 20:26 | |
*** e0ne has joined #openstack-placement | 20:28 | |
*** e0ne has quit IRC | 20:41 | |
*** irclogbot_2 has quit IRC | 21:05 | |
*** irclogbot_2 has joined #openstack-placement | 21:06 | |
*** irclogbot_2 has quit IRC | 21:26 | |
*** irclogbot_2 has joined #openstack-placement | 21:27 | |
*** cdent has quit IRC | 21:28 | |
*** mriedem has quit IRC | 23:20 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!