| *** tosky has quit IRC | 00:18 | |
| *** csatari has joined #openstack-tc | 00:59 | |
| *** TheJulia has joined #openstack-tc | 00:59 | |
| *** cosss_ has joined #openstack-tc | 00:59 | |
| *** coreycb has joined #openstack-tc | 00:59 | |
| *** mrhillsman has joined #openstack-tc | 00:59 | |
| *** mriedem has quit IRC | 01:23 | |
| *** lbragstad has quit IRC | 02:06 | |
| *** jamesmcarthur has joined #openstack-tc | 02:28 | |
| *** jamesmcarthur has quit IRC | 02:32 | |
| *** ricolin has joined #openstack-tc | 03:19 | |
| *** lbragstad has joined #openstack-tc | 04:09 | |
| *** diablo_rojo has quit IRC | 04:09 | |
| *** jamesmcarthur has joined #openstack-tc | 04:29 | |
| *** jamesmcarthur has quit IRC | 04:34 | |
| *** adriant has quit IRC | 05:16 | |
| *** whoami-rajat has joined #openstack-tc | 06:19 | |
| *** e0ne has joined #openstack-tc | 06:24 | |
| *** jamesmcarthur has joined #openstack-tc | 06:30 | |
| *** jamesmcarthur has quit IRC | 06:34 | |
| *** e0ne has quit IRC | 06:35 | |
| *** lbragstad has quit IRC | 06:50 | |
| *** Luzi has joined #openstack-tc | 06:53 | |
| *** zaneb has quit IRC | 06:53 | |
| *** e0ne has joined #openstack-tc | 07:12 | |
| *** e0ne has quit IRC | 07:16 | |
| *** e0ne has joined #openstack-tc | 07:24 | |
| *** e0ne has quit IRC | 07:32 | |
| *** e0ne has joined #openstack-tc | 07:42 | |
| *** jamesmcarthur has joined #openstack-tc | 08:31 | |
| *** e0ne has quit IRC | 08:35 | |
| *** jamesmcarthur has quit IRC | 08:35 | |
| *** e0ne has joined #openstack-tc | 08:37 | |
| *** e0ne has quit IRC | 08:38 | |
| *** tosky has joined #openstack-tc | 08:40 | |
| *** e0ne has joined #openstack-tc | 08:46 | |
| *** e0ne has quit IRC | 08:48 | |
| *** e0ne has joined #openstack-tc | 08:54 | |
| *** whoami-rajat has quit IRC | 08:54 | |
| *** e0ne has quit IRC | 08:56 | |
| *** ricolin has quit IRC | 08:58 | |
| *** jpich has joined #openstack-tc | 08:58 | |
| ttx | o/ | 09:00 |
|---|---|---|
| ttx | Office hour started, in case you have questions | 09:00 |
| ttx | tc-members: we'll have yet another OSF community newsletter this week. For OpenStack news I thought of mentioning elections and the train goals discussion | 09:01 |
| ttx | If you have other ideas please chime in on https://etherpad.openstack.org/p/newsletter-openstack-news | 09:01 |
| gmann | o/ | 09:03 |
| cmurphy | o/ | 09:04 |
| *** ricolin has joined #openstack-tc | 09:06 | |
| ttx | +ops meetup registration open | 09:08 |
| *** ricolin has quit IRC | 09:12 | |
| gmann | ttx: how about Stein-3 release (Feature freeze etc) cmg soon. | 09:16 |
| *** ricolin has joined #openstack-tc | 09:24 | |
| ttx | gmann: we are limited to 2-3 bullet points... maybe a good topic for next time | 09:25 |
| *** e0ne has joined #openstack-tc | 09:25 | |
| *** e0ne has quit IRC | 09:26 | |
| ttx | I added it to "future items" | 09:27 |
| gmann | +1. | 09:27 |
| *** cdent has joined #openstack-tc | 09:35 | |
| *** jpich has quit IRC | 09:37 | |
| *** jpich has joined #openstack-tc | 09:38 | |
| *** jamesmcarthur has joined #openstack-tc | 09:59 | |
| *** jamesmcarthur has quit IRC | 10:04 | |
| *** purplerbot has joined #openstack-tc | 10:41 | |
| *** ricolin_ has joined #openstack-tc | 12:15 | |
| *** ricolin has quit IRC | 12:17 | |
| *** coreycb has quit IRC | 12:36 | |
| *** coreycb has joined #openstack-tc | 12:37 | |
| *** mriedem has joined #openstack-tc | 13:04 | |
| *** whoami-rajat has joined #openstack-tc | 13:22 | |
| *** e0ne has joined #openstack-tc | 13:56 | |
| *** mriedem is now known as mriedem_snow | 14:10 | |
| smcginnis | ttx: Thanks for getting the ops meetup in there. | 14:23 |
| *** lbragstad has joined #openstack-tc | 14:34 | |
| *** e0ne has quit IRC | 14:44 | |
| *** ianychoi has quit IRC | 15:07 | |
| *** ianychoi has joined #openstack-tc | 15:07 | |
| *** e0ne has joined #openstack-tc | 15:09 | |
| *** ijolliffe has joined #openstack-tc | 15:28 | |
| *** e0ne has quit IRC | 15:33 | |
| *** Luzi has quit IRC | 15:41 | |
| *** e0ne has joined #openstack-tc | 15:41 | |
| *** diablo_rojo has joined #openstack-tc | 15:47 | |
| *** mriedem_snow is now known as mriedem | 15:52 | |
| *** jamesmcarthur has joined #openstack-tc | 16:06 | |
| lbragstad | jamesmcarthur do you know if anyone from the foundation is around (on IRC) to answer questions about the user survey question format? | 16:20 |
| jamesmcarthur | lbragstad: we're all on a staff call at the moment, but aprice and I would be happy to discuss with you. Maybe 40 minutes? | 16:20 |
| lbragstad | jamesmcarthur awesome - that sounds good | 16:21 |
| jamesmcarthur | cool - talk to you soon :) | 16:21 |
| aprice | lbragstad: we are here if you want to chat | 16:38 |
| lbragstad | aprice awesome - just wrapping up the keystone meeting and we should be ready to go (#openstack-meeting-alt) | 16:39 |
| aprice | ok sounds good | 16:41 |
| *** ricolin_ has quit IRC | 17:21 | |
| *** jpich has quit IRC | 17:24 | |
| *** e0ne has quit IRC | 17:27 | |
| *** jamesmcarthur has quit IRC | 17:31 | |
| *** jamesmcarthur_ has joined #openstack-tc | 17:32 | |
| *** whoami-rajat has quit IRC | 17:36 | |
| *** spsurya has quit IRC | 17:40 | |
| *** jamesmcarthur_ has quit IRC | 17:42 | |
| *** e0ne has joined #openstack-tc | 17:43 | |
| *** mriedem is now known as mriedem_afk | 17:45 | |
| *** jamesmcarthur has joined #openstack-tc | 17:51 | |
| *** jamesmcarthur has quit IRC | 17:56 | |
| *** jamesmcarthur has joined #openstack-tc | 17:58 | |
| *** jamesmcarthur has quit IRC | 17:58 | |
| *** dtantsur is now known as dtantsur|afk | 17:58 | |
| *** jamesmcarthur has joined #openstack-tc | 18:01 | |
| *** jamesmcarthur has quit IRC | 18:07 | |
| *** jamesmcarthur has joined #openstack-tc | 18:08 | |
| *** jamesmcarthur has quit IRC | 18:12 | |
| *** openstackgerrit has joined #openstack-tc | 18:36 | |
| openstackgerrit | Ed Leafe proposed openstack/governance master: Create separate Placement project https://review.openstack.org/636416 | 18:36 |
| openstackgerrit | Doug Hellmann proposed openstack/governance master: apply 7 day rule to formal-vote items https://review.openstack.org/636418 | 18:46 |
| openstackgerrit | Ed Leafe proposed openstack/governance master: Create separate Placement project https://review.openstack.org/636416 | 18:50 |
| openstackgerrit | Merged openstack/governance master: Add os_mistral role to OpenStack-Ansible https://review.openstack.org/634817 | 18:51 |
| openstackgerrit | Ed Leafe proposed openstack/governance master: Create separate Placement project https://review.openstack.org/636416 | 18:55 |
| *** jamesmcarthur has joined #openstack-tc | 18:56 | |
| *** edleafe has joined #openstack-tc | 18:58 | |
| *** jamesmcarthur has quit IRC | 18:58 | |
| * edleafe waves | 18:58 | |
| smcginnis | o/ | 18:58 |
| dhellmann | edleafe : hey, so I'll start by saying that all of these comments are made with the goal of having a patch that requires the fewest approvals and least discussion, to make things happen quickly | 18:58 |
| smcginnis | Makes sense to break the tags out into a follow on patch to limit concerns. | 18:59 |
| dhellmann | the tags like vulnerability:managed and stable:follows-policy have to be approved by those teams, so I would drop them for now and add them back later | 18:59 |
| dhellmann | tc:approve-release and assert:supports-api-interoperability really only apply to publicly facing APIs meant for users to use | 18:59 |
| edleafe | dhellmann: sure, no worries. I just copy/pasted from the nova project | 18:59 |
| dhellmann | I don't know if placement has that | 18:59 |
| dhellmann | yeah, I just want to explain the reasons for telling you to drop them all :-) | 18:59 |
| dhellmann | starter-kit:compute is probably OK, since placement is going to be a dependency of nova | 19:00 |
| dhellmann | and I think the assertions are probably ok, because we have history with the contributors to know their ability to keep up with those things | 19:00 |
| dhellmann | as smcginnis pointed out on the review | 19:00 |
| dhellmann | so, I'd drop tc:approved-release, vulnerability:managed, stable:follows-policy, and assert:supports-api-interoperability from this patch, and add them later | 19:01 |
| dhellmann | make sense? | 19:01 |
| edleafe | So would it be better to just take them all out except starter-kit? Or do you think that the history argument will be enough for others? | 19:02 |
| dhellmann | it would certainly eliminate that as a source of discussion | 19:02 |
| edleafe | ok, I'm not partial one way or the other, so out they go! | 19:03 |
| dhellmann | although I think you can also clearly justify those others as being valid | 19:03 |
| dhellmann | wfm | 19:03 |
| edleafe | If someone wants to add them later, they can make those arguments then. | 19:03 |
| dhellmann | yeah, that would be the job of the next ptl :-) | 19:03 |
| edleafe | So do you see anything else to fix? Or should I push a new rev? | 19:04 |
| *** dtantsur|afk has quit IRC | 19:04 | |
| *** jroll has quit IRC | 19:04 | |
| *** mugsie has quit IRC | 19:04 | |
| *** fdegir has quit IRC | 19:04 | |
| *** mugsie has joined #openstack-tc | 19:05 | |
| dhellmann | I see you changed the home page from the wiki to the docs. Is that what most of the other teams have? | 19:05 |
| dhellmann | I'm not sure it matters, I'm just checking consistency | 19:05 |
| *** fdegir has joined #openstack-tc | 19:05 | |
| edleafe | I started with the wiki because copy/paste. But since we don't actually use a wiki, I thought that the docs URL was more useful | 19:06 |
| dhellmann | I guess most point to old wiki pages. | 19:06 |
| *** jroll has joined #openstack-tc | 19:06 | |
| dhellmann | yeah. I think it's OK to use the docs. we can fix that in a follow-up if someone objects | 19:06 |
| dhellmann | keep in mind, every time you update the patch the 7 day counter resets to 0 | 19:06 |
| *** jamesmcarthur has joined #openstack-tc | 19:07 | |
| *** dtantsur has joined #openstack-tc | 19:08 | |
| openstackgerrit | Ed Leafe proposed openstack/governance master: Create separate Placement project https://review.openstack.org/636416 | 19:08 |
| dhellmann | edleafe : lgtm, thanks | 19:11 |
| edleafe | dhellmann: thanks for the quick feedback | 19:12 |
| dhellmann | no problem | 19:13 |
| dhellmann | tc-members: please take a look at the patch to add the placement project as a separate team. it's a bit time-sensitive because we want to have it approved before PTL elections, if we can. https://review.openstack.org/#/c/636416 | 19:14 |
| *** AJaeger has joined #openstack-tc | 19:21 | |
| AJaeger | tc members, do you want announcement for more repos here? Please review https://review.openstack.org/635625 | 19:21 |
| *** jamesmcarthur has quit IRC | 19:30 | |
| tonyb[m] | For the record Feb 19 would be the last date for placement to be included in the election by default | 19:37 |
| tonyb[m] | After that and we'd (the officials) would need to do extra stuff and it might not be a thing we can do | 19:38 |
| tonyb[m] | It gets much harder after that date | 19:39 |
| *** jamesmcarthur has joined #openstack-tc | 19:59 | |
| *** mriedem_afk is now known as mriedem | 20:05 | |
| mnaser | fyi i think it's important for the tc to know the current state of the placement service | 20:06 |
| mnaser | imho, not a very good one, it's been extracted, but none of the deployment projects actually use it | 20:06 |
| mnaser | no one has upgrade jobs to test the extraction of the database records from nova to placement database | 20:07 |
| mnaser | we're effectively creating a fork right now with that change, because the only place that placement is used externally is within devstack. | 20:07 |
| mnaser | https://github.com/openstack/kolla-ansible/tree/master/ansible/roles kolla-ansible seems like it doesn't have any placement stuff in place | 20:08 |
| mnaser | in OSA world, we created a role but it hasn't been integrated yet | 20:08 |
| mnaser | puppet-openstack-integration repo doesnt seem to be using anything placement specific | 20:09 |
| edleafe | mnaser: That's why the current placement code is remaining in Nova for the Stein cycle | 20:10 |
| edleafe | We aren't forking anything | 20:11 |
| edleafe | The repos remain unchanged whether the governance changes or not | 20:11 |
| mnaser | right, but what happens when stein goes out is that nova progress will be somewhat blocked | 20:11 |
| mnaser | because we're unfreezing placement repo | 20:11 |
| mnaser | so nova adds a new feature depending on placement 1.$newer_than_stein -- all deployment tools break | 20:11 |
| edleafe | mnaser: Sure, but what does that have to do with governance? | 20:12 |
| edleafe | Nova has already committed to dropping placement code from their code base in Train | 20:12 |
| mnaser | i think the struggle here that i've learned out of this was | 20:13 |
| smcginnis | Deployment tools need to catch up. This plan has been public for quite awhile now. | 20:13 |
| mnaser | a certain group had a priority (extract placement), a bunch of other people ended up with a whole bunch of added work | 20:13 |
| mnaser | deployment tools don't have the resources (well, for the most part) because it's mostly operator ran (at least in our case) | 20:14 |
| *** adriant has joined #openstack-tc | 20:15 | |
| mnaser | i'm starting to reflect back and think that we should involve more people when we take such measures that affect multiple projects and pretty much an entire openstack deployment | 20:15 |
| smcginnis | That's kind of the nature of how all this works, isn't it? | 20:16 |
| mnaser | right, but this is a massively complicated piece we're moving around | 20:16 |
| mnaser | and besides blazar/cyborg, i dont see anyone using the placement service, and blazar/cyborg is always deployed side by side with nova anyways afaik | 20:18 |
| *** AJaeger has left #openstack-tc | 20:19 | |
| mnaser | i dunno, i guess it's just a bit frustrating that we've all been signed up for work that's put our (project) goals behind with no one showing up to do the work | 20:19 |
| jroll | that's kind of the nature of deployment tools, though, right? you have to track all the projects and implement things to deal with their changes | 20:22 |
| dhellmann | it's the nature of having an upstream and a downstream of any sort, even if both are fully open | 20:22 |
| jroll | right | 20:23 |
| mnaser | jroll: except this one is quite major, we're talking about coordinating shutting a service down, deploying a new service, dumping tables and restoring them across different databases (thanks to dansmith for doing most of that part) and bringing the service back up .. with the least amount of downtime | 20:23 |
| mnaser | it's not like running a db migration or anything, it is much more complex than that | 20:23 |
| mnaser | previously, we've had a lot easier scenarios where the projects handled implementing a layer of making that migration much more simple (imagine cells v2 and co) | 20:24 |
| mnaser | this time we're left with that massive (and ultra-scary-bash-script solved solution) | 20:24 |
| dansmith | and in previous cases where we split a project out (i.e. nova-volume -> cinder) there was a bridge release | 20:24 |
| mriedem | as of our last hangout there will be a bridge release now though right? | 20:25 |
| mnaser | i guess stein will be our bridge release based on this planning | 20:25 |
| jroll | mnaser: sure, I get that it's harder than usual. but like you said, dan helped out with the hardest part, the rest is pretty standard fair | 20:25 |
| dansmith | I have never been on board with not having a bridge release for this, and I think the deployment projects proved my point | 20:25 |
| jroll | and yeah, stein is the bridge | 20:25 |
| *** e0ne has quit IRC | 20:25 | |
| dansmith | jroll: previously there was not going to be a bridge, now stein is the bridge | 20:25 |
| mriedem | where's that confounded bridge | 20:25 |
| jroll | keeping *master* of the deployment projects from breaking will be hard, but that's already hard, right? | 20:25 |
| dansmith | mriedem: nice | 20:25 |
| mnaser | surprisingly deployment projects break ourselves more often than upstream projects do | 20:26 |
| jroll | dansmith: right, but I hope we're deliberating the current state, not the past | 20:26 |
| mnaser | i cant remember the last time we were broken by nova/cinder/etc thanks to awesome gating by them | 20:26 |
| dansmith | jroll: I dunno, I was just yanked in via mention and I figured people were upset about the current state, else we'd not be discussing it | 20:26 |
| jroll | I guess I'm curious what the alternative path is, have the placement devs work doubletime so they can do the deployment work too? | 20:26 |
| mnaser | jroll: i'm thinking the alternative would be for us to *maybe* phase work like this over a longer period of time so we can get our ducks in a row | 20:27 |
| mnaser | openstack ansible has wanted to try and get working upgrade jobs for a while and we have struggled to find the resources for them | 20:27 |
| mnaser | now... we have no choice but to drop everything and try to get a) working upgrade jobs and then b) coordinate that upgrade | 20:27 |
| jroll | mnaser: yeah, that seems hard with a full extraction like this | 20:28 |
| mnaser | but i guess part of it is like what dhellmann mentioned it's the nature of upstream/downstream | 20:28 |
| smcginnis | It's been a few months. Hardly seems a "drop everything" situation. | 20:28 |
| jroll | I was genuinely asking about the alternatives, fwiw | 20:28 |
| mnaser | smcginnis: we really don't have much resources on this, no one works full time on openstack ansible at this point | 20:28 |
| dhellmann | the upgrade job is work for train, right? | 20:28 |
| mnaser | between our gates breaking here and there, ci needing fixes, trying to improve our release sitaution ,etc | 20:28 |
| jroll | also, maybe the nova team would be kind enough to drop it later than sooner in train, so you haev some extra time | 20:28 |
| smcginnis | That statement can be applied to many more projects than ansible. | 20:29 |
| mriedem | the actual data migration should be train now | 20:29 |
| mnaser | smcginnis: kolla-ansible hasn't progressed at all, i guess only tripleo has progress in this subject | 20:29 |
| mnaser | well yeah, now that nothing will be dropped, we get a bit of a breather | 20:29 |
| smcginnis | "now that nothign will be dropped" ? | 20:29 |
| mriedem | the placement-in-nova code won't be dropped in stein as of last wed | 20:30 |
| mnaser | smcginnis: the placement code won't be deleted in stein so if you deploy placement the same way as usual, nothing will change | 20:30 |
| mriedem | new deploys in stein/train would be encouraged to use extracted placement | 20:30 |
| mriedem | to avoid the data migration hit | 20:30 |
| mnaser | i'm hoping for us to be able to do greenfield with os_placement in stein but not upgrade, but then that introduces another path of complexity to not let a deployment accidentally redeploy using os_placement if that makes sense | 20:31 |
| dhellmann | do the executables have different names or are we assuming they're packaged separately? how does one switch? | 20:31 |
| *** jamesmcarthur has quit IRC | 20:31 | |
| mnaser | dhellmann: the switch is: deploy new packages (only rdo has them, canoncial has no packages for placement now) | 20:32 |
| mnaser | drop new configuration file | 20:32 |
| dhellmann | so the nova package includes a command with the same name as the placement package to start the placement service? | 20:32 |
| mnaser | extract all database tables from nova database and import to placement database (shuttind down all api services in that period of time) | 20:32 |
| mnaser | no, nova literally just didn't remove the placement code. we have two copies of the placement code now | 20:33 |
| mnaser | a nova-in-tree placement code, an extracted code | 20:33 |
| *** penick has joined #openstack-tc | 20:33 | |
| mnaser | devstack uses the extracted code only | 20:33 |
| dhellmann | so if I install both nova and placement on the same host, I get what? 2 copies of the same executable? | 20:33 |
| dansmith | mnaser: just so I'm clear, you're okay with the governance change to make placement a thing right? | 20:33 |
| mnaser | dhellmann: you end up with two placements. one from nova and one from 'placement'. i'm not sure what the differences are at this point | 20:34 |
| mnaser | nova cores can speak better on the details | 20:34 |
| dhellmann | ok, I guess downstream packagers will have to deal with that and remove the placement executable from the nova packages they build | 20:34 |
| mriedem | there are no api differences, | 20:34 |
| mriedem | any bug fixes have been mirrored in both in stein, | 20:34 |
| dhellmann | when they support the shift, that is | 20:34 |
| mriedem | the only differences are things like CLIs, e.g. placement-manage and placement-status | 20:35 |
| mriedem | placement uses alembic for db schema migrations but that's all under the covers stuff | 20:35 |
| dhellmann | yeah, none of that should cause issues from the outside | 20:35 |
| mriedem | even the config is essentially the same, | 20:36 |
| mriedem | e.g. [placement_database]/connection is still used in the placement.conf | 20:36 |
| mriedem | rather than [database]/connection | 20:36 |
| dhellmann | it sounds to me like you're doing a good job of reducing the pain | 20:36 |
| mnaser | thats the easy bits, its the turn-off-the-world-and-migrate-databases which is going to be very tricky | 20:37 |
| mnaser | dansmith: i almost feel like saying yes here is because we've dug ourself into this situation and "we might just as well get over with this" | 20:37 |
| mnaser | so.. might as well now that we've gotten here. | 20:37 |
| mriedem | mnaser: i thought in berlin you said that you ran dan's db migration script on a pre-prod db and it took about a minute? | 20:37 |
| mnaser | i just want to get the tc more informed about the current *technical* state of things before deciding the governance change | 20:37 |
| dansmith | mnaser: okay I'm super confused, because without this, we have no bridge in stein and this gets pushed off to the unicorn release right? | 20:37 |
| mriedem | https://etherpad.openstack.org/p/BER-placement-extract | 20:38 |
| mriedem | "VEXXHOST ran the database migration script on a 40GB DB and it took ~1 minute." | 20:38 |
| mnaser | mriedem: right, but that's assuming that we don't realize at some point that some magical service needed to be turned off for that to execute cleanly | 20:38 |
| mnaser | say.. if we realize we need to have all nova-computes shutdown to run this | 20:38 |
| dansmith | mriedem: there's still orchestration that has to happen even if the db migration goes quickly | 20:38 |
| mnaser | all i did was run the shell script, not orchestrate a placement extraction | 20:38 |
| mriedem | sure | 20:39 |
| mriedem | if the nova-api isn't down some things could fail in nova trying to schedule etc | 20:39 |
| mriedem | we do still have the safe_connect decorator thing on a lot of the nova placement client which will gracefully handle placement being down | 20:39 |
| mnaser | you'd have to possibly restart all computes if they have a cached service catalog entry | 20:39 |
| mnaser | or we have to somehow update all the placement backend in haproxy, etc, it's a lot of moving parts | 20:40 |
| mnaser | anyways, to your point dansmith, i'm all for having a bridge release | 20:40 |
| mnaser | i just wanted the tc to know what is the technical state of things when we do this governance vote, because that's important to know imho | 20:40 |
| dansmith | alright | 20:40 |
| mriedem | we also acknowledge that placement wasn't going to live within nova forever... | 20:41 |
| mriedem | so at some point this pain was going to come | 20:41 |
| mnaser | right, i agree with that as well. | 20:41 |
| edleafe | mnaser: to put it another way - is there anything that you think will make this whole transition less painful if the governance was not switched? All your points are valid; I'm just not seeing how a governance change would either improve things or make them worse. | 20:42 |
| dansmith | governance isn't being switched here right? it's just, uh, introducing a new project, and nova will delete its thing and point at this new thing later right? | 20:43 |
| mriedem | https://review.openstack.org/#/c/636416/ | 20:43 |
| mriedem | ^ new team under governance | 20:43 |
| mriedem | separate team | 20:43 |
| dansmith | edleafe: trying to re-state it as less of a change | 20:43 |
| dansmith | mriedem: right | 20:43 |
| mnaser | dansmith: no, that change introduces an entire new team, placement, which will have it's own ptl for the next cycle, and it's own election, etc | 20:43 |
| dansmith | mnaser: exactly | 20:43 |
| mriedem | it's essentially the same team | 20:43 |
| mnaser | edleafe: i guess part of it was voicing my frustrations of being a deployment project that got sucked into a bunch of pretty significant and complex work.. and the other part was trying to provide some informative point of view for the tc as we vote on this (as i feel like i've been relatively involved in this) | 20:44 |
| mnaser | but we can ignore my complaints for the sakes of the conversation :) | 20:44 |
| edleafe | mnaser: :) | 20:44 |
| edleafe | mnaser: thanks; I wasn't sure if I was missing something | 20:45 |
| mriedem | the "delete code from nova" is the biggest issue | 20:45 |
| dhellmann | that's not happening until next cycle, right? | 20:45 |
| mriedem | which is why we still have nova-network and cells v1 in nova lo these many years | 20:45 |
| mriedem | correct | 20:45 |
| mriedem | ^ was a change of plans as of last wed | 20:45 |
| dhellmann | did you already say whether it would be early or late? | 20:45 |
| mnaser | i guess that'll be more of a nova <=> deployment tools discussion when that time comes anyways | 20:45 |
| mriedem | dhellmann: nothing firm | 20:46 |
| dhellmann | k | 20:46 |
| mriedem | probably depends on deployment project status | 20:46 |
| dhellmann | makes sense | 20:46 |
| smcginnis | nova-network is on a really long bridge. | 20:46 |
| mriedem | and when nova really needs to unfreeze and start using new placement APIs | 20:46 |
| mriedem | heh | 20:46 |
| mnaser | smcginnis: we're trying to match CERN's LHC | 20:46 |
| dhellmann | heh | 20:47 |
| smcginnis | hah | 20:47 |
| mnaser | but, i think someone from cern mentioned that they're planning to rip it out and replace with neutron out of a recent ML discussion | 20:47 |
| mnaser | so they might have won that size battle :< | 20:47 |
| *** e0ne has joined #openstack-tc | 20:48 | |
| smcginnis | Hopefully placement won't try to match this one - https://www.extremetech.com/extreme/283928-cern-reveals-plans-for-particle-collider-four-times-larger-than-lhc | 20:48 |
| edleafe | smcginnis: Placement could handle that :) | 20:48 |
| mnaser | smcginnis: challenge accepted | 20:48 |
| * mnaser goes to -1 | 20:48 | |
| mnaser | fyi, i don't know who knows much of this, but dhellmann has built such nice tooling for our governance repo | 20:49 |
| mnaser | tox -echeck-review-status | 20:49 |
| mnaser | tc-members: i invite you to read the buffer and try to take time to vote on https://review.openstack.org/#/c/636416/ due to the short-notice to make the lives of our hard working election folks easier :) | 20:52 |
| *** jamesmcarthur has joined #openstack-tc | 20:52 | |
| dhellmann | mnaser : speaking of tools, https://review.openstack.org/#/c/636418/ could use another pair of eyes | 20:54 |
| *** jamesmcarthur has quit IRC | 20:54 | |
| fungi | wow that's some scrollback | 20:57 |
| mnaser | dhellmann: the code looks good to me and makes sense but given our (mostly) reliance on it, i'd take another pair of eyes as well if someone can | 20:58 |
| dhellmann | mnaser : sure, that's fine. the more the merrier | 20:59 |
| dhellmann | we have a couple of formal-vote topics up now, if you want to see the difference in the output | 20:59 |
| mnaser | dhellmann: let me pull it down and see the difference | 21:00 |
| mnaser | i know we have the "too soon" | 21:00 |
| mnaser | but that's only for 4 days after majority and not the 7 days as mentioned | 21:01 |
| dhellmann | mnaser : http://paste.openstack.org/show/744976/ | 21:01 |
| mnaser | so technically after this change it'll still be too soon for the placement change in 4 days (which makes sense) | 21:01 |
| mnaser | ++ the output works so i can +w using smcginnis as a CR+1 | 21:02 |
| mnaser | dhellmann: works for ya? | 21:03 |
| smcginnis | I suppose a CR is more appropriate on that change than a rollcall-vote. Added that. | 21:03 |
| dhellmann | mnaser : sure | 21:03 |
| dhellmann | I think the tool counts CR and RC votes for code changes, but I'd have to look | 21:04 |
| dhellmann | oh, huh, it only looks at Code-Review | 21:05 |
| dhellmann | who wrote this thing?! | 21:05 |
| dhellmann | oh, wow, I clearly need to stop working. It looks at votes for reviews but code_review for dissent | 21:05 |
| * dhellmann stops while he's ahead | 21:05 | |
| *** dtantsur has quit IRC | 21:06 | |
| *** fdegir has quit IRC | 21:06 | |
| *** mugsie has quit IRC | 21:06 | |
| *** dtantsur has joined #openstack-tc | 21:06 | |
| *** mugsie has joined #openstack-tc | 21:06 | |
| mnaser | aha | 21:07 |
| openstackgerrit | Merged openstack/governance master: apply 7 day rule to formal-vote items https://review.openstack.org/636418 | 21:15 |
| fungi | so that the above discussion about placement extraction doesn't get lost, it might be nice to have a post-mortem analysis (on the ml) of what challenges the extraction created with ideas toward how similar future project splits could be less painful | 21:15 |
| *** e0ne has quit IRC | 21:28 | |
| cdent | I was away during all of the above, but had i been around, I would have felt obliged to correct some misapprehensions. However, now after reading the whole thing, I don't think it's useful to do so. We seem to have ended at reasonable point: let's carry on and do our best to learn from it | 21:47 |
| cdent | I'm hoping to that decomposing some of the larger projects might become a thing that is easy (when deemed worthwhile). | 21:48 |
| smcginnis | ++ | 21:52 |
| *** ijolliffe has quit IRC | 21:53 | |
| *** zaneb has joined #openstack-tc | 21:58 | |
| cmurphy | cdent: <3 | 22:00 |
| cdent | cmurphy: you said it so much better than I would | 22:01 |
| cmurphy | I almost didn't say it, it takes so much time to write an email like that | 22:01 |
| cdent | totally worth saying | 22:04 |
| cdent | (I think) | 22:04 |
| smcginnis | Yes, very good write up cmurphy | 22:20 |
| *** jamesmcarthur has joined #openstack-tc | 22:27 | |
| *** cdent has quit IRC | 22:36 | |
| *** ijolliffe has joined #openstack-tc | 23:01 | |
| *** jamesmcarthur has quit IRC | 23:11 | |
| *** whoami-rajat has joined #openstack-tc | 23:12 | |
| *** jamesmcarthur has joined #openstack-tc | 23:32 | |
| *** jamesmcarthur has quit IRC | 23:32 | |
| *** jamesmcarthur has joined #openstack-tc | 23:33 | |
| *** jamesmcarthur has quit IRC | 23:56 | |
| *** ijolliffe has quit IRC | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!