15:00:09 <hberaud[m]> #startmeeting openstack-eventlet-removal
15:00:09 <opendevmeet> Meeting started Tue Mar  4 15:00:09 2025 UTC and is due to finish in 60 minutes.  The chair is hberaud[m]. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:09 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:09 <opendevmeet> The meeting name has been set to 'openstack_eventlet_removal'
15:00:24 <hberaud[m]> Ping list: amorin, JayF, gouthamr, ralonsoh
15:00:39 <hberaud[m]> https://etherpad.opendev.org/p/epoxy-eventlet-tracking
15:00:40 <amorin> pong
15:00:50 <hberaud[m]> let
15:02:11 <hberaud[m]> waiting 1 or 2 min for folks
15:03:12 <hberaud[m]> #topic feature freeze
15:03:14 <hberaud[m]> so
15:03:26 <ralonsoh> hello
15:03:50 <hberaud[m]> we just passed the feature freeze, and final epoxy is one 1 month
15:04:39 <hberaud[m]> from the point of oslo, all the deprecations are in place, we are just waiting for a RFE https://review.opendev.org/c/openstack/releases/+/942904
15:04:50 <hberaud[m]> how are your deliverables at this point?
15:05:18 <ralonsoh> Neutron is not ready, there missing agents
15:05:22 <hberaud[m]> s/is in one/
15:06:09 <hberaud[m]> Do you have an estimate of the remaining work?
15:06:20 <hberaud[m]> (to complete the migration)
15:06:42 <ralonsoh> that will be ready, for sure, in F. We are expecting the oslo.service release without eventlet
15:07:28 <hberaud[m]> the second part of the oslo.service migration will be shipped with F
15:07:39 <hberaud[m]> that's too late for epoxy
15:08:01 <hberaud[m]> and we are waiting for the patches to start our reviews
15:08:14 <ralonsoh> just to be sure, was it expected to deliver all the projects without eventlet in Epoxy?
15:08:37 <hberaud[m]> no no :)
15:08:42 <ralonsoh> ahhhh uffff
15:08:55 <hberaud[m]> it wouldn't have been possible
15:09:14 <hberaud[m]> I think we are in a pretty good shape
15:09:28 <amorin> mistral is almost ready, still some eventlet in tests, oslo.services, coordination, stuff like this
15:09:44 <hberaud[m]> I'd not have expected that amount of things done
15:10:02 <hberaud[m]> that's awesome
15:10:41 <amorin> some of them will disappear during F cycle, because I deprecated them in E
15:10:49 <hberaud[m]> excellent
15:11:10 <hberaud[m]> #topic PTG preparation
15:11:31 <hberaud[m]> We are currently scheduling and preparing the PTG
15:11:51 <hberaud[m]> I think we will at least propose I cross session
15:12:31 <ralonsoh> what projects in particular?
15:12:36 <amorin> that would be nice to have status on each projects, but also on cross-projects, like oslo stuff
15:12:48 <hberaud[m]> I'd suggest to you to start thinking about it and if you identify topics that you want to share with us, then, I'll share an etherpad with you
15:13:02 <hberaud[m]> ack, I take note
15:13:48 <hberaud[m]> I'll ask my oslo teammates to see if they plan to schedule a PTG session for oslo
15:13:56 <amorin> that would also be nice to share tips and tricks proejcts were using to get rid of eventlet
15:14:22 <amorin> sometimes I am wondering how neutron/nova does the work, to avoid reinventing the wheel on mistral project
15:14:23 <hberaud[m]> ralonsoh: all projects are welcome
15:14:41 <hberaud[m]> amorin: good idea
15:15:02 <amorin> like, the wiki page was good, but more oriented to async IO
15:15:14 <amorin> which is not possible in most project for now I believe
15:15:47 <amorin> talking about this: https://wiki.openstack.org/wiki/Eventlet-removal
15:16:12 <hberaud[m]> After some discussion with Designate folks to give them some entry points, I recently wrote a tiny document, which is more a collection of how to split your work etc... maybe it can fit well with this kind of tricks
15:16:27 <hberaud[m]> yeah aboslutely
15:18:00 <hberaud[m]> in all the case I'll communicate you soon all the details about PTG
15:18:43 <amorin> ack
15:19:04 <hberaud[m]> Do you plan to schedule eventlet specific session to eventlet for your deliverables?
15:19:27 <ralonsoh> for sure there will be
15:19:36 <ralonsoh> I need to ask for it to my PTL
15:19:37 <hberaud[m]> If yes, then let me know. This way I'll update the wiki with the various dates and places
15:19:47 <hberaud[m]> thank you
15:19:51 <ralonsoh> I'll tell you as soon as I have this info
15:19:58 <hberaud[m]> thank you
15:20:38 <amorin> nothing will happen on mistral side on PTG
15:20:50 <hberaud[m]> ack
15:21:20 <hberaud[m]> #topic final release candidate deadline
15:22:12 <hberaud[m]> RC1 starts next week, and final release candidates deadline is March 27
15:22:50 <hberaud[m]> so, if some of your deliverables (neutron by example) are rc deliverables, then you still have some time left
15:23:14 <hberaud[m]> I think the same is true for mistal IIRC
15:23:37 <amorin> how is it triggered?
15:24:02 <hberaud[m]> what? The building of rc releases?
15:24:15 <amorin> yup
15:24:30 <hberaud[m]> those are convenience patches from the release teama
15:24:42 <hberaud[m]> they are automatically generated
15:24:51 <amorin> ok, perfect, nothing to do on project side for this
15:24:57 <hberaud[m]> you just have to validate them
15:24:59 <amorin> execpt having green CI / ready code
15:25:07 <hberaud[m]> yes
15:25:58 <hberaud[m]> and sometimes, you might request a RC2 if you observe a last minute bugs
15:26:37 <hberaud[m]> but even RC2 are more or less convenience patches
15:27:22 <hberaud[m]> you just have to keep an eye on proposed releases https://review.opendev.org/q/project:openstack/releases+status:open
15:28:48 <hberaud[m]> patches should be labeled with topic epoxy-rc1-deadline: https://review.opendev.org/q/topic:%22epoxy-rc1-deadline%22
15:29:17 <hberaud[m]> (not yet fulfilled)
15:30:02 <hberaud[m]> #topic Open Discussion
15:30:12 <hberaud[m]> Anything else that you want to discuss today?
15:30:16 <ralonsoh> all good
15:30:36 <hberaud[m]> ralonsoh: I'll check your reproducer by tomorrow
15:30:42 <ralonsoh> thanks!
15:32:32 <hberaud[m]> Then, I think we are done
15:32:39 <hberaud[m]> thanks everyone for joining
15:32:52 <hberaud[m]> see you in 2 weeks https://etherpad.opendev.org/p/epoxy-eventlet-tracking
15:32:57 <hberaud[m]> #endmeeting