14:01:06 <ralonsoh> #startmeeting networking
14:01:07 <opendevmeet> Meeting started Tue Apr 11 14:01:06 2023 UTC and is due to finish in 60 minutes.  The chair is ralonsoh. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:07 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:01:07 <opendevmeet> The meeting name has been set to 'networking'
14:01:08 <mlavalle2> o/
14:01:10 <ralonsoh> hi all
14:01:11 <rubasov> o/
14:02:10 <ralonsoh> ok, let's start
14:02:14 <ralonsoh> #topic announcements
14:02:40 <ralonsoh> we have a new schedule
14:02:45 <ralonsoh> #link https://releases.openstack.org/bobcat/schedule.html
14:02:50 <elvira> o/
14:03:09 <frickler> \o
14:03:25 <ralonsoh> so please check the dates and get used to this new cycle
14:03:28 <ralonsoh> and welcome, of course
14:03:42 <ralonsoh> also please remember the Vancouver etherpad
14:03:45 <ralonsoh> #link https://etherpad.opendev.org/p/neutron-vancouver-2023
14:04:28 <ralonsoh> let me be honest: we need more topics and attendands
14:04:47 <ralonsoh> I know during the vPTG we had a packed agenda
14:04:51 <ralonsoh> and we discussed many topics
14:05:27 <ralonsoh> but if the agenda of vancouver doesn't grow, I don't think I'll attend (I'll be able to)
14:05:57 <ralonsoh> and if you missed that last wee
14:05:59 <ralonsoh> week
14:06:02 <ralonsoh> #link https://lists.openstack.org/pipermail/openstack-discuss/2023-April/033115.html
14:06:29 <ralonsoh> something else?
14:06:54 <ralonsoh> ok, let's jump to the funny topic
14:06:58 <ralonsoh> #topic bugs
14:07:02 <ralonsoh> we have two reports
14:07:07 <ralonsoh> one from mlavalle2
14:07:11 <ralonsoh> https://lists.openstack.org/pipermail/openstack-discuss/2023-April/033184.html
14:07:20 <ralonsoh> and the next one from rubasov
14:07:24 <ralonsoh> https://lists.openstack.org/pipermail/openstack-discuss/2023-April/033279.html
14:07:37 <ralonsoh> there are some pending/not assigned bugs
14:07:46 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2013326
14:07:56 <ralonsoh> slaweq, ^^ is this one related to policies?
14:08:01 <ralonsoh> srbac?
14:08:31 <slaweq> yes
14:08:41 <ralonsoh> ok perfect, will you assign it yo you?
14:09:15 <ralonsoh> I'll do it, yes
14:09:20 <slaweq> yes, thx
14:09:23 <ralonsoh> thanks!
14:09:24 <ralonsoh> next one
14:09:40 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2012332
14:09:44 <ralonsoh> this is a RFE
14:09:59 <ralonsoh> I've added it to the drivers agenda next friday
14:10:11 <ralonsoh> but I hope Liu takes it, I'll ping him
14:10:21 <ralonsoh> ok, next one
14:10:34 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2015065
14:10:51 <ralonsoh> this one is not 100% happening but very very often
14:12:07 <ralonsoh> and if I'm not wrong, is always failing in the same point
14:12:15 <lajoskatona> only in grenade?
14:12:17 <ralonsoh> during the SG creation
14:12:18 <ralonsoh> yes
14:12:22 <ralonsoh> only in this job
14:12:31 <lajoskatona> interesting
14:12:49 <rubasov> yes, I believe it's always failing the same way
14:13:16 <lajoskatona> I can check it
14:13:32 <ralonsoh> cool, thanks a lot (my agenda is full this week, sorry)
14:13:49 <ralonsoh> next one
14:13:51 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2015364
14:13:55 <ralonsoh> related to grenade too
14:14:07 <ralonsoh> this time the -skip-tests, only in OVN
14:14:33 <ralonsoh> these 4 jobs (actually 3 plus a class) are always failing
14:15:39 <ralonsoh> are we installing a different version of OVN? I think so
14:16:39 <ralonsoh> ok, I'll check them with ykarel if he has time
14:16:54 <ykarel> sure i can check this issue
14:16:57 <ralonsoh> or I'll ping infra folks to stop an env once failed
14:17:05 <ralonsoh> I appreciate that, thanks!
14:17:07 <ykarel> but yes looks like something wrong with ovn version
14:17:28 <ralonsoh> that's what I think (but I can't support with facts what I'm saying)
14:17:36 <ykarel> ack i will dig
14:17:46 <ralonsoh> these are passing in OVS (and we are executing the same test set)
14:18:16 <ralonsoh> ok, next one is
14:18:19 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2015449
14:18:31 <ralonsoh> frickler, can you provide some update on this one?
14:19:08 <frickler> well it sounds like intra-subnet traffic may be natted by ovn to the FIP attached to an instance
14:19:22 <rubasov> I had no reproduction yet, but I was unsure about whether we consider the described behavior a bug or an unfortunate undocumented behavior
14:19:30 <frickler> but somebody will need to try to reproduce that I guess
14:19:35 <ralonsoh> that was my first question: ovs or ovn
14:19:49 <ralonsoh> so OVN, I'll make a comment in the bug
14:19:54 <rubasov> by the way I will follow up on the bugs in the "triage in progress" section in my report
14:20:04 <rubasov> no need to take over from the next deputy
14:20:11 <ralonsoh> thanks
14:20:20 <frickler> oh, I was assuming ovn, but maybe not
14:20:31 <ralonsoh> I'll ask that in the BZ
14:20:44 <ralonsoh> and in case of OVS, what FW driver
14:21:17 <ralonsoh> ok, once replied, I'll try to reproduce it
14:21:25 <frickler> for me, as noted in the last comment, fw behavior seems correct, just natting shouldn't happen
14:21:41 <ralonsoh> yeah, looks like
14:22:10 <ralonsoh> ok, once we have the answer, I'll deploy an env and reproduce it
14:22:35 <frickler> also ask for ovn/ovs versions
14:22:53 <ralonsoh> sure
14:23:08 <ralonsoh> ok, and the last one I have is
14:23:10 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2015275
14:23:21 <ralonsoh> I added a comment today
14:23:35 <ralonsoh> related to why the RPC is attended in just one worker
14:23:55 <ralonsoh> this is a problem of how it was designed
14:24:15 <ralonsoh> so it will require a proper refactor
14:24:36 <ralonsoh> does anyone have time/energy/will to do it?
14:25:08 <ralonsoh> ok, I'll assign it to myself
14:25:34 <ralonsoh> ok, and that's all for today
14:25:36 <sahid> ralonsoh: you can assign it to myself
14:26:05 <ralonsoh> sahid, perfect, this is not a trivial refactor
14:26:22 <ralonsoh> thanks for that
14:26:40 <sahid> sure i think it will be interesting to look in that part of the code
14:26:58 <ralonsoh> ok, this week I'm the bug deputy
14:27:05 <ralonsoh> next week is lucasagomes
14:27:27 <ralonsoh> btw, do we have new volunteers to be added to the deputy list?
14:27:39 <ralonsoh> as commented during the PTG
14:27:52 <ralonsoh> you can take your time and ping me in IRC at any time
14:28:01 <ralonsoh> check https://wiki.openstack.org/wiki/Network/Meetings
14:28:24 <ralonsoh> let's move on
14:28:29 <sahid> ralonsoh: please add me on it
14:28:35 <ralonsoh> I'll do it
14:28:38 <ralonsoh> ok
14:28:42 <ralonsoh> we don't have new specs
14:28:47 <ralonsoh> nor os-ken issues
14:29:03 <ralonsoh> so we can jump to the next topic
14:29:11 <ralonsoh> #topic community_goals
14:29:30 <ralonsoh> about sRBAC, I don't think slaweq has any new
14:29:38 <slaweq> actually I have
14:29:42 <ralonsoh> sorry!
14:29:52 <slaweq> last week I started working on switch to new policies by default
14:30:23 <slaweq> I sent patch: https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/879828
14:30:29 <slaweq> and WIP https://review.opendev.org/c/openstack/neutron/+/879827
14:30:50 <slaweq> that last one is going to be a lot of work as there is still almost 2k of UT failing there :/
14:30:54 <slaweq> but I'm working on it
14:31:15 <slaweq> during debugging of those failed UT I found out small issue in new policy for get_ports
14:31:24 <slaweq> and I sent fix https://review.opendev.org/c/openstack/neutron/+/879891
14:32:10 <ralonsoh> "test_port_list_queries_constant" still failing
14:32:11 <slaweq> and this is now puzzling me as 2 UT are failing there
14:32:12 <slaweq> this failure is somehow related to that patch
14:32:13 <slaweq> but I have no idea how
14:32:13 <slaweq> and if I run only that one test locally, it is passing
14:32:24 <slaweq> ralonsoh maybe we can talk about this failing test later offline?
14:32:39 <ralonsoh> yeah, I'll check them today (or tomorrow morning)
14:32:52 <slaweq> thx
14:33:01 <slaweq> and that's my update about S-RBAC for this week
14:33:25 <ralonsoh> thanks, I'll add these patches to the meeting agenda
14:33:29 <ralonsoh> to track them
14:33:57 <ralonsoh> next one is
14:34:04 <ralonsoh> 2) Neutron client deprecation
14:34:22 <ralonsoh> lajoskatona, any new topic?
14:34:22 <lajoskatona> The usual etherpad: https://etherpad.opendev.org/p/python-neutronclient_deprecation
14:34:30 <lajoskatona> not much
14:34:56 <lajoskatona> designate seems to be finished, and octavia is slowly progressing (not me who works on it
14:35:27 <ralonsoh> do you have any pending patch in SDK or OSC?
14:35:29 <lajoskatona> and sahid started to check how nova can be changed to use sdk
14:35:49 <lajoskatona> https://review.opendev.org/q/topic:bug/1999774+status:open
14:36:15 <lajoskatona> these are for neutronclient, but to change the current cli code to use sdk only
14:36:20 <ralonsoh> ok
14:36:26 <slaweq> I have small update related to this one too. I spent today some time checking ralonsoh's patch https://review.opendev.org/c/openstack/python-neutronclient/+/871711 and I finally made it to be green in CI. So maybe we will be able to finally get rid of "neutron" CLI tool in this cycle
14:37:20 <lajoskatona> slaweq: you mean to remove all code that is realted to core Neutron CLI?
14:37:22 <ralonsoh> but ps7 is not removing the CLI code?
14:37:32 <slaweq> lajoskatona not all code really
14:37:52 <slaweq> but at least remove some of bits related to CLI tool and "neutron" command
14:38:06 <slaweq> and to not provide this CLI tool, bash completion and other related things at all
14:38:14 <ralonsoh> yeah, at least the shell class
14:38:16 <slaweq> so users will not have "neutron" command available anymore
14:38:44 <ralonsoh> we can't remove some nclient modules (still used by many other projects)
14:38:47 <slaweq> parts from neutronclient/neutron/v2_0 needs to be still there as e.g. nova is using it
14:38:52 <ralonsoh> but I'm ok removing the shell class
14:38:53 <lajoskatona> I will check it, as I see it also removes some unit tests related to bgp for example but that code remains in neutronclient
14:39:04 <slaweq> but shell.py and everything what depends on it can be removed probably
14:39:23 <lajoskatona> yeah that can be removed (theoretically :P)
14:39:32 <ralonsoh> and all FT/UT related
14:39:37 <slaweq> lajoskatona yes, it removes most of the UT as those tests were importing shell.py module
14:39:48 <slaweq> and those are "CLI" tests really
14:40:03 <lajoskatona> ok, I will check it
14:40:07 <slaweq> thx
14:40:53 <ralonsoh> I need to check again if I sent the mail to make this public
14:41:10 <ralonsoh> It was years ago but again as a heads-up
14:41:18 <ralonsoh> ok, please check this patch
14:41:20 <ralonsoh> and thanks slaweq
14:41:44 <ralonsoh> and that's all in this topic
14:41:48 <slaweq> ++ thx a lot
14:42:02 <ralonsoh> next one
14:42:06 <ralonsoh> #topic on_demand
14:42:11 <ralonsoh> lajoskatona, you have a topic there
14:42:13 <ralonsoh> 2 actually
14:42:47 <lajoskatona> yes, I added here as reminder as we discussed the PTG to list some forum topic ideas
14:43:02 <lajoskatona> that is the first one
14:43:42 <lajoskatona> so if you have forum ideas which can be discussed during the Vancouver summit/PTG please ....
14:44:34 <lajoskatona> I have one but perhaps that is not worth a whole forum: to discuss the deprecation of Neutronclient
14:44:59 <lajoskatona> this whould not be interesting for users, but for deveopers of other projects it can be interesting
14:45:07 <ralonsoh> I need to check with slaweq (I didn't have time) some suggestions he has
14:45:55 <ralonsoh> in any case we still have time to add more topics
14:46:13 <ralonsoh> the second one is related to Xena release
14:46:14 <ralonsoh> https://review.opendev.org/c/openstack/releases/+/878874
14:46:23 <ralonsoh> I've updated (wrongly) the patch
14:46:35 <ralonsoh> note: for a EM branch, it is needed a new release
14:46:44 <ralonsoh> there is one missing patch for Neutron
14:46:55 <lajoskatona> yeah just a headsup that Xena is planned to EMed soon
14:46:59 <ralonsoh> that was +W by slaweq
14:46:59 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/879330
14:47:18 <ralonsoh> so I think we'll be able to push a final releases patch tomorrow morning
14:47:40 <ralonsoh> last day is 20th of April
14:48:03 <ralonsoh> but as commented, tomorrow I'll push the patch that should be the one accepted
14:48:09 <ralonsoh> and that's all for today
14:48:11 <lajoskatona> cool, thanks
14:48:16 <ralonsoh> remember the CI meeting is in 10 mins
14:48:27 <mlavalle2> video or irc?
14:48:28 <ralonsoh> do you have any other topic?
14:48:31 <slaweq> thx ralonsoh
14:48:35 <slaweq> it's irc this week
14:48:38 <ralonsoh> cool
14:48:39 <mlavalle2> ack
14:48:43 <lajoskatona> ack
14:49:04 <ralonsoh> so thank all
14:49:07 <ralonsoh> #endmeeting