14:01:04 <ralonsoh> #startmeeting networking 14:01:04 <opendevmeet> Meeting started Tue Apr 18 14:01:04 2023 UTC and is due to finish in 60 minutes. The chair is ralonsoh. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:04 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:04 <opendevmeet> The meeting name has been set to 'networking' 14:01:05 <ralonsoh> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, sahid, slawek, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, amotoki 14:01:05 <mlavalle> o/ 14:01:09 <sahid_> o/ 14:01:09 <mtomaska> o/ 14:01:15 <slaweq> o/ 14:01:16 <obondarev> o/ 14:01:17 <ralonsoh> hello all 14:01:20 <bcafarel> o/ 14:01:38 <rubasov> o/ 14:01:40 <ykarel> o/ 14:02:09 <ralonsoh> ok, I think we can start now 14:02:10 <lajoskatona> o/ 14:02:14 <ralonsoh> #topic announcements 14:02:19 <ralonsoh> #link https://releases.openstack.org/bobcat/schedule.html 14:02:35 <ralonsoh> in 3 weeks, we'll have the bobcat-1 milestone 14:02:49 <ralonsoh> and in 2 months the Vancouver summit! 14:03:29 <ralonsoh> I would like to refresh the link for the summit 14:03:32 <ralonsoh> #link https://etherpad.opendev.org/p/neutron-vancouver-2023 14:03:53 <ralonsoh> of course, as usual, please check the openinfra videos 14:03:56 <ralonsoh> #link https://openinfra.dev/live/#all-episodes 14:04:20 <ralonsoh> and finally, the oslo.db 13.0.0 release 14:04:22 <ralonsoh> #link https://review.opendev.org/c/openstack/releases/+/880659 14:04:28 <opendevreview> yatin proposed openstack/neutron master: [DNM] check 880586 https://review.opendev.org/c/openstack/neutron/+/880704 14:04:34 <ralonsoh> that should have full support for sqlalchemy 2.0 14:04:41 <ralonsoh> so just a heads-up 14:04:52 <lajoskatona> is that already bumped in requirements? 14:04:56 <ralonsoh> not yet 14:05:00 <lajoskatona> ok, thanks 14:05:17 <ralonsoh> but here is the link 14:05:18 <ralonsoh> #link https://review.opendev.org/c/openstack/requirements/+/880729 14:05:34 <ralonsoh> (and Neutron is not failing) 14:05:43 <ralonsoh> but remember this is not bumping sqlalchemy 14:05:51 <ralonsoh> only oslo.db with support for 2.0 14:06:11 <lajoskatona> ack, thanks for the headsup 14:06:16 <ralonsoh> something else I'm missing here? 14:06:54 <ralonsoh> ok, let's move on 14:06:57 <ralonsoh> #topic bugs 14:07:10 <ralonsoh> last week I was the bug deputy 14:07:14 <ralonsoh> #link https://lists.openstack.org/pipermail/openstack-discuss/2023-April/033393.html 14:07:24 <ralonsoh> and there are some bugs to be discussed/assigned 14:07:40 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2016413 14:07:53 <ralonsoh> I think this is a low hanging fruit one 14:07:56 <ralonsoh> I'll add the tag 14:08:24 <ralonsoh> so please check it if you want to collaborate with documentation 14:08:49 <ralonsoh> next one is 14:08:54 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2016198 14:09:03 <ralonsoh> that is related to https://bugs.launchpad.net/neutron/+bug/2016197 14:09:22 <ralonsoh> should we bring this topic to the drivers meeting? 14:09:53 <mlavalle> it's a bug, isn't it? 14:09:59 <ralonsoh> the first one yes 14:10:15 <ralonsoh> but the suggestion is to limit the port creation in a network without subnets 14:10:23 <mlavalle> ahh, 14:10:25 <ralonsoh> that I don't think should be done because of a race condition 14:10:27 <mlavalle> I get it 14:10:57 <ralonsoh> so LP#2016198 should be addressed, but LP#2016197 should not be the solution 14:11:05 <ralonsoh> IMO 14:11:07 <lajoskatona> let's discuss it during the drivers meeting 14:11:15 <ralonsoh> I'll move this topic then 14:11:20 <lajoskatona> ok, thanks 14:11:24 <ralonsoh> last one is 14:11:26 <mlavalle> yes, that's probably a good idea 14:11:28 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2015844 14:11:44 <ralonsoh> in a nutshell, OpenStack is removing windows support 14:12:07 <ralonsoh> this code is not tested in the CI 14:12:08 <slaweq> regarding recent discussion about winstackers project that it will be gone I think we should remove completely this windows bits from neutron 14:12:36 <ralonsoh> first deprecate and then remove? 14:12:38 <slaweq> but we should probably follow deprecations policy so deprecate it in this cycle 14:12:42 <ralonsoh> perfect 14:12:48 <sahid_> if windows is not supported anmore no need to maintain all that code 14:12:49 <slaweq> and remove in 2024.2 at least 14:13:03 <ralonsoh> yes, that should be in 2 cycles 14:13:10 <ralonsoh> in D 14:13:13 <mlavalle> ++ 14:13:16 <ralonsoh> +1 14:13:17 <lajoskatona> +1 14:13:22 <slaweq> we need to keep it deprecated for at least one SLURP release 14:13:28 <slaweq> which will be C in this case 14:13:28 <ykarel> +1 14:13:31 <obondarev> +1 14:13:36 <slaweq> I can deprecate it 14:13:38 <mtomaska> +1 14:13:41 <ralonsoh> ok, decided, I'll comment that in the LP 14:13:50 <ralonsoh> slaweq, thanks, please propose the patches 14:13:57 <slaweq> I will propose patch, sure 14:13:57 <ralonsoh> and assign that to you 14:14:07 <ralonsoh> we can consider this LP closed with the deprecation 14:14:17 <ralonsoh> and in 1 year, we'll create a new one to remove the code 14:14:36 <slaweq> ++ 14:14:36 <mlavalle> LOL, close a LP by removing the related code 14:14:52 <ralonsoh> yes, just to track it 14:14:56 <slaweq> mlavalle we should follow that practice more often :P 14:15:01 <ralonsoh> but I would prefer not to have this LP open 1 year 14:15:11 <ralonsoh> ok, that was fast 14:15:23 <ralonsoh> this week lucasgomes is the deputy, next week will be jlibosva 14:15:33 <ralonsoh> I know lucasagomes is aware 14:15:44 <ralonsoh> I'll ping Jakub next week 14:15:56 <ralonsoh> something else you want to discuss here? 14:16:08 <ralonsoh> yes, one more 14:16:38 <ralonsoh> https://ubuntu.com/security/CVE-2023-1668 14:16:44 <ralonsoh> this is a problem in OVS 14:16:52 <ralonsoh> I know there is a list of patches to fix that 14:17:09 <ralonsoh> and the mail from ovs-discuss 14:17:10 <ralonsoh> https://www.openwall.com/lists/oss-security/2023/04/06/1 14:17:18 <ralonsoh> that is affecting ML2/OVS and ML2/OVN 14:17:52 <ralonsoh> well, I don't know if that is possible in OVN 14:18:00 <ralonsoh> but we can create this rule in OVS FW 14:18:16 <ralonsoh> in any case, this is a heads-up 14:18:44 <ralonsoh> ok, something else? 14:19:13 <ralonsoh> ok, I'll skip the spec reviewal as we don't have any active one 14:19:21 <ralonsoh> #topic community_goals 14:19:28 <ralonsoh> 1) Consistent and Secure Default RBAC 14:19:39 <ralonsoh> #link https://review.opendev.org/c/openstack/neutron/+/879827 14:19:51 <ralonsoh> I think we don't have active bugs, slaweq ? 14:20:09 <slaweq> nothing new this week 14:20:20 <ralonsoh> btw, https://review.opendev.org/c/openstack/neutron/+/880461/ 14:20:24 <slaweq> I'm still working on patch https://review.opendev.org/c/openstack/neutron/+/879827/ 14:20:28 <ralonsoh> is failing in neutron-tempest-plugin-openvswitch-enforce-scope-new-defaults 14:20:42 <ralonsoh> yes, this is the 4th patch of the series 14:20:47 <ralonsoh> but is failing the 1st 14:20:54 <slaweq> ohh, I will need to check it 14:21:15 <ralonsoh> ok, I think this is just a red herring 14:21:22 <slaweq> it failed because of issue with installation of packages 14:21:24 <ralonsoh> something with the mirrors, nothing else 14:21:26 <ralonsoh> yes 14:21:31 <slaweq> nothing relevant to patch itself 14:21:33 <ralonsoh> (much better) 14:21:46 <lajoskatona> fingers crossed 14:21:54 <ralonsoh> so we are almost ready to merge the sRBAC by default in Neutron 14:22:02 <slaweq> and main patch to enable new policies by default is almost there, just about 300 UT to fix still 14:22:09 <slaweq> and 200 FT 14:22:26 <ralonsoh> nahhh almost nothing... 14:22:31 <ralonsoh> (ping for help if needed) 14:22:54 <slaweq> but once all that will be done, we will also have a bit better coverage of policy testing in UT as all such tests will make API requests with proper context 14:23:01 <slaweq> not without context at all 14:23:32 <ralonsoh> so we need to review all UTs to use the proper context? 14:23:39 <slaweq> I will definitely bother You all with review requests once it will be green 14:23:46 <ralonsoh> perfect 14:24:01 <slaweq> ralonsoh not all but many which are using api to e.g. create resources 14:24:08 <slaweq> like many tests for extensions 14:24:13 <slaweq> or ml2 plugin 14:24:16 <slaweq> or db modules 14:24:34 <slaweq> I'm on it and it's high priority for me to finish it ASAP 14:25:00 <ralonsoh> thank you 14:25:21 <lajoskatona> thanks slaweq 14:25:42 <ralonsoh> ok, the second topic is 14:25:45 <ralonsoh> 2) Neutron client deprecation 14:25:51 <ralonsoh> I 've identified 3 active patches 14:25:59 <ralonsoh> #link https://review.opendev.org/c/openstack/python-neutronclient/+/880629 14:26:04 <ralonsoh> #link https://review.opendev.org/c/openstack/python-neutronclient/+/875728 14:26:09 <lajoskatona> yes, I started to work on fwaas CLI 14:26:10 <ralonsoh> #link https://review.opendev.org/c/openstack/python-neutronclient/+/868321 14:26:31 <lajoskatona> 880629 is wip (this is for fwaas) 14:27:03 <ralonsoh> so we have the OSC/SDK code merged for fwaas? 14:27:04 <lajoskatona> the patch for octavia is also in good shape: https://review.opendev.org/c/openstack/octavia/+/866327 (thanks octavia team) 14:27:28 <lajoskatona> ralonsoh: yes it is already done (https://review.opendev.org/c/openstack/openstacksdk/+/592303 ) 14:27:35 <ralonsoh> perfect 14:28:00 <lajoskatona> so if yo have time you can even check the octavia patch 14:28:14 <ralonsoh> ^^ fantastic the patch in octavia, good to see traction outside Neutron team 14:28:35 <lajoskatona> agree, we had the same good example from designate 14:29:06 <lajoskatona> that's it from me for the neutronclient topic 14:29:12 <ralonsoh> thanks! 14:29:23 <ralonsoh> and that's all! 14:29:27 <ralonsoh> #topic on_demand 14:29:40 <ralonsoh> I have one topic but was discussed in the bug section 14:29:54 <ralonsoh> so please, if you have something, this is the moment 14:30:24 <slaweq> I wanted to ask people about ideas for forum sessions in vancouver 14:30:36 <slaweq> do You have any ideas what we can propose there? 14:31:25 <ralonsoh> the forum proposals ends this Friday 14:31:34 <lajoskatona> not sure but the RBAC as default would be interesting even for operators 14:31:46 <ralonsoh> actually it is 14:31:53 <ralonsoh> that was a request from operators 14:31:58 <slaweq> lajoskatona but what You want exactly discussing regarding RBAC? 14:32:33 <ralonsoh> yeah, maybe this is a topic for a presentation, not a forum 14:32:37 <lajoskatona> what kind of new personas they have to use and what can be expected? things like that 14:33:05 <lajoskatona> but presentation proposal is closed :-) so we have the forum only now >/( 14:33:06 <mlavalle> a general operators feedback session? 14:33:20 <slaweq> lajoskatona but this actually is something broader than Neutron only thing 14:33:30 <slaweq> so maybe S-RBAC team will propose some session 14:33:37 <lajoskatona> ok 14:33:53 <lajoskatona> mlavalle: that can be interesting anytime I suppose 14:33:59 <slaweq> I don't think we should have any neutron specific discussion about it with operators, at least not something for separate session 14:34:00 <mlavalle> yeap 14:34:22 <mlavalle> and this would be the first time face to face in a long time 14:34:25 <slaweq> mlavalle yes, I was also thinking about "general feedback session" 14:35:01 <mlavalle> another one might be discussion on scalability issues 14:35:03 <slaweq> and ask operators about what features/plugins/drivers they are using, what are their pain points, etc. 14:35:23 <lajoskatona> +1 14:35:55 <slaweq> and my idea was to ask operators question like "If You could choose one thing which will be fixed in Neutron, what it would be" to see if there would be any pattern there, maybe all of them needs the same thing to be there :) 14:36:13 <ralonsoh> ^^ right 14:36:31 <mlavalle> yes, we can start with a menu of teaser questions and then see where the discussion goes 14:37:23 <mlavalle> to uncover pattersn, as you say slaweq 14:38:04 <ralonsoh> ok, tomorrow we'll present, in this channel, some proposals 14:38:11 <ralonsoh> and then you'll approve then or not 14:38:17 <ralonsoh> and then we'll register them 14:38:25 <ralonsoh> ok? 14:38:32 <lajoskatona> sounds good 14:38:35 <slaweq> ++ 14:38:40 <ralonsoh> we are running out of time and we need to ground something 14:38:43 <ralonsoh> perfect then 14:38:51 <mlavalle> what time? don't forget us, the ones on this side of the pond 14:39:03 <ralonsoh> of course, at this time more or less 14:39:16 <mlavalle> say around 1400 utc? 14:39:25 <ralonsoh> ok 14:39:28 <mlavalle> cool 14:39:32 <mlavalle> good time for me 14:40:09 <sahid_> lajoskatona, rubasov o/ 14:40:20 <ralonsoh> ok, I think that's all for today 14:40:29 <sahid_> quick question regarding https://review.opendev.org/c/openstack/neutron/+/872905/15 14:40:42 <sahid_> oh sorry i think the meeting was finished :-) 14:40:47 <ralonsoh> not yet 14:40:49 <ralonsoh> one sec 14:41:04 <ralonsoh> please remember the CI meeting is in 20 mins in this channel 14:41:10 <ralonsoh> #endmeeting