#openstack-meeting log

21:00:08 <slaweq> #startmeeting networking
21:00:09 <openstack> Meeting started Mon Mar 30 21:00:08 2020 UTC and is due to finish in 60 minutes.  The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:10 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:12 <openstack> The meeting name has been set to 'networking'
21:00:14 <slaweq> hi
21:00:17 <haleyb> hi
21:00:23 <njohnston> o/
21:00:24 <lajoskatona> Hi
21:00:38 <amotoki> hi
21:02:29 <slaweq> ok, lets start
21:02:41 <slaweq> #topic Announcements
21:03:10 <slaweq> this weeks is final release for non-client libraries for Ussuri
21:04:10 <slaweq> last friday I sent email with list of patches which I found for os-ken, ovsdbapp and neutron-lib: http://lists.openstack.org/pipermail/openstack-discuss/2020-March/013628.html
21:04:33 <slaweq> I think that those are only 3 non-client libraries projects in neutron stadium
21:04:39 <slaweq> or am I missing something?
21:05:33 <mlavalle> o/
21:05:41 <njohnston> AFAIK that is it
21:05:58 <amotoki> I think that's all for non-client libs
21:06:15 <slaweq> thx for confirmation
21:06:19 <slaweq> so we don't have many patches still in queue for those projects
21:06:46 <slaweq> actually it seems for me that neutron-lib and os-ken are good to go even now :)
21:07:00 <slaweq> but I would wait until e.g. thursday with releases
21:07:05 <slaweq> amotoki: what do You think?
21:07:16 <slaweq> can You take care of it this week?
21:07:31 <amotoki> slaweq: I can take care of it
21:07:36 <slaweq> amotoki: thx a lot
21:07:59 <slaweq> for ovsdbapp I will ask tomorrow ovn folks if https://review.opendev.org/#/c/683355/ is something which they would like to get in
21:08:05 <slaweq> if not, it should be good to go also
21:08:26 <amotoki> agree. I am looking at it, but you are faster than me
21:08:54 <slaweq> thx amotoki - it will be great if You can check once again if I didn't miss anything
21:09:09 <amotoki> sure
21:09:33 <slaweq> ok, lets move on with announcements
21:09:59 <slaweq> next week is Ussuri-3 milestone which is also feature freeze and final date for client libraries
21:11:11 <slaweq> so we need to revisit list of our BPs for ussuri and move some of them to Victoria
21:11:51 <slaweq> speaking about Victoria:
21:11:56 <slaweq> Victoria (virtual) PTG planning etherpad: https://etherpad.openstack.org/p/neutron-victoria-ptg
21:12:05 <slaweq> PTG in Vancouver is cancelled https://www.openstack.org/events/opendev-ptg-2020/
21:12:07 <slaweq> If You want to help to organize virtual PTG, please add Your name in https://etherpad.openstack.org/p/Virtual_PTG_Planning
21:12:55 <amotoki> re: CLI/SDK, OSC usually requires another openstacksdk release, so it would be nice to land SDK patches soon if you have any.
21:12:58 <mlavalle> I have a question on this regard
21:13:09 <mlavalle> I mean the virtual PTG
21:13:20 <slaweq> sure mlavalle
21:13:51 <mlavalle> are we (Neutron team) going to do some organizing effort of our own?
21:14:12 <mlavalle> or are we going to let the community, as indicated above, do the orgnization for us?
21:14:13 <slaweq> mlavalle: I don't know yet to be honest
21:14:36 <njohnston> mlavalle: How to accomplish the virtual PTG will be discussed among those people on the etherpad slaweq posted
21:14:45 <slaweq> exactly
21:14:55 <njohnston> I believe all options are on the table
21:14:58 <slaweq> I think meetings should be this and/or next week
21:15:22 <diablo_rojo_phon> Yep. I'll be sending out an email with dates/times later today :)
21:15:23 <njohnston> So anyone who wants to be a part of the process, post your name!
21:15:27 <mlavalle> the reason I am asking is that I would like to help organize the Neutron meeting
21:15:27 <diablo_rojo_phon> Everyone is invited!
21:15:52 <amotoki> is there any announcement on this? I might miss it.
21:15:52 <slaweq> hi diablo_rojo_phon :)
21:16:29 <diablo_rojo_phon> Hello slaweq :) Thought I'd interject :)
21:17:02 <slaweq> sure, thx, I think You are the best person to answer those questions :)
21:17:29 <slaweq> amotoki: regarding announcements, I know only about this one https://www.openstack.org/events/opendev-ptg-2020/
21:17:33 <mlavalle> so what's the best way to help the Neutron PTG?
21:18:05 <mlavalle> follow the community discussion, see what happens at that level and later decide what to do for this team?
21:18:25 <slaweq> mlavalle: IMHO we should wait a bit longer to see what community will decide, and then discuss that in our team as well
21:18:38 <mlavalle> ok
21:18:53 <mlavalle> I'll follow the community conversation for the time being, then
21:19:16 <slaweq> mlavalle: thx, and thank You for volunteering for help with this :)
21:19:31 <mlavalle> :-)
21:19:37 <amotoki> agree. we need to consider both neutron team meetup and cross-project meetup, so it would be nice to follow the community best practice and coversations.
21:20:38 <slaweq> ok, I think we can move on
21:20:54 <slaweq> or do You have any other announcements for the team today?
21:21:20 <amotoki> let me repeat this again
21:21:22 <njohnston> yes thank you mlavalle, and also than you diablo_rojo for getting this together for the whole community!
21:21:23 <amotoki> re: CLI/SDK, OSC usually requires another openstacksdk release, so it would be nice to land SDK patches soon if you have any.
21:21:30 <slaweq> ahh, right
21:21:32 <slaweq> thx amotoki
21:22:06 <mlavalle> sorry amotoki for getting in the way
21:22:16 <amotoki> no problem
21:22:35 <slaweq> amotoki: do You know about any SDK patches related to neutron which are waiting for review?
21:23:03 <amotoki> slaweq: I don't check them at the moment unfortunately....
21:23:42 <slaweq> ok, I will check them tomorrow morning and I will ping You if there will be any which needs review
21:24:51 <slaweq> and now I think we can move on
21:24:53 <slaweq> #topic Blueprints
21:25:15 <slaweq> list of BPs targeted for Ussuri-3 is on https://launchpad.net/neutron/+milestone/ussuri-3
21:25:23 <slaweq> I checked this list today
21:25:56 <slaweq> and, as I said earlier, IMO we will need to move some of them to Victoria
21:26:30 <slaweq> lets move through the list one by one
21:26:31 <slaweq> https://blueprints.launchpad.net/neutron/+spec/secgroups-custom-ethertypes
21:26:49 <slaweq> I would move this to Victoria as there is not too much done for it still
21:26:55 <slaweq> do You agree?
21:27:04 <njohnston> yes, I agree
21:27:54 <slaweq> next one
21:27:56 <slaweq> https://blueprints.launchpad.net/neutron/+spec/enginefacade-switch
21:28:37 <slaweq> It will not be done before next week for sure, but IMHO we can continue of it still, even after FF as it's not completly new feature
21:28:52 <slaweq> I sent some patches related to it recently and I want to continue this work
21:28:59 <slaweq> what do You think about it?
21:29:06 <mlavalle> yes, let's continue it
21:29:15 <mlavalle> I'll try to help as well
21:29:19 <njohnston> I agree
21:29:22 <slaweq> thx mlavalle
21:30:33 <slaweq> ok, next one
21:30:35 <slaweq> https://blueprints.launchpad.net/neutron/+spec/neutron-ovn-merge
21:31:02 <slaweq> there is still this one patch https://review.opendev.org/#/c/702247/ needed to mark it as completed
21:31:18 <slaweq> but I think that we can do it before U-3
21:31:41 <slaweq> I will ask lucasgomes tomorrow if he can address comments from this patch
21:32:06 <slaweq> haleyb: do You also think it's doable to get this done in Ussuri still?
21:32:23 <haleyb> slaweq: yes, since it's just the one patch
21:33:12 <slaweq> great, thx
21:33:30 <slaweq> next one is my favorite recently :) https://blueprints.launchpad.net/neutron/+spec/metadata-over-ipv6
21:33:37 <slaweq> spec was finally merged recently
21:33:43 <slaweq> thx a lot for all help with this
21:34:01 <slaweq> rubasov already started some patches with implementation https://review.opendev.org/#/q/topic:metadata-ipv6
21:34:12 <slaweq> but I don't think we can do this for Ussuri still
21:34:31 <slaweq> maybe with request for FFE but I'm not sure if that is so urgent
21:34:45 <mlavalle> I advice is not to rush it
21:35:10 <slaweq> and as it may potentially open e.g. routers for some traffic from outside if we will configure new IP addresses wrongly in namespace I would rather wait for next cycle
21:35:38 <njohnston> +1
21:36:02 <slaweq> great, so lets move it to Victoria :)
21:36:21 <mlavalle> and even though in code might not be big, it is really a big change
21:36:32 <mlavalle> so let's test it well
21:36:46 <slaweq> mlavalle: I totally agree with You
21:36:47 <njohnston> +100
21:37:14 <slaweq> ok, next one then
21:37:23 <slaweq> https://blueprints.launchpad.net/neutron/+spec/rbac-address-scope-and-subnet-pools - patches are ready for review and I think that neutron part can be still done before end of next week
21:37:33 <slaweq> patches can be found at https://review.opendev.org/#/q/owner:%22Igor+Malinovskiy+%253Cimalinovskiy%2540mirantis.com%253E%22+status:open
21:37:52 <slaweq> all those patches has got high review priority set already
21:37:57 <mlavalle> ok, added them to my pile
21:38:41 <slaweq> amotoki: there are also patches for OSC there, please take a look if You will have some time :)
21:39:22 <slaweq> ok, next one
21:39:24 <amotoki> slaweq: sure. I will review this series of patches. I believe I have time as horizn gate breakage
21:39:29 <amotoki> has been fixed
21:39:43 <slaweq> thx a lot
21:39:56 <slaweq> and mlavalle thx for reviews also :)
21:39:59 <slaweq> ok, next one
21:40:01 <slaweq> https://blueprints.launchpad.net/neutron/+spec/tagging-on-post
21:40:11 <slaweq> for this one we are also missing only OSC part
21:40:26 <mlavalle> yeah, it shouldn't be a big change
21:40:29 <slaweq> should we keep it for Ussuri-3 or should we move it to Victoria to finish this OSC part?
21:40:54 <mlavalle> tomorrow is quarter end at work so after that I think I will have time to work on it
21:41:07 <mlavalle> so let's keep it in U
21:41:15 <slaweq> mlavalle: great, thx a lot
21:41:48 <slaweq> ok, next one
21:41:50 <slaweq> https://blueprints.launchpad.net/neutron/+spec/stateless-security-groups
21:41:56 <mlavalle> you will give me a FF if I need it, right?
21:42:05 <slaweq> I would say: move to Victoria, what do You think?
21:42:26 <slaweq> mlavalle: Yes, but is it me who should do it for OSC part?
21:42:26 <mlavalle> I meant the OSC stuff for tagging
21:42:47 <mlavalle> ahh, don't know. maybe amotoki can help
21:43:00 <slaweq> or we will have to ask mordred about it if needed
21:43:03 <mlavalle> I'll still do my best over the next few days
21:43:21 <slaweq> thx a lot
21:43:31 <slaweq> lets sync next week and we will see what is the status
21:43:39 <mlavalle> yeap
21:43:40 <amotoki> re tagging one, I wonder SDK support is needed. SDk and CLI changes would be smaller.
21:44:52 <njohnston> For stateless security groups, the main code is merged as well as the SDK bits, so AFAIK what is left is just the fullstack test and the OSC change which has no negative votes Thanks very much for the loan of titan15; I am all done with it.
21:44:54 <slaweq> there may be SDK change needed also, I'm not sure
21:44:59 <lajoskatona> slaweq: stateless sec-groups is nearly finished as I see
21:45:13 <njohnston> So I still think stateless security groups can be achieved this cycle
21:45:41 <slaweq> njohnston: lajoskatona: ok, that's great, thx for update about stateless SG
21:45:48 <lajoskatona> yeah, fullstack is now working ,some cosmetic sis needed as I remember from this morning
21:45:54 <njohnston> https://review.opendev.org/#/q/topic:bp/stateless-security-groups for links to the 2 pending stateless SG changes
21:46:14 <slaweq> amotoki: about tagging - lets wait for mlavalle what he will do in next days with this
21:46:24 <slaweq> we can sync e.g. on friday's drivers meeting too
21:46:36 <mlavalle> ok
21:46:39 <amotoki> slaweq: yeah
21:46:56 <mlavalle> so what is higher priority: OSC -> SDK
21:47:00 <mlavalle> ?
21:47:12 <mlavalle> where do you want me to start?
21:47:18 <slaweq> mlavalle: SDK as OSC will probably need changes from SDK
21:47:25 <mlavalle> ok
21:47:25 <slaweq> OSC is generally using SDK
21:47:32 <mlavalle> will do
21:48:00 <slaweq> mlavalle: if You will send some patches, please ping me, I will review them
21:48:08 <amotoki> let me check the current impl in SDK today or tomorrow. I implemented network tag support in SDK.
21:48:15 <slaweq> (I have +2 power in SDK repo)
21:48:18 <mlavalle> ok
21:48:50 <slaweq> ok, can we move on?
21:49:20 <amotoki> nothing more from me
21:49:21 <mordred> mlavalle: ping me too - and yeah, I'll support a FF if you need it
21:49:34 <slaweq> thx mordred :)
21:49:46 <mlavalle> mordred: thanks!
21:49:46 <slaweq> ok, so next one
21:49:52 <slaweq> https://blueprints.launchpad.net/neutron/+spec/fip-pf-description - only OSC patch isn’t merged https://review.opendev.org/#/c/714512/
21:50:19 <slaweq> so this should be still doable before Ussuri-3
21:50:41 <mlavalle> yeap
21:51:03 <slaweq> I will update requirements in this patch tomorrow morning so it should pass gate
21:51:14 <slaweq> next one is:
21:51:21 <slaweq> https://bugs.launchpad.net/neutron/+bug/1863113 - George framework
21:51:23 <openstack> Launchpad bug 1863113 in neutron "[RFE] Introduce new testing framework for Neutron - OVN integration - a.k.a George" [Undecided,New] - Assigned to Jakub Libosvar (libosvar)
21:51:36 <slaweq> I would move it to Victoria as there is no progress on it so far
21:52:08 <slaweq> and the last one on my list:
21:52:10 <slaweq> https://bugs.launchpad.net/neutron/+bug/1841067
21:52:11 <openstack> Launchpad bug 1841067 in neutron "[RFE] SR-IOV agent depends on mac addresses for getting bound ports" [Medium,Confirmed] - Assigned to Adrian Chiris (adrian.chiris)
21:52:11 <mlavalle> great idea
21:52:33 <slaweq> I will also move it to Victoria as was already propsoed by owner of this RFE
21:52:50 <slaweq> and that's all from BPs from me today
21:53:12 <slaweq> sorry for taking so long time for this topic today
21:53:25 <njohnston> slaweq: There is a fair amount of work done for George, I would check with jlibosva before punting it: https://review.opendev.org/#/q/topic:george+(status:open+OR+status:merged)
21:53:47 <slaweq> njohnston: oh, thx
21:53:52 <slaweq> I wasn't aware of it
21:54:00 <slaweq> I checked only this PoC patch linked in RFE
21:54:09 <mlavalle> BPs is a topic that deserves a good chunk of time at this stage of the cycle
21:54:19 <mlavalle> so it was time well spent
21:54:22 <slaweq> njohnston: can You add link to those patches in RFE?
21:54:30 <slaweq> mlavalle: thx :)
21:54:43 <njohnston> slaweq: Sure thing!
21:54:57 <slaweq> thx
21:55:17 <slaweq> ok, lets skip community goals for today as we are almost out of time
21:55:24 <slaweq> and lets go to bugs
21:55:28 <slaweq> #topic Bugs
21:55:36 <slaweq> hongbin was on bug deputy. Summary http://lists.openstack.org/pipermail/openstack-discuss/2020-March/013659.html
21:55:47 <slaweq> and haleyb is on de
21:55:54 <slaweq> haleyb is deputy this week
21:56:11 <haleyb> yes, so no bugs please :)
21:56:26 <slaweq> from hongbin's summary:
21:56:38 <slaweq> 2 critical issues: one already fixed, second (https://bugs.launchpad.net/neutron/+bug/1869034) has got fix in progress - so this is good
21:56:39 <openstack> Launchpad bug 1869034 in neutron "Postgresql periodic job failing every day" [Critical,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
21:56:53 <slaweq> and there are 2 bugs which needs triaging:
21:57:12 <slaweq> for ovn subteam: https://bugs.launchpad.net/neutron/+bug/1868697 [ovn] transaction error: {"details":"Transaction causes multiple rows in \"MAC_Binding\" table to have identical values
21:57:13 <openstack> Launchpad bug 1868697 in neutron "[ovn] transaction error: {"details":"Transaction causes multiple rows in \"MAC_Binding\" table to have identical values" [Undecided,New]
21:58:07 <slaweq> tomorrow I will ask OVN folks to take a look at this
21:58:16 <slaweq> and second one: https://bugs.launchpad.net/neutron/+bug/1869129
21:58:17 <openstack> Launchpad bug 1869129 in neutron "neutron accepts CIDR in security groups that are invalid in ovn" [Undecided,New]
21:58:24 <slaweq> it is also related to ovn driver
21:58:30 <slaweq> but also to API
21:59:02 <slaweq> IMO it should be fixed on driver's side, but please take a look at it and write some comment with Your thoughts about that one
21:59:21 <slaweq> and I think that this is all for today
21:59:34 <slaweq> sorry mlavalle that we couldn't get to performance subteam topic today
21:59:47 <mlavalle> we'll do it next time
21:59:52 <mlavalle> np
21:59:56 <slaweq> thx for attending the meeting and have a great week all :)
22:00:04 <njohnston> o/
22:00:05 <mlavalle> o/
22:00:07 <amotoki> o/
22:00:07 <slaweq> stay safe
22:00:09 <slaweq> o/
22:00:13 <lajoskatona> o/
22:00:14 <slaweq> #endmeeting