14:08:16 <lajoskatona> #startmeeting networking 14:08:16 <opendevmeet> Meeting started Tue Feb 22 14:08:16 2022 UTC and is due to finish in 60 minutes. The chair is lajoskatona. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:08:16 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:08:16 <opendevmeet> The meeting name has been set to 'networking' 14:08:20 <mlavalle> o/ 14:08:22 <lajoskatona> o/ 14:08:23 <ralonsoh> hi 14:08:27 <slaweq> hi 14:08:28 <isabek> Hi 14:08:32 <amotoki> hi 14:08:35 <rubasov> o/ 14:08:54 <ykarel> o/ 14:09:02 <bcafarel> \o 14:09:19 <obondarev> hi 14:09:25 <lajoskatona> #topic Announcements 14:09:37 <lajoskatona> Release countdown, R-5: http://lists.openstack.org/pipermail/openstack-discuss/2022-February/027342.html 14:09:47 <lajoskatona> this means: "February 24th, 2022 is the Yoga-3 milestone, also known as feature freeze." 14:10:17 <lajoskatona> so this week we have to finalize the patches for Yoga features 14:11:27 <slaweq> lajoskatona should we focus only on reviewing patches which have review-priority set now? 14:11:41 <lajoskatona> slaweq: yes 14:11:42 <slaweq> or are there any others which should be prioritize too? 14:12:18 <lajoskatona> I increased review priority for some, but I have to check the list because perhaps not all can be reviewd in the few days we have 14:12:50 <slaweq> ok 14:12:52 <slaweq> thx 14:13:25 <lajoskatona> if you know any which we should focus on please add to the priority list 14:14:48 <lajoskatona> ok, if nothing more to add to Announcements..... 14:15:07 <lajoskatona> #topic Bugs 14:15:14 <lajoskatona> Report from slaweq: http://lists.openstack.org/pipermail/openstack-discuss/2022-February/027351.html 14:15:31 <lajoskatona> it was really a quiet week :-) 14:15:38 <slaweq> indeed :) 14:16:14 <lajoskatona> slaweq: is there any bug to focus on? 14:17:37 <slaweq> not really, I think all is covered already 14:17:52 <lajoskatona> ok, as I see all bugs are assigned 14:17:58 <lajoskatona> ok, thanks 14:18:32 <lajoskatona> hongbin can't be bug deputy in the future (he resigns from his coreship also) I take this week, and next week haleyb will be. 14:18:57 <lajoskatona> hongbin also asked me to remove him from the core reviewer group 14:19:28 <ralonsoh> sad to read this 14:19:35 <lajoskatona> yeah 14:19:57 <mlavalle> :-( 14:20:07 <slaweq> :( 14:21:04 <lajoskatona> liuyulong: is there anything you would like to discuss for L3? 14:21:24 <liuyulong> One RFE: https://bugs.launchpad.net/neutron/+bug/1961011 14:22:22 <liuyulong> Currently, I'm focusing on the retirements of subprocess of L3 routers. 14:23:03 <liuyulong> And, I noticed the radvd, which is used to send out RA simply. 14:23:24 <lajoskatona> liuyulong: nice this is again to give more responsibility to ovs-agent 14:24:10 <liuyulong> So, I raise such approach to remove radvd for L3 router. 14:24:43 <liuyulong> lajoskatona, yes, but for ovs-agent, it many times run empty loops. 14:25:05 <liuyulong> rpc_loop is idle when there is no new ports events. 14:25:30 <lajoskatona> ok: we can discuss in details on Friday during the drivers meeting, but sounds promising 14:25:37 <liuyulong> So, I think the time window is sufficient for these agent extensions. 14:26:29 <trident> Any possibility https://bugs.launchpad.net/neutron/+bug/1952907 could get some attention / input on potential solutions? We are seeing pretty severe effects from it. 14:28:11 <lajoskatona> trident: the proposed solutions doesn't work for you? I just quickly run through the comments 14:29:17 <ralonsoh> actually there are three patches 14:29:21 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/712474 14:29:23 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/821433 14:29:27 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/821434 14:29:31 <liuyulong> Yes, many splits patches 14:31:32 <lajoskatona> Hmmmm 14:32:01 <trident> So far I have not been able to find a combination that works. Seems a bit like a rabbit hole making it work for all version of keepalived. 14:32:38 <lajoskatona> as these patches hang for month, shall we discuss how to push it during the drivers meeting? 14:33:05 <lajoskatona> trident: so it seems that the proposals are not enough? 14:33:06 <liuyulong> There are many vrrp config options for keepalived, vrrp_garp_master_delay, rrp_garp_master_repeat, vrrp_garp_interval. 14:33:20 <liuyulong> It's too long time for me to remember the use of each. 14:33:41 <ralonsoh> agree, we can re-read first the patches and comment them on friday 14:34:12 <ralonsoh> I'll add this topic in the drivers meeting 14:34:54 <liuyulong> But for my personal option, vrrp config options should be a better solution for this. 14:35:10 <ralonsoh> but should be widely supported by keepalived versions 14:35:19 <ralonsoh> but we can check that on friday 14:35:22 <lajoskatona> agree 14:35:30 <trident> Sounds great! Will keep researching and testing until then and will try to get Damian who filed the initial report into the meeting as well. 14:35:43 <lajoskatona> ok, than let's check it on Friday, and reread the patches and cfg options :-) 14:36:07 <lajoskatona> trident: thanks that would be really good 14:37:00 <lajoskatona> #topic On Demand Agenda 14:37:32 <lajoskatona> I added one topic, for Outreachy 14:38:11 <lajoskatona> I sent a mail that I have an idea which should be possible to "outsource" to an intern with mentoring perhaps 14:38:40 <lajoskatona> It's about checking the overlapping tests that we have in neutron-tempest-plugin and in tempest 14:39:01 <slaweq> ++ great idea :) 14:39:26 <lajoskatona> If there's other ideas that would be great 14:40:03 <lajoskatona> as I see this can be not deep dive to networking and Neutron code, but really useful for the candidate 14:40:32 <lajoskatona> and useful for the community also to reduce code duplicatin and perhaps even test execution time 14:40:50 <mlavalle> It's a good introduction to the API and different networking scenarios 14:41:07 <elvira> And it would require them to get familiar with the openstack environment which is already something 14:41:26 <lajoskatona> exactly 14:41:44 <elvira> would that greatly improve the CI? 14:42:03 <lajoskatona> the only issue is that the QA team has to be aware of it 14:42:14 <lajoskatona> so I have to ask them also about it 14:43:10 <lajoskatona> elvira: I don't expect big improvement but if we have less code/test duplication that is even good 14:43:33 <lajoskatona> ok, thanks for the feedback 14:44:36 <lajoskatona> That's it from me for today's meeting 14:44:37 <liuyulong> What about docs, typos, unit test cases, pep8 fixes ? 14:45:49 <lajoskatona> liuyulong: those are issues also, but perhaps harder to do in 3 months timeframe and measure if it is finished 14:47:28 <liuyulong> Alright, maybe they can do some statistical work, such as bugs or CI failure rates. 14:47:30 <slaweq> lajoskatona maybe some RFE? I would need to check what are currently approved but not worked on 14:47:52 <mlavalle> RFEs are a better idea 14:48:00 <mlavalle> slaweq: ++ 14:48:21 <lajoskatona> do we have low hanging RFEs? 14:48:35 <ralonsoh> bugs, not RFEs 14:48:39 <slaweq> lajoskatona I will check it this week and will reply to Your email 14:48:43 <slaweq> when is deadline for that? 14:49:23 <mlavalle> We can go back over the next few days and look at what we have approved and see if any of those can be tackled in the 3 months period by a newby 14:49:27 <lajoskatona> Next Friday I think 14:49:29 <mlavalle> slaweq: March 4th 14:49:42 <slaweq> ok 14:49:59 <liuyulong> Soma bugs are achievable in 3 months, RFE may need many works of coding, testing and docs. 14:50:15 <liuyulong> sorry, somall bugs 14:51:45 <mlavalle> So deadline is next week's Friday 14:51:48 <lajoskatona> good, we can collect ideas on the wiki: https://wiki.openstack.org/wiki/Network/Meetings 14:52:07 <lajoskatona> and vote next meeting? 14:52:43 <slaweq> ok 14:52:57 <mlavalle> yeah, let's add it as a discussion topic next meeting 14:53:06 <lajoskatona> +1 14:54:05 <lajoskatona> ok, thanks for the ideas and common thinking 14:54:30 <mlavalle> I would encourage folks to read Dmitry's blog post that lajoskatona included in his email 14:55:02 <mlavalle> we don't want to just give "busy work" to the interns. We want to give them a meaningful experience 14:55:54 <mlavalle> lajoskatona: maybe post the blog post here? 14:55:56 <lajoskatona> yes that's really important to make them feel "it is a good team with meaningful things to do" :-) 14:56:11 <lajoskatona> https://owlet.today/posts/how-to-propose-the-best-project-for-outreachy/ 14:56:15 <mlavalle> ++ 14:57:09 <lajoskatona> ok, if nothing more I think we can close the meeting 14:57:23 <slaweq> o/ 14:57:25 <lajoskatona> #endmeeting