14:00:10 <slaweq> #startmeeting networking
14:00:11 <openstack> Meeting started Tue Jul  2 14:00:10 2019 UTC and is due to finish in 60 minutes.  The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:14 <openstack> The meeting name has been set to 'networking'
14:00:18 <mlavalle> o/
14:00:22 <njohnston> o/
14:00:29 <slaweq> welcome everyone
14:00:40 <lajoskatona> o/
14:00:44 <boden> howdy
14:00:48 <liuyulong> hi
14:01:19 <tidwellr> hi
14:01:21 <amotoki> hi
14:01:24 <slaweq> #topic Announcements
14:01:49 <rubasov> hi
14:01:58 <bcafarel> hello
14:02:05 <slaweq> Next milestone, T-2 is week of July 22 - 26 - that is only 3 weeks from now
14:02:19 <ralonsoh> hi
14:02:56 <slaweq> Deadline for CFP for Shanghai summit is today: http://lists.openstack.org/pipermail/openstack-discuss/2019-May/006262.html
14:03:33 <slaweq> so it's last minute to propose talk if You want to have one there :)
14:03:49 <slaweq> Any other announcements from the team?
14:04:00 <mlavalle> not from me
14:04:23 <slaweq> ok
14:04:24 <slaweq> #topic Blueprints
14:04:44 <slaweq> Train-2 BPs are on https://launchpad.net/neutron/+milestone/train-2
14:05:06 <slaweq> any updates?
14:05:15 <mlavalle> I don't have any today
14:05:41 <mlavalle> other than say that the spec for Improve Extra Route API was approved last week
14:05:58 <mlavalle> so it's implementation ony now
14:06:29 <mlavalle> for Allow multiple segments per host for routed networks
14:06:30 <tidwellr> I have code for address scope affinity changes https://review.opendev.org/#/c/667511/
14:06:31 <slaweq> yes, I also updated couple of BPs and added to them links to specs, RFEs and/or patches in review
14:07:14 <mlavalle> tidwellr: is it good to go
14:07:16 <tidwellr> really I'm waiting on neutron-lib 1.28.0 before sprinting ahead
14:07:21 <mlavalle> zuul seems happy with it
14:07:36 <mlavalle> so still wip
14:07:37 <tidwellr> there are some todo's related to use of constants
14:08:04 <tidwellr> but other than that it's ready for reviews
14:08:13 <mlavalle> ok
14:09:15 <liuyulong> Is there any BP coming out from this: https://review.opendev.org/#/c/658414/ ?
14:09:31 <slaweq> thx tidwellr for work on this and update today :)
14:09:45 <mlavalle> liuyulong: eventually, several
14:10:43 <slaweq> I think that haleyb proposed (or will propose) new meeting to sync between neutron and networking-ovn so it can be discussed there
14:10:56 <slaweq> are You ok with this?
14:11:12 <haleyb> mlavalle: i am in the process of adding a meeting which can start next Tuesday at this time, once a month to start
14:11:35 <mlavalle> haleyb: thanks, that's what we agreed on last week
14:11:52 <haleyb> https://review.opendev.org/#/c/668013/ - i just need to remove my -W
14:12:43 <mlavalle> +1ed it
14:13:16 <haleyb> tidwellr: if you want to be a chair in ^^^ let me know :)
14:13:41 <tidwellr> let's co-chair
14:14:02 <haleyb> tidwellr: ack, i'll update it, i had added lucas as well
14:14:13 <haleyb> in case there is a bus...
14:15:23 <tidwellr> sounds good
14:16:15 <slaweq> ok, lets move forward then
14:16:45 <slaweq> I assume that there is no updates about other blueprints, right?
14:17:07 <mlavalle> not from me
14:17:16 <slaweq> ok
14:17:18 <slaweq> #topic Community goals
14:17:32 <slaweq> firs on our list is
14:17:37 <slaweq> *first
14:17:38 <slaweq> Run under Python 3 by default
14:17:58 <slaweq> we still have some stadium projects to switch to py3
14:18:21 <slaweq> etherpad with todo list https://etherpad.openstack.org/p/neutron_stadium_python3_status
14:18:42 <slaweq> recently my patch to switch neutron-fwaas tempest job to py3 was merged
14:19:00 <slaweq> so only functional tests job in fwaas repo left
14:19:28 <slaweq> but there are some other projects to take care still :)
14:19:29 <mlavalle> +1
14:19:35 <slaweq> njohnston any other updates on this?
14:20:22 <njohnston> No, I think we have volunteers for each of the remaining projects
14:20:23 <bcafarel> slaweq: no objection to moving around sections in that etherpad? (moving completed projects down)
14:20:43 <njohnston> I'll check on bgpvpn, that may be done now
14:20:46 <bcafarel> just to easily find remaining work
14:20:46 <slaweq> bcafarel: yes, that is good idea
14:20:55 <slaweq> bcafarel: I will do it tomorrow
14:21:05 <bcafarel> thanks
14:21:36 <slaweq> thx for this idea :)
14:21:48 <slaweq> njohnston: thx for taking care of bgpvpn
14:22:05 <slaweq> next goal on the list is
14:22:06 <slaweq> Support IPv6-Only Deployments (slaweq)
14:22:12 <slaweq> I didn't talk with gmann yet
14:22:34 <slaweq> but I was looking into storyboard related to this and it looks that there is no any work related to this yet
14:22:59 <slaweq> I will keep an eye on it to know if we will have something to do here
14:23:49 <mlavalle> Thanks!
14:23:56 <slaweq> and the last goal on the list is:
14:23:58 <slaweq> Enabling PDF generation support for project documentation (amotoki)
14:24:02 <slaweq> amotoki: any updates?
14:24:13 <amotoki> I tried the initial build of PDF doc last week.
14:24:32 <amotoki> https://review.opendev.org/#/c/667345/
14:25:02 <amotoki> As of now, we have a trouble when rendering inline neutron.conf sample file but if we drop it it succeeded.
14:25:22 <amotoki> links to raw sample files are also broken
14:25:37 <amotoki> so we need to discuss the general direction as the whole community. it is not specific to neutron.
14:25:52 <amotoki> i will keep my eyes on the discussoin.
14:25:56 <amotoki> that's all from me.
14:26:35 <slaweq> thx amotoki for the update and for taking care of this
14:26:52 <slaweq> anything else You want to talk regarding to community goals?
14:28:02 <slaweq> ok, lets move on
14:28:03 <slaweq> #topic Bugs
14:28:16 <rubasov> the deputy report: http://lists.openstack.org/pipermail/openstack-discuss/2019-July/007455.html
14:28:25 <slaweq> thx rubaso
14:28:30 <slaweq> You were faster :)
14:28:52 <rubasov> all gate failure bugs were fixed, including this one that did not make the report: http://lists.openstack.org/pipermail/openstack-discuss/2019-July/007455.html
14:29:28 <rubasov> may I call the attention of sqlalchemy experts to this bug: https://bugs.launchpad.net/neutron/+bug/1834484
14:29:29 <openstack> Launchpad bug 1834484 in neutron "[QoS] qos_plugin._extend_port_resource_request is killing port retrieval performance" [High,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
14:30:01 <ralonsoh> I still need to polish this: http://logs.openstack.org/98/667998/3/check/neutron-tempest-plugin-scenario-linuxbridge/7315f86/controller/logs/screen-q-agt.txt.gz?level=ERROR#_Jul_02_09_44_20_643049
14:30:27 <ralonsoh> I don't know why the server is not retrieving correctl the value
14:31:14 <ralonsoh> that's all from my side (https://review.opendev.org/#/q/topic:bug/1834484+(status:open+OR+status:merged))
14:31:18 <njohnston> is this under a context writer with the new engine facade?
14:31:35 * njohnston reconsiders, will debug with ralonsoh offline
14:31:38 <ralonsoh> njohnston, should be, I'm still reviwing this problem
14:31:44 <rubasov> ralonsoh: thank you for the quick patch
14:31:49 <ralonsoh> np!
14:31:57 <slaweq> I noticed that only linuxbridge job is failing due to this issue
14:32:29 <slaweq> but I'm not sure if same error is visible also in other jobs, but it not cause test failures or there is no similar errors in other jobs
14:32:36 <slaweq> ralonsoh: did You check logs of other jobs?
14:32:42 <lajoskatona> slaweq: as I see only the linuxbridge job is running scenario tests, or am I missing somthing?
14:32:47 <ralonsoh> I did and I don't see this error
14:33:04 <lajoskatona> slaweq: I checked and for other tempest jobs I can't see the sql error
14:33:05 <slaweq> there is also dvr multinode job which is non voting
14:33:07 <ralonsoh> LB agent is the only one using the common agent loop and this call
14:33:29 <slaweq> ok
14:34:28 <slaweq> anything else You want to talk about rubasov?
14:34:44 <rubasov> all other bugs have assignees from last week
14:34:50 <slaweq> I think other bugs from last week are in pretty good state now, right?
14:35:04 <rubasov> slaweq: yep
14:35:05 <slaweq> ok, thx
14:35:13 <mlavalle> I have a question for rubasov
14:35:21 <slaweq> mlavalle: sure, go on
14:35:24 <rubasov> mlavalle: shoot
14:35:45 <mlavalle> we don't need to discuss the rfe https://bugs.launchpad.net/neutron/+bug/1833674, right?
14:35:47 <openstack> Launchpad bug 1833674 in neutron "[RFE] Improve profiling of port binding and vif plugging" [Wishlist,In progress] - Assigned to Bence Romsics (bence-romsics)
14:36:08 <rubasov> mlavalle: I don't think so
14:36:17 <rubasov> mlavalle: it means to cover what we agreed on the ptg
14:36:27 <rubasov> mlavalle: I created it for tracking reasons
14:36:33 <mlavalle> cool. I'll mark it approved then
14:38:21 <slaweq> ok, are You ok mlavalle now?
14:38:37 <mlavalle> yes, thanks :-)
14:38:46 <slaweq> great
14:38:58 <slaweq> this week our bug deputy is bcafarel
14:39:13 <slaweq> and next week will be me (so I don't need to ping anyone :P)
14:39:21 <bcafarel> bug deputy on deck o/
14:39:46 <slaweq> thx bcafarel
14:39:54 <slaweq> so we are in good hands this week :)
14:40:00 <bcafarel> :)
14:40:06 <slaweq> any other bugs You want to talk about today?
14:40:10 <mlavalle> not me
14:40:23 <rubasov> not me
14:40:46 <slaweq> moving on then
14:40:49 <slaweq> next topic
14:40:51 <slaweq> #topic neutron-lib
14:40:57 <slaweq> boden: Your turn :)
14:41:00 <boden> hi
14:41:31 <boden> quickly; we mentioned neutron-lib 1.28.0 earlier... as far as I'm concerned it's ready to be released... see https://review.opendev.org/#/c/666870/
14:42:09 <boden> also I have a question about stable branches and neutron-lib.. probably a question for amotoki
14:42:48 <boden> is there a way to back-port a bug fix for a "stable" version of neutron-lib?? for example a bug in the version of neutron-lib that was used with stein or queens
14:43:27 <boden> anybody know?
14:43:31 <amotoki> neutron-lib has stable branches so it would be same as usual.
14:43:57 <amotoki> if we don't have any version number for them, it needs a discussion
14:44:02 <boden> amotoki, but stable branches in neutron-lib point to some arbitrary commit... they are not at all "linked" to the tags we release to PyPI
14:44:30 <boden> e.g. look at stable/stein and 1.25.0 for neutron-lib... they are different and 1.25.0 is wahts used by neutron stable/stein
14:44:57 <slaweq> boden: is tag 1.25 ahead of behind stable/stein?
14:45:10 <slaweq> s/of/or
14:45:10 <boden> behind.. I think, but would need to verify
14:45:29 <bcafarel> only opendev migration patch for now
14:45:35 <bcafarel> (+ the usual commits for branch creation)
14:45:42 <boden> I think stable/stein branch for lib was just created when we did it for all the other net projects... not related to the actual 1.25.0 release on pypi
14:45:43 <amotoki> 1.25.0 is the last release for stein and stable/stein is cut from there,
14:45:52 <slaweq> so IMO shouldn't be problem if we will release e.g. 1.25.1 from stable/stein branch
14:46:04 <amotoki> so we can backport any bug fix to stable/stein branch. we can cut 1.25.1 release then
14:46:56 <amotoki> stable/stein branch was created for usual maintenance too, so I believe there is no thing different.
14:47:07 <boden> how about I look into the details of 1.25.0 vs stable/stein after this meeting and I can ping slaweq and amotoki then? no need to digress here I suppose
14:47:31 <amotoki> boden: sure. makes sense
14:47:31 <slaweq> boden: good for me
14:47:36 <boden> thanks guys!
14:47:49 <boden> nothing else on neutron-lib from me.. just business as usual
14:48:29 <bcafarel> upper-constraints in requirements does state neutron-lib===1.25.0
14:48:36 <amotoki> this is only difference between 1.25.0 and stable/stein now http://paste.openstack.org/show/753763/
14:48:41 <amotoki> the diff is super small.
14:49:21 <boden> bcafarel.. hmm, can we bump the upper-constraints in stable/stein?
14:50:07 <amotoki> we can bump u-c in stable branches. bumping the minimum requirements needs to be more carefl.
14:50:10 <amotoki> *careful
14:50:32 <bcafarel> looks like we can from git log I see clients and oslo.cache updates
14:50:57 <bcafarel> amotoki: ack that one makes sense
14:52:31 <slaweq> ok, lets move on then
14:52:33 <slaweq> #topic On demand agenda
14:52:46 <slaweq> there is nothing in this section for today
14:52:56 <njohnston> I have one question
14:52:56 <slaweq> but I just remembered one thing which I wanted to ask
14:53:06 <slaweq> and as we talked about requirements, it's kind of related
14:53:15 <slaweq> we have bug https://bugs.launchpad.net/neutron/+bug/1835044
14:53:17 <openstack> Launchpad bug 1835044 in neutron "[Queens] Memory leak in pyroute2 0.4.21" [Undecided,New]
14:53:23 <slaweq> which is only in stable/queens release
14:53:25 <ralonsoh> https://review.opendev.org/#/c/668676/
14:53:35 <ralonsoh> There is a leak in pyroute
14:53:40 <slaweq> can we bump pyroute2 version in stable/queens to 0.5.2?
14:53:43 <slaweq> do You know?
14:53:43 <ralonsoh> both in py2.7 and py3.6
14:53:56 <ralonsoh> That's what I want to know
14:54:01 <ralonsoh> I really don't know
14:54:02 <ralonsoh> but
14:54:14 <amotoki> AFAIK we only update the upper-constraints in most cases as long as it works with both versions.
14:54:15 <ralonsoh> if the Neutron CI tests pass, I see no problem
14:54:33 <ralonsoh> that's the point: I have a patch to test it
14:54:33 <slaweq> ok
14:54:40 <ralonsoh> https://review.opendev.org/#/c/668677/
14:54:53 <mlavalle> if needed, we can ask the stable branches team
14:54:59 <ralonsoh> of course, I don't know the repercussion in other out-of-tree projects
14:55:07 <ralonsoh> but of course, we need to handle this problem
14:55:18 <slaweq> I just wasnt't sure what are policies for doing such big bump of version in stable branch
14:55:56 <amotoki> isn't it better to start a mailing list discussion including the requirements and stable teams?
14:55:57 <ralonsoh> let's see first if the neutron patch (CI testing) works
14:56:08 <bcafarel> it may not be such big jump in pyroute, high level changes between 0.4.21 and 0.5.2 maybe useful
14:56:08 <ralonsoh> ok, I'll send a mail for this
14:56:18 <slaweq> amotoki: good idea
14:56:20 <slaweq> ralonsoh: thx
14:56:41 <slaweq> ok, lets talk about it in review and on ML
14:56:46 <slaweq> njohnston: You had question also
14:56:48 <njohnston> mlavalle: In Denver you mentioned sending an email to reinvigorate the fwaas community (or show that the project is unmanned), I was wondering if you sent that and I missed it?
14:56:49 <slaweq> go on
14:56:58 <haleyb> i think we can bump things to a degree, just have to make release version minor bump, eg 12.0.6 -> 12.1.0
14:57:13 <haleyb> didn't mean to step on question
14:57:14 <mlavalle> njohnston: no, I will do it soon
14:57:21 <njohnston> mlavalle: thanks!
14:57:32 <njohnston> that's all from me
14:57:38 <slaweq> thx njohnston
14:57:46 <slaweq> we have 3 minutes left
14:57:53 <slaweq> anyone else wants to talk about something?
14:58:09 <bcafarel> haleyb: similar to the recent queens release situation then
14:58:33 <haleyb> bcafarel: i think, but i didn't know about that until it happened
14:58:42 <bcafarel> same here :)
14:58:48 <haleyb> i just don't think we can go past u/c
14:59:25 <bcafarel> I think it will probably be "check all projects that depend on pyroute and run ci on them"
14:59:29 <slaweq> ok, we are almost out of time now
14:59:38 <slaweq> thx for attending and have a great week
14:59:40 <slaweq> o/
14:59:42 <bcafarel> o/
14:59:42 <slaweq> #endmeeting