15:00:00 <ralonsoh> #startmeeting neutron_qos
15:00:01 <openstack> Meeting started Tue Jul  2 15:00:00 2019 UTC and is due to finish in 60 minutes.  The chair is ralonsoh. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:04 <openstack> The meeting name has been set to 'neutron_qos'
15:00:07 <ralonsoh> Hello again!
15:00:18 <rubasov> hello
15:00:19 <slaweq> hi ralonsoh
15:00:29 * slaweq will be only lurking here
15:00:32 <ralonsoh> the meeting today is going to be very short
15:00:50 <ralonsoh> davidsha, are you ready for the Open Discussion?
15:00:59 <rubasov> I do not have updates for today's meeting and may need to drop off early
15:01:01 <ralonsoh> (let's wait 30 secs more)
15:01:08 <ralonsoh> rubasov, np
15:01:11 <davidsha> ralonsoh: yup!
15:01:17 <ralonsoh> let's start
15:01:20 <ralonsoh> let's start
15:01:24 <ralonsoh> #topic RFEs
15:01:28 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1578989
15:01:29 <openstack> Launchpad bug 1578989 in neutron "[RFE] Strict minimum bandwidth support (egress)" [Wishlist,Fix released] - Assigned to Bence Romsics (bence-romsics)
15:01:39 <ralonsoh> The only missing piece here is #link https://review.opendev.org/#/c/629253/
15:01:51 <ralonsoh> Lajos told me to say this in the meeting
15:02:06 <rubasov> that was merged I think
15:02:24 <ralonsoh> 10 days ago..... so I don't understand
15:02:26 <ralonsoh> hmmmm
15:02:31 <ralonsoh> ok, rubasov, any update?
15:02:49 <rubasov> so I think we have no more open changes for that rfe
15:03:01 <ralonsoh> I think we can remove this RFE from the agenda
15:03:04 <ralonsoh> what do you think?
15:03:18 <rubasov> yep, the neutron side is all done
15:03:25 <ralonsoh> if you have any new update, you can write it in the etherpad agenda (https://etherpad.openstack.org/p/neutron_qos_meeting_chair)
15:03:31 <ralonsoh> or just you can ping me
15:03:57 <rubasov> okay, we may come back with related news in nova when that happens
15:04:01 <ralonsoh> cool!
15:04:07 <ralonsoh> and again, thank you!
15:04:24 <rubasov> thanks to lajoskatona for these last patches
15:04:29 <ralonsoh> sure
15:05:03 <ralonsoh> ok, let's move on
15:05:04 <ralonsoh> #topic Bugs
15:05:14 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1826695
15:05:15 <openstack> Launchpad bug 1826695 in neutron "[L3][QoS] cache does not removed when router is down or deleted" [Medium,In progress] - Assigned to LIU Yulong (dragon889)
15:05:19 <ralonsoh> #link https://review.opendev.org/#/c/656105/
15:05:23 <ralonsoh> not too much activity
15:05:35 <ralonsoh> I'll ping liuyulong in the review
15:06:27 <ralonsoh> ok, let's move to the next one
15:06:27 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1834484
15:06:29 <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)
15:06:49 <ralonsoh> #link https://review.opendev.org/#/q/topic:bug/1834484+(status:open+OR+status:merged)
15:06:53 <liuyulong> Actually no much review activity from the community. We run such code locally for some times long.
15:07:34 <ralonsoh> liuyulong, but the CI is still failing
15:07:48 <ralonsoh> anyway, I'll try to review this patch today
15:08:11 <liuyulong> Let me rebase it once more.
15:08:17 <ralonsoh> liuyulong, thanks!
15:08:56 <ralonsoh> ok, about the second bug, 1834484
15:09:03 <ralonsoh> the problem in https://review.opendev.org/#/c/667998/
15:09:31 <ralonsoh> I found the issue and I'm both writting the problem description in the patch and implementing the solution
15:09:40 <ralonsoh> I'll push it ASAP
15:09:43 <rubasov> ralonsoh: thank you for working on that
15:10:06 <ralonsoh> rubasov, np! We had some testers in my company complaining about
15:10:09 <ralonsoh> about this
15:10:15 <ralonsoh> but it was easy to solve
15:10:35 <ralonsoh> (and now we have more information in the port DB object and OVO)
15:10:54 <ralonsoh> --> the network qos policy, always good to have it!
15:11:25 <ralonsoh> ok, no more bugs in the list (good!!)
15:11:32 <ralonsoh> Am I missing something?
15:11:50 <ralonsoh> cool
15:11:52 <ralonsoh> #topic Open Discussion
15:12:04 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1476527 - Adoption of Neutron Classifier API into QoS
15:12:05 <openstack> Launchpad bug 1476527 in neutron "[RFE] Add common classifier resource" [Wishlist,Triaged] - Assigned to Igor D.C. (igordcard)
15:12:07 <ralonsoh> davidsha,
15:12:50 <davidsha> I've been a bit locked waiting for feedback, but I might pre-empt one approach by making a patch thats moves Neutron classifier's resources into neutron/neutron-lib
15:13:06 <ralonsoh> yeeees
15:13:08 <davidsha> and rebase the main QoS patch on top of it
15:13:14 <ralonsoh> I support this idea!
15:13:37 <davidsha> Thanks, I'll try to get that up asap!
15:13:43 <ralonsoh> please, submit this patch and let the community comment about this
15:13:54 <davidsha> will do!
15:14:06 <ralonsoh> IMO, although that means increasing the complexity of QoS (just a bit, and only for DSCP)
15:14:27 <ralonsoh> this feature is difficult to be integrated outside Neutron, as a side project
15:14:50 <ralonsoh> the DB changes (related to DSCP policies) should be integrated in Neutron
15:14:52 <ralonsoh> and
15:15:01 <davidsha> Hopefully though the complexity would make it easier if we ever wanted to extend classification to other qos rules in the future, do it right once and all that
15:15:15 <ralonsoh> sure, that's right!
15:15:20 <ralonsoh> of course:
15:15:37 <ralonsoh> you should provide a code ready NOT to use the classification
15:15:40 <ralonsoh> that means
15:15:50 <ralonsoh> the code should with and without classification
15:15:51 <davidsha> of course
15:16:05 <ralonsoh> DO NOT BRAKE ANYTHING!!!
15:16:09 <ralonsoh> hehehehe
15:16:14 <davidsha> Only a little ;P
15:16:18 <ralonsoh> broke
15:16:30 <ralonsoh> break
15:16:35 <ralonsoh> sorry!! my bad english
15:16:39 <davidsha> np!
15:17:08 <ralonsoh> ok, so waiting for your n-lib patch, integrating the classification groups in the QoS
15:17:15 <ralonsoh> and the DB modifications in Neutron
15:17:44 <ralonsoh> am I missing anything here?
15:17:53 <davidsha> just to be sure, the prference is for Neutron classifier to go into Neutron-lib?
15:18:11 <ralonsoh> well, the API yes
15:18:26 <ralonsoh> the neutron classifier is going to be used outside Neutron?
15:19:03 <ralonsoh> the classifier should be a Neutron service, depending on QoS
15:19:03 <davidsha> kk, njohnston mentioned a while back that putting all of it into neutron-lib made sense to him, just wanted to see if that was what you meant
15:19:33 <ralonsoh> yes, but what are you going to put in n-lib?
15:20:11 <davidsha> I suppose just the API definitions, you need the db models and migrations in neutron.
15:20:34 <ralonsoh> yes, the DB model will be only in Neutro
15:20:44 <ralonsoh> and the service definition
15:20:50 <davidsha> kk
15:21:24 <ralonsoh> something else??
15:22:10 <davidsha> nope, thats all for me
15:22:47 <ralonsoh> ok, thank you all folks, see you in IRC and in two weeks here
15:22:58 <davidsha> thanks!
15:22:59 <ralonsoh> #endmeeting