15:01:09 <ralonsoh> #startmeeting neutron_qos
15:01:10 <openstack> Meeting started Tue May 23 15:01:09 2017 UTC and is due to finish in 60 minutes.  The chair is ralonsoh. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:11 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:14 <openstack> The meeting name has been set to 'neutron_qos'
15:01:20 <ralonsoh> #topic RFEs
15:01:33 <ralonsoh> Today the agenda is pretty full
15:01:57 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1560961
15:02:01 <openstack> Launchpad bug 1560961 in neutron "[RFE] Allow instance-ingress bandwidth limiting" [Wishlist,Fix released] - Assigned to Slawek Kaplonski (slaweq)
15:02:07 <ralonsoh> The API code is merged
15:02:15 <ralonsoh> Congrats slaweq !
15:02:26 <davidsha> Congrats!
15:02:32 <ralonsoh> OVS implementation: #link https://review.openstack.org/#/c/457816/
15:02:38 <slaweq> hello
15:02:40 <slaweq> and thx
15:02:41 <slaweq> :)
15:02:52 <ralonsoh> this patch is going well
15:02:53 <ralonsoh> but
15:02:59 <ralonsoh> #link https://review.openstack.org/#/c/466899/
15:03:06 <slaweq> yes, but first I need to do one more small patch
15:03:21 <ralonsoh> This second patch to change the delete handler is kind of rejected
15:03:23 <slaweq> https://review.openstack.org/#/c/466899/
15:03:39 <ralonsoh> Ok, I don't see Ihar here....
15:03:49 <slaweq> I talked with ihrahys yesterday night
15:03:54 <ralonsoh> and?
15:03:56 <slaweq> I know what to do there
15:04:12 <ralonsoh> to add several entry points?
15:04:14 <slaweq> I will add there also calls to "delete_ingress_bandwidth_limit_rule"
15:04:21 <slaweq> and other such rules
15:04:23 <ralonsoh> pffff I don't like it
15:04:34 <slaweq> so basically I will do what mlavalle proposed in comment
15:04:40 <ralonsoh> I know
15:04:48 <ralonsoh> but I prefer to have a unique entry point
15:05:01 <ralonsoh> and the backend will accept or not the parameter
15:05:08 <ralonsoh> but it's not big deal
15:05:25 <slaweq> I'm also not big fan of this but what I can do :)
15:05:31 <ralonsoh> ok, I'll review your following patches
15:05:40 <ralonsoh> no problem, do this way
15:05:45 <slaweq> thx
15:05:52 <ralonsoh> next one
15:05:53 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1578989
15:05:54 <openstack> Launchpad bug 1578989 in neutron "[RFE] Strict minimum bandwidth support (egress)" [Wishlist,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
15:06:06 <ralonsoh> I need to talk to Miguel Lavalle
15:06:13 <ralonsoh> I can't see him now
15:06:45 <ralonsoh> so i'll postpone this conversation and I'll submit the conversation results
15:06:55 <ralonsoh> next one
15:06:57 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1596611
15:06:58 <openstack> Launchpad bug 1596611 in neutron "[RFE] Create L3 IPs with qos (rate limit)" [Wishlist,Triaged] - Assigned to LIU Yulong (dragon889)
15:07:13 <ralonsoh> First, the spec
15:07:14 <ralonsoh> #link https://review.openstack.org/#/c/374506/
15:07:41 <ralonsoh> It's almost approved, although there are some gaps in the implementation details
15:07:51 <ralonsoh> but if cores accept it, that's ok
15:08:04 <ralonsoh> please, take a final look at the spec
15:08:15 <slaweq> I will try to look on it today
15:08:20 <ralonsoh> second, the TC changes
15:08:20 <ralonsoh> #link https://review.openstack.org/#/c/453458/
15:08:34 <ralonsoh> I don't agree with having two TC wrappers
15:08:38 <ralonsoh> I'm against it
15:09:10 <ralonsoh> and I was assertive on this idea
15:09:19 <slaweq> what's the point to have two classes for same puprose?
15:09:26 <ralonsoh> one sec
15:09:43 <ralonsoh> "l3 tc lib has much differences from original tc_lib. The file is stored separately is more clearly."
15:09:54 <ralonsoh> the answer from the reviews
15:10:00 <slaweq> what are those differences?
15:10:23 <ralonsoh> the way the qdiscs and the filters are created and filtered
15:10:32 <slaweq> I will check it today also but I'm not sure if it's good idea
15:10:51 <slaweq> yes, but can't it be added in separate methods in existing class?
15:10:59 <ralonsoh> I appreciate your reviews
15:11:33 <ralonsoh> IMO, it's just a matter of merging both wrappers
15:11:58 <ralonsoh> and taking care of the functionalities added by both classes, the current one and the new one
15:12:00 <slaweq> I agree with You but I will take a look on it tonight
15:12:07 <ralonsoh> perfect
15:12:20 <ralonsoh> and 3rd, the API
15:12:21 <ralonsoh> #link https://review.openstack.org/#/c/424466/
15:12:29 <ralonsoh> IMO, that needs a big refactor
15:12:46 <ralonsoh> this API is breaking the current QOS API
15:13:01 <ralonsoh> my comments are in the review
15:13:13 <ralonsoh> btw
15:13:28 <ralonsoh> it's anyone, apart from slaweq and me here??
15:13:33 <ralonsoh> davidsha
15:13:50 <davidsha> Sorry was looking at the patches!
15:13:54 <ralonsoh> no problem
15:14:17 <ralonsoh> so, just to recap this RFE: there is a lot of work to be done in the API coding
15:14:31 <ralonsoh> next one
15:14:32 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1560961
15:14:33 <openstack> Launchpad bug 1560961 in neutron "[RFE] Allow instance-ingress bandwidth limiting" [Wishlist,Fix released] - Assigned to Slawek Kaplonski (slaweq)
15:14:44 <ralonsoh> code merged! congrats
15:15:01 <ralonsoh> I'll remove this one from the list
15:15:04 <slaweq> but it's only API patch merged :)
15:15:14 <ralonsoh> I know
15:15:21 <slaweq> :)
15:15:28 * mlavalle got confused with the meeting start time :-(
15:15:32 <ralonsoh> Which backend are you going to do first
15:15:35 <ralonsoh> ?
15:15:41 <slaweq> but as I said before it's moving forward :)
15:15:46 <slaweq> ovs is first one
15:15:52 <davidsha> Hey mlavalle
15:15:54 <ralonsoh> perfect
15:15:59 <ralonsoh> hi mlavalle, no problem
15:16:03 <slaweq> https://review.openstack.org/#/c/457816/
15:16:10 <slaweq> this is patch for ovs ingress bw limit
15:16:25 <slaweq> for lb I have something done partially in my local devstack
15:16:31 <ralonsoh> oh, yes! I missed that in the meeting etherpad. One sec....
15:17:15 <ralonsoh> ok, I had this spec twice!
15:17:22 <ralonsoh> hehehee
15:17:33 <ralonsoh> perfect, next one
15:17:35 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1639220
15:17:36 <openstack> Launchpad bug 1639220 in neutron "[RFE] Introduce Network QoS policy "is_default" behaviour" [Low,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
15:17:43 <mlavalle> slaweq: should I expect a revision to https://review.openstack.org/#/c/457816/ before reviewing again?
15:18:11 <ralonsoh> yes, there should be a new one
15:18:14 <slaweq> mlavalle: yes
15:18:19 <mlavalle> ++
15:18:35 <ralonsoh> first to remove the delete handling code and more stuff
15:18:37 <slaweq> I will need to rebase it when https://review.openstack.org/#/c/466899/ will be finished
15:19:01 <slaweq> mlavalle: I talked with ihrahys yesterday and we deviced to make it as You proposed :)
15:19:07 <mlavalle> ahh, that's the new one, ok, will also take a look
15:19:19 <slaweq> mlavalle: but this new one is not finished yet
15:19:21 <mlavalle> slaweq: yeah,m I followed the conversation in channel yesterday
15:20:00 <ralonsoh> ok, going to the "is_default" one
15:20:05 <slaweq> ok
15:20:11 <ralonsoh> the code is waiting....
15:20:25 <mlavalle> reviews, maybe?
15:20:33 <ralonsoh> it has one +2 and one +1
15:20:40 <ralonsoh> yes, final ones
15:20:46 <ralonsoh> that's all
15:21:13 <ralonsoh> ok, let's move to the next one
15:21:18 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1505627
15:21:19 <openstack> Launchpad bug 1505627 in neutron "[RFE] QoS Explicit Congestion Notification (ECN) Support" [Wishlist,Triaged] - Assigned to Reedip (reedip-banerjee)
15:21:21 <mlavalle> ok, will take a look
15:21:33 <ralonsoh> reedip?
15:22:23 <ralonsoh> I'll try to ping reedip latter today. The spec is stopped since march
15:22:25 <mlavalle> The comment I have from the drivers meeting is that they want an opinion from the QoS gang as to whether this is something you are going to pursue
15:23:03 <ralonsoh> well, at least should be interesting to have a detailed spec to propose
15:23:20 <mlavalle> that would be a good next step
15:23:23 <slaweq> ++
15:23:29 <ralonsoh> I'll join next drivers meeting to ask this
15:24:00 <ralonsoh> This is mainly reedip's work
15:24:08 <mlavalle> it is very late for you: 2300UTC
15:24:17 <ralonsoh> no, problem
15:24:20 <mlavalle> ok
15:24:36 <slaweq> mlavalle: I think it's 2200 UTC
15:24:42 <ralonsoh> and the last one
15:24:43 <mlavalle> yes, slaweq is right
15:24:58 <ralonsoh> yes, it's 2200UTC
15:25:12 <ralonsoh> next one
15:25:13 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1686035
15:25:14 <openstack> Launchpad bug 1686035 in neutron "[RFE] More detailed reporting of available QoS rules" [Wishlist,Triaged] - Assigned to Slawek Kaplonski (slaweq)
15:25:26 <ralonsoh> I have one concern: ihar's comment
15:25:43 <slaweq> I wrote some specs for this once https://review.openstack.org/#/c/466500/
15:25:51 <mlavalle> yeah, same concern here
15:25:51 <ralonsoh> about doing this feature more generic: (https://review.openstack.org/#/c/386555/7/guidelines/capabilities.rst)
15:26:11 <slaweq> I don't know how to do it more generic in fact
15:26:20 <ralonsoh> Yes, but this spec is focused only in QoS
15:26:24 <slaweq> but I wanted to be on next drivers meeting to talk about it also
15:26:38 <ralonsoh> I know, it's very difficult to make a generic API for all extensions
15:26:58 <ralonsoh> slaweq: that will be great
15:27:25 <ralonsoh> I have some concerns about the differentiation between the admin user and regular user
15:27:41 <slaweq> I even don't know exactly what other resources should be available in same way (backend support details)
15:27:45 <ralonsoh> the differences in the information sent
15:28:14 <ralonsoh> if you join next day the drivers meeting, can you ask this too?
15:28:36 <ralonsoh> having more feedback will be very helpful
15:28:43 <slaweq> yes, I will ask them
15:29:00 <slaweq> I wanted to write some first draft of specs to discuss it on drivers meeting
15:29:40 <ralonsoh> yes, I understand, and that's ok. You have smething to present
15:29:51 <slaweq> exactly :)
15:30:10 <ralonsoh> one more thing before jumping to the other section
15:30:13 <mlavalle> slaweq: I will try to provide feedback before the drivers meeting
15:30:20 <slaweq> thx mlavalle
15:30:49 <ralonsoh> mlavalle: after this meeting, do you have time to talk about https://bugs.launchpad.net/neutron/+bug/1578989?
15:30:51 <openstack> Launchpad bug 1578989 in neutron "[RFE] Strict minimum bandwidth support (egress)" [Wishlist,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
15:31:08 <mlavalle> sure
15:31:13 <ralonsoh> perfect
15:31:30 <ralonsoh> ok, let's go to bugs
15:31:33 <ralonsoh> #topic Bugs
15:31:39 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1676877
15:31:40 <openstack> Launchpad bug 1676877 in neutron "Increase "TestQosPlugin.test_update_policy_rule" coverage" [Medium,In progress] - Assigned to Reedip (reedip-banerjee)
15:32:11 <ralonsoh> reedip has this one on-hold
15:32:39 <ralonsoh> as I said, I'll try to ping reedip a bit latter
15:32:55 <ralonsoh> next one
15:32:56 <ralonsoh> https://bugs.launchpad.net/neutron/+bug/1657381
15:32:57 <openstack> Launchpad bug 1657381 in neutron "QoS drivers need to implement a precommit for the actions" [Medium,In progress] - Assigned to Rodolfo Alonso (rodolfo-alonso-hernandez)
15:33:51 <ralonsoh> The code is delivered and just to be reviewed
15:34:06 <ralonsoh> Please, this patch is almost finished
15:34:16 <slaweq> ok, I will check it
15:34:48 <ralonsoh> and the last one
15:34:50 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/1649517
15:34:50 <openstack> Launchpad bug 1649517 in neutron "qos policy attached to network, qos_policy_id is reflecting on neutron net-show , but not on the port with neutron port-show" [Wishlist,In progress] - Assigned to Reedip (reedip-banerjee)
15:35:05 <ralonsoh> I belongs to reedip
15:35:14 <ralonsoh> I'll ping him
15:35:16 <slaweq> reedip took this one from me but he didn't do anything there yet
15:35:53 <ralonsoh> I'll try to ping him to know which patches is going to continue
15:36:10 <ralonsoh> if he is overloaded, I'll ask him to transfer some of them
15:36:41 <ralonsoh> I'll try to have the answer this week
15:36:59 <slaweq> ok
15:37:05 <ralonsoh> and that's all!
15:37:24 <ralonsoh> there are no more topics
15:37:37 <ralonsoh> #topic Open Discussion
15:37:37 <davidsha> ack, If he needs to offload that work I can take that bug.
15:37:50 <ralonsoh> davidsha: that's perfect
15:38:01 <ralonsoh> as I said, I'll try to ping him this week
15:38:41 <ralonsoh> does anyone any other topic to talk about?
15:38:46 <ralonsoh> or any question?
15:39:02 <davidsha> I'm good.
15:39:07 <mlavalle> me too
15:39:14 <slaweq> me too
15:39:24 <mlavalle> davidsha: where are you located?
15:39:24 <ralonsoh> perfect
15:39:30 <davidsha> mlavalle: Ireland
15:39:47 <mlavalle> yeah, with that last name, that's what I thought ;-)
15:39:52 <davidsha> :P
15:40:01 <ralonsoh> so please, take a look at "L3 IP limit" and "ingress limiting BW"
15:40:10 <ralonsoh> both RFEs are important
15:40:15 <davidsha> To be fair there is more Irish outside of Ireland than in it!
15:40:15 <slaweq> ralonsoh: yes sir! :P
15:40:23 <ralonsoh> hehehe
15:40:25 <davidsha> ralonsoh: Will do!
15:40:31 <mlavalle> davidsha: yeah, I know, That' why I asked
15:40:34 <slaweq> davidsha: in Ireland there is more poles probably :D
15:41:15 <mlavalle> LOL
15:41:28 <ralonsoh> hehehehe
15:41:49 <ralonsoh> ok, I'm going to end the meeting. See you next week and in IRC
15:42:10 <ralonsoh> #endmeeting