| *** rfolco has quit IRC | 00:22 | |
| *** rfolco has joined #openstack-meeting | 00:22 | |
| *** toabctl has quit IRC | 00:24 | |
| *** tetsuro has quit IRC | 00:24 | |
| *** rfolco has quit IRC | 00:24 | |
| *** tetsuro has joined #openstack-meeting | 00:24 | |
| *** rfolco has joined #openstack-meeting | 00:24 | |
| *** armax has quit IRC | 00:24 | |
| *** rfolco has quit IRC | 00:25 | |
| *** tetsuro has quit IRC | 00:25 | |
| *** rfolco has joined #openstack-meeting | 00:25 | |
| *** tetsuro has joined #openstack-meeting | 00:26 | |
| *** toabctl has joined #openstack-meeting | 00:27 | |
| *** rfolco has quit IRC | 00:28 | |
| *** rfolco has joined #openstack-meeting | 00:29 | |
| *** jawad_axd has joined #openstack-meeting | 00:30 | |
| *** rfolco has quit IRC | 00:30 | |
| *** rfolco has joined #openstack-meeting | 00:31 | |
| *** rfolco has quit IRC | 00:33 | |
| *** rfolco has joined #openstack-meeting | 00:33 | |
| *** jawad_axd has quit IRC | 00:34 | |
| *** rfolco has quit IRC | 00:38 | |
| *** mattw4 has quit IRC | 00:38 | |
| *** ricolin has joined #openstack-meeting | 00:48 | |
| *** jawad_axd has joined #openstack-meeting | 00:51 | |
| *** jawad_axd has quit IRC | 00:55 | |
| *** jamesmcarthur has joined #openstack-meeting | 00:57 | |
| *** jmasud has joined #openstack-meeting | 00:57 | |
| *** jawad_axd has joined #openstack-meeting | 01:12 | |
| *** jawad_axd has quit IRC | 01:16 | |
| *** jamesmcarthur has quit IRC | 01:17 | |
| *** masahito has joined #openstack-meeting | 01:20 | |
| *** Dinesh_Bhor has quit IRC | 01:22 | |
| *** Dinesh_Bhor has joined #openstack-meeting | 01:26 | |
| *** Lucas_Gray has quit IRC | 01:29 | |
| *** ociuhandu has joined #openstack-meeting | 01:31 | |
| *** jawad_axd has joined #openstack-meeting | 01:32 | |
| *** armax has joined #openstack-meeting | 01:35 | |
| *** ociuhandu has quit IRC | 01:35 | |
| *** jawad_axd has quit IRC | 01:37 | |
| *** igordc has quit IRC | 01:43 | |
| *** rh-jelabarre has joined #openstack-meeting | 02:02 | |
| *** jawad_axd has joined #openstack-meeting | 02:14 | |
| *** jawad_axd has quit IRC | 02:18 | |
| *** jawad_axd has joined #openstack-meeting | 02:35 | |
| *** jawad_axd has quit IRC | 02:40 | |
| *** kaisers has quit IRC | 02:40 | |
| *** kaisers has joined #openstack-meeting | 02:55 | |
| *** links has joined #openstack-meeting | 03:02 | |
| *** tetsuro has quit IRC | 03:05 | |
| *** tetsuro_ has joined #openstack-meeting | 03:05 | |
| *** gyee has quit IRC | 03:05 | |
| *** apetrich has quit IRC | 03:12 | |
| *** jamesmcarthur has joined #openstack-meeting | 03:16 | |
| *** jamesmcarthur has quit IRC | 03:27 | |
| *** jamesmcarthur has joined #openstack-meeting | 03:27 | |
| *** diablo_rojo has quit IRC | 03:33 | |
| *** diablo_rojo has joined #openstack-meeting | 03:34 | |
| *** psachin has joined #openstack-meeting | 03:37 | |
| *** armax has quit IRC | 04:11 | |
| *** jamesmcarthur has quit IRC | 04:14 | |
| *** masahito has quit IRC | 04:16 | |
| *** jamesmcarthur has joined #openstack-meeting | 04:33 | |
| *** jamesmcarthur has quit IRC | 04:41 | |
| *** psachin has quit IRC | 04:47 | |
| *** rh-jelabarre has quit IRC | 04:50 | |
| *** diablo_rojo has quit IRC | 05:13 | |
| *** brinzhang has joined #openstack-meeting | 05:26 | |
| *** rcernin has quit IRC | 05:39 | |
| *** tetsuro_ has quit IRC | 06:04 | |
| *** tetsuro has joined #openstack-meeting | 06:19 | |
| *** stephen-ma has joined #openstack-meeting | 06:22 | |
| *** tetsuro has quit IRC | 06:25 | |
| *** stephen-ma has quit IRC | 06:40 | |
| *** masahito has joined #openstack-meeting | 06:47 | |
| *** slaweq has joined #openstack-meeting | 06:58 | |
| *** persia has quit IRC | 07:06 | |
| *** persia has joined #openstack-meeting | 07:08 | |
| *** tetsuro has joined #openstack-meeting | 07:12 | |
| *** tetsuro has quit IRC | 07:17 | |
| *** masahito has quit IRC | 07:23 | |
| *** masahito has joined #openstack-meeting | 07:24 | |
| *** psachin has joined #openstack-meeting | 07:28 | |
| *** jawad_axd has joined #openstack-meeting | 07:28 | |
| *** tetsuro has joined #openstack-meeting | 07:29 | |
| *** masahito has quit IRC | 07:32 | |
| *** masahito has joined #openstack-meeting | 07:33 | |
| *** tetsuro_ has joined #openstack-meeting | 07:41 | |
| *** ralonsoh has joined #openstack-meeting | 07:42 | |
| *** tetsuro has quit IRC | 07:44 | |
| *** tetsuro has joined #openstack-meeting | 07:46 | |
| *** tetsuro__ has joined #openstack-meeting | 07:47 | |
| *** tetsuro_ has quit IRC | 07:49 | |
| *** tetsuro has quit IRC | 07:50 | |
| *** masahito has quit IRC | 08:00 | |
| *** masahito has joined #openstack-meeting | 08:00 | |
| *** masahito has quit IRC | 08:05 | |
| *** masahito has joined #openstack-meeting | 08:06 | |
| *** maciejjozefczyk has joined #openstack-meeting | 08:07 | |
| *** masahito has quit IRC | 08:09 | |
| *** masahito has joined #openstack-meeting | 08:10 | |
| *** rpittau|afk is now known as rpittau | 08:16 | |
| *** vishalmanchanda has quit IRC | 08:21 | |
| *** tetsuro has joined #openstack-meeting | 08:23 | |
| *** tesseract has joined #openstack-meeting | 08:25 | |
| *** tetsuro__ has quit IRC | 08:25 | |
| *** jawad_axd has quit IRC | 08:27 | |
| *** tetsuro has quit IRC | 08:32 | |
| *** tetsuro_ has joined #openstack-meeting | 08:32 | |
| *** jawad_axd has joined #openstack-meeting | 08:38 | |
| *** alecuyer has left #openstack-meeting | 08:43 | |
| *** masahito has quit IRC | 08:48 | |
| *** masahito has joined #openstack-meeting | 08:50 | |
| *** psachin has quit IRC | 08:53 | |
| *** tetsuro_ has quit IRC | 08:53 | |
| *** masahito has quit IRC | 09:00 | |
| *** masahito has joined #openstack-meeting | 09:05 | |
| *** masahito has quit IRC | 09:06 | |
| *** masahito has joined #openstack-meeting | 09:07 | |
| *** lucasagomes has joined #openstack-meeting | 09:11 | |
| *** e0ne has joined #openstack-meeting | 09:14 | |
| *** Liang__ has joined #openstack-meeting | 09:32 | |
| *** masahito has quit IRC | 09:38 | |
| *** masahito has joined #openstack-meeting | 09:39 | |
| *** masahito has quit IRC | 09:39 | |
| *** masahito has joined #openstack-meeting | 09:47 | |
| *** masahito has quit IRC | 09:53 | |
| *** hjensas has quit IRC | 09:54 | |
| *** lpetrut has joined #openstack-meeting | 10:14 | |
| *** rpittau is now known as rpittau|bbl | 11:06 | |
| *** M0nk3Ee has joined #openstack-meeting | 11:10 | |
| *** Lucas_Gray has joined #openstack-meeting | 11:14 | |
| *** ociuhandu has joined #openstack-meeting | 11:28 | |
| *** Wryhder has joined #openstack-meeting | 11:31 | |
| *** Lucas_Gray has quit IRC | 11:32 | |
| *** Wryhder is now known as Lucas_Gray | 11:32 | |
| *** Lucas_Gray has quit IRC | 11:37 | |
| *** Lucas_Gray has joined #openstack-meeting | 11:44 | |
| *** raildo has joined #openstack-meeting | 12:09 | |
| *** rfolco has joined #openstack-meeting | 12:10 | |
| *** ociuhandu has quit IRC | 12:10 | |
| *** links has quit IRC | 12:15 | |
| *** ysandeep has joined #openstack-meeting | 12:15 | |
| *** ociuhandu has joined #openstack-meeting | 12:19 | |
| *** rh-jelabarre has joined #openstack-meeting | 12:22 | |
| *** Lucas_Gray has quit IRC | 12:27 | |
| *** Lucas_Gray has joined #openstack-meeting | 12:31 | |
| *** TomStappaerts has joined #openstack-meeting | 13:04 | |
| *** rpittau|bbl is now known as rpittau | 13:08 | |
| *** jamesmcarthur has joined #openstack-meeting | 13:22 | |
| *** ociuhandu has quit IRC | 13:28 | |
| *** enriquetaso has joined #openstack-meeting | 13:29 | |
| *** felixhuettner has joined #openstack-meeting | 13:33 | |
| *** jamesmcarthur has quit IRC | 13:35 | |
| *** jamesmcarthur has joined #openstack-meeting | 13:37 | |
| *** stephen-ma has joined #openstack-meeting | 13:37 | |
| *** jamesmcarthur has quit IRC | 13:42 | |
| *** eharney has joined #openstack-meeting | 13:44 | |
| *** Lucas_Gray has quit IRC | 13:46 | |
| *** Lucas_Gray has joined #openstack-meeting | 13:47 | |
| *** davidsha has joined #openstack-meeting | 13:51 | |
| *** artom has joined #openstack-meeting | 13:55 | |
| slaweq | #startmeeting neutron_drivers | 14:00 |
|---|---|---|
| openstack | Meeting started Fri Jan 31 14:00:02 2020 UTC and is due to finish in 60 minutes. The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:00 |
| slaweq | hi | 14:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:00 |
| *** openstack changes topic to " (Meeting topic: neutron_drivers)" | 14:00 | |
| openstack | The meeting name has been set to 'neutron_drivers' | 14:00 |
| njohnston | o/ | 14:00 |
| felixhuettner | o/ | 14:00 |
| *** jawad_axd has quit IRC | 14:00 | |
| ralonsoh | hi | 14:01 |
| slaweq | I don't think we will have quorum today as yamamoto and haleyb sent me info that they can't be today and amotoki is sick | 14:01 |
| *** jawad_axd has joined #openstack-meeting | 14:01 | |
| ralonsoh | hmmm you are right... | 14:01 |
| ralonsoh | so what do you suggest? | 14:01 |
| slaweq | lets wait few more minutes for mlavalle and maybe we can discuss about this min bw issue https://bugs.launchpad.net/neutron/+bug/1861442 at least | 14:02 |
| openstack | Launchpad bug 1861442 in neutron "QOS minimum bandwidth rejection of non-physnet network and updates should be driver specific" [Undecided,New] | 14:02 |
| slaweq | or we will cancel the meeting | 14:02 |
| slaweq | ahh, and welcome our new member of drivers team - njohnston :) | 14:02 |
| ralonsoh | perfect, I would like to talk about this issue in particular | 14:02 |
| ralonsoh | welcome!! | 14:03 |
| *** ociuhandu has joined #openstack-meeting | 14:03 | |
| njohnston | \o/ thanks all! | 14:03 |
| *** mlavalle has joined #openstack-meeting | 14:04 | |
| mlavalle | o/ | 14:04 |
| slaweq | hi mlavalle | 14:04 |
| *** jawad_ax_ has joined #openstack-meeting | 14:04 | |
| njohnston | so in quorum 50% or >50%? | 14:04 |
| slaweq | I would say that >50% but I don't think it was written somewhere | 14:05 |
| slaweq | mlavalle: do You think we can say that we have quorum if we have 50% of team? | 14:05 |
| *** jawad_a__ has joined #openstack-meeting | 14:05 | |
| mlavalle | no, I think quorum should be simple majority | 14:06 |
| njohnston | we always had an odd number of members so it did not come up, at least in the past 4 years or so if I remember correctly | 14:06 |
| *** jawad_axd has quit IRC | 14:06 | |
| slaweq | I don't think we have any rush today | 14:06 |
| *** jamesmcarthur has joined #openstack-meeting | 14:06 | |
| mlavalle | it is better for the sake of the validity of any decision we make that we have a moajority of the team | 14:06 |
| slaweq | so lets skip discussing RFEs and go straigh to open discussion | 14:06 |
| *** obondarev has joined #openstack-meeting | 14:07 | |
| slaweq | #topic open discussion | 14:07 |
| *** openstack changes topic to "open discussion (Meeting topic: neutron_drivers)" | 14:07 | |
| slaweq | we have one bug to discuss: https://bugs.launchpad.net/neutron/+bug/1861442 | 14:07 |
| openstack | Launchpad bug 1861442 in neutron "QOS minimum bandwidth rejection of non-physnet network and updates should be driver specific" [Undecided,New] | 14:07 |
| ralonsoh | TomStappaerts, hi ^^ | 14:07 |
| TomStappaerts | Hi | 14:07 |
| slaweq | TomStappaerts: ralonsoh do You want to explain what You want to discuss exactly? | 14:08 |
| njohnston | ralonsoh your question is on point | 14:08 |
| ralonsoh | I'm just a viewer here | 14:08 |
| ralonsoh | the bug is not mine | 14:08 |
| *** ociuhandu has quit IRC | 14:08 | |
| *** jawad_ax_ has quit IRC | 14:08 | |
| slaweq | TomStappaerts: it's You who reported it, right? | 14:09 |
| TomStappaerts | Yes | 14:09 |
| TomStappaerts | My problem is that we block stuff in the qos_plugin without calling upon the qos_drivers to ask them whether they support it. With the message "this is unsupported regardless of driver" | 14:09 |
| TomStappaerts | But in our SDN we can support minimum bandwidth without needing a physnet | 14:10 |
| *** jawad_a__ has quit IRC | 14:10 | |
| ralonsoh | sorry for interrupting, just a bit of context | 14:10 |
| ralonsoh | #link https://bugs.launchpad.net/neutron/+bug/1819029 | 14:11 |
| openstack | Launchpad bug 1819029 in neutron "QoS policies with minimum-bandwidth rule should be rejected on non-physnet ports/networks" [Undecided,Fix released] - Assigned to Slawek Kaplonski (slaweq) | 14:11 |
| *** Lucas_Gray has quit IRC | 14:11 | |
| ralonsoh | that was added to, in case of enforcing a minBW, to have a reference of the available BW in the physnet | 14:11 |
| TomStappaerts | Yes thx ralonsoh, so at the time it was blocked on purpose. My argument is that the plugin should ask the available drivers whether they support it instead of plainly blocking the possibility. | 14:11 |
| slaweq | so how You are than reporting to the placement what resources You have available? | 14:12 |
| njohnston | Right, how does the driver indicate the bandwidth capacity of a non-physnet network? | 14:13 |
| TomStappaerts | So it seems like I overlooked that part of the reference implementation. Our implementation is based on best effort | 14:13 |
| *** jamesmcarthur has quit IRC | 14:13 | |
| TomStappaerts | Btw I totally understand if you say it makes no sense in normal world :) | 14:14 |
| *** Lucas_Gray has joined #openstack-meeting | 14:14 | |
| ralonsoh | but, eventually this minBW must be enforced in a physical device | 14:14 |
| ralonsoh | (that's why we included this limitation, a part from the placement info population) | 14:14 |
| mlavalle | yeah, this is indpendent of placement | 14:15 |
| njohnston | So you are trying to get the effect of kin BW by prioritization as opposed to quotaing? | 14:15 |
| njohnston | *min | 14:15 |
| slaweq | ok, I see | 14:16 |
| slaweq | I'm looking into the code now and IMO we made simply mistake by merging this validation in https://github.com/openstack/neutron/blob/master/neutron/services/qos/qos_plugin.py#L222 | 14:16 |
| slaweq | it should be in driver.is_rule_supported() method | 14:16 |
| slaweq | and than it would be "per driver" to decide that | 14:17 |
| slaweq | it is called in https://github.com/openstack/neutron/blob/bd26435b5bf42ade45af1059f57b423346663cf5/neutron/services/qos/drivers/manager.py#L137 | 14:17 |
| slaweq | and it's base definition is in neutron-lib https://opendev.org/openstack/neutron-lib/src/branch/master/neutron_lib/services/qos/base.py#L71 | 14:17 |
| slaweq | IMO it is simple bug which should be fixed by moving this validation to the driver | 14:18 |
| TomStappaerts | That is what I was thinking yes | 14:18 |
| slaweq | but I wonder what ralonsoh and maybe rubasov have to say about it | 14:18 |
| ralonsoh | I'm reviewing the code right now | 14:19 |
| TomStappaerts | But in is_rule_supported you do not have access to the network object as of now | 14:19 |
| ralonsoh | ?? | 14:20 |
| *** lajoskatona has joined #openstack-meeting | 14:20 | |
| *** Lucas_Gray has quit IRC | 14:20 | |
| ralonsoh | I don't understand the last comment | 14:20 |
| TomStappaerts | https://opendev.org/openstack/neutron-lib/src/branch/master/neutron_lib/services/qos/base.py#L71 Would add verification that it does not support a rule when the network does not have physnet as first segment, right? | 14:20 |
| ralonsoh | ahhh you say you don't have the network object in this method | 14:21 |
| slaweq | TomStappaerts: yes, now it only accepts rule as argument but IMO we can add additional argument(s) to it | 14:21 |
| ralonsoh | exactly | 14:21 |
| rubasov | hello | 14:21 |
| slaweq | hi rubasov | 14:21 |
| ralonsoh | rubasov, https://bugs.launchpad.net/neutron/+bug/1861442 | 14:21 |
| openstack | Launchpad bug 1861442 in neutron "QOS minimum bandwidth rejection of non-physnet network and updates should be driver specific" [Undecided,New] | 14:21 |
| rubasov | looking at it | 14:21 |
| ralonsoh | slaweq, you made a good point there | 14:21 |
| ralonsoh | the rule should be checked in https://github.com/openstack/neutron/blob/bd26435b5bf42ade45af1059f57b423346663cf5/neutron/services/qos/drivers/manager.py#L137 | 14:22 |
| ralonsoh | there, each driver decide what to do | 14:22 |
| ralonsoh | my concern is what is happening with the information reported to the placement | 14:22 |
| ralonsoh | those ports won't have any BW assigned | 14:22 |
| ralonsoh | just in the other way | 14:22 |
| ralonsoh | when those ports are going to be assigned to a host | 14:23 |
| ralonsoh | this host won't have (reported) any BW | 14:23 |
| *** Lucas_Gray has joined #openstack-meeting | 14:23 | |
| slaweq | ralonsoh: yes, but IMO it's driver's problem, not neutron core - if driver says that it supports this kind of rule, it should be fine for neutron | 14:23 |
| ralonsoh | yes | 14:24 |
| rubasov | I may not be understanding the whole problem, but the comment in the bug report wanted to say that we were not considering support for min-bw qos for tunneled networks | 14:24 |
| njohnston | that is a very good point. but what awareness does placement have of non-physnet networks? I don’t totally understand the use case. | 14:24 |
| rubasov | just to reduce the scope of what we were doing | 14:24 |
| rubasov | so if TomStappaerts is doing something like that now then it may not be relevant anymore | 14:25 |
| slaweq | rubasov: basically my question to You was: why we are doing validation of phys_net in https://github.com/openstack/neutron/blob/master/neutron/services/qos/qos_plugin.py#L229 rather than in driver.is_rule_supported() method | 14:25 |
| *** Lucas_Gray has quit IRC | 14:26 | |
| rubasov | it seems to me I made a bug | 14:27 |
| *** ociuhandu has joined #openstack-meeting | 14:27 | |
| TomStappaerts | rubasov: You're saying I should not be trying to re-use reference qos_plugin for min bw on tunneled networks? (Sorry I don't quite understand :) ) | 14:28 |
| slaweq | rubasov: :) that's what I was thinkning now when looking at it | 14:28 |
| rubasov | TomStappaerts: I'm trying to say less | 14:29 |
| rubasov | TomStappaerts: that we did not consider the consequences of doing that, therefore you may get surprises | 14:29 |
| *** jmasud has quit IRC | 14:30 | |
| ralonsoh | rubasov, what happens if you have ports with minBW and the placement does not have any host with BW reported? | 14:30 |
| ralonsoh | you can't spawn a VM on those hosts | 14:31 |
| *** jmasud has joined #openstack-meeting | 14:31 | |
| rubasov | ralonsoh: vm scheduling should fail, shouldn't it? | 14:31 |
| ralonsoh | that's the point | 14:31 |
| ralonsoh | we can assign ports with minBW qos | 14:31 |
| ralonsoh | if we remove this limitation (valid or not) | 14:31 |
| ralonsoh | and then, if the placement does not have BW reported, won't schedule to those hosts | 14:32 |
| TomStappaerts | Ah ok, yes that is something I haven't considered yet, as I am blocked at this point now. I am more than fine with you guys saying it doesn't make much sense | 14:32 |
| slaweq | that's very good point, thx rubasov and ralonsoh | 14:33 |
| *** TomStappaerts has quit IRC | 14:33 | |
| ralonsoh | ?? | 14:33 |
| ralonsoh | bye... | 14:33 |
| *** TomStappaerts has joined #openstack-meeting | 14:33 | |
| rubasov | so basically I think slaweq summed it up correctly in this: it is simple bug which should be fixed by moving this validation to the driver | 14:34 |
| ralonsoh | how do you disable this placement check? | 14:34 |
| ralonsoh | when scheduling, rubasov | 14:35 |
| *** brinzhang has quit IRC | 14:37 | |
| rubasov | ralonsoh: by not adding a min-bw rule on the port's policy (but I may be misunderstanding your question) | 14:37 |
| ralonsoh | sorry, but at this point this is the perfect example of interlocking | 14:38 |
| njohnston | Can we say that placement only tracks physnet networks, so if the network is not a physnet we treat it as if there is no min bw policy in placement interactions? | 14:39 |
| slaweq | tbh if You will spawn instance without qos_policy and than attach qos to the port than it should be fine, no? | 14:39 |
| ralonsoh | slaweq, yes | 14:40 |
| slaweq | but that's "hack" to workaround this limitation | 14:40 |
| ralonsoh | njohnston, I don't know, I need to check that | 14:40 |
| ralonsoh | placement is only for scheduling | 14:40 |
| ralonsoh | then the port is responsibility of Neutron only | 14:40 |
| TomStappaerts | https://github.com/openstack/neutron/blob/master/neutron/services/qos/qos_plugin.py#L129 Seems to handle that case | 14:41 |
| ralonsoh | TomStappaerts, this is a temporary code until the placement and Neutron fully supports this feature | 14:42 |
| rubasov | currently there's no way to represent bw of non-physnet networks in placement, we need to come up with a way to do that | 14:42 |
| ralonsoh | exactly: some kind of manual report (or configured report) | 14:43 |
| TomStappaerts | Is that something that fits into the overall scheme of things, or would it be better for me to werite my own qos_plugin | 14:44 |
| slaweq | TomStappaerts: IMO this validation part should be fixed by moving it to the driver's code | 14:44 |
| slaweq | that's a bug as rubasov confirmed also | 14:44 |
| rubasov | this is implicit in the use of CUSTOM_PHYSNET_ traits | 14:45 |
| slaweq | so that's we should IMO treat this LP | 14:45 |
| slaweq | and about reporting non-physnet networks bw to placement - that's IMO different issue and this should be new RFE | 14:45 |
| rubasov | slaweq, TomStappaerts: yes, I'm okay with moving the check to the driver | 14:45 |
| slaweq | what do other people think about it? Does it makes sense? | 14:45 |
| rubasov | slaweq: exactly | 14:46 |
| ralonsoh | at least, for now, we should add documentation for this: if not physnet --> no scheduling | 14:46 |
| ralonsoh | it makes sense | 14:46 |
| ralonsoh | +1 | 14:46 |
| njohnston | makes sense | 14:46 |
| njohnston | +1 | 14:46 |
| ralonsoh | thanks rubasov and TomStappaerts | 14:46 |
| TomStappaerts | Thx guys | 14:47 |
| rubasov | thanks | 14:47 |
| slaweq | mlavalle: any thoughts? | 14:47 |
| mlavalle | +1 | 14:47 |
| njohnston | so we can revisit this when we have quorum but hopefully this discussion will help set the stage for then | 14:47 |
| slaweq | ok, I will sum up this discussion in the comment to LP and will also ask offline other drivers to check if they have any concerns to treat it as regular bug | 14:49 |
| slaweq | IMO in this case this should be enough | 14:49 |
| njohnston | sounds good! | 14:49 |
| slaweq | ok, we have few more minutes and one more "not really rfe" thing do discuss | 14:50 |
| slaweq | https://review.opendev.org/#/c/702806/ | 14:50 |
| slaweq | and related bug https://bugs.launchpad.net/neutron/+bug/1855260 | 14:50 |
| openstack | Launchpad bug 1855260 in neutron "SmartNIC OVS representor port is not removed from OVS on instance tear down" [Undecided,In progress] - Assigned to waleed mousa (waleedm) | 14:50 |
| slaweq | ralonsoh: can You elaborate on this a bit as You were involved in that already? | 14:50 |
| ralonsoh | if we assume that the OVS agent is going to plug/unplug the interfaces | 14:51 |
| ralonsoh | then OK | 14:51 |
| ralonsoh | but as Sean pointed out, the L1 operations should be done by Nova/os-vif | 14:51 |
| ralonsoh | but | 14:51 |
| ralonsoh | because those are smartnic ports | 14:51 |
| ralonsoh | and this is happening when the OVS agent is offloaded | 14:51 |
| ralonsoh | then I'm not 100% sure about Sean's affirmation | 14:51 |
| ralonsoh | I need to check with Sean this last point | 14:52 |
| ralonsoh | conclusion: I know nothing | 14:52 |
| slaweq | ok :) | 14:52 |
| slaweq | but current solution with ovs agent responsible to plug/unplug interfaces is already in neutron (iirc in train at least), right? | 14:53 |
| ralonsoh | yes for smartnic ports | 14:53 |
| ralonsoh | so yes, the patch is needed once we assumed this role | 14:53 |
| slaweq | yes, so IMO we should go with fix for the reported bug now as quick and backportable solution | 14:53 |
| ralonsoh | sure | 14:54 |
| slaweq | and later maybe propose new rfe to change the logic | 14:54 |
| ralonsoh | +1 | 14:54 |
| njohnston | +1 | 14:54 |
| mlavalle | =1 | 14:54 |
| slaweq | I know that sean-k-mooney and You had some opinions about it so I would like to ask what do You think about such solution | 14:54 |
| mlavalle | +1 | 14:54 |
| ralonsoh | I'll review the patch assuming the current state: ovs agent handle those ports | 14:56 |
| slaweq | ok, thx a lot | 14:57 |
| sean-k-mooney | just reading scrollback now | 14:57 |
| slaweq | so I think we had pretty productive meeting even without discussion any rfe :) | 14:57 |
| slaweq | hi sean-k-mooney :) | 14:57 |
| sean-k-mooney | o/ | 14:57 |
| *** jamesmcarthur has joined #openstack-meeting | 14:58 | |
| sean-k-mooney | slaweq: the question is should the neutron l2 agent be managin l1 | 14:58 |
| ralonsoh | yes | 14:58 |
| sean-k-mooney | IMO i dont think that desing sacles well and i dont think it should. that said im not going to block extending os-vif to support this | 14:59 |
| slaweq | sean-k-mooney: yes, I understand Your point and IMO we should think about better solution maybe (I'm not smartnics expert so I may be wrong) | 14:59 |
| sean-k-mooney | so the choice is really up to the neuton team to decide | 14:59 |
| slaweq | but now I would like to go with quicker solution to fix existing bug and backport fix to stable release | 14:59 |
| sean-k-mooney | well my many concner is upgrades | 14:59 |
| slaweq | and later we can think about redesign | 14:59 |
| slaweq | ok, we are out of time now, thx for the meeting guys | 15:00 |
| slaweq | and have a great weekend | 15:00 |
| slaweq | o/ | 15:00 |
| slaweq | #endmeeting | 15:00 |
| ralonsoh | bye! | 15:00 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/" | 15:00 | |
| sean-k-mooney | o/ | 15:00 |
| openstack | Meeting ended Fri Jan 31 15:00:32 2020 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:00 |
| mlavalle | o/ | 15:00 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/neutron_drivers/2020/neutron_drivers.2020-01-31-14.00.html | 15:00 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/neutron_drivers/2020/neutron_drivers.2020-01-31-14.00.txt | 15:00 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/neutron_drivers/2020/neutron_drivers.2020-01-31-14.00.log.html | 15:00 |
| slaweq | sean-k-mooney: we can continue on neutron channel if You want to discuss about it | 15:00 |
| njohnston | o/ | 15:00 |
| sean-k-mooney | sure | 15:01 |
| *** felixhuettner has quit IRC | 15:01 | |
| *** stephen-ma has quit IRC | 15:05 | |
| *** obondarev has quit IRC | 15:05 | |
| *** jamesmcarthur has quit IRC | 15:11 | |
| *** jamesmcarthur has joined #openstack-meeting | 15:12 | |
| *** armax has joined #openstack-meeting | 15:13 | |
| *** zbr has quit IRC | 15:15 | |
| *** ociuhandu has quit IRC | 15:17 | |
| *** jamesmcarthur has quit IRC | 15:17 | |
| *** ociuhandu has joined #openstack-meeting | 15:19 | |
| *** ociuhandu has quit IRC | 15:19 | |
| *** ociuhandu has joined #openstack-meeting | 15:20 | |
| *** jawad_axd has joined #openstack-meeting | 15:21 | |
| *** zbr has joined #openstack-meeting | 15:22 | |
| *** eharney has quit IRC | 15:27 | |
| *** jamesmcarthur has joined #openstack-meeting | 15:27 | |
| *** pcaruana has quit IRC | 15:30 | |
| *** jamesmcarthur has quit IRC | 15:32 | |
| *** jawad_axd has quit IRC | 15:34 | |
| *** lpetrut has quit IRC | 15:35 | |
| *** belmoreira has quit IRC | 15:37 | |
| *** lajoskatona has left #openstack-meeting | 15:44 | |
| *** jamesmcarthur has joined #openstack-meeting | 15:45 | |
| *** bnemec has joined #openstack-meeting | 15:49 | |
| *** pcaruana has joined #openstack-meeting | 15:49 | |
| *** TomStappaerts has quit IRC | 15:51 | |
| *** TomStappaerts has joined #openstack-meeting | 15:51 | |
| *** anastzhyr has joined #openstack-meeting | 15:53 | |
| *** jamesmcarthur_ has joined #openstack-meeting | 15:58 | |
| *** jamesmcarthur has quit IRC | 16:01 | |
| *** ayoung has joined #openstack-meeting | 16:04 | |
| *** rpittau is now known as rpittau|afk | 16:05 | |
| *** e0ne has quit IRC | 16:07 | |
| *** ysandeep is now known as ysandeep|afk | 16:18 | |
| *** Adri2000 has quit IRC | 16:28 | |
| *** jamesmcarthur_ has quit IRC | 16:47 | |
| *** rsimai_away has quit IRC | 16:52 | |
| *** mattw4 has joined #openstack-meeting | 16:56 | |
| *** jamesmcarthur has joined #openstack-meeting | 17:00 | |
| *** jamesmcarthur has quit IRC | 17:00 | |
| *** jamesmcarthur has joined #openstack-meeting | 17:00 | |
| *** jmasud has quit IRC | 17:01 | |
| *** diablo_rojo has joined #openstack-meeting | 17:04 | |
| *** ociuhandu has quit IRC | 17:06 | |
| *** TomStappaerts has quit IRC | 17:10 | |
| *** TomStappaerts has joined #openstack-meeting | 17:10 | |
| *** davidsha has quit IRC | 17:10 | |
| *** diablo_rojo has quit IRC | 17:13 | |
| *** gyee has joined #openstack-meeting | 17:13 | |
| *** lucasagomes has quit IRC | 17:21 | |
| *** tesseract has quit IRC | 17:38 | |
| *** ociuhandu has joined #openstack-meeting | 17:43 | |
| *** jamesmcarthur has quit IRC | 17:48 | |
| *** TomStappaerts has quit IRC | 17:49 | |
| *** jamesmcarthur has joined #openstack-meeting | 17:50 | |
| *** ociuhandu has quit IRC | 17:58 | |
| *** ociuhandu has joined #openstack-meeting | 17:59 | |
| *** jmasud has joined #openstack-meeting | 18:11 | |
| *** maciejjozefczyk has quit IRC | 18:14 | |
| *** eharney has joined #openstack-meeting | 18:21 | |
| *** ociuhandu has quit IRC | 18:22 | |
| *** jamesmcarthur has quit IRC | 18:38 | |
| *** jmasud has quit IRC | 18:40 | |
| *** jamesmcarthur has joined #openstack-meeting | 18:40 | |
| *** jmasud has joined #openstack-meeting | 18:46 | |
| *** jamesmcarthur has quit IRC | 18:48 | |
| *** ralonsoh has quit IRC | 18:51 | |
| *** jmasud_ has joined #openstack-meeting | 19:00 | |
| *** jmasud has quit IRC | 19:02 | |
| *** igordc has joined #openstack-meeting | 19:03 | |
| *** eharney has quit IRC | 19:09 | |
| *** eharney has joined #openstack-meeting | 19:21 | |
| *** eharney has quit IRC | 19:26 | |
| *** eharney has joined #openstack-meeting | 19:26 | |
| *** ayoung has quit IRC | 19:26 | |
| *** pcaruana has quit IRC | 19:30 | |
| *** jmasud_ has quit IRC | 19:38 | |
| *** jamesdenton has quit IRC | 19:40 | |
| *** ayoung has joined #openstack-meeting | 19:48 | |
| *** jmasud has joined #openstack-meeting | 19:57 | |
| *** efried has quit IRC | 19:58 | |
| *** jmasud has quit IRC | 20:13 | |
| *** dklyle has quit IRC | 20:15 | |
| *** Lucas_Gray has joined #openstack-meeting | 20:15 | |
| *** jmasud has joined #openstack-meeting | 20:16 | |
| *** dklyle has joined #openstack-meeting | 20:19 | |
| *** dmacpher has quit IRC | 20:34 | |
| *** dmacpher has joined #openstack-meeting | 20:34 | |
| *** rfolco has quit IRC | 20:43 | |
| *** eharney has quit IRC | 20:49 | |
| *** Liang__ has quit IRC | 20:51 | |
| *** mattw4 has quit IRC | 20:54 | |
| *** mattw4 has joined #openstack-meeting | 20:58 | |
| *** enriquetaso has quit IRC | 21:02 | |
| *** jmasud has quit IRC | 21:08 | |
| *** eharney has joined #openstack-meeting | 22:01 | |
| *** e0ne has joined #openstack-meeting | 22:04 | |
| *** e0ne_ has joined #openstack-meeting | 22:06 | |
| *** e0ne has quit IRC | 22:06 | |
| *** e0ne_ has quit IRC | 22:12 | |
| *** e0ne has joined #openstack-meeting | 22:12 | |
| *** Lucas_Gray has quit IRC | 22:15 | |
| *** e0ne has quit IRC | 22:16 | |
| *** e0ne has joined #openstack-meeting | 22:16 | |
| *** Lucas_Gray has joined #openstack-meeting | 22:16 | |
| *** mattw4 has quit IRC | 22:17 | |
| *** ociuhandu has joined #openstack-meeting | 22:20 | |
| *** raildo has quit IRC | 22:22 | |
| *** e0ne has quit IRC | 22:25 | |
| *** ociuhandu has quit IRC | 22:26 | |
| *** mattw4 has joined #openstack-meeting | 22:30 | |
| *** rh-jelabarre has quit IRC | 22:31 | |
| *** Lucas_Gray has quit IRC | 22:58 | |
| *** Lucas_Gray has joined #openstack-meeting | 23:03 | |
| *** eharney has quit IRC | 23:20 | |
| *** mlavalle has quit IRC | 23:29 | |
| *** enriquetaso has joined #openstack-meeting | 23:29 | |
| *** rfolco has joined #openstack-meeting | 23:48 | |
| *** igordc has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!