| *** aginwala has joined #openstack-neutron-ovn | 00:00 | |
| *** jckasper has quit IRC | 00:01 | |
| *** fzdarsky has quit IRC | 00:04 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 00:06 | |
| *** brad_behle has joined #openstack-neutron-ovn | 01:18 | |
| *** fzdarsky has quit IRC | 01:18 | |
| *** nghosh has quit IRC | 01:42 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:57 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:57 | |
| *** aginwala has quit IRC | 02:06 | |
| *** aginwala has joined #openstack-neutron-ovn | 02:09 | |
| *** aginwala has quit IRC | 02:09 | |
| *** aginwala has joined #openstack-neutron-ovn | 02:10 | |
| *** s3wong has quit IRC | 02:23 | |
| *** azbiswas has quit IRC | 02:36 | |
| *** azbiswas has joined #openstack-neutron-ovn | 02:36 | |
| *** aginwala has quit IRC | 02:40 | |
| *** azbiswas has quit IRC | 02:41 | |
| *** salv-orl_ has quit IRC | 03:04 | |
| *** gangil has quit IRC | 03:18 | |
| *** gizmoguy has quit IRC | 03:19 | |
| *** gizmoguy has joined #openstack-neutron-ovn | 03:20 | |
| *** yamamoto has joined #openstack-neutron-ovn | 03:41 | |
| *** azbiswas has joined #openstack-neutron-ovn | 04:03 | |
| *** brad_behle has quit IRC | 04:17 | |
| *** azbiswas has quit IRC | 04:27 | |
| *** azbiswas has joined #openstack-neutron-ovn | 04:28 | |
| *** azbiswas has quit IRC | 04:32 | |
| *** jckasper has quit IRC | 04:33 | |
| *** jckasper has joined #openstack-neutron-ovn | 04:33 | |
| *** azbiswas has joined #openstack-neutron-ovn | 04:56 | |
| *** azbiswas has quit IRC | 05:01 | |
| *** allan_h has joined #openstack-neutron-ovn | 05:04 | |
| *** allan_h has quit IRC | 05:11 | |
| *** gal_ is now known as gsagie | 05:44 | |
| *** aginwala has joined #openstack-neutron-ovn | 06:24 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 06:34 | |
| *** numans has joined #openstack-neutron-ovn | 06:46 | |
| *** fzdarsky has quit IRC | 06:52 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 06:54 | |
| *** azbiswas has joined #openstack-neutron-ovn | 07:08 | |
| *** armax has quit IRC | 07:09 | |
| *** arosen121 has joined #openstack-neutron-ovn | 07:21 | |
| *** arosen12 has quit IRC | 07:24 | |
| *** arosen12 has joined #openstack-neutron-ovn | 07:28 | |
| *** gongysh has joined #openstack-neutron-ovn | 07:29 | |
| *** arosen121 has quit IRC | 07:30 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 07:44 | |
| *** azbiswas_ has joined #openstack-neutron-ovn | 07:49 | |
| *** azbiswas has quit IRC | 07:51 | |
| *** pcaruana has joined #openstack-neutron-ovn | 08:05 | |
| *** fzdarsky has quit IRC | 08:08 | |
| *** gongysh has quit IRC | 08:09 | |
| *** numans has quit IRC | 08:22 | |
| *** numans has joined #openstack-neutron-ovn | 08:33 | |
| *** gangil has joined #openstack-neutron-ovn | 08:56 | |
| *** gangil has joined #openstack-neutron-ovn | 08:56 | |
| *** palexster has quit IRC | 09:27 | |
| *** palexster has joined #openstack-neutron-ovn | 09:27 | |
| *** aginwala has quit IRC | 09:33 | |
| *** arosen12 has quit IRC | 09:40 | |
| *** openstackgerrit has quit IRC | 10:02 | |
| *** openstackgerrit has joined #openstack-neutron-ovn | 10:03 | |
| *** salv-orl_ has joined #openstack-neutron-ovn | 10:03 | |
| *** roeyc has joined #openstack-neutron-ovn | 10:03 | |
| *** salv-orlando has quit IRC | 10:06 | |
| *** gal_ has joined #openstack-neutron-ovn | 10:06 | |
| *** gsagie has quit IRC | 10:10 | |
| *** azbiswas_ has quit IRC | 10:24 | |
| *** palexster has quit IRC | 10:45 | |
| *** palexster has joined #openstack-neutron-ovn | 10:46 | |
| *** gal_ is now known as gsagie | 10:50 | |
| *** gsagie has quit IRC | 10:55 | |
| *** roeyc has quit IRC | 11:00 | |
| *** yamamoto has quit IRC | 11:03 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 11:20 | |
| *** roeyc has joined #openstack-neutron-ovn | 11:41 | |
| *** yamamoto has joined #openstack-neutron-ovn | 11:48 | |
| *** yamamoto_ has joined #openstack-neutron-ovn | 11:49 | |
| *** yamamoto has quit IRC | 11:53 | |
| *** yamamoto_ has quit IRC | 11:57 | |
| *** fzdarsky has quit IRC | 11:57 | |
| *** salv-orl_ has quit IRC | 12:03 | |
| *** yamamoto has joined #openstack-neutron-ovn | 12:11 | |
| *** roeyc has quit IRC | 12:14 | |
| *** roeyc has joined #openstack-neutron-ovn | 12:17 | |
| *** rtheis has joined #openstack-neutron-ovn | 12:18 | |
| *** yamamoto has quit IRC | 12:22 | |
| *** yamamoto has joined #openstack-neutron-ovn | 12:23 | |
| *** azbiswas has joined #openstack-neutron-ovn | 12:25 | |
| *** elurkki_ is now known as elurkki | 12:25 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 12:26 | |
| *** yamamoto has quit IRC | 12:27 | |
| *** azbiswas has quit IRC | 12:30 | |
| *** gangil has quit IRC | 12:34 | |
| *** yamamoto has joined #openstack-neutron-ovn | 12:46 | |
| *** gsagie has joined #openstack-neutron-ovn | 13:12 | |
| *** thumpba has joined #openstack-neutron-ovn | 13:15 | |
| *** allan_h has joined #openstack-neutron-ovn | 13:23 | |
| *** allan_h has quit IRC | 13:34 | |
| *** yamamoto has quit IRC | 13:42 | |
| *** yamamoto has joined #openstack-neutron-ovn | 13:45 | |
| *** yamamoto has quit IRC | 13:47 | |
| *** yamamoto has joined #openstack-neutron-ovn | 13:47 | |
| *** yamamoto has quit IRC | 13:47 | |
| Sam-I-Am | moo. | 13:48 |
|---|---|---|
| *** yamamoto has joined #openstack-neutron-ovn | 13:49 | |
| *** yamamoto has quit IRC | 13:59 | |
| *** azbiswas has joined #openstack-neutron-ovn | 14:04 | |
| *** jckasper has quit IRC | 14:05 | |
| *** thumpba has quit IRC | 14:09 | |
| *** jckasper has joined #openstack-neutron-ovn | 14:28 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 14:37 | |
| *** gongysh has joined #openstack-neutron-ovn | 14:45 | |
| *** yamamoto has joined #openstack-neutron-ovn | 14:59 | |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 15:02 | |
| *** azbiswas has quit IRC | 15:03 | |
| *** allan_h has joined #openstack-neutron-ovn | 15:20 | |
| *** gongysh has quit IRC | 15:22 | |
| *** azbiswas has joined #openstack-neutron-ovn | 15:32 | |
| *** fzdarsky_ has quit IRC | 15:43 | |
| *** fzdarsky has quit IRC | 15:43 | |
| *** numans has quit IRC | 15:44 | |
| *** allan_h has quit IRC | 15:45 | |
| *** thumpba has joined #openstack-neutron-ovn | 15:49 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 15:50 | |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 15:50 | |
| *** azbiswas_ has joined #openstack-neutron-ovn | 15:51 | |
| *** yamamoto has quit IRC | 15:52 | |
| *** azbiswas has quit IRC | 15:54 | |
| *** fzdarsky has quit IRC | 15:56 | |
| *** fzdarsky_ has quit IRC | 15:56 | |
| *** salv-orl_ has joined #openstack-neutron-ovn | 16:03 | |
| *** salv-orlando has quit IRC | 16:06 | |
| *** Guest99081 is now known as ioni | 16:07 | |
| *** ioni has quit IRC | 16:07 | |
| *** ioni has joined #openstack-neutron-ovn | 16:07 | |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 16:07 | |
| *** fzdarsky has joined #openstack-neutron-ovn | 16:07 | |
| *** armax has joined #openstack-neutron-ovn | 16:10 | |
| openstackgerrit | Merged openstack/networking-ovn: Network availability zone deployment support https://review.openstack.org/281284 | 16:11 |
| *** allan_h has joined #openstack-neutron-ovn | 16:11 | |
| *** armax has quit IRC | 16:14 | |
| *** pcaruana has quit IRC | 16:15 | |
| *** regXboi has joined #openstack-neutron-ovn | 16:16 | |
| *** yamamoto has joined #openstack-neutron-ovn | 16:21 | |
| *** armax has joined #openstack-neutron-ovn | 16:22 | |
| *** fzdarsky has quit IRC | 16:24 | |
| *** fzdarsky_ has quit IRC | 16:24 | |
| *** Mic22 has joined #openstack-neutron-ovn | 16:30 | |
| *** brad_behle has joined #openstack-neutron-ovn | 16:35 | |
| azbiswas_ | russellb: An update regarding https://bugs.launchpad.net/networking-ovn/+bug/1545189 | 16:38 |
| openstack | Launchpad bug 1545189 in networking-ovn "Port create results OVSDB transaction timeout: O(N) ACL recreation in North DB" [High,Confirmed] - Assigned to Amitabha Biswas (azbiswas) | 16:38 |
| azbiswas_ | Our folks encountered this timeout a couple of times in a real deployment scenario - so it can happen in realistic scenarios. | 16:39 |
| azbiswas_ | I'm testing with a patch that eliminates the timeout, have gotten to a much larger number of ports than before. | 16:40 |
| *** fzdarsky has joined #openstack-neutron-ovn | 16:40 | |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 16:41 | |
| russellb | azbiswas_: sounds like good progress! | 16:41 |
| *** fzdarsky_ has quit IRC | 16:41 | |
| *** fzdarsky__ has joined #openstack-neutron-ovn | 16:41 | |
| *** fzdarsky__ has quit IRC | 16:41 | |
| azbiswas_ | yeah, but now realizing there are other loops in that area (without db transactions) that are consuming a bunch of time | 16:42 |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 16:42 | |
| azbiswas_ | so debugging that | 16:42 |
| azbiswas_ | I'm at 500 ports on a provider network | 16:42 |
| *** aginwala has joined #openstack-neutron-ovn | 16:51 | |
| russellb | azbiswas_: related, there's going to be a really nice simplification and optimization of provider networks merged soon | 16:58 |
| russellb | azbiswas_: which you could consider testing with if you'd like. | 16:58 |
| russellb | azbiswas_: you need 2 patches to test it, though. first you need https://patchwork.ozlabs.org/patch/582095/ for OVN. and then https://review.openstack.org/#/c/266186/ for networking-ovn, though that one needs a rebase | 16:59 |
| russellb | for 500 ports on a provider network, it's going to be a really big difference | 16:59 |
| azbiswas_ | sounds good | 17:00 |
| azbiswas_ | I'll try out the combined patches | 17:00 |
| *** s3wong has joined #openstack-neutron-ovn | 17:00 | |
| russellb | for 500 ports, old/existing code results in 7000 logical flows in OVN | 17:01 |
| russellb | new method with patches, 1555 logical flows | 17:01 |
| azbiswas_ | sounds about right | 17:01 |
| azbiswas_ | oh!, that's huge | 17:01 |
| azbiswas_ | that should reduce some load on that ovn-controller as well. | 17:02 |
| *** s3wong has quit IRC | 17:06 | |
| *** aginwala has quit IRC | 17:07 | |
| azbiswas_ | russellb: It's the _acl_remote_match_ip that's causing the O(N^2) behavior in the _update_acls_for_security_group routine. I'm not sure if we can get rid of that. With N=500 in a provider network, that's big chunk of processing time in creating 1 new port. | 17:35 |
| azbiswas_ | I'm going to file a separate bug for that O(N^2) behavior and deal with it at a later point since it's a pre-existing condition in the code. | 17:46 |
| russellb | yes, that's because of remote_group_id | 17:47 |
| russellb | it's painful .... | 17:47 |
| russellb | there's probably a way to improve it somewhat | 17:47 |
| azbiswas_ | I'll look into possibly optimizing it after I submit the fix for ovsdb timeout now that i've a better idea of the ACL implementation. | 17:57 |
| regXboi | russellb: speaking of things - now that I've got the incremental flow processing working, I'm likely going to abandon it as impractical | 18:03 |
| regXboi | :( | 18:03 |
| *** numans has joined #openstack-neutron-ovn | 18:04 | |
| russellb | azbiswas_: nice work :) | 18:04 |
| russellb | regXboi: that happens sometimes, but the silver lining is usually a better understanding of the code, which can lead to new ideas for improvement | 18:04 |
| regXboi | let me lay out the reasoning and the problem that leads me to thinking about abandonment | 18:05 |
| *** roeyc has quit IRC | 18:06 | |
| regXboi | incremental processing should not skip a logical flow record that hits a continue statement in lflow_run | 18:06 |
| regXboi | so right now, the incremental processing resets its starting point to that record | 18:06 |
| regXboi | *BUT* the amount of time for the prereq for a record to be processed can be extreme - I ran a test case this morning where it was over 80 seconds | 18:07 |
| regXboi | and that's just not worth it :( | 18:07 |
| russellb | i think the general idea still has merit | 18:08 |
| russellb | i don't have the code in my head right now to really comment | 18:08 |
| russellb | i started looking at doing something similar in ovn-northd and it wasn't as easy in practice as it was in my head :-) | 18:09 |
| regXboi | oh, the idea has merit, but it may run afoul on the rocks of cold reality | 18:09 |
| regXboi | I understand - the whole persistent flow storage patch is ... well ... ugly | 18:09 |
| regXboi | FYI, what's the procedure for abandoning a patch - just send an email saying "abandoned" and why? | 18:10 |
| russellb | yeah, roughly | 18:10 |
| regXboi | ok | 18:10 |
| russellb | nothing formal | 18:11 |
| russellb | if you make an account on https://patchwork.ozlabs.org/project/openvswitch/list/ you can mark it as abandoned there too | 18:11 |
| russellb | or a project admin will do it later | 18:11 |
| * regXboi makes a note | 18:11 | |
| regXboi | oh yeah - I did find that the ovn e2e test cases create race conditions in terms of what the different hypervisors see | 18:11 |
| russellb | https://patchwork.ozlabs.org/project/openvswitch/list/?submitter=68244 | 18:11 |
| *** azbiswas_ is now known as azbiswas | 18:12 | |
| russellb | OVN meeting in #openvswitch in 3 minutes | 18:12 |
| *** aginwala has joined #openstack-neutron-ovn | 18:13 | |
| *** zhouhan has joined #openstack-neutron-ovn | 18:13 | |
| *** gangil has joined #openstack-neutron-ovn | 18:20 | |
| *** gangil has joined #openstack-neutron-ovn | 18:20 | |
| *** aginwala has quit IRC | 18:23 | |
| *** aginwala has joined #openstack-neutron-ovn | 18:27 | |
| *** gsagie has quit IRC | 18:31 | |
| *** chandrav has joined #openstack-neutron-ovn | 18:33 | |
| *** s3wong has joined #openstack-neutron-ovn | 18:49 | |
| *** fzdarsky has quit IRC | 18:50 | |
| *** fzdarsky_ has quit IRC | 18:51 | |
| Sam-I-Am | russellb: after the meeting, can i bug you about ovn architecture bits? | 18:54 |
| russellb | it's going to cost you | 18:57 |
| russellb | (yes you can) | 18:57 |
| russellb | feel free to bring it up in #openvswitch too | 18:57 |
| russellb | may be others still around and interested, depending on topic | 18:57 |
| Sam-I-Am | russellb: i'm trying to improve the architectural diagram | 19:00 |
| Sam-I-Am | what connects to what where | 19:00 |
| Sam-I-Am | looking at a recent email on dev, the nb and sb processes spawn off unique ovsdb-server instances? | 19:01 |
| *** fzdarsky has joined #openstack-neutron-ovn | 19:01 | |
| *** fzdarsky_ has joined #openstack-neutron-ovn | 19:02 | |
| russellb | Sam-I-Am: not yet, but technically it could be done that way | 19:04 |
| russellb | Sam-I-Am: and soon, our ovn-ctl helper may set things up that way by default | 19:04 |
| Sam-I-Am | where would conf.db end up in that case? | 19:04 |
| Sam-I-Am | a third instance? | 19:04 |
| russellb | yes, if that host was also running ovs-vswitchd | 19:05 |
| Sam-I-Am | i see a conf.db on the database node here | 19:05 |
| Sam-I-Am | its not running vswitchd | 19:05 |
| *** azbiswas has quit IRC | 19:05 | |
| *** azbiswas has joined #openstack-neutron-ovn | 19:06 | |
| russellb | if it's not running vswitchd, then conf.db isn't necessary, but it's not harmful either | 19:06 |
| Sam-I-Am | hmmm | 19:06 |
| zhouhan | russellb: now the option --remote=db:Open_vSwitch,Open_vSwitch,manager_options requires conf.db | 19:07 |
| russellb | zhouhan: that's right | 19:07 |
| zhouhan | but that would be just a small change to remove it | 19:07 |
| russellb | oh, hm | 19:07 |
| Sam-I-Am | russellb: on the compute node with vswitch... it talks to ovn-controller | 19:07 |
| russellb | yeah, but it's convenient to be able to just update the db for manager config | 19:08 |
| russellb | zhouhan: i wish that, and the ssl config, were in a separate db schema | 19:08 |
| russellb | Sam-I-Am: correct | 19:08 |
| zhouhan | yes | 19:08 |
| Sam-I-Am | what role does ovsdb play on the compute node if it just has conf.db ? | 19:08 |
| Sam-I-Am | seems like ovs gets most of its bits from ovn-controller | 19:09 |
| russellb | conf.db is the database required for ovs-vswitchd | 19:09 |
| Sam-I-Am | well, ovn-controller pushes them down | 19:09 |
| russellb | ovn-controller is a client of that db | 19:09 |
| russellb | but so is nova | 19:09 |
| russellb | nova creates ports in the db for VMs | 19:09 |
| *** azbiswas has quit IRC | 19:10 | |
| russellb | there's 2 pieces to configuring OVS, the DB, conf.d (bridges, ports, related config) and OpenFlow for configuring forwarding rules | 19:10 |
| russellb | sorry if i'm saying stuff you know | 19:11 |
| Sam-I-Am | i always find i dont know much :/ | 19:11 |
| Sam-I-Am | so ovs-vsctl reads out of conf.db? | 19:12 |
| russellb | yep! | 19:12 |
| russellb | ovs-vsctl is basically the client utility for conf.db (or the Open_vSwitch OVSDB schema) | 19:13 |
| russellb | similarly, ovn-nbctl is the client utility for the OVN_Northbound OVSDB schema | 19:13 |
| russellb | and ovn-sbctl for OVN_Southbound OVSDB schema | 19:13 |
| russellb | they're just handy wrappers around interacting with a particular db schema | 19:13 |
| russellb | a specific instance of ovsdb-server can host one or more databases based on one of those schemas | 19:15 |
| Sam-I-Am | ovn-controller talks to conf.db via the ovsdb-server and back up to ovsdb-server on the db node? | 19:15 |
| *** fzdarsky_ has quit IRC | 19:15 | |
| Sam-I-Am | more specifically, the sb database on the db node | 19:15 |
| russellb | ovn-controller talks to the local ovsdb-server for conf.db (Open_vSwitch db) as that database is node-local config | 19:15 |
| russellb | ovn-controller also talks to ovsdb-server on whatever host is running the central OVN_Southbound database | 19:16 |
| russellb | yes | 19:16 |
| *** fzdarsky has quit IRC | 19:16 | |
| russellb | ovn-controller is basically a client of 3 interfaces, those 2, plus it uses the OpenFlow programming interface of ovs-vswitchd | 19:16 |
| Sam-I-Am | mmmmm | 19:18 |
| Sam-I-Am | russellb: i'll put a working copy of this diagram up somewhere | 19:19 |
| *** sramamurthy has joined #openstack-neutron-ovn | 19:19 | |
| russellb | ok | 19:21 |
| russellb | ascii right? | 19:22 |
| russellb | :-p | 19:22 |
| Sam-I-Am | russellb: lolololololol | 19:22 |
| russellb | i've switched to 3-d ascii diagrams | 19:24 |
| *** aginwala has quit IRC | 19:25 | |
| Sam-I-Am | russellb: http://apps1-dfw.osdt.ionosphere80.com/misc/scenario-ovn-architecture1.png | 19:35 |
| Sam-I-Am | thats sort of where i'm going | 19:35 |
| russellb | looking good | 19:36 |
| russellb | minor thing ... red line between networking-ovn and ovn-northd should be between networking-ovn and the OVN_Northbound db | 19:36 |
| russellb | and ovn-northd should have lines to both OVN_Northbound and OVN_Southbound | 19:36 |
| russellb | that's going to be great | 19:37 |
| *** aginwala has joined #openstack-neutron-ovn | 19:38 | |
| Sam-I-Am | russellb: done! | 19:41 |
| Sam-I-Am | kind of wondering if i need to break out ovs-vswitchd into more bits | 19:42 |
| Sam-I-Am | like... its openflow controller and db stuff is really control plane... but the routing/switching/acl is data plane | 19:42 |
| russellb | yeah, depends how deep down the rabbit hole you want to go | 19:46 |
| russellb | vswitch, vrouter, and access control seems good enough to me | 19:46 |
| russellb | explaining the breakdown between control plane and data plane is probably too much detail for this diagram | 19:47 |
| Sam-I-Am | i think i called it 'ovs data plane' because 'ovs switch' felt redundant | 19:51 |
| Sam-I-Am | bbiab, appt | 19:52 |
| russellb | cya | 19:52 |
| *** numans has quit IRC | 19:58 | |
| *** lsp42 has quit IRC | 20:00 | |
| *** brad_behle_ has joined #openstack-neutron-ovn | 20:06 | |
| *** gangil has quit IRC | 20:11 | |
| *** lsp42 has joined #openstack-neutron-ovn | 20:11 | |
| *** brad_behle has quit IRC | 20:12 | |
| *** aginwala has quit IRC | 20:16 | |
| *** azbiswas has joined #openstack-neutron-ovn | 20:19 | |
| *** lsp42 has quit IRC | 20:20 | |
| *** allan_h has quit IRC | 20:22 | |
| sramamurthy | Hi I am seeing a problem with provider networks and acls | 20:23 |
| sramamurthy | I wanted to bounce it here prior to filing a bug | 20:24 |
| sramamurthy | The ovn plugin is configuring acls agains the provider network | 20:25 |
| sramamurthy | but it should really configure it on the 2-port lswitch | 20:26 |
| sramamurthy | corresponding to the port on the provider network | 20:26 |
| *** aginwala has joined #openstack-neutron-ovn | 20:27 | |
| sramamurthy | cmd = self._ovn.add_acl( lswitch=utils.ovn_name(port['network_id']), lport=port['id'], | 20:28 |
| sramamurthy | The switch derived from the port must be specialcased for | 20:28 |
| sramamurthy | provider network ports | 20:28 |
| sramamurthy | to point to the 2-port lswitch | 20:28 |
| sramamurthy | ?? | 20:29 |
| *** allan_h has joined #openstack-neutron-ovn | 20:35 | |
| *** allan_h has quit IRC | 20:36 | |
| russellb | sramamurthy: there is an open bug for that | 20:40 |
| russellb | sramamurthy: and we decided to not fix it, in favor of some OVN work to get rid of the need for the 2-port lswitch | 20:40 |
| russellb | once we get back to only having a single OVN lswitch per provider network, it'll be fixed as a side effect | 20:40 |
| sramamurthy | yea the current design is awkward because, the lswitch for the provider net doesnt have any ports but has ACLs referring to them | 20:42 |
| sramamurthy | the 2 port switch has ports but no ACLs, and even if we put the ACLs there for the port | 20:43 |
| sramamurthy | the ACLs would refer to ports not in the lswitch | 20:43 |
| russellb | yes, ACLs are put on the wrong lswitch | 20:43 |
| russellb | the ACLs shouldn't be referring to lports by name anymore | 20:43 |
| russellb | so if we moved them, they should work fine | 20:44 |
| russellb | just not worth the effort since that lswitch is going away | 20:44 |
| sramamurthy | Is it worth patching the plugin to make the SG work on provider networks in the interim ? | 20:44 |
| russellb | i'd review/merge it if someone wanted to do it | 20:44 |
| russellb | but i'm hoping we'll merge the alternative in the next week | 20:45 |
| sramamurthy | I can volunteer for it | 20:45 |
| russellb | ok | 20:45 |
| sramamurthy | if the merge is 1 week away then it may not be worth the effort | 20:45 |
| russellb | https://patchwork.ozlabs.org/patch/582095/ for OVN. and then https://review.openstack.org/#/c/266186/ for networking-ovn, though that one needs a rebase | 20:46 |
| russellb | ^^^ the patches in question | 20:46 |
| sramamurthy | thanks.. | 20:46 |
| russellb | sramamurthy: sorry for the trouble | 20:49 |
| sramamurthy | @russelb, thanks for the quick resolution.. | 20:50 |
| *** lsp42 has joined #openstack-neutron-ovn | 20:53 | |
| *** thumpba has quit IRC | 20:53 | |
| *** lsp42 has quit IRC | 20:57 | |
| russellb | sramamurthy: https://bugs.launchpad.net/networking-ovn/+bug/1539227 | 21:09 |
| openstack | Launchpad bug 1539227 in networking-ovn "Security group not enforced on physical networks" [Critical,In progress] - Assigned to Han Zhou (zhouhan) | 21:09 |
| *** lrichard_ has quit IRC | 21:13 | |
| *** lsp42 has joined #openstack-neutron-ovn | 21:24 | |
| *** lsp42 has quit IRC | 21:28 | |
| *** lsp42 has joined #openstack-neutron-ovn | 21:37 | |
| *** gangil has joined #openstack-neutron-ovn | 21:41 | |
| *** gangil has joined #openstack-neutron-ovn | 21:41 | |
| *** lsp42 has quit IRC | 21:49 | |
| *** lsp42 has joined #openstack-neutron-ovn | 22:00 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 22:03 | |
| *** allan_h has joined #openstack-neutron-ovn | 22:03 | |
| *** salv-orl_ has quit IRC | 22:05 | |
| *** rtheis has quit IRC | 22:05 | |
| *** aginwala has quit IRC | 22:14 | |
| *** aginwala has joined #openstack-neutron-ovn | 22:16 | |
| *** lsp42 has quit IRC | 22:16 | |
| *** lsp42 has joined #openstack-neutron-ovn | 22:30 | |
| *** zhouhan has quit IRC | 22:35 | |
| *** lsp42 has quit IRC | 22:35 | |
| *** allan_h has quit IRC | 22:35 | |
| *** lsp42 has joined #openstack-neutron-ovn | 22:45 | |
| *** yamamoto has quit IRC | 22:49 | |
| *** brad_behle_ has quit IRC | 22:50 | |
| *** lsp42 has quit IRC | 22:56 | |
| *** aginwala has quit IRC | 23:02 | |
| *** lsp42 has joined #openstack-neutron-ovn | 23:03 | |
| *** aginwala has joined #openstack-neutron-ovn | 23:06 | |
| *** jhfeng has joined #openstack-neutron-ovn | 23:07 | |
| *** lsp42 has quit IRC | 23:09 | |
| *** lsp42 has joined #openstack-neutron-ovn | 23:14 | |
| *** jckasper has quit IRC | 23:16 | |
| *** regXboi has quit IRC | 23:17 | |
| *** yamamoto has joined #openstack-neutron-ovn | 23:17 | |
| *** jckasper has joined #openstack-neutron-ovn | 23:17 | |
| *** jckasper has quit IRC | 23:21 | |
| *** yamamoto has quit IRC | 23:31 | |
| *** chandrav has quit IRC | 23:43 | |
| *** openstackgerrit has quit IRC | 23:47 | |
| *** openstackgerrit_ is now known as openstackgerrit | 23:47 | |
| *** openstackgerrit_ has joined #openstack-neutron-ovn | 23:47 | |
| *** openstackgerrit_ is now known as openstackgerrit | 23:48 | |
| *** openstackgerrit_ has joined #openstack-neutron-ovn | 23:49 | |
| *** jckasper has joined #openstack-neutron-ovn | 23:51 | |
| *** lsp42 has quit IRC | 23:53 | |
| *** jckasper has quit IRC | 23:55 | |
| *** openstackgerrit_ has quit IRC | 23:55 | |
| *** openstackgerrit_ has joined #openstack-neutron-ovn | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
