| *** azbiswas has joined #openstack-neutron-ovn | 00:00 | |
| *** azbiswas_ has joined #openstack-neutron-ovn | 00:03 | |
| *** azbiswas has quit IRC | 00:07 | |
| *** azbiswas_ has quit IRC | 00:15 | |
| *** manand has quit IRC | 00:19 | |
| *** azbiswas has joined #openstack-neutron-ovn | 00:33 | |
| *** azbiswas has quit IRC | 01:10 | |
| *** azbiswas has joined #openstack-neutron-ovn | 01:15 | |
| *** jckasper has joined #openstack-neutron-ovn | 01:42 | |
| *** jhfeng has joined #openstack-neutron-ovn | 01:43 | |
| *** azbiswas has quit IRC | 01:46 | |
| *** jckasper has quit IRC | 01:47 | |
| *** yamamoto has joined #openstack-neutron-ovn | 01:53 | |
| *** jhfeng has quit IRC | 02:13 | |
| *** yamamoto has quit IRC | 02:23 | |
| *** yamamoto has joined #openstack-neutron-ovn | 02:23 | |
| *** yamamoto has quit IRC | 02:29 | |
| *** arosen12 has quit IRC | 02:34 | |
| *** gangil has quit IRC | 02:53 | |
| *** s3wong has quit IRC | 02:53 | |
| *** yamamoto has joined #openstack-neutron-ovn | 03:11 | |
| *** brad_behle has joined #openstack-neutron-ovn | 03:33 | |
| *** gangil has joined #openstack-neutron-ovn | 03:53 | |
| *** gangil has joined #openstack-neutron-ovn | 03:53 | |
| *** gangil has quit IRC | 04:05 | |
| *** numans has joined #openstack-neutron-ovn | 04:17 | |
| *** openstack has joined #openstack-neutron-ovn | 04:29 | |
| *** markmcclain has joined #openstack-neutron-ovn | 04:29 | |
| *** jamespd has joined #openstack-neutron-ovn | 04:29 | |
| *** openstack has quit IRC | 04:30 | |
| *** openstack has joined #openstack-neutron-ovn | 04:32 | |
| *** openstack has quit IRC | 04:34 | |
| *** openstack has joined #openstack-neutron-ovn | 14:07 | |
| mestery | russellb: morning! | 14:07 |
|---|---|---|
| russellb | morning | 14:08 |
| russellb | what'd i break | 14:08 |
| mestery | lol | 14:08 |
| mestery | Nothing :) | 14:08 |
| mestery | Just been swamped this week | 14:08 |
| mestery | Wanted to say hi on Friday morning | 14:08 |
| mestery | FRIDAY! | 14:08 |
| mestery | Woot! | 14:08 |
| russellb | FRIDAY | 14:08 |
| russellb | it has been a little quiet here this week | 14:08 |
| mestery | YEah, man, tons going on | 14:08 |
| *** fzdarsky is now known as fzdarsky|afk | 14:09 | |
| mestery | I do have a question for you: is there any reason to assume that OVN will NOT work with VRRP? | 14:09 |
| russellb | yeah, quiet in a good way, lots of heads down | 14:09 |
| * mestery nods | 14:09 | |
| russellb | um | 14:09 |
| russellb | VRRP in what context | 14:09 |
| russellb | with l3 agent? | 14:09 |
| russellb | or future ovn? | 14:09 |
| mestery | So, guests running VRRP themselves in either provider networks or private networks | 14:09 |
| Sam-I-Am | moooOOOO | 14:09 |
| Sam-I-Am | l3ha > dvr :) | 14:09 |
| mestery | E.g. a guest does their own VRRP | 14:09 |
| mestery | *tenant | 14:09 |
| Sam-I-Am | if you're going to implement something | 14:09 |
| russellb | i see | 14:09 |
| russellb | it may require some patches not merged yet | 14:10 |
| russellb | i'd have to look at exactly how VRRP works again | 14:10 |
| mestery | Yeah, that's what I was thinking. | 14:10 |
| Sam-I-Am | its the sound you make after eating a pound of bbq | 14:10 |
| russellb | like, you might need allowed-address-pairs support or something | 14:10 |
| mestery | Right, that's what I was thinking too | 14:10 |
| Sam-I-Am | vrrp uses multicast addresses | 14:10 |
| mestery | We need to get that allowed address pair stuff in | 14:10 |
| russellb | i suspect allowed-address-pairs is needed... | 14:10 |
| * russellb guessing | 14:11 | |
| russellb | but any limitation would be something in that area | 14:11 |
| russellb | so certainly nothing fundamental | 14:11 |
| russellb | just short term stuff we need to get finished | 14:11 |
| russellb | i guess a related question is, what is needed to use it with current neutron+ovs | 14:11 |
| mestery | Good question | 14:12 |
| russellb | what API details | 14:12 |
| mestery | So, it turns out that internally, we had some folks using Linuxbridge + L2pop | 14:12 |
| mestery | And you can't do it there | 14:12 |
| mestery | Because of the l2pop | 14:12 |
| *** jckasper has joined #openstack-neutron-ovn | 14:12 | |
| mestery | I don't htink OVN prevents it in the same way though | 14:12 |
| russellb | orly | 14:12 |
| mestery | Yeah | 14:12 |
| russellb | we do something similar to l2pop though | 14:12 |
| mestery | Orly | 14:12 |
| mestery | That's right | 14:12 |
| russellb | what about l2pop breaks it | 14:12 |
| mestery | o_o | 14:12 |
| * mestery grabs the email | 14:12 | |
| mestery | Meh, no details, just that due to LB+L2pop+VXLAN guests can do their own VRRP | 14:13 |
| russellb | hand wave :) | 14:13 |
| mestery | Must be something with how L2pop is distributing the Mac:IP mappings | 14:13 |
| mestery | So you're right | 14:13 |
| mestery | Maybe it won't work for us | 14:14 |
| mestery | I should try it | 14:14 |
| russellb | we can certainly make it work | 14:14 |
| Sam-I-Am | why would tenants do their own vrrp? set up their own little network node boxes? | 14:14 |
| russellb | just need detailed requirements | 14:14 |
| mestery | Yes | 14:14 |
| russellb | i' | 14:14 |
| russellb | turns out we have a lot of tempest disabled right now :( | 14:15 |
| russellb | by accident | 14:15 |
| mestery | Oh no! | 14:15 |
| Sam-I-Am | russellb: dont look too closely at devstack | 14:17 |
| *** yamamoto_ has quit IRC | 14:17 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Update supported API extensions for tempest. https://review.openstack.org/288142 | 14:22 |
| russellb | we'll see if that actually passes tempest | 14:26 |
| russellb | i'm a little scared | 14:26 |
| *** fzdarsky|afk is now known as fzdarsky | 14:30 | |
| Sam-I-Am | heh | 14:32 |
| Sam-I-Am | well, everything will work fine until we remove external_network_bridge :) | 14:34 |
| *** openstackgerrit_ has quit IRC | 14:36 | |
| *** openstackgerrit_ has joined #openstack-neutron-ovn | 14:37 | |
| *** thumpba has quit IRC | 14:46 | |
| *** jckasper has quit IRC | 14:48 | |
| *** jckasper has joined #openstack-neutron-ovn | 14:48 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:05 | |
| *** yamamoto has quit IRC | 15:05 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:06 | |
| *** thumpba has joined #openstack-neutron-ovn | 15:18 | |
| *** thumpba has quit IRC | 15:20 | |
| *** thumpba has joined #openstack-neutron-ovn | 15:20 | |
| *** thumpba has quit IRC | 15:20 | |
| *** jckasper has quit IRC | 15:22 | |
| *** thumpba_ has joined #openstack-neutron-ovn | 15:22 | |
| *** thumpba_ has quit IRC | 15:22 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Update supported API extensions for tempest. https://review.openstack.org/288142 | 15:22 |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Fix QoS unit tests. https://review.openstack.org/288524 | 15:22 |
| *** thumpba has joined #openstack-neutron-ovn | 15:22 | |
| *** jckasper has joined #openstack-neutron-ovn | 15:22 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Run cross-tenant traffic tempest test https://review.openstack.org/288115 | 15:23 |
| *** jhfeng has joined #openstack-neutron-ovn | 15:31 | |
| *** palexster has quit IRC | 15:34 | |
| *** palexster has joined #openstack-neutron-ovn | 15:34 | |
| *** yamamoto has quit IRC | 15:39 | |
| *** regXboi has joined #openstack-neutron-ovn | 15:40 | |
| regXboi | russellb: ping | 15:40 |
| russellb | pong, on phone though | 15:40 |
| regXboi | russellb: it looks like some ovn tests cases are failing in master right now - I'm going to see about pushing a patch through to address them | 15:41 |
| russellb | unit tests? | 15:41 |
| russellb | if so, see https://review.openstack.org/288524 | 15:41 |
| regXboi | no I mean OVN cases | 15:42 |
| regXboi | from github | 15:42 |
| russellb | oh, ovn itself | 15:42 |
| regXboi | yes | 15:42 |
| regXboi | I just cloned openvswitch/ovs.git and the nbctl, vtep and one sbctl cases are failing | 15:43 |
| * russellb runs tests | 15:43 | |
| * regXboi re-runs tests | 15:43 | |
| * regXboi trying to make sure he doesn't have unexpected cruft hanging around | 15:44 | |
| russellb | all tests passing for me | 15:45 |
| regXboi | huh | 15:45 |
| regXboi | ok, then I must have some cruft running around | 15:46 |
| russellb | $ make check TESTSUITEFLAGS="-k ovn" | 15:46 |
| regXboi | yeah | 15:46 |
| regXboi | I'm doing that | 15:46 |
| russellb | passes here | 15:46 |
| russellb | travis-ci is faililng, but seems all IPv6 related | 15:46 |
| russellb | like IPv6 is suddenly completely disabled or something, so unrelated | 15:46 |
| regXboi | ok then it must be cruft on my test server, though I have no idea what or why | 15:47 |
| russellb | darn software | 15:47 |
| *** numans has quit IRC | 15:47 | |
| *** shangxdy has joined #openstack-neutron-ovn | 15:53 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:55 | |
| regXboi | russellb: I'm pretty sure something strangeo is going on here - the error message on the first ovn-nbctl test is ovn-nbctl: : database connection failed (Address family not supported by protocol) | 16:03 |
| regXboi | which means it's not even getting a socket | 16:03 |
| *** yamamoto has quit IRC | 16:05 | |
| lrichard_ | all tests passed for me this morning as well | 16:05 |
| lrichard_ | ^fwiw | 16:06 |
| regXboi | ugh | 16:07 |
| regXboi | ok I found it | 16:08 |
| regXboi | if you have OVN_NB_DB and or OVS_SB_DB set in your environment, then tests might fail unexpectedly | 16:09 |
| regXboi | er OVN_SB_DB | 16:09 |
| regXboi | russellb: is it worth doing something in the test code to avoid that possibility? | 16:09 |
| russellb | in meeting | 16:10 |
| regXboi | ack: will ping later | 16:10 |
| *** jckasper has quit IRC | 16:11 | |
| *** jckasper has joined #openstack-neutron-ovn | 16:12 | |
| *** jckasper has quit IRC | 16:16 | |
| *** jckasper has joined #openstack-neutron-ovn | 16:17 | |
| openstackgerrit | Richard Theis proposed openstack/networking-ovn: Doc: Add missing Neutron API extensions https://review.openstack.org/288563 | 16:20 |
| *** tongli has joined #openstack-neutron-ovn | 16:26 | |
| *** jckasper has quit IRC | 16:27 | |
| *** jckasper has joined #openstack-neutron-ovn | 16:28 | |
| *** jckasper has quit IRC | 16:29 | |
| *** armax has quit IRC | 16:29 | |
| *** jckasper has joined #openstack-neutron-ovn | 16:29 | |
| *** armax has joined #openstack-neutron-ovn | 16:29 | |
| mestery | russellb: Turns out we do need allowed address pairs to support VRRP in guest instances with OVN | 16:37 |
| mestery | A blog arosen wrote indicates as such: http://superuser.openstack.org/articles/implementing-high-availability-instances-with-neutron-using-vrrp | 16:37 |
| mestery | :) | 16:37 |
| russellb | k | 16:37 |
| russellb | will have that soon | 16:37 |
| russellb | up for review i think | 16:37 |
| regXboi | russellb: you still in mtg? | 16:37 |
| russellb | yes | 16:37 |
| regXboi | ok | 16:37 |
| * regXboi wanders away again | 16:37 | |
| mestery | russellb: Yes, looks like numan needs to re-roll to address a few things, so coolio! | 16:42 |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Run cross-tenant traffic tempest test https://review.openstack.org/288115 | 16:56 |
| russellb | moarrrrr testing | 16:56 |
| *** fzdarsky is now known as fzdarsky|afk | 16:59 | |
| *** yamamoto has joined #openstack-neutron-ovn | 17:05 | |
| *** yamamoto has quit IRC | 17:15 | |
| mestery | russellb: \o/ Testing! | 17:40 |
| *** jhfeng has quit IRC | 17:52 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 17:52 | |
| *** yamamoto has joined #openstack-neutron-ovn | 17:53 | |
| *** jhfeng has joined #openstack-neutron-ovn | 17:54 | |
| *** salv-orl_ has quit IRC | 17:55 | |
| *** azbiswas has joined #openstack-neutron-ovn | 17:57 | |
| *** jhfeng has quit IRC | 17:58 | |
| regXboi | ugh... so putting in the delete detection just didn't work | 18:06 |
| * regXboi goes back to the drawing board | 18:06 | |
| *** shangxdy has quit IRC | 18:09 | |
| russellb | software is hard | 18:12 |
| *** yarkot_ has joined #openstack-neutron-ovn | 18:13 | |
| *** yamamoto has quit IRC | 18:14 | |
| *** yamamoto has joined #openstack-neutron-ovn | 18:15 | |
| *** tongli has quit IRC | 18:21 | |
| *** yarkot_ has quit IRC | 18:21 | |
| *** yamamoto has quit IRC | 18:25 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Try re-enabling some IPv6 tests. https://review.openstack.org/288629 | 18:29 |
| russellb | "SKIPPED: subnet pools extension not enabled" | 18:32 |
| russellb | i thought we had that ... | 18:32 |
| russellb | mestery: ^ | 18:32 |
| mestery | huh? It should be there | 18:32 |
| russellb | i'll look closer to see why it thinks that | 18:32 |
| mestery | russellb:https://github.com/openstack/networking-ovn/blob/master/networking_ovn/plugin.py#L111 | 18:33 |
| russellb | 43 if not test.is_extension_enabled('subnetpools', 'network'): | 18:33 |
| russellb | 44 msg = "subnet pools extension not enabled." | 18:33 |
| russellb | 45 raise cls.skipException(msg) | 18:33 |
| russellb | not sure why it's checking "subnetpools" instead of "subnet_allocation" | 18:33 |
| russellb | i bet it's wrong | 18:33 |
| russellb | and nobody has noticed | 18:33 |
| russellb | because normal ml2+ovs runs with "all" | 18:34 |
| mestery | Yeah | 18:34 |
| russellb | so this would always return true | 18:34 |
| * russellb patches tempest | 18:34 | |
| russellb | MOARRRR TESTING | 18:34 |
| mestery | russellb: https://github.com/openstack/neutron/blob/master/neutron/plugins/ml2/plugin.py#L122 | 18:34 |
| mestery | subnet_allocation there too | 18:34 |
| russellb | yep | 18:34 |
| russellb | tempest is just wrong | 18:34 |
| russellb | thanks | 18:34 |
| mestery | Though they ahve a "default-subnetpools" extension now | 18:34 |
| mestery | cool | 18:35 |
| *** fzdarsky|afk is now known as fzdarsky | 18:36 | |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Run subnet_allocation tempest tests. https://review.openstack.org/288633 | 18:38 |
| *** pcaruana has quit IRC | 18:41 | |
| *** s3wong has joined #openstack-neutron-ovn | 18:53 | |
| mestery | russellb: You want https://review.openstack.org/#/c/288142/3 merged? Sam-I-Am can help if so | 18:57 |
| mestery | Sam-I-Am: Also https://review.openstack.org/#/c/288524/1 | 18:58 |
| *** gangil has joined #openstack-neutron-ovn | 18:59 | |
| *** gangil has joined #openstack-neutron-ovn | 18:59 | |
| russellb | sure, merge all the things | 19:01 |
| russellb | those 2 plus https://review.openstack.org/#/c/288115/3 | 19:01 |
| russellb | are the ones ready | 19:01 |
| mestery | ++ | 19:05 |
| russellb | rally job broken | 19:12 |
| russellb | if anyone is bored :-p | 19:12 |
| mestery | :( | 19:12 |
| *** fzdarsky is now known as fzdarsky|afk | 19:14 | |
| *** yamamoto has joined #openstack-neutron-ovn | 19:16 | |
| *** yamamoto has quit IRC | 19:21 | |
| * regXboi sadly decides that a simap isn't enough for what he needs to do, he has to do a full blown hmap :( | 19:39 | |
| russellb | 2016-03-04 19:26:19.861 | {2} tempest.api.network.test_subnetpools_extensions.SubnetPoolsTestJSON.test_create_list_show_update_delete_subnetpools [0.284324s] ... ok | 19:42 |
| russellb | yay passes | 19:42 |
| * russellb 's blood pressure starts returning to normal as all these new tests just work | 19:42 | |
| Sam-I-Am | russellb: here's a new one for you | 19:44 |
| * russellb prepares for blood pressure to rise again | 19:44 | |
| Sam-I-Am | russellb: how come 'neutron net-show' doesnt report the provider network parameters? | 19:44 |
| russellb | WELL | 19:45 |
| russellb | because the place in the db for that info is ML2 specific | 19:45 |
| russellb | we could figure it back out from the OVN db, but that code never got added | 19:45 |
| russellb | better solution is: https://review.openstack.org/242393 | 19:45 |
| russellb | if needed more urgently, we could fix it by reading info from the OVN db | 19:46 |
| Sam-I-Am | its something people are going to want to see when they do net-show | 19:47 |
| openstackgerrit | Merged openstack/networking-ovn: Fix QoS unit tests. https://review.openstack.org/288524 | 19:47 |
| Sam-I-Am | especially for nets with segmentation ids | 19:47 |
| russellb | right, totally makes sense | 19:47 |
| russellb | i honestly forgot to fix it | 19:47 |
| russellb | and then i thought that neutron patch was going to go in faster than it has | 19:47 |
| russellb | so it has just languished | 19:48 |
| russellb | but carl picked up that neutron patch to help push it through <3 | 19:48 |
| russellb | because carl_baldwin is awesome like that | 19:48 |
| * carl_baldwin blushes | 19:49 | |
| openstackgerrit | Merged openstack/networking-ovn: Update supported API extensions for tempest. https://review.openstack.org/288142 | 19:50 |
| Sam-I-Am | woot? | 19:50 |
| openstackgerrit | Russell Bryant proposed openstack/networking-ovn: Re-enable some IPv6 tests. https://review.openstack.org/288629 | 19:51 |
| russellb | GUYS | 19:51 |
| russellb | that brings our tempest config down to only disabling a couple of features because they're not implemented | 19:51 |
| russellb | and nothing because "fails for some weird reason i dunno" | 19:52 |
| russellb | and both of the things disabled have patches up for review | 19:52 |
| Sam-I-Am | is there something i should be reviewing? | 19:52 |
| Sam-I-Am | sort of scatterbrained right now | 19:52 |
| russellb | nothing urgent, no | 19:52 |
| openstackgerrit | Merged openstack/networking-ovn: Run cross-tenant traffic tempest test https://review.openstack.org/288115 | 20:08 |
| *** yamamoto has joined #openstack-neutron-ovn | 20:16 | |
| *** regXboi has quit IRC | 20:16 | |
| *** yamamoto has quit IRC | 20:21 | |
| *** salv-orlando has quit IRC | 20:33 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 20:33 | |
| *** yamamoto has joined #openstack-neutron-ovn | 20:37 | |
| *** yamamoto has quit IRC | 20:37 | |
| *** yamamoto has joined #openstack-neutron-ovn | 20:39 | |
| *** yamamoto has quit IRC | 20:39 | |
| *** thumpba has quit IRC | 20:54 | |
| *** thumpba has joined #openstack-neutron-ovn | 20:55 | |
| *** thumpba has quit IRC | 21:00 | |
| Sam-I-Am | russellb: moo? | 21:00 |
| Sam-I-Am | russellb: how would i pull the provider net information out of the ovn db? | 21:01 |
| russellb | one sec phone | 21:01 |
| russellb | Sam-I-Am: you'd have to check to see if there's a localnet port on a logical switch, and pull the info off it if so | 21:13 |
| russellb | Sam-I-Am: or we could stash the details as external_ids on the logical switch itself to make it even easier | 21:14 |
| Sam-I-Am | hmm | 21:14 |
| russellb | did that make sense? | 21:16 |
| Sam-I-Am | yeah, i'd need to get one in the broken state first | 21:17 |
| russellb | it'll always be in that state | 21:19 |
| russellb | if you use prov nets | 21:19 |
| Sam-I-Am | i dont know if sarcasm? | 21:24 |
| russellb | not sarcasm | 21:25 |
| russellb | that info will never be in the output of net-show | 21:25 |
| russellb | for a provider net | 21:25 |
| russellb | (i think) | 21:25 |
| Sam-I-Am | oh, i thought you were referring to provider nets as 'in the broken state' | 21:25 |
| russellb | oh, no | 21:25 |
| Sam-I-Am | we really do need that info in net-show though | 21:25 |
| russellb | they should be working :) | 21:25 |
| russellb | yep | 21:25 |
| russellb | understood | 21:26 |
| russellb | like, nowish? | 21:26 |
| Sam-I-Am | didnt realize it wasnt there until it... broke | 21:26 |
| russellb | i could do a quick hack to get it there now | 21:26 |
| Sam-I-Am | well, the soonerish the better, but no need to get hacky | 21:26 |
| russellb | well i call it a hack only because i'd be using the OVN as the source of the data | 21:26 |
| russellb | it'll be finnnne | 21:26 |
| russellb | can you create a bug for me? | 21:26 |
| Sam-I-Am | sure | 21:26 |
| russellb | i'll probably run out of time today | 21:26 |
| Sam-I-Am | russellb: https://bugs.launchpad.net/networking-ovn/+bug/1553368 | 21:29 |
| openstack | Launchpad bug 1553368 in networking-ovn "Networks should contain provider attributes" [Undecided,New] | 21:29 |
| Sam-I-Am | russellb: extend hours. done! | 21:29 |
| Sam-I-Am | <- problem solver | 21:30 |
| russellb | heh, is that how IBM works? make everyone work 80 hour weeks to get it done? | 21:30 |
| russellb | :-p | 21:30 |
| Sam-I-Am | russellb: shhhhhh | 21:30 |
| *** yamamoto has joined #openstack-neutron-ovn | 21:39 | |
| *** salv-orlando has quit IRC | 21:47 | |
| *** yamamoto has quit IRC | 21:48 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 21:48 | |
| *** rtheis has quit IRC | 21:51 | |
| *** salv-orlando has quit IRC | 21:51 | |
| *** salv-orlando has joined #openstack-neutron-ovn | 21:52 | |
| *** pcaruana has joined #openstack-neutron-ovn | 21:58 | |
| *** azbiswas has quit IRC | 22:14 | |
| Sam-I-Am | russellb: have a good weekend :) | 22:47 |
| *** salv-orl_ has joined #openstack-neutron-ovn | 23:52 | |
| *** salv-orlando has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!