15:00:07 <regXboi> #startmeeting neutron_northbound 15:00:07 <odl_meetbot> Meeting started Fri May 8 15:00:07 2015 UTC. The chair is regXboi. Information about MeetBot at http://ci.openstack.org/meetbot.html. 15:00:07 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:07 <odl_meetbot> The meeting name has been set to 'neutron_northbound' 15:00:15 <regXboi> #chair flaviof edwarnicke 15:00:15 <odl_meetbot> Current chairs: edwarnicke flaviof regXboi 15:00:16 <adetalhouet> #info adetalhouet 15:00:24 <regXboi> #topic roll call and agenda bashing 15:00:37 <flaviof> mestery: need a hand on that. I know I need to edit requirements.txt, right? 15:00:40 <flaviof> mestery: https://bugs.launchpad.net/devstack/+bug/1448376 15:00:48 <regXboi> #link https://wiki.opendaylight.org/view/NeutronNorthbound:Meetings#Agenda_for_Next_Meeting_.285.2F8.29 agenda in the usual place 15:00:57 <regXboi> #info regXboi 15:01:10 <regXboi> anybody have additional topics? 15:01:50 * regXboi waits a few minutes 15:02:02 <regXboi> oh yes, attendees, please #info in 15:02:19 <flaviof> #info flaviof 15:02:34 * regXboi looks pointedly at mestery 15:02:46 <regXboi> regXboi looks for edwarnicke 15:02:48 <mestery> #info mestery (lurking) 15:03:24 <edwarnicke> #info edwarnicke 15:03:43 <regXboi> last call for agenda topics 15:04:21 <regXboi> #topic action items from last meeting 15:04:22 <edwarnicke> regXboi: Do we have status of the odl-neworking driver in OS and secgroups/secgroup_rules/router_interfaces ? 15:05:05 <regXboi> edwarnicke: sorry for being very blunt - but given that M5 was a week ago and we've not (afaik) done anything for it - can I afford to care? 15:05:18 <mestery> lol 15:05:32 <edwarnicke> regXboi: We sort of need to 15:05:47 <edwarnicke> regXboi: Because we still aren't getting necessary objects from OS 15:05:55 <edwarnicke> regXboi: We have code to handle them 15:05:59 <edwarnicke> regXboi: We just aren't getting it 15:06:15 <edwarnicke> regXboi: And without those we can't even do L2 secgroups correctly 15:06:16 <regXboi> so - I understand that SB projects may care 15:06:22 <edwarnicke> regXboi: Much less L3 correctly 15:06:36 <edwarnicke> regXboi: So we either get them... or we bump real neutron support that works correctly to Be 15:06:39 <regXboi> but *at this precise moment* - that isn't the top of my priority list 15:06:53 <edwarnicke> regXboi: We actually do need to care 15:07:03 <edwarnicke> regXboi: Because we need to test and fix bugs in the code that catches that stuff 15:07:23 <edwarnicke> regXboi: But lets file that down agenda, don't want to disturb your flow :) 15:07:34 <regXboi> no... you've already blown up the meeting 15:07:38 <regXboi> I'd I'm not in the mood for it 15:07:40 <regXboi> so .... 15:07:46 <edwarnicke> ? 15:07:59 <regXboi> you are essentially saying that we can't declare any M5 items as closed? 15:08:05 <regXboi> that's what I'm hearing 15:08:13 <edwarnicke> regXboi: I don't think *precisely* that 15:08:22 <edwarnicke> regXboi: We *do* have code for those things to 'catch' on the ODL side 15:08:38 <edwarnicke> regXboi: So if there are issues where we don't match up with the OS side 15:08:42 <edwarnicke> regXboi: Those *are* bugs 15:08:58 <edwarnicke> But I think we should call out for M5 clearly what we do and don't have the capacity to test against OS 15:09:10 <edwarnicke> mestery: Were do we stand on getting those objects from OS? 15:09:16 <edwarnicke> Where 15:09:31 <mestery> edwarnicke: We can't *test* this stuff, so the patch isn't passing jenkins on the OpenStack side, thus it's not gonna merge anytime soon. 15:09:49 <mestery> flaviof and I have been working on it 15:09:58 <edwarnicke> mestery: I'm confused 15:10:18 <edwarnicke> mestery: So do we have a patch that sends us secgrp, secgrp_rule, and router_interface? 15:10:38 <mestery> edwarnicke: We have one proposed, yes, but it's not passing our Jenkins on the OpenStack side yet. 15:10:50 <edwarnicke> mestery: OK... I am now unconfused :) 15:10:51 <flaviof> edwarnicke: https://review.openstack.org/#/q/status:open+project:stackforge/networking-odl,n,z 15:11:04 <edwarnicke> mestery: What is failing? 15:11:19 <mestery> edwarnicke: Lets not go down that rathole here and side-track regXboi's fancy meeting 15:11:30 <regXboi> #link https://review.openstack.org/#/c/178325/ patch for passing security groups to OS 15:11:37 <regXboi> no actually - this is the last topic 15:11:37 <edwarnicke> mestery: Could you at least point to the failing tests? 15:11:41 <regXboi> and we might as well cover it 15:11:44 <edwarnicke> mestery: I might break free some cycles to help 15:11:51 <edwarnicke> mestery: And I have a history of unblocking things ;) 15:11:56 <mestery> edwarnicke: ODL doesn't like to run in the OpenStack gate 15:12:03 <regXboi> #info (note: this is *not* action items, this is other topics) 15:12:03 <mestery> Hell, I can't get it to run on my systems either lately 15:12:08 <mestery> flaviof is likely more help as he has it working 15:12:23 <edwarnicke> mestery: Ah... OK... so that's something I am much more likely to be helpful on 15:12:44 <regXboi> #action edwarnicke, mestery, flaviof to work on ODL running in the OS gate so that we can unblock OS patches 15:12:54 * regXboi delegating 15:13:07 <edwarnicke> mestery: flaviof Got time right after this to jump on a hangout and kill this issue dead? 15:13:17 <regXboi> #info and maybe back to the regularly scheduled agenda 15:13:18 <mestery> Not me 15:13:29 <edwarnicke> mestery: OK. When? 15:13:41 <flaviof> edwarnicke: other than mestery and I think nobody is doing much on OS with ODL... this is the biggest issue, imo 15:13:42 <edwarnicke> mestery: Because historically, I'm pretty good at solving this sort of problem :) 15:13:48 <mestery> flaviof: ++ 15:13:54 <edwarnicke> flaviof: That's not actually true 15:14:10 * mestery grumbles 15:14:10 <edwarnicke> flaviof: But I would say its accurate you guys are the only ones trying to get it running in OS CI 15:14:24 <edwarnicke> flaviof: And in CI at ODL currently 15:14:39 <edwarnicke> But I know quite a few folks who stack against ODL at least daily :) 15:15:06 <regXboi> so - that statement (while true) - is not really a helpful one 15:15:09 <edwarnicke> mestery: So when would be a good time to get on a hangout and kill the 'not running' issues you are seeing 15:15:19 <flaviof> edwarnicke: I'd like to hear more about that.... this has proven to be a bit fragile. 15:15:28 <flaviof> right now I can't stack. 15:15:28 <edwarnicke> flaviof: I feel ya 15:15:57 <regXboi> ok folks 15:16:06 <edwarnicke> OK... so flaviof can't stack, and mestery can't get ODL to run, correct? 15:16:09 <regXboi> I'm taking back the meeting - you all can work on that afterwards 15:16:17 * edwarnicke yields 15:16:29 <regXboi> #topic action items from last meeting (for real this time) 15:16:42 <flaviof> regXboi: ack. sorry for the sidetrack 15:16:42 <regXboi> #info item 1: trello board reminder 15:17:06 <regXboi> #link https://trello.com/b/LhIIQ8Z0/odl-neutronnorthbound trello board 15:17:26 <regXboi> #info item 2: documentation - do we have a patch for this? we desparately need it for M5 15:17:35 <adetalhouet> yes: MdsalUtils#getOtherConfig(Node node, Type type, String key) 15:17:39 <adetalhouet> oops 15:17:46 <adetalhouet> there we go: https://git.opendaylight.org/gerrit/#/c/19850/ 15:17:48 * regXboi understands 15:17:59 <regXboi> #link https://git.opendaylight.org/gerrit/#/c/19850/ documentation patch 15:18:06 <regXboi> adetalhouet: thanks 15:18:09 <adetalhouet> I started something, needs to be updated 15:18:35 <regXboi> #action all to review documentation patch at https://git.opendaylight.org/gerrit/#/c/19850/ and provide updates 15:18:43 <regXboi> #item 3: stability branch 15:18:45 <adetalhouet> So feedback are welcome on this, due to lack of inspiration from my part 15:18:56 <regXboi> #info item 3 stability branch 15:19:06 <regXboi> #info noe: this bleeds into the next topic so... 15:19:11 <regXboi> #topic M5 15:19:21 <regXboi> #info we are > week behind on this so... 15:19:28 <regXboi> can we declare code freeze? 15:19:50 <regXboi> modulo the possibility of bugs that have to be fixed w.r.t to OS? 15:19:58 <regXboi> er s/possibility/probability/ 15:20:12 <edwarnicke> regXboi: How is the open patch queue? 15:20:16 * edwarnicke goes to look 15:20:30 <regXboi> #link https://git.opendaylight.org/gerrit/#/q/status:open+project:neutron open patch queue 15:20:38 <regXboi> #info only one "do not merge" patch for testing 15:21:01 <edwarnicke> regXboi: Yep 15:21:01 <regXboi> note: we *need* that patch, but I have zero time to work on it for the next week :( 15:21:12 <edwarnicke> Guys, how are we doing on the neutron mdsal stuff? 15:21:26 <edwarnicke> regXboi: Do we just declare what we have is what we have at this point? 15:21:28 * regXboi checks trello board 15:21:50 * edwarnicke is distinctly *not* putting his thumb on that scale, just wants us to go into M5 eyes open 15:22:23 <regXboi> other than the NeutronTranscriberProvider and FW stuff it looks like it is all complete 15:22:28 <adetalhouet> Missing this one: NeutronTranscriberProvider on all the NeutronFirewall* 15:22:53 <regXboi> who can take those last four? 15:23:05 <adetalhouet> Hang on, I added the Provider but might be a mistake 15:23:12 <edwarnicke> regXboi: Could you #link in the trello board 15:23:27 <regXboi> edwarnicke: scan back - already did 15:23:32 <adetalhouet> Ok, I'll end this card 15:23:47 <adetalhouet> Just regarding the NeutronFirewall*, what do we do? 15:23:56 <regXboi> afaik nobody is using it 15:23:59 * edwarnicke is apparently not quite awake yet 15:24:11 <regXboi> I'm inclined to not do it 15:24:13 <adetalhouet> nothing has been implemented yet as it is deprecated 15:24:20 <adetalhouet> alright 15:24:38 <adetalhouet> https://trello.com/c/cxsEMMyV/5-and-support-for-yang-model-to-transcriber 15:24:41 <regXboi> any objections to closing the card? 15:24:46 <edwarnicke> Nope 15:24:55 <regXboi> #agreed declaring code freeze 15:25:02 <regXboi> adetalhouet: go ahead and close the card 15:25:14 <regXboi> so... that means we can cut a stability branch 15:25:36 <regXboi> edwarnicke: you and I were on the hook for that - when do you have time? 15:26:09 <flaviof> regXboi: about https://trello.com/c/U2awRdUs/1-feature-based-testing-of-dummy-provider ... 15:26:18 <flaviof> regXboi: did you see the errors in https://jenkins.opendaylight.org/releng/view/neutron/job/neutron-daily-openstack-master/ ? 15:26:42 <regXboi> flaviof: I've had no time to look at any of this stuff 15:26:51 <flaviof> regXboi: ack. 15:26:53 <regXboi> I'm stealing time just to review patches 15:27:08 <flaviof> i hear ya 15:27:51 <regXboi> flaviof: the "do not merge" patch was an attempt to test the dummy provider without OS 15:27:52 <edwarnicke> regXboi: Going to look for instructions for pulling the stable/lithium branch 15:28:02 <edwarnicke> (I know how to do it... but I may not know the *best* way to do it :) ) 15:28:29 <regXboi> edwarnicke: what time zone are you in today? 15:28:38 <edwarnicke> regXboi: PST 15:28:49 <edwarnicke> regXboi: I'm in Portland until the OpenStack Summit 15:28:54 <edwarnicke> Then in Vancouver, BC 15:28:57 <edwarnicke> Then back to Seattle 15:29:07 <regXboi> edwarnicke: I have a scrum call at the top of the hour - but will find you when its done and we can get this branch cut 15:29:19 <edwarnicke> regXboi: Sure 15:29:24 <regXboi> note: this call could go as long as 90 minutes - that's how bad it is 15:29:41 <regXboi> ok... we've covered documentation 15:29:50 <regXboi> and integration and test was earlier as well 15:30:08 <regXboi> so... I can fill out the M5 template by the end of the day (warts and all) 15:31:10 <regXboi> at this point, *now* the dev stack comments from earlier are front and center because we need to fix them so that we have I&ST 15:31:39 <regXboi> #topic Working Session on Openstack CI issues 15:31:46 <edwarnicke> regXboi: Ack 15:32:15 <regXboi> are we working off the errors in https://jenkins.opendaylight.org/releng/view/neutron/job/neutron-daily-openstack-master/ ? 15:32:49 <regXboi> flaviof, mestery - is that a reasonable link to add for the minutes? 15:33:04 <mestery> I'll defer to flaviof, but it looks ok 15:33:06 <flaviof> regXboi: I'd think not. I think that is related to dummy_neutron 15:33:29 <regXboi> #link https://jenkins.opendaylight.org/releng/view/neutron/job/neutron-daily-openstack-master current job has errors 15:33:40 <regXboi> #info flaviof thinks they are related to the dummy provider 15:33:59 <flaviof> regXboi: actually, that is not a bad place to start, because it is 100% neutron. 15:34:02 <regXboi> ok... I'll try and take a look in this next week 15:34:29 <regXboi> #action regXboi to do best effort to look at issues neutron-daily-openstack-master is having 15:34:44 <flaviof> as for sg, mestery is relentlessly working on https://review.openstack.org/#/c/178325/ 15:35:07 <regXboi> #info flaviof and mestery continue to work on https://review.openstack.org/#/c/178325/ 15:35:27 <regXboi> flaviof: that gist above - is it the current issue with respect to SG? 15:36:11 <flaviof> #link https://wiki.opendaylight.org/view/OVSDB:Running_Tempest_Tests a good place for folks to look in order to help testing OS+ODL 15:36:33 <regXboi> yes, but do we have the list of issues on the trello board? 15:36:43 <regXboi> *should* we have the list of issues on the trello board? 15:36:46 <regXboi> :) 15:37:09 <flaviof> regXboi: yes: https://trello.com/b/ddIvDQE0/ovs-openstack 15:37:24 <regXboi> um 15:37:29 <flaviof> they are at the bottom of the "to do" 15:37:49 <regXboi> imho, those should really be moved to the neutron board, shouldn't they? 15:38:01 <flaviof> it is hard to separate ovsdb net-virt from neutron. my gut tells me 99% is pure neutron issues 15:38:10 <flaviof> regXboi: absolutely. 15:38:24 <flaviof> I'd think they should be 'updated' and moved to neutron. 15:38:30 <regXboi> can I action you to do that? 15:38:32 <flaviof> it is a very moving target. 15:38:55 <flaviof> now that openstack ci moved to Li, the tests results are very different. 15:39:13 <regXboi> #action flaviof to update OS issues and move from https://trello.com/b/ddIvDQE0/ovs-openstack to https://trello.com/b/LhIIQ8Z0/odl-neutronnorthbound 15:39:16 <flaviof> #link https://jenkins.opendaylight.org/releng/view/ovsdb/job/ovsdb-daily-openstack-master/46/artifact/logs/testr_results.html latest tempest results 15:39:23 <flaviof> regXboi: ack 15:39:28 <regXboi> flaviof: thanks 15:40:23 <flaviof> one more link.... 15:40:34 <flaviof> #link https://wiki.opendaylight.org/view/OVSDB:Tempest_Tests_daily_Results OVSDB:Tempest Tests daily Results 15:41:27 <regXboi> flaviof: thanks 15:41:29 <regXboi> (again) 15:41:34 <flaviof> regXboi: np! 15:41:42 <regXboi> ok... anything else? 15:42:01 <flaviof> that is all from me. 15:43:36 <regXboi> anybody else - going once? 15:44:13 <regXboi> going twice? 15:44:34 <regXboi> #action regXboi to update the meeting page today :) 15:44:52 <regXboi> and sold... 15:44:56 <regXboi> #endmeeting