21:00:48 #startmeeting Networking 21:00:49 Meeting started Mon May 5 21:00:48 2014 UTC and is due to finish in 60 minutes. The chair is markmcclain. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:50 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:52 The meeting name has been set to 'networking' 21:01:03 hi 21:01:17 mestery is at a conference today, so I'll be filling in 21:01:23 #link https://wiki.openstack.org/wiki/Network/Meetings Agenda 21:01:38 #topic Announcements 21:01:58 #link Juno Blueprint instructions are here: https://wiki.openstack.org/wiki/Blueprints#Neutron 21:02:00 hiya 21:02:07 yo 21:02:11 hi 21:02:11 hi 21:02:12 We already have a large of number of them filed 21:02:22 #link https://review.openstack.org/#/q/status:open+project:openstack/neutron-specs,n,z 21:03:04 Also the team is converging on a date/location for the mid-cycle sprint those plans should be shared sometime next week 21:03:12 #topic Bugs 21:03:53 everyone's favorite bug is back 21:03:53 https://bugs.launchpad.net/tempest/+bug/1253896 21:03:55 Launchpad bug 1253896 in tempest/havana "Attempts to verify guests are running via SSH fails. SSH connection to guest does not work." [Critical,Confirmed] 21:03:55 #link https://bugs.launchpad.net/tempest/+bug/1253896 21:04:03 but can we open a different one? 21:04:13 1253896 means just: "network does not work" 21:04:23 yeah we should 21:05:18 markmcclain: Is somebody looking at resurface of this bug. I think i've seen something related and afazekas has already a patch for it 21:05:25 well two actually 21:05:45 right there is this proposed fix 21:05:45 https://review.openstack.org/#/c/92018/ 21:06:15 it passed the heat slow tests 21:06:28 but failed grenade testing 21:07:14 the revert was also proposed: https://review.openstack.org/#/c/92147/ 21:07:37 I'd have to look more, but it seems that 92018 makes more sense than reverting 21:08:23 I cannot see why it make more sense. sorry. 21:08:58 am I missing something? 21:09:22 if the approach is remove what introduced the failure than it does not make sense... 21:10:08 * salv-orlando when I had a bad knee I also decided to stop walking to solve the problem 21:10:55 Just amputate ;) 21:10:56 gongysh, markmcclain: the patches are on gerrit, and we can keep the discussion going on there, I think 21:11:06 sure 21:11:16 agreed 21:11:43 We have one other critical bug 21:11:45 #link https://bugs.launchpad.net/neutron/+bug/1314313 21:11:46 Launchpad bug 1314313 in neutron "Firewall fails to become active within 300 seconds" [Critical,Triaged] 21:12:12 this is the one I find confusing according to comments from the fwaas team 21:12:28 salv-orlando: confusing? 21:12:37 one comment says that we should change the test in order to ensure the router is active before creating the firewall (while leaves me with a slight question mark) 21:12:40 salv-orlando: i thought i echoed your findings 21:12:58 and another comment says that instead the firewall should be created fine even if the router is pending create 21:13:07 SumitNaiksatam: see the comment after yours 21:13:50 This is showing often in the gate queue, so I would like to sort it out before we get closer to J-1 deadline 21:13:50 salv-orlando: the commend from SridarK is also reinforcing what we are saying 21:14:09 salv-orlando: was echoing SumitNaiksatam 's comment - the router create should be fine even if it happens after the firewall create 21:14:21 SridarK: ah the ROUTER create 21:14:37 salv-orlando: In further digging thru the logs - it seems that does happen 21:15:06 but why that does not get reflected back to the Plugin is not very clear 21:15:15 ok, I think I can be reassured you guys are working on the issue, we can discuss futher in openstack-neutron 21:15:33 sounds good 21:15:38 salv-orlando: SumitNaiksatam sounds good - will add more updates to the bug 21:15:43 any other bugs the team needs to discuss? 21:15:47 SridarK: for that I think the analysis in my comment and the logs I pointed out might help 21:16:03 markmcclain: Something is bugging me, but it's not a bug 21:16:08 salv-orlando: yes was looking at the logs 21:16:21 I was looking at the various sub teams and we do not have a bug management team... 21:16:26 is it possible to change the timeout for that test? 21:16:36 salv-orlando: ah good point 21:16:59 might be good to reform that again 21:17:16 should we put on topic list to ensure we cover next? 21:17:19 shouldnt each sub team be triaging the bugs? 21:17:21 yeah, we have some obsolete team. 21:17:36 triaging the bugs related to their sub team focus 21:17:47 SumitNaiksatam: ideally yes, but it still helps to have a team that looks across the teams 21:18:06 SumitNaiksatam: let me answer with another question - did the firewall team triage bug 1314313 21:18:07 Launchpad bug 1314313 in neutron "Firewall fails to become active within 300 seconds" [Critical,Triaged] https://launchpad.net/bugs/1314313 21:18:09 If the bug team tags stuff that might help 21:18:09 ;) 21:18:14 markmcclain: that should be the core team, right? 21:18:27 I know we monitor the ipv6 tag 21:18:33 SumitNaiksatam: doesn't have to be 21:18:38 I did not want to be cheeky (well, a bit), but it would help have a clear interface for qa and infra teams 21:18:56 I think we have 2 sessions that should possibly touch on this topic 21:19:25 if we don't get to it we can revisit in 2 wks when we're all here 21:19:46 salv-orlando: the bug was responded to as soon as it was noticed 21:20:04 salv-orlando: I agree a clear interface would help everyone 21:20:09 +1 21:20:21 sure. That's the problem indeed (as soon as it's noticed is often too late). 21:20:29 and nobody but the fixer's have visibility 21:21:00 Any other bugs to discuss? 21:21:51 #topic Docs 21:21:52 emagana: hi 21:22:01 markmcclain: hi there! 21:22:24 just a couple of announcements 21:22:57 as you know the Docs team follows a different schedule for their release and that will just happen this week 21:23:13 right 21:23:47 a terrible number of bugs have been fixed in the past weeks.. great kudos to the docs team and anyone here helping on closing bugs 21:24:08 I will update our wiki with the leftovers and keep in eye on new ones 21:24:09 awesome glad we've been able to close a bunch of them 21:24:19 good idea 21:24:41 Anything else to add? 21:24:46 That is all I've got 21:24:51 Thanks for the update 21:24:57 #topic Summit 21:25:11 #link http://junodesignsummit.sched.org/overview/type/neutron Summit Schedule 21:25:28 #info If you have a conflict contact Kyle 21:25:38 Also it's time to start adding etherpad links 21:25:46 #link https://wiki.openstack.org/wiki/Summit/Juno/Etherpads#Neutron 21:26:09 it really helps to have these precreated and prepopulated prior to the summit 21:26:37 doing so will greatly benefit everyone including those of us driving sessions 21:27:14 markmcclain: Our team wanted to understand better a couple of sessions that have been scheduled but no blueprint has been associate to them.. 21:27:42 http://summit.openstack.org/cfp/details/433 & http://summit.openstack.org/cfp/details/432 21:27:59 emagana: I'd suggest contacting the session proposer and ask that they post their etherpad 21:28:07 * markmcclain has few etherpads to post 21:28:22 you are the lead for those ones, can you point us to the BPs? :-) 21:29:06 emagana: yes I 21:29:16 I'll post the etherpads and blueprings 21:29:21 *blueprints 21:29:48 markmcclain: was blueprint a pre-requisite for proposing sessions? 21:30:11 I've had to refactor the designs a bit to maintain compatibility for certain items for a longer period of time 21:30:11 I didn't think so 21:30:22 SumitNaiksatam: That was my understanding as well 21:30:47 +1 for don't think so 21:31:14 SumitNaiksatam: it's not a hard requirement and Kyle and I discussed the outline of the sessions before they were added to schedule 21:31:30 I'll post items in advance so everyone can be prepared 21:31:31 yeah those two sessions are not bureaucratically compliant. 21:31:57 markmcclain: ok 21:32:00 * salv-orlando loves bureaucracy. Loves stamps too. 21:32:18 salv-orlando: do you have a permit for stamps? 21:32:31 haha 21:32:33 * stamps are here everywhere. 21:33:05 One last item is we'll have an informal dev gather on Thursday… we can discuss this during Open Discussion 21:33:16 #topic Tempest 21:33:30 mlavalle: anything to update on this week? 21:33:41 Hi, Yes a few things 21:33:55 what's new? 21:34:03 I have posted to the meeting agenda the last 6 api tests than need to be merged 21:34:17 after those, our api tests will be complete 21:34:24 great! 21:34:43 I ask the team to take a look and help to move them towards merge 21:35:14 I added the tempest https://etherpad.openstack.org/p/TempestAndNeutronJuno etherpad to the summit page 21:35:15 yeah definitely.. would be nice have these merged by next week 21:35:22 thanks for adding it 21:35:40 I am proposing 4 key actions for Juno 21:35:57 1) place empahsis on scenario tests, where we have one weak area 21:36:13 2) Fill any gaps that might have been left in api in icehouse 21:36:36 3) See with markmcclain if there are special testing needs for nova parity 21:36:51 4) See if any other subprojects need help 21:37:03 I think those are reasonable 21:37:09 asking the team to take a look on the etherpad and give us their inpits 21:37:20 let's work offline on #3 21:37:26 cool 21:37:30 that's all I have 21:38:05 mlavalle: thank for the update 21:38:08 mlavalle: any known bug blocking these tests: check-grenade-dsvm-partial-ncpu-icehouse ? 21:38:08 #topic L3 21:38:14 hi 21:38:23 carl_baldwin: looks like the team has been busy with summit prep 21:38:24 I’ll just highlight a couple of things. 21:38:33 emagana: mhh, let's talk offline 21:38:36 Right, we have a number of summit sessions to prepare for. 21:38:39 mlavalle: ok 21:39:08 I’ll highlight DVR progress. The distributed component is mostly functional now. Patches are being updated. 21:39:26 The central component which will host default SNAT is still a WIP. 21:39:35 … but moving right along. 21:40:06 it's good to have code now 21:40:27 Yes, it is. 21:40:53 Go to the subteam report on the neutron meeting page to see all of the summit discussions that are planned. 21:41:04 thanks for keeping it updated 21:41:17 glad to help. That is all I have. 21:41:34 carl_baldwin: thanks for sharing 21:41:49 carl_baldwin: nice to hear code are coming. 21:41:53 #topic Advanced Services 21:41:56 SumitNaiksatam: hi 21:42:01 markmcclain: hi 21:42:12 looks like this spec needs folks to read it: https://review.openstack.org/#/c/92200 21:42:29 yeah 21:42:37 we have been discussing for some time now 21:42:45 the high level design have been captured in that 21:42:53 or rather the building blocks 21:43:05 other folks are working in parallel on detailed specs 21:43:21 this and other topics in weekly IRC 21:43:30 https://wiki.openstack.org/wiki/Meetings/AdvancedServices 21:43:56 thats it on this 21:44:37 ok… thanks for updating 21:44:45 #topic IPv6 21:44:47 sc68cal: hi 21:44:51 Hello 21:45:07 I updated my part of the agenda with links to some specs 21:45:22 * sc68cal is away from desktop 21:45:39 Our main focus is getting specs drafted 21:45:50 thanks for updating the agenda 21:46:05 However there was a duplicate review for handling entries in dnsmasq that had me concerned 21:46:18 I sadly don't have the link handy 21:46:45 But there are some tricky corner cases that I hope to document at a later date 21:46:53 That's all I have for now 21:47:15 ok.. I'll see if I can track that down after the meeting 21:47:31 Thanks for the update 21:47:34 #topic ML2 21:47:49 rkukura: agenda looks old anything new? 21:47:58 Nothing new to report 21:48:03 ok 21:48:07 #topic Group Policy 21:48:14 We just need to get the etherpads ready for the summit 21:48:27 rkukura: don't we all :) 21:48:39 So moving onto group policy 21:48:49 I wanted to raise an issue 21:48:50 https://review.openstack.org/#/c/89469/9/specs/juno/group-based-policy-abstraction.rst 21:49:16 I raised a -1 on this spec and then a new version was pushed without my concern ever really being addressed 21:50:06 when reviewing specs please read the comment history to make sure review concerns are addressed before adding +2s 21:50:17 markmcclain: i did respond to your comment 21:50:38 right but it was merged before I had a chance to acknowledge and respond 21:50:51 markmcclain: the approval was not in my control 21:51:05 markmcclain: at my end i tried reaching out to you over IRC as well 21:51:15 SumitNaiksatam: agreed it is not under your control 21:51:27 this is more a reminder that spec reviews should consider review history 21:51:44 SumitNaiksatam: last week was an off week for entire project 21:51:51 markmcclain: +1 21:52:05 I'll follow up w an email on my concerns I still have 21:52:16 would you be posting that on the review? 21:52:18 markmcclain: sure 21:53:28 rudrarugge: I'll follow up via email 21:53:43 #topic Open Discussion 21:53:48 So next week is the summit 21:54:01 traditionally we've held a informal happy hour 21:54:46 looking at holding early evening on Thursday (since there are no Neutron meetings that day) 21:55:28 So that area around the venue does not great bars or restartaunts 21:55:41 usually requires a short cab or subway ride 21:56:18 quick show of hands who'd be interested in something pre-dinner 5-7ish 21:56:22 You're the local, I'm willing to travel 21:56:32 For a good place 21:56:32 +1 21:56:42 markmcclain: +1 to happy hour 21:56:43 I think the neutron team should be sober 21:56:56 salv-orlando: -1 21:57:13 salv-orlando: we should all watch salv-orlando :-) 21:57:18 +1 21:57:32 +1 21:57:32 +1 21:57:33 +1 21:57:37 +1 21:57:38 +1 21:57:39 : 21:57:45 :) 21:57:48 +1 21:57:55 +1 21:58:15 ok.. I'll call a couple of places and we can announce it next week 21:58:29 +1 just to make salv-orlando happy I'll be designated driver 21:58:53 markmcclain: has a decision been made on the mid-cycle meeting location? 21:58:57 pcm_: well, cab driver would be the designated driver in this case :-) 21:59:14 mlavalle: not yet should be announced next week once final location is confirmed 21:59:21 mlavalle: will US based though 21:59:42 tnanks 22:00:15 Since I've live in Atlanta I've gotten pinged for restaurant recommendations 22:00:17 http://atlanta.eater.com/archives/2014/04/08/the-38-essential-atlanta-restaurants-april-2014.php 22:00:39 It's a good list and many aren't too far from the GWCC 22:00:50 markmcclain: I think people were expecting personal recommendations, not a link 22:01:13 Tough crowd.. 22:01:17 you should write a " Atlanta food and drinks for the openstack developer " booklet 22:01:27 salv-orlando: haha… Bacchanalia is the #1 place in town 22:01:37 but prob not on many people budgets :) 22:02:11 westside district has lots of good places and is not too far from convention site 22:03:01 but you will need car/taxi walking the 2mis between that westside neighborhood and cnn center not ideal at night :) 22:03:18 also a few evening events are in midtown neighborhood 22:03:33 that neighborhood is walkable 22:04:07 but not back to downtown hotels, but subway/cab is available 22:04:14 markmcclain: you're not doing a good service to tourism in your city 22:04:19 the mayor is disappointed 22:04:25 * markmcclain is a terrible tourist 22:05:29 We're at time for this week. There will not be an IRC meeting next week since we'll all be meeting Wednesday morning. I look forward to seeing everyone next week 22:05:32 #endmeeting