Tuesday, 2015-09-22

*** elo has joined #akanda00:07
*** mugu has quit IRC00:15
*** mugu has joined #akanda00:16
*** elo has quit IRC00:29
*** elo has joined #akanda00:48
*** elo has quit IRC01:18
*** elo has joined #akanda01:21
openstackgerritAdam Gandelman proposed stackforge/akanda-appliance: Introduces advanced service drivers to akanda-appliance  https://review.openstack.org/22611901:22
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Various fixes to get drivers functional (WIP)  https://review.openstack.org/22388401:23
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Router driver and required changes to instance manager (WIP)  https://review.openstack.org/21522701:23
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Base akanda driver  https://review.openstack.org/21522601:23
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Start of the LBAAS driver  https://review.openstack.org/22536901:23
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Driver factory  https://review.openstack.org/21522001:23
adam_gfwew. good night01:23
sarobthanks adam_g!01:24
*** elo has quit IRC02:00
*** elo has joined #akanda02:03
*** elo has quit IRC02:37
*** dhellmann has quit IRC02:40
*** dhellmann has joined #akanda02:41
*** elo has joined #akanda03:02
*** jordantardif has quit IRC03:30
*** elo has quit IRC03:34
*** elo has joined #akanda04:08
*** stanchan has joined #akanda04:42
*** smcginnis has quit IRC05:12
openstackgerritAdam Gandelman proposed stackforge/akanda-appliance: Introduces advanced service drivers to akanda-appliance  https://review.openstack.org/22611905:49
openstackgerritAdam Gandelman proposed stackforge/akanda-rug: Start of the LBAAS driver  https://review.openstack.org/22536905:51
*** elo has quit IRC06:06
*** elo has joined #akanda06:06
*** stanchan has quit IRC06:48
*** smcginnis has joined #akanda06:53
*** stanchan has joined #akanda06:59
*** elo has quit IRC07:13
*** elo has joined #akanda08:13
*** elo has quit IRC08:15
*** elo has joined #akanda09:15
*** elo has quit IRC09:17
*** elo has joined #akanda10:17
*** elo has quit IRC10:18
*** elo has joined #akanda11:19
*** elo has quit IRC11:20
*** elo has joined #akanda13:21
*** elo has quit IRC13:22
ryanpetrelloadam_g in your CI testing of akanda, are you all doing anything specific to set mtu?13:34
ryanpetrelloto avoid the issue I've been seeing?13:34
*** elo has joined #akanda14:03
*** elo has quit IRC14:08
*** jordantardif has joined #akanda15:23
*** puranamr has joined #akanda15:27
*** puranamr has quit IRC15:27
drwahllooks like the stable/kilo branch is failing tests15:46
drwahl>>  assert isinstance(resp, <module 'webob' from '/home/jenkins/jobs/akanda-rug-kilo-trusty-python-github-package/workspace/.tox/py27/local/lib/python2.7/site-packages/webob/__init__.pyc'>.exc.HTTPNotImplemented)15:47
drwahljust an example of one of the failures15:47
drwahlhere's the list of the failed tests: http://pastebin.com/Q2pUx0G915:49
ryanpetrelloyep, there's a bug filed: https://bugs.launchpad.net/bugs/149811415:49
openstackLaunchpad bug 1498114 in akanda kilo "stable/kilo py27 fails with python version conflict: Requirement.parse('oslo.i18n<=1.3.1,>=1.0.0'), set(['oslo.utils'])" [Undecided,Confirmed]15:49
*** stanchan has quit IRC15:54
*** puranamr has joined #akanda16:17
*** elo has joined #akanda16:20
jordantardifnoticed a bug last night in the rug that is leading to tons of stray ports16:26
jordantardifbasicly if nova 500's when trying to boot the vm16:26
jordantardifall the ports that were created before do not get cleaned up16:26
jordantardifand then the rug trys to boot another vm with new ports, and so on16:27
jordantardifbasicly if this fails https://github.com/stackforge/akanda-rug/blob/master/akanda/rug/api/nova.py#L100-L10716:27
jordantardifthe ports need to get cleaned up that were created before that16:27
jordantardifto reproduce just put bad credentials in the neutron section of nova.conf16:29
jordantardifi can file a bug if one doesn't exist for this already16:29
*** seanmwinn has joined #akanda16:30
ryanpetrelloyep, please file a bug16:32
ryanpetrelloand tell them about it in #akanda16:32
jordantardifi think im in #akanda ..16:33
jordantardif:)16:33
drwahlhey, #akanda. ^^16:33
drwahl:)16:33
ryanpetrelloblurgh16:33
ryanpetrellohttp://giant.gfycat.com/NiftyFarEnglishsetter.gif16:33
drwahlhttp://edge.alluremedia.com.au/m/k/2015/09/trump.gif16:34
adam_gryanpetrello, no, im not exactly  sure why that issue only affects some environments. it must be specific to the env under devstack16:44
*** adam_g has left #akanda16:45
*** adam_g has joined #akanda16:45
*** puranamr has quit IRC17:10
*** puranamr has joined #akanda17:31
*** puranamr has quit IRC17:33
*** puranamr has joined #akanda17:37
davidlenwelladam_g:  it looks like the api config thing is a feature that only exists in plus.. not FOSS .. http://nginx.org/en/docs/http/ngx_http_upstream_conf_module.html17:54
adam_gdavidlenwell, ah cool17:59
*** stanchan has joined #akanda18:01
*** stanchan has quit IRC18:01
davidlenwellAlso checked on the neatron cleanup thing we discussed.. currently octavia is trusting neutron to do the right thing when they set the status of an lb to DELETED18:01
davidlenwelldoes that answer your query? do you need me to do any more digging?18:01
davidlenwellI'll start .. but probably not finish until tonight .. working on a good pbase devstack config for having lbaasv2  ref architecture up18:02
davidlenwelladam_g:  here are some details about how to get health status out of nginx plus.. https://www.nginx.com/resources/admin-guide/load-balancer/#health_active18:04
adam_gdavidlenwell, i think it does, ill give it a shot18:04
davidlenwellon the trusting neutron to do the right thing .. I'm not sure we can trust it pre-M .. might require some testing .. if you look at the conversation in the other channel he was saying that the neutron-sync stuff they are using is new to <18:06
davidlenwellM18:06
*** puranamr has quit IRC18:09
*** stanchan has joined #akanda18:21
stanchanhello all18:28
stanchanso, just an update.  devstack + akanda worked with ubuntu 14.04.18:28
stanchango figure18:29
stanchanI'm going to try multinode now18:29
markmcclainstanchan: cool... so should be fairly straight forward since no Akanda bits run on the extra nodes18:36
*** puranamr has joined #akanda19:27
*** puranamr has quit IRC19:28
*** elo has quit IRC19:57
ryanpetrellomarkmcclain adam_g21:18
ryanpetrellow/ the new akanda-rug21:18
ryanpetrelloI'm spinning this up in vagrant w/ vxlan stuff and trying to test the akanda VM21:18
ryanpetrelloseems to work21:18
ryanpetrellobut I can't get tenant traffic to work from within the router VM21:18
ryanpetrelloit feels like a security group issue - does the akanda VM need a security group set?21:19
adam_gryanpetrello, it shouldnt, the router ports get port security turned off21:19
ryanpetrelloactually, my port in question is DOWN with vif_type binding_failed21:20
adam_gryanpetrello, but markmcclain was reporting something similar last week, and im about to need tenant traffic flowing correctly myself21:20
ryanpetrelloalright21:23
ryanpetrelloabout to call it a day on my end21:23
ryanpetrelloif you find anything interesting, let me know :)21:23
*** elo has joined #akanda21:43
*** jordantardif1 has joined #akanda21:54
*** jordantardif has quit IRC21:58
*** davidlenwell has quit IRC22:08
*** markmcclain has quit IRC22:08
*** davidlenwell has joined #akanda22:09
*** ChanServ sets mode: +o davidlenwell22:09
*** jordantardif has joined #akanda22:10
*** jordantardif1 has quit IRC22:14
*** markmcclain has joined #akanda22:16
*** puranamr has joined #akanda22:26
*** drwahl_ has quit IRC22:29
*** puranamr has quit IRC22:29
*** seanmwinn has quit IRC22:50
*** elo has quit IRC23:05
*** stanchan has quit IRC23:26
*** jordantardif has quit IRC23:39

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!