16:00:27 #startmeeting openstack_ansible_meeting 16:00:28 Meeting started Tue Jul 17 16:00:27 2018 UTC and is due to finish in 60 minutes. The chair is mgariepy. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:32 The meeting name has been set to 'openstack_ansible_meeting' 16:00:33 #topic rollcall 16:00:40 Could someone review https://review.openstack.org/#/c/581459/ to get it finally merged. While someone hasn't made any changes 16:00:42 i'm here ! 16:00:54 O/ sorta here 16:01:53 o/ 16:02:14 ping cloudnull, DimGR, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, scarlisle, luckyinva, ntt, javeriak, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, bgmccollum, darrenc, JRobinson__, colinmcnamara, thorst, adreznec, eil397, 16:02:15 qwang,nishpatwa_, cathrichardson, drifterza, hwoarang, cshen, ullbeking, Tahvok, mnaser, nicolasbock 16:02:18 hrm 16:02:45 * tosky is around 16:02:52 o/ 16:03:21 if I can help, I'm here too :) 16:03:41 team team team 16:03:43 :) 16:03:56 o/ 16:04:04 yeah more ppl :) 16:04:32 another minute or 2 16:06:15 #topic last week highlights 16:07:01 odyssey4me: Distro builds for Xenial are not testing Rocky, and that's making things a little complicated. Should we try and recruit more help for Bionic and port it back to Queens, then eliminate Xenial from Rocky? ref: http://eavesdrop.openstack.org/irclogs/%23openstack-ansible/%23openstack-ansible.2018-06-29.log.html#t2018-06-29T13:34:40 16:07:33 odyssey4me, are you around to talk about it ? 16:08:07 I think this was last week 16:08:09 mgariepy: added, ty 16:08:28 hmm. 16:08:47 maybe would make sense since it's from 18 days ago. 16:09:08 mgariepy: I've updated my points 16:09:30 David Wilde proposed openstack/openstack-ansible-tests master: Update get-pip to version 3.3 https://review.openstack.org/583298 16:09:33 evrardjp: working on independant inventories as stalled until testing is more stable for roles and the integrated. Focus is on landing M3. 16:09:49 o/ 16:10:09 hwoarang: 's point on the agenda is still valid 16:10:16 yep i updated it today 16:10:30 hwoarang: Finished aio_distro_basekit scenario - Please review https://review.openstack.org/#/c/579770/ 16:11:05 hwoarang: Leap 15 is blocked due to lxc, sudo and mariadb bugs. Will work on that after holidays (and once aio_distro_basekit is in). 16:11:20 anything to add for those ? 16:11:51 no that would be all 16:11:54 evrardjp: M3 is next week. https://review.openstack.org/#/c/574006/ is not solved yet (more patches needed for keystone and cinder-backup). The other branches have been released. 16:11:55 thank you 16:12:16 * mgariepy is in training for hosting meeting 16:12:20 have some question - do we have some plans about implementing support of ubuntu 18.04 for rocky? 16:13:12 Dmitriy, can we talk about this during the open discussion ? 16:13:34 #topic bug triage 16:13:41 tosky: https://review.openstack.org/582331 just passed, no inventory changes ;) 16:13:45 mgariepy: oh, sorry. it seems, that I've missed topic change... 16:13:58 #link https://bugs.launchpad.net/openstack-ansible/+bug/1782195 16:13:58 Launchpad bug 1782195 in openstack-ansible "neutron l3 agent fails to spawn HA router" [Undecided,New] 16:15:00 odyssey4me: oooh, great, thanks 16:15:08 anyone is running 17.0.6 with ha router that can confirm it's broken on all the distro ? 16:15:29 that looks valid 16:15:42 cloudnull: has an env that could be used, am I right cloudnull ? 16:16:14 mgariepy: in the meantime it looks valid, and I'd say that's quite an important bug, so high 16:16:24 ok. 16:16:27 agree 16:16:29 but I'd like to wait for confirmation before marking confirmed 16:16:39 mgariepy: what do you think? 16:17:06 seems like path is missing. i'm not familiar with ha router 16:17:24 yeah it seems it's that 16:17:26 anyone runnign 17.X with ha router on other distro ? 16:18:06 we can ask the user to run cat /proc/$PID/env 16:18:16 jenglisch_, ^^ 16:18:29 for which pid ? 16:18:41 for the l3 agent, which i believe it would tell us the path 16:18:52 path as in $PATH 16:19:40 however. 16:19:41 https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/rootwrap.conf.j2 16:19:48 i don't think thats interesting, since rootwrap will wrap it 16:19:49 http://paste.openstack.org/show/726132/ 16:20:17 ok in progress. can you update the bug with the info jenglisch_ ? 16:20:30 we can chat about debugging more after meeting :) 16:20:33 next one 16:20:41 #link https://bugs.launchpad.net/openstack-ansible/+bug/1782147 16:20:41 Launchpad bug 1782147 in openstack-ansible "Deploy Sahara using unversioned endpoints" [Undecided,New] 16:20:59 I opened that bug 16:21:07 cloudnull: https://review.openstack.org/583298 16:21:40 I think the description should say it all, and I was planning to abuse the gates with a blind patch that changes the default values for entry points for sahara 16:21:49 but first I needed working gates for sahara :) 16:22:28 ok so medium, triagated and assigned to tosky ? 16:22:36 are you ok with that ? 16:22:37 I would say yes 16:22:51 and if I hit some issues, I will nag here as usual 16:23:03 evrardjp: https://review.openstack.org/583276 16:23:18 I trust you tosky 16:23:23 perfect thanks tosky :) 16:23:34 #link https://bugs.launchpad.net/openstack-ansible/+bug/1782051 16:23:34 Launchpad bug 1782051 in openstack-ansible "Install PIP Failed " [Undecided,New] 16:24:35 the last comment says that it is a duplicate of https://bugs.launchpad.net/openstack-ansible/+bug/1779534 16:24:35 Launchpad bug 1779534 in openstack-ansible "pip install offline fails with new version of get-pip.py" [High,Fix released] - Assigned to Jesse Pretorius (jesse-pretorius) 16:24:38 since the ticket has a link to another ticket, i would mark as ducplicate 16:24:45 I think that one was oppened by tux_ 16:24:53 cloudnull was working with him yesterday 16:24:58 he can give more details 16:25:14 we can do an actual mark as duplicate 16:26:04 yeah 16:26:08 done. 16:26:16 * mgariepy was looking for it 16:26:27 #link https://bugs.launchpad.net/openstack-ansible/+bug/1782012 16:26:28 Launchpad bug 1782012 in openstack-ansible "openstack ansible with ceph doesnot generate keyring file" [Undecided,New] 16:26:36 Victor Palma proposed openstack/openstack-ansible-ops master: adding kolide fleet https://review.openstack.org/583046 16:27:26 i think the user was running this play with tags or something 16:27:34 i just deployed 17.0.6 with ceph and it was fine 16:27:36 or the ceph-playbook failed ? 16:27:38 I doubt device can be used like that 16:27:46 it seems a support request 16:28:04 I don't see how it was run 16:28:18 (at the first run) 16:29:10 incomplete, then and ask for more info ? 16:30:09 I'd say that. 16:30:31 mnaser: 's point is valid: was that run with tags? 16:31:39 added a comments 16:31:49 next one. 16:31:54 cool thanks mgariepy 16:31:57 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781619 16:31:57 Launchpad bug 1781619 in openstack-ansible "Neutron role does not enable osprofiler" [Undecided,New] 16:32:28 that's indeed a gap 16:32:47 cloudnull: is neutron ready for osprofiler ? 16:33:00 can we just edit the paste file? 16:33:14 confirmed. medium 16:33:50 yup 16:33:57 or low 16:34:04 low ;) 16:34:29 should we assing someone ? 16:34:33 it's kinda a feature request but at the same time we are already kinda providing the feature so... low makes sense to me 16:34:43 o.O 16:35:00 d34dh0r53, interested ? 16:35:05 mgariepy: why not :) 16:35:08 anyone interested? 16:35:12 jrosser: maybe? 16:35:14 yeah, I'll look at it 16:35:18 thanks a lot 16:35:19 d34dh0r53: should be easy 16:35:24 low-hanging-fruit : ) 16:35:25 dont mind :) 16:35:37 https://github.com/openstack/neutron/blob/master/etc/api-paste.ini i guess upstream it doesn't exist though 16:35:47 haha great so you liberate d34dh0r53 for more bionic things! :p 16:36:07 does this file get copied in from upstream? 16:36:07 mnaser: yes I checked upstream first -- that's why I asked that question 16:36:07 can you take it please jrosser ? 16:36:20 next. 16:36:22 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781321 16:36:22 Launchpad bug 1781321 in openstack-ansible "ceph-ansible error No first item, sequence was empty" [Undecided,New] 16:36:28 it not only needs to be changed in our repo, but also upstream, and making sure it works : ) 16:36:36 we can confirm it works 16:37:59 mgariepy: well it seems a configuration issue -- maybe we should mark it incomplete, and wait for user feedback to know if this is fixed/solved already? 16:38:01 last comment says now it work . 16:38:10 yeah 16:38:24 yep 16:38:49 so invalid then 16:39:06 I'd say incomplete 16:39:11 but up to you to decide 16:39:15 support request 16:39:31 ok then invalid 16:39:57 hmm, ceph shouldn't require br-storage, but that may not have been the issue 16:40:13 depends on the config i would guess ? 16:40:28 yeah, config issue I think 16:40:42 so, next ? 16:40:46 yep 16:40:47 yes I'd say so 16:40:56 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781189 16:40:56 Launchpad bug 1781189 in openstack-ansible "Task “Get Modern PIP” doesn't honour HTTP_PROXY" [Undecided,New] 16:42:39 anyone can confirm ? 16:43:34 * jrosser looking.... 16:43:55 By default, if an environment variable _proxy is set on the target host, requests will be sent through that proxy. This behaviour can be overridden by setting a variable for this task (see setting the environment), or by using the use_proxy option. 16:43:56 that's typically a jrosser thing :p 16:44:06 HTTP_PROXY vs http_proxy ? 16:44:11 well that worked for us with aio+jenkins today 16:44:20 do you use HTTP_PROXY or http_proxy 16:45:06 looks like lowercase 16:45:23 but thats not totally scientific, just looking at the jenkins log 16:45:43 invalid, support request, ask the user to use lowercase ? 16:45:53 or incomplete 16:45:57 i'd ask to see if lowercase works 16:47:01 ok. incomplete + comment. 16:47:28 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781166 16:47:28 Launchpad bug 1781166 in openstack-ansible "Firewall as a service v2 is not compatible with LXB" [Undecided,New] 16:49:17 maybe johnsom can tell us more about that ^ 16:49:29 What did I do now? 16:49:30 he will tell the user to move to octavia though. 16:49:47 johnsom: you didn't do anything wrong, no need for a lawyer! 16:49:49 :) 16:49:52 fwaas and linux bridges? 16:49:56 xgerman_, are you arround ? 16:50:02 what do you think of https://bugs.launchpad.net/openstack-ansible/+bug/1781166 johnsom ? 16:50:02 Launchpad bug 1781166 in openstack-ansible "Firewall as a service v2 is not compatible with LXB" [Undecided,New] 16:50:03 https://review.openstack.org/#/c/548195/ 16:50:07 this isnt lbaas stuff 16:50:07 yes 16:50:11 Yeah, fwaas is not my area. JP looking 16:50:15 mnaser: OMG 16:50:23 yeah I am tired. 16:50:26 sorry johnsom ! 16:50:30 is fwaasv2 compatible with lxb ? 16:50:31 I read LBaaS v2 16:50:39 no 16:50:46 https://bugs.launchpad.net/openstack-ansible/+bug/1781166 16:50:48 mgariepy: no — it only works with ovs 16:50:49 Ha, NP. Octavia does support linux bridges BTW, OVS and others too. 16:50:59 shameless plug 16:51:10 johnsom: that's not shameless when it's educational! 16:51:11 ok confirm medium 16:51:13 or better the l2 part only works with ovs — l3 doesn;t care 16:51:21 and need docs and patches i guess. 16:51:44 sorry for the spam in the channel -- my bad. 16:51:58 can't confirm 16:52:09 why ? 16:52:13 xgerman_, just did. 16:52:35 l2 part is not working with lxb. 16:52:56 neutron made us believe they are deprecating lxb ;-) 16:54:31 mgariepy: I meant I can't confirm myself :p 16:54:34 sorry 16:54:38 I am really tired 16:54:48 Luigi Toscano proposed openstack/openstack-ansible-os_sahara master: Deploy Sahara with unversioned endpoints https://review.openstack.org/583310 16:54:56 i would add a needs docs and some patch to make sure it's not enabled with lxb 16:55:06 confirm, medium. 16:55:14 lgtm for the classification 16:55:17 anyone would like to work on this? 16:55:58 next? 16:56:29 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781148 16:56:29 Launchpad bug 1781148 in openstack-ansible "vpnaas agent is removed since Queens and configuration changed" [Undecided,New] 16:56:41 looks valid 16:56:44 medium? 16:57:19 yep, 16:57:22 lgtm. 16:57:31 ok next 16:57:42 #link https://bugs.launchpad.net/openstack-ansible/+bug/1781126 16:57:42 Launchpad bug 1781126 in openstack-ansible "ceph-ansible (libselinux-python) aren't installed" [Undecided,New] 16:58:16 that is weird, as this should be installed 16:58:26 in the container ? 16:58:46 http://git.openstack.org/cgit/openstack/openstack-ansible/tree/playbooks/ceph-install.yml#n88 16:59:02 for me we need more details. 16:59:14 so it's the interpreter that is broken 16:59:19 oh 16:59:25 mnaser, anything on that ? 16:59:27 maybe my patch to remove python3 fixed it 16:59:37 deployed you ceph cluster into container ? 17:00:09 incomplete and more ask for more info ? 17:00:18 The containers don’t have selinux 17:00:28 So it shouldn’t try to ask to install that 17:01:19 mnaser: you mean that task is not relevant when containerized? 17:01:31 I believe so. 17:01:35 It should be skipped. 17:01:49 we should add a conditional then 17:01:59 the taks is relevant, i thinkg the issue is that ansible don't have the bindings. 17:02:02 but anyway, it's not the problem of the user I'd say 17:02:10 it's a config change. 17:02:21 not a pkg install. 17:02:27 oh 17:02:36 you mean the necessary libs aren't symlinked into the venv? 17:03:29 i'm not sure what is happenning there 17:04:31 it could as well be a problem on the remote node... it's weird we don't see that in gates 17:04:32 when you create a file with ansible if your target has selinux enabled (or ansible think it has) it will want to manage se attrib. 17:04:36 or I haven't seen one 17:04:48 yeah 17:04:52 that's when it needs the selinux binding. 17:05:02 I have to go 17:05:09 i would have to go also. 17:05:13 thanks mgariepy for chairing the meeting1 17:05:21 !* 17:05:22 evrardjp: Error: "*" is not a valid command. 17:05:44 thank you! 17:05:53 thanks all of you. 17:06:19 #endmeeting