16:01:26 #startmeeting openstack_ansible_meeting 16:01:27 Meeting started Tue Aug 8 16:01:26 2017 UTC and is due to finish in 60 minutes. The chair is evrardjp. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:31 The meeting name has been set to 'openstack_ansible_meeting' 16:01:38 lvdombrkr: there wouldn't be thats why the mapping should point to the interface that maps into the container - e.g. brex3 maps to eth12 on the container so it creates that link there. 16:01:52 hi 16:01:59 yo 16:01:59 o/ 16:02:05 woot 16:02:11 We have quorum to start the meeting. 16:02:17 #topic this week's bugs 16:02:23 o/ 16:02:28 #link https://bugs.launchpad.net/openstack-ansible/+bug/1709291 16:02:28 Launchpad bug 1709291 in openstack-ansible "rsyslog_server: /var/log/log-storage is not synchronized" [Undecided,New] 16:02:41 Melissa Kam proposed openstack/openstack-ansible-os_tempest stable/ocata: Support installing plugins from packages https://review.openstack.org/491842 16:03:06 I think this is worth investigating, because it looks we are missing logs. 16:03:24 Has anyone an instance available to verify this? 16:03:40 not sure I'm actually here, at airport 16:03:48 andymccr: its not clear for me... 16:04:27 and wifi only works for 30 min, so if we want to talk about the newton utf8 config template thing... 16:04:49 o/ 16:05:14 prometheanfire: in the middle of bug triage 16:05:35 thanks asettle :) 16:05:38 asettle: which is why I bring it up :P 16:05:54 prometheanfire: got a link? :) 16:06:01 The next bug is a docs bug, which I've just looked at 16:06:43 let me find it 16:07:05 sure, https://bugs.launchpad.net/bugs/1705469 16:07:05 Launchpad bug 1705469 in openstack-ansible "newton config-template change breaks previously working configuration" [High,Confirmed] - Assigned to Matthew Thode (prometheanfire) 16:07:05 confirmed medium 16:07:06 so 16:07:09 next one 16:07:20 #link https://bugs.launchpad.net/openstack-ansible/+bug/1709068 16:07:20 Launchpad bug 1709068 in openstack-ansible "Docs: Available overrides out-of-date" [Undecided,New] 16:07:31 You have an opinion asettle? 16:07:53 low, confirmed 16:07:53 We can add the suggested line easily enough I think, not sure about the proposal bot though 16:08:02 Proposal bot is too far, agree with spotz 16:08:02 prometheanfire: we can have your bug triaged now 16:08:15 spotz: do you want to take that? 16:08:27 asettle: ok with your triage, let me update the bug 16:08:31 my suggesiton for the docs bug is to remove the list of overrides and just have a paragraph informing people how ot find it 16:08:32 evrardjp: already done 16:08:39 :) 16:08:40 cool 16:08:44 next 16:08:44 proposal bot is not something infra wants more of 16:08:50 bah, grab the bug then asettle:) 16:08:56 odyssey4me: good to know. 16:09:09 so prometheanfire 16:09:12 #link https://bugs.launchpad.net/bugs/1705469 16:09:12 Launchpad bug 1705469 in openstack-ansible "newton config-template change breaks previously working configuration" [High,Confirmed] - Assigned to Matthew Thode (prometheanfire) 16:09:19 right 16:09:20 this is already triaged 16:09:43 oh? guess it is, sorry, been out for the last week 16:09:59 prometheanfire in our early discussions when this was discovered, and I'm pretty sure in the bug log too, it was requested that tests be added to confirm now and forever that the config template is doing the right things 16:10:17 yes. 16:10:28 this still hasn't been done, and another issue was raised by hughsaunders regarding other values 16:10:37 so the issue is not solved, nor are there tests 16:10:40 so the bug is still opened. 16:11:10 ok, will work on it when I get back then 16:11:10 so either we revert the last two patches to get it all back to the previous state where all was well, except unicode things - or we move forward with a complete fix 16:11:39 For context of last week's bug triage: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2017/openstack_ansible_meeting.2017-08-01-16.00.log.html#l-144 16:12:00 the unicode things are fixed in ocata onwards, so I do kinda feel like trying to change newton now is perhaps a little too far to step 16:12:04 I'm inclined to back out, but I'm fairly sure it breaks on 16.04 16:12:14 odyssey4me: ya, agreed 16:12:47 so my suggestion is to back out of https://github.com/openstack/openstack-ansible-plugins/commit/782460644902f391cadb83e61c882956c0e569ee and https://github.com/openstack/openstack-ansible-plugins/commit/64353d8ff0d0c2d803aba1f2af607e7ffb13c662 16:13:00 then if there is a problem, a complete fix can be added 16:13:25 yep 16:13:32 I am fine with the revert if the revert doesn't break more things. 16:13:34 I'll put that in the bug and submit reviews 16:13:56 evrardjp: it's more of a situation where you do one step forward, one back 16:14:03 fix one thing, break something else 16:14:07 yeah. Fine for the revert. 16:14:18 and for the inclusion of tests in upper branches. 16:14:19 thanks prometheanfire 16:14:22 thanks! 16:14:36 next 16:14:38 #link https://bugs.launchpad.net/openstack-ansible/+bug/1708360 16:14:38 Launchpad bug 1708360 in openstack-ansible "Openstack-Ansible Install Fails with OVS" [Undecided,New] 16:14:56 ovs... 16:15:41 do we have resources to check that? 16:15:51 someone here? 16:16:18 else I'd leave the bug as is. Let me mark on the bug we don't have enough resources to triage this bug for now. 16:17:29 yeah id love if somebody was interested and had time to fix up our ovs bits 16:17:33 and add reasonable gates too 16:18:09 I think we let enough time for ppl to step up on this bug, if nobody answers, it will have to wait. 16:18:21 this is true 16:18:26 we did the install with ovs 16:18:27 it fails 16:18:32 but we had to manually work around it 16:19:08 exactly the same bug admin0 ? 16:19:16 Did you document it for others? 16:19:23 Or a patch? 16:19:57 Else we don't have any code we can show to this bug reporter... 16:20:04 solution was to not use neutron nodes in controllers 16:20:10 but move to hypervisors/compute direct 16:20:15 i will update that 16:20:53 next 16:20:55 #link https://bugs.launchpad.net/openstack-ansible/+bug/1707967 16:20:55 Launchpad bug 1707967 in openstack-ansible "ANSIBLE_ROLE_FETCH_MODE = 'git-clone' should not source openstack-ansible.rc" [Undecided,New] 16:21:03 Fair enough 16:21:07 Confirmed 16:21:14 What's the level though? 16:21:22 It can be fairly disruptive. 16:21:40 if loading plugins for old versions when bootstrapping new version :) 16:21:55 high? 16:22:07 I'm trying to do a quick AIO install (stable/newton) but it keeps failing on the galera-install.yml trying to do the pip_install... Any thoughts on what I can do to troubleshoot? 16:22:08 yeah, but also low hanging fruit 16:22:14 yeah. 16:22:15 I'm hoping that someone will pick that up 16:22:49 new contributors? ^^^^^^^^^^^^^^^^^^^^^^^^^^^ 16:22:56 easy to pick ! 16:23:06 I'd like to have the slack @here thing. 16:23:15 or a buzz warning. 16:23:33 anyway 16:23:33 next 16:23:35 #link https://bugs.launchpad.net/openstack-ansible/+bug/1707963 16:23:35 Launchpad bug 1707963 in openstack-ansible "Keystone role needs an LDAP scenario test" [Wishlist,New] 16:23:54 is there anyone wanting this very badly? :p 16:24:06 images.linuxcontainers.org is tossing out 400's randomly 16:24:08 :/ 16:24:10 most are i tink 16:24:12 think 16:24:20 because very comany that does private cloud, they have an existing ldap 16:24:29 if there can be a test/ldap .. its good 16:24:32 that doesn't mean they integrate ldap. 16:24:34 :p 16:24:42 but let's mark it confirmed and wishlist 16:24:56 sorry - forgot it was meeting time :/ 16:24:57 if those companies want it, they will contribute or make it higher prio. 16:25:08 mhayden if you change the patch to cater for a list, then put both the us and uk sources in the list? ;) 16:25:08 strattao: don't worry. 16:25:12 Merged openstack/openstack-ansible-openstack_hosts master: Rename package install task https://review.openstack.org/491607 16:25:20 odyssey4me: considering that 16:25:22 ok confirmed wishlist. 16:25:35 next 16:25:38 #link https://bugs.launchpad.net/openstack-ansible/+bug/1707142 16:25:38 Launchpad bug 1707142 in openstack-ansible "pip_install: "The following packages cannot be authenticated"" [Undecided,New] 16:28:43 The problem is the reproducibility of this bug. 16:29:05 well, it looks weird - but I haven't had a chance to try to reproduce 16:29:12 If there is someone with that image, shoot. 16:29:19 someone has to feedback 16:29:24 yeah. 16:29:48 start a virtualbox vagrant VM with box "geerlingguy/ubuntu1604 — somone forgot to pull in right pacakges in the box i guess 16:30:17 i guess the ca-certificates package is missing 16:30:38 that could be it, but then they have a dependency problem with their package if ca-certificates isn't included. 16:30:40 :p 16:30:49 not our fault in that case. 16:30:51 Anyway. 16:31:04 next 16:31:06 #link https://bugs.launchpad.net/openstack-ansible/+bug/1706446 16:31:06 Launchpad bug 1706446 in openstack-ansible "Keystone bootstrap failed when using LDAP" [Undecided,New] 16:31:14 \o/ LDAP finally :D 16:31:18 needs ldap testing to verify 16:31:38 if someone can put together a repeatable ldap test, we'll be able to verify 16:32:04 looking at last week's conversation, nothing needs updating. 16:32:13 next 16:32:16 #link https://bugs.launchpad.net/openstack-ansible/+bug/1705420 16:32:16 Launchpad bug 1705420 in openstack-ansible "os-neutron-install.yml fails during LBaaS v2 deployment" [Undecided,New] 16:32:39 what do you think of this one? 16:33:02 is there any neutron db migration expert out there? 16:34:07 ok let's continue our path then... 16:34:10 #link https://bugs.launchpad.net/openstack-ansible/+bug/1705374 16:34:10 Launchpad bug 1705374 in openstack-ansible "realpath() of /etc/uwsgi/nova-placement-uwsgi.ini failed" [Undecided,New] 16:34:18 uwsgi! andymccr? 16:34:19 :D 16:34:48 hmm 16:34:50 not sure how that could happen 16:34:58 the first task in the uwsgi thing is literally to setup /etc/uwsgi dir :P 16:36:04 hmm 16:36:10 weird "could not find the requested service nova-compute" 16:36:44 it looks like a hurricane went through the deploy 16:36:45 hmm 16:36:56 quick q, is there a way to force repo nodes to rebuild/resync everything recursively they should contain? I tried adding a 16.04 node as a repo server and it won't rsync the data across to the other repo nodes 16:36:59 we could move uwsgi bits before the service setup that may be part of it 16:38:02 It sounds ok, but I am thinking at the impact for upgrades. 16:40:33 that error still seems weird 16:40:38 it wont have an upgrade impact but assign it to me 16:40:42 im basically fixing it up as part of the wsgi bits anyway 16:40:59 ok thanks. 16:41:16 next 16:41:26 ... is pending an answer 16:41:38 next 16:41:40 https://bugs.launchpad.net/openstack-ansible/+bug/1703616 16:41:40 Launchpad bug 1703616 in openstack-ansible "Error " This is not a recognized Fernet token" when installing os-trove" [Undecided,New] 16:41:51 do we have someone to help triage this trove bug? 16:42:09 if not let's move on 16:42:16 5 16:42:18 4 16:42:19 3 16:42:21 2 16:42:23 1 16:42:26 next. 16:42:30 https://bugs.launchpad.net/openstack-ansible/+bug/1698871 16:42:30 Launchpad bug 1698871 in openstack-ansible "[master] [os_gnocchi]Gnocchi role tests failing on installing pip packages" [Undecided,New] - Assigned to Miguel Alejandro Cantu (miguel-cantu) 16:42:49 I think we should remove the assignee 16:43:08 Has someone worked on gnocchi recently ? 16:43:25 stevelle was our expert:( 16:43:40 alex was taking over the work, but... 16:43:56 ok let's leave it that way 16:44:09 next 16:44:11 and last 16:44:15 #link https://bugs.launchpad.net/openstack-ansible/+bug/1697782 16:44:15 Launchpad bug 1697782 in openstack-ansible "Mounting of ceph-backed cinder volumes is broken after Ocata upgrade" [Undecided,New] - Assigned to Logan V (loganv) 16:44:21 logan-: do you have an update there? 16:44:37 I think you were trying to verify this on your env right? 16:45:16 Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Download LXC image with async via aria2 https://review.openstack.org/491798 16:47:08 ok let's close it up for today 16:47:12 thanks everyone! 16:47:42 No additional bug someone wants to raise? 16:47:55 (I often forgot to ask that...sorry!) 16:48:10 End meeting in 10 16:48:18 5 16:48:19 4 16:48:19 3 16:48:20 2 16:48:22 1 16:48:24 #endmeeting