16:01:07 #startmeeting openstack_ansible_meeting 16:01:07 Meeting started Tue Apr 17 16:01:07 2018 UTC and is due to finish in 60 minutes. The chair is evrardjp. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:10 The meeting name has been set to 'openstack_ansible_meeting' 16:01:24 evrardjp reminded on the ML we need more help on the releases/bug handling/monitoring jobs. 16:01:28 sup 16:01:29 woops forgot the topic 16:01:33 o/ 16:01:38 o/ 16:01:38 #topic what happened since last meeting 16:01:41 evrardjp reminded on the ML we need more help on the releases/bug handling/monitoring jobs. 16:01:48 evrardjp reminded we have pending specs to review 16:01:56 evrardjp says he didn't ask for newton repo EOL yet 16:02:03 evrardjp reminds there is an etherpad for vancouver forum: https://etherpad.openstack.org/p/YVR-openstack-ansible-brainstorming 16:02:10 evrardjp and odyssey4me discussed the releases, as we now have issues with them 16:02:12 odyssey4me: I will look into it. Dumping on the rabbitmq containers vinterface shows no traffic reaching it, looks like a possible bridging issue between the br-mgmt and vinterface. Tons of other traffic from all other containers and the infra host itself. 16:02:46 I will let you a few minutes to process these 16:03:06 if you have any other news to add, please add it to the wiki page, as usual. 16:03:19 #topic bug triage 16:03:52 #link https://bugs.launchpad.net/openstack-ansible/+bug/1764470 16:03:53 Launchpad bug 1764470 in openstack-ansible "upgraded get-pip does not install pip in venvs in repo container" [Undecided,New] 16:04:57 the 17.0.0 in 17.0.1 is my mistake but it's not fixable, so people should check out next version 16:05:57 odyssey4me: do you think there is a place where we don't constraint properly? the first part of the bug seems that it's installing 10.0.0 16:06:36 it doesn't show the usage of -c either 16:06:42 maybe that an old bug 16:06:51 I think that's what we fixed 16:07:09 evrardjp that's likely the issue that was resolved in later patch to constrain get-pip properly? 16:07:17 yes 16:07:22 that's what I'd say 16:07:30 yep 16:07:37 odyssey4me: fixed it! I deployed the compute host in a VM (for testing), I needed to enable promiscuous mode on the vswitch. Thanks for the guidance. 16:07:45 Merged openstack/openstack-ansible stable/pike: Do not log passwords https://review.openstack.org/561883 16:07:46 Merged openstack/openstack-ansible stable/pike: Do not log passwords https://review.openstack.org/561896 16:07:47 Merged openstack/openstack-ansible stable/queens: Do not log passwords https://review.openstack.org/561880 16:07:49 Merged openstack/openstack-ansible stable/queens: Do not log passwords https://review.openstack.org/561879 16:08:05 I mark it as invalid for now 16:08:10 queens: https://review.openstack.org/#/c/558127/ should be in 17.0.1 and 17.0.2 16:08:30 o/ 16:08:55 o/ 16:08:59 but that looks like repo build 16:09:07 /tmp/openstack-venv-builder/venvs/keystone/bin/python 16:09:40 I think that's confusing and we don't see the task in question 16:10:23 ok let's continue 16:10:27 #link https://bugs.launchpad.net/openstack-ansible/+bug/1764016 16:10:28 Launchpad bug 1764016 in openstack-ansible "openstack-venv-builder Fails PIP not found" [Undecided,New] 16:11:32 do we have dual stack openstack-ansible? 16:11:51 maybe we should just drop python2, and deal with the fallout 16:12:07 I am tired of these python2 and python3 issues 16:12:30 evrardjp: you see the py3 thread on -dev? 16:12:35 or have multiple jobs 16:12:35 just this morning 16:12:41 yes I saw it 16:13:48 like throughout the whole stack? 16:13:56 just drop py2 everywhere 16:14:15 swift is py2 only 16:14:24 yeah there are a few i think 16:14:39 here I think the python thing is a red-herring though 16:14:46 probably 16:15:54 ok found the problem 16:16:00 linked to previous bug 16:16:03 https://github.com/openstack/openstack-ansible-repo_build/blob/master/templates/venv-build-script.sh.j2#L86 16:16:07 we don't constraint there 16:16:30 or I missed something in the PIP_INSTALL_OPTIONS 16:17:23 anyone to confirm there? 16:17:24 ok, I'll check into that constraint - it might be right 16:17:35 I've commented in the bug 16:17:39 ok thanks. 16:17:59 We might need to step into pip10 sooner rather than later, and deal with all we're gonna see. 16:18:15 but right now I think it's best to constraint if a constraint is required. 16:19:07 I do not see your comment odyssey4me did you comment on the previous bug? 16:19:10 yeah, I'll see if I can replicate the issue - py3 issues have come up quite a bit lately 16:19:29 yes, I commented on the previous bug - these look similar though so I'll look into them both 16:19:38 ok 16:19:43 can you assign them both to you? 16:19:54 ++ about py3 only (hopefully) 16:19:56 ja, will do that now 16:19:59 it looks like it's an easy fix to add constraints there, and we can bump to pip10 when we're ready 16:20:03 cool 16:20:06 thanks 16:20:10 next 16:20:12 #link https://bugs.launchpad.net/openstack-ansible/+bug/1763422 16:20:12 Launchpad bug 1763422 in openstack-ansible "config_template is not fully python3 compatible" [Undecided,New] 16:20:37 andymccr: you started to work on this, right? 16:21:00 evrardjp: si si, i believe it is though 16:21:12 but that bug says otherwise :) 16:21:15 leave that with me and i'll try recreate 16:21:25 https://review.openstack.org/#/c/561900/ 16:21:28 :) 16:21:35 yeah i need to fix that up 16:21:41 but the py3 test is currently passing 16:21:45 whcih could just mean we are missing some test case 16:21:58 probably missing a lot of those 16:22:09 but with a separate repo it would be hopefully easier to deal with 16:22:13 and faster to iterate 16:22:59 I assign this bug to you andymccr 16:23:03 yeah please 16:23:23 I am not marking it as confirmed, as the job will help us discover if it's true or false, and if extra coverage is necessary 16:23:41 importance medium, okay for you? 16:23:48 for everyone I mean 16:23:58 sure 16:24:01 ok next 16:24:07 yes we have a worksround, it's ok 16:24:15 #link https://bugs.launchpad.net/openstack-ansible/+bug/1763361 16:24:16 Launchpad bug 1763361 in puppet-sahara "Sahara: add support for the new [trustee] configuration section" [Undecided,New] - Assigned to zhongshengping (chdzsp) 16:24:23 that's me o/ 16:24:44 woot tosky ! 16:24:51 thanks for being there for your bug 16:24:52 (bug both for puppet-sahara and openstack-ansible_os_sahara 16:25:33 yeah, seems sensible - should be a pretty straightforward patch as the same mechanism is used for neutron/nova IIRC 16:25:47 I think that I put all the details in the bug, tl;dr we changed a bit the configuration, and while the old one works with some magic code, it's time to change 16:25:49 tosky: is that only for Rocky? 16:26:05 evrardjp: technically it could be backported to Queens, but not a strong requirement 16:26:05 or does Queens will get this backported? 16:26:28 the new configuration works with Queens too, so it's up to you 16:26:29 nope but if it's a better practice, more future proof, let's do it 16:26:40 thanks for the heads up 16:26:48 I'd like to mark this as confirmed 16:27:08 that's what I wanted to ask in fact - I see that it's assigned already, but it was not marked yet 16:27:54 I've never seen a patch from the person who self--assigned it. 16:28:04 odyssey4me: https://review.openstack.org/#/c/558378/1/templates/sahara.conf.j2 16:28:18 you reviewed it though :p 16:28:38 I am fine with that. 16:28:43 ah ok :p clearly I need more coffee 16:29:00 these are morning patches, before you have your second one. :D 16:29:16 anyway thanks tosky it's really nice to have your feedback 16:29:33 anyone wants to pick this one up? 16:29:45 I will keep an eye on the patch - we receive notifications for new patches on openstak-ansible-os_sahara on #openstack-sahara 16:29:55 oh great! 16:30:02 :) 16:30:13 let's talk about that after the bug triage then 16:30:55 next 16:31:06 #link https://bugs.launchpad.net/openstack-ansible/+bug/1763336 16:31:07 Launchpad bug 1763336 in openstack-ansible "stale tempest config remains in the workspace after re-running os_tempest role" [Undecided,New] 16:31:25 I haven't got the chance to try that on my machine 16:32:00 Andy Smith proposed openstack/openstack-ansible-tests master: (WIP) Update to use oslo.messaging services for RPC and Notify https://review.openstack.org/559127 16:32:02 but it looks like the task https://github.com/openstack/openstack-ansible-os_tempest/blob/540a6be329d78df6a195bf1749364c6f514cefe9/tasks/tempest_post_install.yml#L49 16:32:03 oh wow, nice find there jrosser 16:32:11 should maybe happen after https://github.com/openstack/openstack-ansible-os_tempest/blob/540a6be329d78df6a195bf1749364c6f514cefe9/tasks/tempest_post_install.yml#L61 16:32:19 and yes good find 16:32:45 yeah i ws wtf for a bit till i found that 16:33:47 I don't know in which order does those file resolve though? 16:34:06 yeah, I can't recall - it's been a while since I've read the tempest docs 16:34:10 things may have changed 16:34:11 probably worth simplifying and putting them on the same file 16:34:28 making sure these pesky errors don't show up :) 16:34:54 confirmed high/ 16:34:57 ? 16:35:18 yep 16:35:29 anyone up for it? 16:35:55 ok let's continue then 16:35:58 #link https://bugs.launchpad.net/openstack-ansible/+bug/1763058 16:35:59 Launchpad bug 1763058 in openstack-ansible "openstack_user_config.yml.example says log_hosts is required" [Undecided,New] 16:36:21 fair 16:36:25 confirmed and low 16:36:31 and low-hanging-fruit 16:36:36 fine for everyone? 16:36:46 yep 16:36:57 good thing for a first time contributor to hack on :> 16:37:11 yup :) 16:37:28 low-hanging-fruit are the deal people, enjoy! :) 16:37:32 ok moving on 16:37:38 #link https://bugs.launchpad.net/openstack-ansible/+bug/1762742 16:37:39 Launchpad bug 1762742 in openstack-ansible "Some delegated tasks are failing when deployer is node part of a deployment (eg. controller node)" [Undecided,New] 16:38:46 should we consider this completely done now? 16:39:11 I think that's done, yes. 16:39:13 looks solved to me 16:39:21 are we waiting for a pike release? 16:39:26 nope 16:39:34 last week everything was released 16:39:53 or did I miss that patch for inclusion before release? 16:40:21 ah https://review.openstack.org/#/c/560920/ didn't make it into the release 16:41:04 yep, it looks like https://review.openstack.org/#/c/561167/ was missed 16:42:23 want me to re-issue a release? 16:43:21 I am fine with that, or we wait till EOM 16:43:45 well, the next sha bump is about to merge - I guess you could do a release immediately from that 16:44:04 pike is still what a lot of people are using, so perhaps that's a good idea 16:44:06 up to you people :) 16:44:13 ok 16:44:39 will do an extra release for it 16:44:50 next 16:44:53 #link https://bugs.launchpad.net/openstack-ansible/+bug/1758144 16:44:54 Launchpad bug 1758144 in openstack-ansible "resolv.conf in containers set too late" [Undecided,New] - Assigned to Logan V (loganv) 16:45:54 I will let logan work on this one more week 16:46:11 unless someone else wants to help right now :) 16:46:21 next 16:46:23 #link https://bugs.launchpad.net/openstack-ansible/+bug/1755821 16:46:24 Launchpad bug 1755821 in openstack-ansible "config_template fails to parse template if it contains a comment with leading spaces" [High,New] - Assigned to Kevin Carter (kevin-carter) 16:46:39 same I will leave it cloudnull for now 16:46:53 and last one is for cloudnull too: 16:46:56 #link https://bugs.launchpad.net/openstack-ansible/+bug/1743032 16:46:57 Launchpad bug 1743032 in openstack-ansible "Galera cluster maintenance in OpenStack-Ansible" [Undecided,New] - Assigned to Kevin Carter (kevin-carter) 16:47:15 let's keep them one more week, and then unassign next week 16:47:32 that's all for bug triage 16:47:40 #topic open discussion 16:47:49 anything to add? 16:48:49 just a curiosity: I was checking the jobs executed for openstack-ansible-os_sahara 16:49:07 I see that tempest is executed, but it's not clear which tests were executed 16:49:48 odyssey4me: COA exam is moving to pike soon 16:49:57 did I miss them, or otherwise, would it be possible to collect them too? iirc there are few useful roles in the tempest repository 16:50:20 tosky: generally each role defines its own whitelist 16:50:40 https://github.com/openstack/openstack-ansible-os_sahara/blob/master/tests/os_sahara-overrides.yml#L24 16:51:03 spotz: good 16:51:14 spotz: queens would be even better! :) 16:51:40 is there anything we can do to help making it move to P or Q ? 16:51:57 tosky: any increased coverage is welcomed 16:52:20 evrardjp: just to be clear, I'm talking about the testr* files here http://logs.openstack.org/85/560985/6/check/sahara-tests-tempest/8794f37/ 16:52:30 which are not visible for example in http://logs.openstack.org/78/558378/1/check/openstack-ansible-functional-ubuntu-xenial/cedddae/ 16:53:01 http://logs.openstack.org/78/558378/1/check/openstack-ansible-functional-ubuntu-xenial/cedddae/logs/openstack/infra1/ 16:53:11 oh, ok 16:53:14 thanks! 16:53:21 I didn't dig too deep 16:53:36 the real location depends on where tempest is installed (so check the inventory to find it) 16:53:47 ok I think we are good for today's meeting 16:53:54 thanks everyone! 16:54:04 thanks evrardjp 16:54:10 #endmeeting