14:00:02 #startmeeting tripleo 14:00:03 Meeting started Tue Jan 19 14:00:02 2021 UTC and is due to finish in 60 minutes. The chair is marios. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:06 The meeting name has been set to 'tripleo' 14:00:16 #topic agenda 14:00:16 * Review last minutes & action items 14:00:16 * One off agenda items 14:00:16 * Bugs & Blueprints 14:00:16 * Projects releases or stable backports 14:00:18 * Specs 14:00:21 * open discussion 14:00:23 Anyone can use the #link, #action and #info commands, not just the moderatorǃ 14:00:26 Hello folks, who is around today? o/ 14:00:52 hello tripleo anyone here :) 14:01:02 hi 14:01:14 o/ 14:01:16 o/ 14:01:21 o/ 14:01:39 so lets get going and folks can catch up we have a few things in the etherpad 14:01:49 #topic review last meeting logs & action items 14:02:01 #link http://eavesdrop.openstack.org/meetings/tripleo/2021/tripleo.2021-01-05-14.00.html 14:02:06 o/ 14:02:08 checking for any AI from last week ... 14:02:33 o/ 14:02:38 «o/ 14:02:49 couple of the community ones had AI (bump rocky eol thread and the pip resolver) but we can discuss after 14:03:03 anything else from last meeting someone wants to hilight before we move on? 14:03:31 k moving onto 14:03:32 #topic one off agenda items 14:03:32 #link https://etherpad.openstack.org/p/tripleo-meeting-items 14:03:47 so let's take them in order 14:03:52 o/ 14:03:54 #info [Tengu] conditionnally replacing cURL by a python script for the http healthcheck? 14:03:58 #link https://review.opendev.org/c/openstack/tripleo-common/+/771422 14:04:04 Tengu: please go ahead ^ 14:04:08 ah :) 14:04:24 sooo yeah. the whole story here is: cURL doesn't support CIDR notation for the no_proxy env var. 14:04:50 this leads to errors on some (many...) corp deploys - the healthchecks are failing because they are trying to go through the proxy... 14:05:06 so after many thoughts and discussions, it appears we'll need to get around this once for all. 14:05:23 so I'm proposing a patch that will try to be smart: default to the standard, fast, tiny efficient "cURL" call 14:05:49 but if it founds a "no_proxy" or NO_PROXY (or a specific env key), it will switch to a "small" python script using python requests 14:06:04 that way..... we should hopefully solve that issue once for all, in a mostly elegant way. 14:06:25 I'd want to raise awareness about this need, and see if anyone has strong concerns or, better, a better idea :) 14:06:37 fact is: neither cURL nor wget supports CIDR notation in the no_proxy 14:06:55 Tengu: thanks, nicely explained in the commit message there too 14:06:59 python-requests does support it, apparently so does some golang lib... but I'd rather go for python here. 14:07:14 especially since python is installed in the containers - not golanf 14:08:06 so review request here for https://review.opendev.org/c/openstack/tripleo-common/+/771422 14:08:08 (note: the patch has some errors I'm working on right now, mostly cosmetics - hence the -w) 14:08:10 anything else on that one? 14:08:32 nope. just wanting to ensure ppl see this change, since it might influence a bit how healthchecks are working 14:08:32 k moving on 14:08:41 Tengu: thank you for bringing that 14:08:48 yw :) 14:08:51 #info ephemeral Heat work is ongoing and we have a tracking etherpad(s) 14:08:58 #link https://etherpad.opendev.org/p/tripleo-ephemeral-heat 14:09:02 #link https://etherpad.opendev.org/p/podman-generate-kube 14:09:06 #link https://etherpad.opendev.org/p/podman-generate-kube 14:09:17 (I think the meetbot doesn't pick it up if there are spaces) ^^ 14:09:32 slagle: ^^ that yours ... want to say something about it or mainly review requests/info 14:09:40 i just wanted to point out the above etherpads 14:09:45 * Tengu follows as well, since working on this as well. 14:09:59 to raise awareness 14:10:16 slagle: great thanks 14:10:33 lets give it a sec anyone have any comments or anything else to hilight about this item 14:10:41 spec is not yet merged, but we're going ahead as written 14:10:42 * marios looks at etherpads 14:10:51 Francesco Pantano proposed openstack/tripleo-ansible master: WIP: Introduce tripleo_cephadm role https://review.opendev.org/c/openstack/tripleo-ansible/+/770674 14:11:14 slagle: is the spec blocked on something i thought we were done there 14:11:21 slagle: i haven't checked for couple days though 14:11:29 mostly cosmetics I think at this point 14:11:59 not blocked per se, a few last minute comments which have been addressed 14:12:15 slagle: k thank you 14:12:17 bogdando: do you plan to re-review, or do you want your -1 to remain sticky? 14:13:11 k lets move on for now 14:13:15 yea, i think his concern was addressed 14:13:24 thanks 14:13:32 #info tripleo ceph client spec is implemented and is default for scenario004 14:13:36 #info 004 ceph-ansible for server config but not client config 14:13:41 #info 001 is all ceph-ansible right now, but will move to new ceph-client and cephadm during this sprint 14:13:49 #link https://blueprints.launchpad.net/tripleo/+spec/tripleo-ceph-client 14:13:52 #link https://blueprints.launchpad.net/tripleo/+spec/tripleo-ceph-client 14:14:13 marios: yeah so basically the first spec is done, we're actively working on the second one (cephadm integration) 14:14:20 +1 14:14:22 deprecating ceph-ansible and moving to cephadm 14:14:27 fultonj: gfidente: fmount: o/ hi folks anything you want to hilight 14:14:36 fultonj: thanks 14:14:46 sorry fmount :) ^ 14:14:55 i don't think so, just a status highlight 14:15:14 marios: just to share our status, the goal is to reach something working by the end of this milestone 14:15:14 scenario001 will be on cephadm and off of ceph-ansible hopefully within 2 weeks 14:15:16 right 14:15:22 fmount: fultonj: gfidente: k let me paste all the things here sec 14:15:29 #link https://blueprints.launchpad.net/tripleo/+spec/tripleo-ceph-client 14:15:35 #link https://review.opendev.org/q/topic:%22cephadm_integration%22 14:15:41 #info we aim to have scenario001 running without ceph-ansible during this sprint 14:15:46 #info older branches will continue to use nautilus/ceph-ansible 14:15:51 #info main branch for wallaby 001 will use ocotopus/cephadm (hopefully 004 too by end of W) 14:15:55 #info found and patched new bugs from new client 14:16:01 #link https://launchpad.net/bugs/1912109 14:16:03 Launchpad bug 1912109 in tripleo "FATAL | tripleo client role ceph_external_multi_config is unfined" [High,In progress] - Assigned to John Fulton (jfulton-org) 14:16:09 #link https://launchpad.net/bugs/1912103 14:16:10 Launchpad bug 1912103 in tripleo "permission error when linking tripleo-inventory in config-download ceph-ansible subdirecotry" [High,In progress] - Assigned to John Fulton (jfulton-org) 14:16:15 #link https://bugs.launchpad.net/tripleo/+bug/1912218 14:16:16 Launchpad bug 1912218 in tripleo "tripleo-ansible roles/tripleo_ceph_client putting ctlplane IPs in mon_host instead of storage network IPs when using composable networks" [High,In progress] - Assigned to John Fulton (jfulton-org) 14:16:37 nice thank you folks, something more you want to add here. or anyone else has comments about it 14:16:43 fmount: fultonj: gfidente: ^ 14:16:58 not me 14:17:03 me neither 14:17:05 not me either thanks 14:17:16 k lets give it a sec there's a lot there 14:18:07 k lets move on then 14:18:21 #info Inventory re-factoring and support for inventory from neutron resources 14:18:26 #link https://review.opendev.org/c/openstack/tripleo-common/+/766109 14:18:34 #link https://review.opendev.org/c/openstack/tripleo-common/+/765902 14:18:49 hjensas: o/ around? is that yours i suspect ^ 14:19:04 hjensas: something more you want to add or just review requests 14:19:42 k i think these are all from hjensas going to just paste them here and see if there are comments 14:19:53 #info A tool to convert Heat nic-config templates to Ansible J2 NIC config templates 14:20:01 #link https://review.opendev.org/c/openstack/tripleo-heat-templates/+/767466 14:20:10 #info A good part of the patches to manage network's and ports has been merged 14:20:20 #link https://review.opendev.org/q/topic:%22network-data-v2%22+(status:open%20OR%20status:merged) 14:20:55 so review requests and status here ^^^ seems like good progress on the ports v2 spec 14:21:04 anyone have something further to add to this topic 14:21:48 k moving on 14:21:56 #info ci & upgrades having an ongoing discussion around upgrade jobs & how to improve, what to remove etc 14:22:03 #link https://etherpad.opendev.org/p/tripleo-upgrade-testing-wallaby 14:22:20 so upgrades and ci teams have had a couple of meetings around this already and it is an ongoing discussion. 14:22:59 mainly socialising that for now we don't have clear answers yet. highly likely that at least some of the upstream upgrade jobs will be removed or at least changed 14:23:19 any comments about that? odyssey4me or anyone else from upgrades around want to add something more^^ 14:23:59 k moving on 14:24:08 #info close to merging upgrades job related excess content-providers https://review.opendev.org/q/topic:reduce-content-providers 14:24:14 #info last patch in series gives us the reduction - we'll see it in all the repos after https://review.opendev.org/c/openstack/tripleo-ci/+/769131 merges 14:24:21 #info last patch in series gives us the reduction - we'll see it in all the repos after https://review.opendev.org/c/openstack/tripleo-ci/+/769131 merges 14:24:44 sry .. so this is just to say we're close on that, we were blocked by an ussuri and train bug 14:24:56 it came up this week as infra was complaining about all the extra content providers 14:25:03 so we are addressing it, should be done very soon now 14:25:39 k and couple of community items left on the etherpad... moving on 14:25:48 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019723.html [cinder][tripleo][release] Victoria Cycle-Trailing Release Deadline Herve Beraud 14:26:12 so this one.. we hadn't released victoria for os-refresh-config/os-collect-config/tripleo-ipsec 14:26:37 this started a discussion, first around if anyone is even using these and would they care if we moved it to release-management: none 14:26:47 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019751.html [tripleo] move os-refresh-config, os-collect-config, tripleo-ipsec to 'release-management': none Marios Andreou 14:27:06 and then also possibly about moving the os-*-config under heat governance, assuming they are interested 14:27:17 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019777.html [heat][tripleo] os-apply-config os-refresh-config os-collect-config project governance Marios Andreou 14:27:43 so, it seems that they aren't hugely interested (no reply so far) and that is fine. i guess the next step is to decide what we will do with them 14:27:49 any comments or thoughts about this topic? 14:28:00 we ended up making the victoria release for now anyway with 14:28:09 #link https://review.opendev.org/c/openstack/releases/+/769915 14:28:35 * odyssey4me is catching up on scrollback 14:28:35 we can consider moving them to independent too, as another option. i think this one had some support on the mailing list threads i pointed to above 14:29:27 OK, folks can reply on the ML if they have thoughts about those 14:29:38 moving onto next 14:30:01 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019860.html [TripleO] moving stable/rocky for tripleo repos to unmaintained (+ then EOL) OK? 14:30:04 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019860.html [TripleO] moving stable/rocky for tripleo repos to unmaintained (+ then EOL) OK? 14:30:11 sorry i keep pasting the space hence the duplicates 14:30:38 this one.. i asked the question ~10 december so as i said in the last meeting here, i bumped the thread in preparation for taking action here 14:31:09 i haven't had any response and in particular no objections to this 14:31:50 we may even go directly to EOL. there isn't an official mechanism for declaring it 'unmaintained' so that doesn't make sense as an option 14:32:21 if there are no other comments here then moving on, i will post the patch to make it EOL within the next couple weeks and send to the ML as usual 14:32:37 after that we plan to remove all the upstream CI for it 14:32:48 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019839.html [release] Release countdown for week R-12 Jan 18 - Jan 22 Herve Beraud 14:32:54 #info wallaby-2 milestone is this week. we should release or will be switched to https://releases.openstack.org/reference/release_models.html#cycle-with-rc 14:32:58 #info wallaby-2 milestone is this week. we should release or will be switched to https://releases.openstack.org/reference/release_models.html#cycle-with-rc 14:33:19 this one... so apparently we should make a wallaby release for wallaby-2 which is this week 14:33:37 i don't know if it really makes sense for us to do that though as nothing is ready for release. 14:33:46 o/ Hello from nova-land folks. Have a question about using standalone in CI 14:33:50 And running tempest plugin tests 14:33:57 as written in that thread, if we don't we might be asked to move to the cycle-with-rc model instead 14:34:10 artom: o/ hey we are in our irc meeting now 14:34:23 Whoops :( 14:34:25 artom: we have an open topic slot in a bit though 14:34:27 artom: np :) 14:34:28 * artom apologises and waits 14:34:55 any comments on releasing for wallaby-2? 14:35:26 slagle: hi, yes, we can chat 14:35:41 imo it doesn't make sense, we'd be releasing just for 'the rules' but what we release isn't at all complete, perhaps there is no expectation for it to be though it is wallaby-2 so ... (and we are trailing release anyway ) 14:36:20 ok, then moving on ... 14:36:25 folks are very quiet today :D 14:36:36 #topic Bugs and blueprints 14:36:43 #link https://bugs.launchpad.net/tripleo/ 14:36:43 #link https://storyboard.openstack.org/#!/project/openstack/tripleo-ansible 14:36:45 #link https://launchpad.net/tripleo/+milestone/wallaby-2 14:36:48 #link https://launchpad.net/tripleo/wallaby 14:37:08 anyone wants to hilight bugs or blueprints? 14:37:23 ruck or rover around anything you want to point to /any gate blockers etc right now? 14:38:21 ok then moving on 14:38:27 #topic Project releases or stable backports 14:38:29 #info tripleo wallaby repos https://releases.openstack.org/teams/tripleo.html#wallaby 14:38:58 so as just mentioned... we have to decide whether to make a release for wallaby-2 14:39:28 unless i hear otherwise, i may just go ahead and make the release for the tripleo repos (if that is a requirement of us, per http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019839.html 14:39:56 another thing to hilight here is that victoria release https://review.opendev.org/c/openstack/releases/+/769915 i mentioned earlier also 14:40:08 anything else on this topic, any release requests etc 14:40:26 slagle: it seems that the spec is pending for the update to address my comments? 14:40:48 bogdando: i'll bring it up again in a sec in the open topics bit sec 14:40:49 bogdando: let's sync up post-meeting 14:40:56 ack 14:41:02 ha or even now cos... lets move on to 14:41:05 #topic specs 14:41:05 #info https://review.opendev.org/q/project:openstack/tripleo-specs 14:41:06 #info https://opendev.org/openstack/tripleo-specs/src/branch/master/specs/wallaby 14:41:14 ok bogdando slagle want to chat here a bit? 14:41:53 bogdando: oh, just the backup/restore comment? 14:41:55 I think unaddressed comments remaining for the DB backup/restore feature and podman networking mode impact on UC minions feature 14:42:49 also couldnot quickly figure out what was the decision for adding a dependency work or keep it parallel 14:42:50 i can say we do not intend to scale Heat with minions in this model 14:42:58 does that work for you? 14:43:02 yes please 14:43:06 ok 14:43:29 aside of that there may be changs needed to the minion deployment code 14:43:47 to not deploy heat for example 14:44:27 i think the only other one apart from ephemeral heat that we didn't merge yet is the frrouter one 14:44:30 makes sense, i will add that 14:44:36 #link https://review.opendev.org/c/openstack/tripleo-specs/+/758249 14:44:39 thanks slagle 14:44:58 it has enough votes but there are some comments there dsneddon are you planning to update it ^^^ or should we merge 14:45:15 i think i will merge that this week if there are no more updates 14:45:21 thanks bogdando slagle 14:45:28 anything else about specs from anyone 14:45:57 #topic open discussion 14:46:00 Anything else that folks want to bring up to the meeting? 14:46:03 Merged openstack/tripleo-operator-ansible master: add role to list available tripleo validations https://review.opendev.org/c/openstack/tripleo-operator-ansible/+/755365 14:46:45 any comments /feedback on this meeting are very much appreciated 14:46:50 seems like it took us a while today 14:46:58 but perhaps that's cos we have them every 2 weeks 14:47:53 well OK. if there is nothing more then let's close it. 14:48:02 #info next meeting tuesday 02 Feb 14:48:20 thank you everyone for participating and bringing topics today 14:48:28 #endmeeting tripleo