Wednesday, 2020-03-11

openstackgerritMerged opendev/system-config master: Simplify goaccess command  https://review.opendev.org/71220100:10
openstackgerritMerged opendev/system-config master: Pull goaccess report into zuul logs  https://review.opendev.org/71220200:10
openstackgerritMerged opendev/system-config master: nodepool-builder: Add webserver  https://review.opendev.org/71157400:31
openstackgerritMerged opendev/zone-opendev.org master: Add nb01.opendev.org  https://review.opendev.org/69311800:57
openstackgerritIan Wienand proposed openstack/project-config master: nodepool-elements: skip tox install with no pip-and-virtualenv  https://review.opendev.org/71222502:29
openstackgerritIan Wienand proposed opendev/zone-opendev.org master: nb01 acme challenge: add missing trailing period  https://review.opendev.org/71222702:43
openstackgerritMerged openstack/project-config master: nodepool-elements: skip tox install with no pip-and-virtualenv  https://review.opendev.org/71222502:47
openstackgerritMerged opendev/zone-opendev.org master: nb01 acme challenge: add missing trailing period  https://review.opendev.org/71222702:49
*** DSpider has joined #opendev04:57
openstackgerritOpenStack Proposal Bot proposed openstack/project-config master: Normalize projects.yaml  https://review.opendev.org/71223506:03
openstackgerritMerged openstack/project-config master: Normalize projects.yaml  https://review.opendev.org/71223507:29
AJaegerinfra-root, reviews of infra-manual are appreciated, please07:45
*** AJaeger has quit IRC09:54
openstackgerritMerged opendev/irc-meetings master: Reschedule Nova Team meeting  https://review.opendev.org/71205210:10
*** AJaeger has joined #opendev10:14
*** ttx has joined #opendev11:19
*** ttx has quit IRC11:19
*** ttx has joined #opendev11:28
openstackgerritMerged openstack/project-config master: Add rss link for cyborg-specs  https://review.opendev.org/71187512:08
openstackgerritSorin Sbarnea proposed zuul/zuul-jobs master: Add ensure-snap role  https://review.opendev.org/71241413:17
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Networking VPP team meeting  https://review.opendev.org/71244814:14
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Senlin team meeting  https://review.opendev.org/71245014:17
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Trove team meeting  https://review.opendev.org/71245114:18
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Upgrades SIG meeting  https://review.opendev.org/71245214:18
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Doc team meeting  https://review.opendev.org/71245314:20
corvusAJaeger: even after your changes, there's a ton of information in infra-manual that's openstack specific14:20
corvusAJaeger, fungi: this is why i'd argue that we did not get the namespace "wrong" when we sorted infra-manual into openstack/14:20
corvusAJaeger: but that's fine, our goals may have changed :)14:21
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove FEMDC SIG meeting  https://review.opendev.org/71245414:21
corvusAJaeger: so if our new goal is really to make an opendev infra manual, how much openstack do we want to remove?14:21
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Glance Bug Squad meeting  https://review.opendev.org/71245514:22
corvusAJaeger: take 711817 for example -- it removes some really specific stuff about the openstack release process, but it still describes how to create a "stable/" branch.  even the idea of creating a branch with a prefix of "stable/" is an openstackism.  non-openstack projects may feel free to create branches named whatever they want.  they could create stable branches called "stable-2.16" or they could14:23
corvuscreate branches called "release-3.0".14:23
corvustake 711818 for example -- there's a 5 step process for retiring a project -- 2 of them are openstack specific14:24
fungicorvus: my motive for positioning the rename as a correction from the earlier rename migration is so that we have an argument not to retire it in place and fork to a new namespace like tc policy requires14:25
openstackgerritThierry Carrez proposed opendev/irc-meetings master: Remove Group-Based policy team meeting  https://review.opendev.org/71245714:25
corvusfungi: i think what i'm getting at is that maybe we should fork it to a new namespace and not retire it14:25
fungithat would also work14:25
fungithough i think the goal was for all the openstack-specific details to end up in the contributor guide14:26
corvusthere still seems to be enough content to warrant an openstack-specific manual14:26
corvusfungi: let's look at 711818 then14:26
clarkbwe discussed this with the TC (diablo_rojo mostly) and thr feeling eas that the contributor guide should have the openstack bits14:26
corvusfungi: how should that move into the cg?14:26
clarkbajaeger has been pushing changes that direction too as a sprt of remove from here add to therebalance14:27
fungieither as a retirement topic in a project drivers section of the contributor guide, or maybe things like that make more sense in the project-team guide14:28
corvusfungi: so duplicate the content as a new 5 step process in the cg?14:28
corvusfungi: then make the process 3 steps in infra-manual?14:28
fungibasically the openstack guide, whatever it is, would just have steps 1 and 5 and then link to opendev's retirement recommendations in between14:28
corvusfungi: does the cg generally do that?14:29
fungialso arguably, step 1 can just go away. that's cruft from when openstack used to do requirements list syncing14:29
fungi(i think)14:29
corvusi'm scanning through the cg right now, and i don't see it linking to the infra-manual very much14:30
fungiit mostly only does for account setup right now, last i looked14:30
fungibut the cg maintainers have expressed a desire to just link to things which the infra manual covers, not duplicate content14:31
corvuslooks like it links to the infra-manual in 3 places14:31
fungianyway, i get your point, leaving the infra manual intact and then starting a new opendev manual we can lift general bits into lets us and the cg and pt-guide maintainers chip away at it until there's nothing left14:32
corvusi'm not arguing strongly for that; i'd be fine moving infra-manual to opendev, but we're getting into the harder parts of it now, and i'd like to find out how committed we are to this idea :)14:33
clarkbI dont see how starting from scratch is any easier. I think the process is largely the same either way, we'll start with exoeting doc, remove bits and rewrite as necessary then either push into blank slate or update existing content14:36
AJaegerSorry, can't chime into this discussion fully. Just reading backscroll: My goal so far was to point to existing information and serve *all* hosted projects. So, I'm happy to generalize - but also point out OpenStack specific bits.14:36
AJaegerthe contributor guide has many of these parts not today, so for some content we might need to create completely new pages in contributor guide14:37
fungiso far the cg hasn't linked to the infra manual much because the latter was already a fairly complete treatment of topics so the cg was focused more on filling in the non-infra-related gaps, but were up for hosting openstack-specific infra recommendations "somewhere" in the cg as they moved out of the infra manual14:37
fungier, the cg maintainers were up for hosting14:37
corvusokay, then i think i will point out in review where i think we should find more openstacky homes for places, even if it means that, say, 711818 may need to grow a depends-on to a PTG change or something14:39
fungithat seems like a reasonable way forward. for the record i did not expect this to be quick or easy, there's a lot of entanglement14:40
AJaeger I want to avoid jumping between guides, so for some changes like 711818, how are we going to describe retirement?14:40
AJaegerHAve a section in PTG that says: step 1, see steps 2-4 in Infra Manual, then do step 5 here?14:40
clarkbAJaeger: ya I think that is how14:41
corvusAJaeger: or duplicate it?14:41
AJaegerAnd in OpenDev manual: A link back to PTG: For extra OpenStack bits, go to PTG?14:41
corvusi don't like that14:41
AJaegerIf we duplicate, it gets quickly out of sync ;(14:41
corvusi feel like if we start doing that, *every* section is going to say "for openstack go here"14:41
corvuswhich really kinda undercuts the equality principle we're going for14:42
fungiAJaeger: well, probably drop step 1 entirely, have a section in the ptg say openstack follows opendev's recommended retirement process <link> and additionally you should <openstack-specific things>14:42
AJaegerI'm happy to say "For Airship, go here; for StarlingX go here". But if indeed every section would say that - we need to duplicate...14:42
mordredridiculous stupid idea ... what if we decomposed some of the sphinx into chunks we could include in different pages - and then set up a mulit-repo build job which grabbed the opendev guide when building the opentack CG and included relevant chunks in the right places14:42
corvusmordred: that would be complicated, but i kinda like it14:42
AJaegerfungi: yes, that works for PTG; But what to add in OpenDev manual? Just the "middle" steps or a link back?14:43
fungiAJaeger: my goal is for us to get out of the business of tracking or pointing contributors at project-specific guides and recommendation. they should be talking to those projects they're collaborating in14:43
mordredthen the opendev guide could not have any openstack/airship go here sections - but there could be a complete openstack or airship guide that didn't involve jumping elsewhere14:43
corvusmordred: ++14:43
AJaegerwe tried these kind of guides with the Install Guide in docs team - and at one point gave up, it was too complicated ;(14:44
fungithough i do like mordred's idea about transcluding bits of the opendev manual instead of linking. i had suggested that to diablo_rojo for the cg14:44
corvuswe could try it with 711818 and see what it would look like14:44
AJaegerIt has its merits - but sounds more involved than I want to sign up for ;( (happy to review still, I won't block)14:44
openstackgerritMerged openstack/infra-manual master: Fix Peer Review section formatting  https://review.opendev.org/71181914:45
openstackgerritMerged openstack/infra-manual master: Remove section "Updating devstack-vm-gate-wrap.sh"  https://review.opendev.org/71182014:45
fungiwhat i would like to avoid is having the opendev manual maintainers responsible for reviewing openstack-specific and airship-specific and ara-specific content, and being a bottleneck for them maintaining their specific workflow documentation14:45
mordredcorvus: might mean the CG would be unbuildable without a zuul git checkout layout ... so locally people would need the same layout14:45
mordredbut we coudl do include ../../../../opendev/opendev-manual/doc/source/some/thing14:46
AJaegerWe could also sync (copy) files around into another guide? IF it's just a few pages...14:46
mordredin the CG - and the fact that that content is in a different repo is opaque to sphinx14:46
fungialso i don't want the opendev systems maintainers to have to be intimately familiar with every possible project-specific workflow, which hosting them all in our central manual would imply14:46
AJaeger(via proposal job). But required-projects etc. works, just makes local building harder14:46
AJaegerTime for my daily cycle ride, I'll be back in less than an hour!14:47
clarkbthat seems like an optimization to me. We should be able to proceed with url linking then transform that into direct rst includes if someone has time to reorg the rst that way and update the jobs?14:47
AJaegerGlad to see this discussion here!14:47
corvusclarkb: yes14:47
corvusi really hesitate to say this, but submodules may help address local build issues14:48
corvus(though to be fair, they would create other local build issues)14:48
mordredyeah. this may be one of the very few times when that's maybe a valuable tool14:48
mordredbut I also agree with clarkb - I think we can move forward with linking - then it's really clear what parts we want to decompose out and direct include if we get a process for that we like14:48
corvusyeah, i can go with that plan14:49
corvusand we can tool around with updating that to direct includes14:49
mordredmake sure each link from CG->OpenDev is to a self-contained section14:49
mordredor at least a thing with a linkable header14:49
fungithat's basically how i suggested doing it in conversations with diablo_rojo too. do the easy linking thing first but look at possible ways to embed content from the opendev manual into appropriate places in the cg instead as time allows14:54
fungi(embed at build time, not copy into cg source)14:54
openstackgerritSorin Sbarnea proposed openstack/project-config master: Add periodic-1w pipeline  https://review.opendev.org/71247315:09
*** AJaeger has quit IRC15:14
openstackgerritClark Boylan proposed opendev/system-config master: Return goaccess html as zuul artifact  https://review.opendev.org/71247815:24
*** AJaeger has joined #opendev15:56
AJaegerback again, will read backscroll now...15:58
AJaegerSo, reading backscroll I wonder what the next step is - did anybody volunteer for one?16:03
openstackgerritThierry Carrez proposed opendev/infra-specs master: POC Xwiki farm at wiki.opendev.org  https://review.opendev.org/71005716:04
clarkbAJaeger: what I took away from it is we can continue with what you were doing, but stop linking from infra-manual to $projectguide. Instead only link from $projectguide to infra-manual. Then if at some point someone is able to reorg the rst to make it directly includable we can do that instead of the links16:15
AJaegerOk, let me look at the open reviews and see what I can come up with.16:20
openstackgerritClark Boylan proposed opendev/system-config master: Return goaccess html as zuul artifact  https://review.opendev.org/71247816:23
openstackgerritClark Boylan proposed opendev/system-config master: Return goaccess html as zuul artifact  https://review.opendev.org/71247816:43
openstackgerritMonty Taylor proposed opendev/system-config master: Split zuul_return into its own file and skip linting  https://review.opendev.org/71250717:05
openstackgerritMonty Taylor proposed opendev/system-config master: Split zuul_return into its own file and skip linting  https://review.opendev.org/71250717:07
openstackgerritsebastian marcet proposed opendev/puppet-openstackid master: Added python packages install support  https://review.opendev.org/71251117:13
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Update Feature branches, releases sections for OpenStack  https://review.opendev.org/71181717:20
AJaegerclarkb, fungi, corvus, mordred, please review the change above ^17:20
AJaegerIs that more what you have in mind? This is a larger rework as well...17:20
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Update Feature branches, releases sections for OpenStack  https://review.opendev.org/71181717:24
AJaegerbbl17:25
openstackgerritMonty Taylor proposed opendev/system-config master: Move ansible-lint config into config file  https://review.opendev.org/71251917:36
openstackgerritMonty Taylor proposed opendev/system-config master: Move ansible-lint config into config file  https://review.opendev.org/71251917:52
openstackgerritMonty Taylor proposed opendev/system-config master: Add a fake zuul_return module  https://review.opendev.org/71252517:52
openstackgerritMerged opendev/irc-meetings master: Remove FEMDC SIG meeting  https://review.opendev.org/71245417:56
openstackgerritSorin Sbarnea proposed openstack/project-config master: Add periodic-1w pipeline  https://review.opendev.org/71247317:58
openstackgerritMerged opendev/irc-meetings master: Remove Senlin team meeting  https://review.opendev.org/71245017:59
openstackgerritSorin Sbarnea proposed openstack/project-config master: Add periodic-1w pipeline  https://review.opendev.org/71247318:24
openstackgerritSorin Sbarnea proposed openstack/project-config master: Add periodic-weekly pipeline  https://review.opendev.org/71247318:25
*** AJaeger has quit IRC18:34
*** AJaeger has joined #opendev18:37
AJaegerclarkb: do you have a proposal for the translation paragraph? https://review.opendev.org/#/c/711827/18:48
clarkbAJaeger: we might even drop it all together until our tooling story shapes up18:53
AJaegerclarkb: that is the other option - have it only in i18n repo...18:54
AJaegerclarkb, corvus, fungi, mordred, https://review.opendev.org/712534 updates the project-team-guide - now working on infra-manual...19:03
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Remove OpenStack specific steps from project retirement  https://review.opendev.org/71181819:06
AJaegerand here's the corresponding manual change ^19:06
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Move Translation setup to i18n document  https://review.opendev.org/71182719:07
AJaegerclarkb: updated as well ^19:07
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Remove OpenStack specific steps from project retirement/rename  https://review.opendev.org/71181819:16
AJaegerok, I'm happy for now - please review ;)19:16
openstackgerritAndreas Jaeger proposed openstack/infra-manual master: Remove OpenStack specific steps from project retirement/rename  https://review.opendev.org/71181819:20
clarkbAJaeger: those changes lgtm19:24
AJaegerthanks19:24
AJaegerclarkb: https://review.opendev.org/#/c/712534/ as well?19:26
AJaegerfungi, you're core on the project-team-guide, please review https://review.opendev.org/#/c/712534/19:26
AJaegerclarkb: is https://review.opendev.org/711817 also ok?19:27
openstackgerritMonty Taylor proposed opendev/system-config master: Return goaccess html as zuul artifact  https://review.opendev.org/71247819:28
openstackgerritMerged openstack/project-config master: Add periodic-weekly pipeline  https://review.opendev.org/71247319:28
clarkbAJaeger: looking19:33
AJaegerthanks, clarkb19:34
clarkbAJaeger: I think https://review.opendev.org/#/c/711817/3 works. It removes the very openstack specific stuff, then provides examples for people that may want to use certain processes19:40
clarkbprobably want corvus to rereview19:40
openstackgerritMonty Taylor proposed zuul/zuul-jobs master: Use a fake zuul_return and an .ansible-lint file  https://review.opendev.org/71254719:46
openstackgerritMonty Taylor proposed zuul/zuul-jobs master: Use a fake zuul_return and an .ansible-lint file  https://review.opendev.org/71254719:51
openstackgerritMonty Taylor proposed opendev/base-jobs master: Allow ozj to use base-minimal  https://review.opendev.org/71254820:05
mordredclarkb, AJaeger: ^^20:05
clarkbmordred: ha check the error on it20:06
clarkbmaybe we should push that test from ozj into base-jobs20:06
clarkbsince I think it is the base jobs that use the role right?20:06
clarkb(and that reduces tenant project web matrix)20:06
mordredwell - sort of - this is the role that sets up the old /etc/ci/mirror_info.sh script20:07
mordredso I think it's in openstack not opendev on purpose20:07
clarkboh20:07
clarkbI wonder if it needs base-minimal to be tested effectively? I guess because it is trying to be a subset of base-legacy?20:07
clarkbor do we set that up in regular base?20:08
mordredI think it goes in regular base - looking20:08
clarkbif it is in regula base then I think putting it in opendev/base-jobs is fine20:09
clarkbsince its part of the base jo bthere20:09
mordredclarkb: it's in opendev base20:09
clarkbputting the test in opendev/base-jobs I mean20:09
mordredyeah. but ...20:09
mordredmirror-info role is also in base-jobs20:10
mordredso - maybe this is just a leftover cleanup20:10
clarkboh ya because the role in ozj can't be accessed in opendev/base-jobs based on the error we just had20:10
clarkbya I think we can probably rm it from ozj20:10
AJaegerthere's a mirror-info role in openstack-zuul-jobs and one in zuul-jos as well.20:11
mordredI'm just gonna do that20:11
clarkbthe error in base-jobs is basically saying it doesn't know about ozj roles20:11
AJaegerso, agree with what you said ;)20:11
AJaegeryes, we should remove ozj, it even has two variables less20:12
AJaegercorvus, could you put the infra-manual opendev changes on your review queue, please?20:16
corvusAJaeger: +3s20:18
AJaegerthx20:19
openstackgerritMerged openstack/project-config master: Revert "Disable limestone"  https://review.opendev.org/71125720:25
openstackgerritMerged openstack/infra-manual master: Update Feature branches, releases sections for OpenStack  https://review.opendev.org/71181720:26
openstackgerritMerged openstack/infra-manual master: Remove OpenStack specific steps from project retirement/rename  https://review.opendev.org/71181820:29
openstackgerritMerged openstack/infra-manual master: Move Translation setup to i18n document  https://review.opendev.org/71182720:29
openstackgerritMerged opendev/system-config master: Return goaccess html as zuul artifact  https://review.opendev.org/71247820:40
openstackgerritMonty Taylor proposed opendev/system-config master: Remove ansible-lint arguments  https://review.opendev.org/71255420:55
openstackgerritMonty Taylor proposed opendev/system-config master: Go back to using find with ansible-lint  https://review.opendev.org/71255421:30
openstackgerritMonty Taylor proposed zuul/zuul-jobs master: Use a fake zuul_return and an .ansible-lint file  https://review.opendev.org/71254721:40
openstackgerritClark Boylan proposed opendev/system-config master: Remove old 404 checker job  https://review.opendev.org/71257522:23
openstackgerritClark Boylan proposed opendev/system-config master: Add goaccess reporting to more static.o.o sites  https://review.opendev.org/71257622:23
*** Shrews has quit IRC22:31
*** msdisme has quit IRC22:31
openstackgerritClark Boylan proposed opendev/system-config master: Add goaccess reporting to more static.o.o sites  https://review.opendev.org/71257622:33
*** mnaser has quit IRC22:36
*** mnaser has joined #opendev22:39
*** Shrews has joined #opendev22:40
openstackgerritMerged opendev/system-config master: Go back to using find with ansible-lint  https://review.opendev.org/71255422:42
*** zbr has quit IRC22:50
*** DSpider has quit IRC22:54
*** zbr has joined #opendev22:55
fungii just noticed, we don't configure the openstackstatus bot to join this channel... anyone opposed to me adding it?23:06
fungii'll propose a change23:09
openstackgerritJeremy Stanley proposed opendev/system-config master: Configure statusbot to join #opendev  https://review.opendev.org/71258223:09
fungiet voila ^23:09
clarkbfungi: there is already a change iirc23:19
clarkbfungi: https://review.opendev.org/#/c/711245/23:20
fungiaha, thanks23:27
fungii was about to add some entries which will merge-conflict with it23:28
openstackgerritJeremy Stanley proposed openstack/project-config master: Add OpenInfra Labs IRC channels to bots  https://review.opendev.org/71258623:45
openstackgerritJeremy Stanley proposed opendev/system-config master: Add IRC bots to OpenInfra Labs channels  https://review.opendev.org/71258823:45
openstackgerritMerged opendev/system-config master: Add statusbot to #opendev  https://review.opendev.org/71124523:50

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!