Tuesday, 2018-10-30

*** eernst has quit IRC00:00
openstackgerritMerged openstack-infra/system-config master: Mirror puppet6 for bionic and stretch  https://review.openstack.org/61219800:03
*** gyee has quit IRC00:04
openstackgerritIan Wienand proposed openstack/diskimage-builder master: Remove python3 legacy jobs  https://review.openstack.org/61404700:07
openstackgerritMerged openstack-infra/zuul-jobs master: Fix noqa warning  https://review.openstack.org/61378500:10
*** mriedem_away has quit IRC00:15
*** ansmith has joined #openstack-infra00:17
openstackgerritIan Wienand proposed openstack/diskimage-builder master: Remove python3 legacy jobs  https://review.openstack.org/61404700:17
openstackgerritMerged openstack-dev/pbr master: Correct documentation hyperlink for environment-markers  https://review.openstack.org/61357600:24
*** dmsimard has quit IRC00:28
openstackgerritIan Wienand proposed openstack/diskimage-builder master: Add ubuntu-systemd-container operating-system element  https://review.openstack.org/56374800:31
openstackgerritIan Wienand proposed openstack/diskimage-builder master: [wip] containers test  https://review.openstack.org/61405100:31
*** longkb has joined #openstack-infra00:46
*** xinliang has joined #openstack-infra00:54
openstackgerritMerged openstack-infra/nodepool master: Cleanup down ports  https://review.openstack.org/60982901:12
*** hongbin has joined #openstack-infra01:23
openstackgerritIan Wienand proposed openstack-infra/system-config master: Add puppet6 to apt-updates  https://review.openstack.org/61405401:28
*** bhavikdbavishi has joined #openstack-infra01:36
*** diablo_rojo has quit IRC01:56
*** scas has joined #openstack-infra01:56
*** agopi is now known as agopi|brb01:57
*** agopi|brb has quit IRC02:01
*** bhavikdbavishi has quit IRC02:08
*** armax has quit IRC02:34
*** agopi|brb has joined #openstack-infra02:42
*** rcarrillocruz has quit IRC02:43
*** ramishra has joined #openstack-infra02:50
*** ykarel|away has joined #openstack-infra02:55
*** ykarel|away is now known as ykarel02:56
*** mrsoul has quit IRC02:57
*** zhangfei has joined #openstack-infra02:58
*** zhangfei has quit IRC03:06
*** psachin has joined #openstack-infra03:10
*** bhavikdbavishi has joined #openstack-infra03:11
openstackgerritNguyen Hai proposed openstack/gertty master: fix tox python3 overrides  https://review.openstack.org/57975703:23
*** cfriesen has quit IRC03:23
*** ramishra has quit IRC03:29
*** diablo_rojo has joined #openstack-infra03:32
*** ramishra has joined #openstack-infra03:45
*** udesale has joined #openstack-infra03:49
*** janki has joined #openstack-infra04:14
*** diablo_rojo has quit IRC04:21
*** diablo_rojo_ has joined #openstack-infra04:21
*** hongbin has quit IRC04:23
*** ccamacho has quit IRC04:26
*** dmsimard has joined #openstack-infra04:27
openstackgerritMerged openstack-infra/system-config master: Add puppet6 to apt-updates  https://review.openstack.org/61405404:36
*** diablo_rojo_ has quit IRC04:42
*** dmsimard has quit IRC04:45
*** diablo_rojo has joined #openstack-infra04:50
*** lpetrut has joined #openstack-infra04:52
openstackgerritIan Wienand proposed openstack-infra/nodepool master: Logs stats for nodepool automated cleanup  https://review.openstack.org/61407404:57
*** diablo_rojo has quit IRC05:00
openstackgerritIan Wienand proposed openstack-infra/nodepool master: Logs stats for nodepool automated cleanup  https://review.openstack.org/61407405:01
*** dmsimard has joined #openstack-infra05:10
*** d0ugal has joined #openstack-infra05:18
AJaegerconfig-core, could you review https://review.openstack.org/613979 https://review.openstack.org/613630 https://review.openstack.org/614014 , please?05:19
*** ykarel has quit IRC05:20
*** lpetrut has quit IRC05:28
*** d0ugal has quit IRC05:30
*** jaosorior has quit IRC05:32
*** jaosorior has joined #openstack-infra05:32
*** pcaruana has joined #openstack-infra05:36
openstackgerritMerged openstack-infra/project-config master: Add noop to instack-undercloud  https://review.openstack.org/61363005:41
openstackgerritMerged openstack-infra/project-config master: Add noop to instack  https://review.openstack.org/61397905:41
openstackgerritMerged openstack-infra/project-config master: Add publish to pypi jobs for gabbi-tempest  https://review.openstack.org/61401405:41
*** ykarel has joined #openstack-infra05:42
*** d0ugal has joined #openstack-infra05:44
*** pcaruana has quit IRC05:47
*** kjackal has joined #openstack-infra05:57
AJaegerthanks, fungi !05:59
*** jbadiapa has quit IRC06:10
*** carl_cai has joined #openstack-infra06:13
*** ramishra has quit IRC06:16
*** mnaser has quit IRC06:16
*** hamzy_ has joined #openstack-infra06:17
*** mnaser has joined #openstack-infra06:17
*** hamzy has quit IRC06:18
*** ramishra has joined #openstack-infra06:18
*** linshuicheng[m]1 has quit IRC06:18
*** vdrok has quit IRC06:22
*** vdrok_ has joined #openstack-infra06:23
tonybmordred: If you get a few minutes can you look at https://review.openstack.org/#/c/612474/  the last 2 openstacksdk releases cause the docs for python-openstackclient to fail.  git bisect says it's https://review.openstack.org/#/c/600292/ which is your change06:23
*** anteaya has quit IRC06:23
tonybmordred: I'm not smart enough to work out why.06:23
tonybmordred: the digging I've done is in the comments of 61247406:23
*** beisner has quit IRC06:30
openstackgerritTobias Henkel proposed openstack-infra/nodepool master: WIP: Add request handler timer to stats  https://review.openstack.org/61407906:31
*** beisner has joined #openstack-infra06:31
*** quiquell|off is now known as quiquell06:34
*** slaweq has quit IRC06:40
*** tonyb has quit IRC06:50
*** ramishra_ has joined #openstack-infra06:59
*** roman_g has quit IRC06:59
*** ramishra has quit IRC07:01
*** d0ugal has quit IRC07:01
*** xek has joined #openstack-infra07:01
openstackgerritIan Wienand proposed openstack-infra/nodepool master: Logs stats for nodepool automated cleanup  https://review.openstack.org/61407407:17
*** aojea has joined #openstack-infra07:18
*** dpawlik has joined #openstack-infra07:26
*** dpawlik has quit IRC07:27
*** dpawlik has joined #openstack-infra07:27
*** ykarel is now known as ykarel|lunch07:35
*** jbadiapa has joined #openstack-infra07:36
openstackgerritAndreas Jaeger proposed openstack-infra/infra-manual master: Add info on needing publish-to-pypi template  https://review.openstack.org/61401707:45
*** shardy has joined #openstack-infra07:46
*** slaweq has joined #openstack-infra07:50
*** pcaruana has joined #openstack-infra07:57
*** pcaruana is now known as pcaruana|elisa|07:59
*** alexchadin has joined #openstack-infra08:02
*** ccamacho has joined #openstack-infra08:04
*** kopecmartin|off is now known as kopecmartin08:08
*** quiquell has quit IRC08:12
*** egonzalez has quit IRC08:15
*** egonzalez has joined #openstack-infra08:17
*** carl_cai has quit IRC08:22
*** jpich has joined #openstack-infra08:26
*** ykarel|lunch is now known as ykarel08:28
*** e0ne has joined #openstack-infra08:41
*** lpetrut has joined #openstack-infra08:48
*** rossella_s has joined #openstack-infra08:48
*** d0ugal has joined #openstack-infra08:50
*** priteau has joined #openstack-infra08:51
*** slaweq has quit IRC08:53
*** shardy_ has joined #openstack-infra08:54
*** jpena|off is now known as jpena08:57
*** xinliang has quit IRC08:59
*** ccamacho has quit IRC09:01
*** sshnaidm|bbl is now known as sshnaidm|off09:02
*** d0ugal has quit IRC09:05
*** carl_cai has joined #openstack-infra09:06
*** quiquell has joined #openstack-infra09:16
openstackgerritTobias Henkel proposed openstack-infra/nodepool master: WIP: Add request handler timer to stats  https://review.openstack.org/61407909:19
*** kjackal has quit IRC09:20
*** kjackal_v2 has joined #openstack-infra09:20
*** derekh has joined #openstack-infra09:25
*** hashar has joined #openstack-infra09:26
*** gbutnaru has joined #openstack-infra09:37
*** ralonsoh has joined #openstack-infra10:03
ralonsohmordred, hi10:04
ralonsohmordred, I'm trying to modify an existing zuul playbook but I don't know what I'm doing wrong10:05
ralonsohmordred, https://review.openstack.org/#/c/482226/34/playbooks/openstack-tox-functional-ovs-with-sudo/pre.yaml10:05
*** slaweq has joined #openstack-infra10:08
*** priteau has quit IRC10:11
*** ccamacho has joined #openstack-infra10:11
*** bhavikdbavishi has quit IRC10:20
*** jtomasek has quit IRC10:30
openstackgerritMerged openstack-infra/irc-meetings master: Remove JJB meeting  https://review.openstack.org/61269710:31
*** e0ne has quit IRC10:32
*** shrasool has joined #openstack-infra10:33
openstackgerritSam Betts proposed openstack-infra/devstack-gate master: Move location of devstack-gate unstack to post-tempest  https://review.openstack.org/53072610:36
*** dpawlik has quit IRC10:37
*** dpawlik has joined #openstack-infra10:37
*** icey has quit IRC10:40
*** shrasool has quit IRC10:47
*** pbourke has quit IRC10:49
*** pbourke has joined #openstack-infra10:50
ssbarneai need to add a new job that needs docker to be available to run? any examples? triggered by "make" as in https://review.openstack.org/#/c/613672/2/Makefile10:50
ssbarneausually i would have used tox but there is a bug in molecule that prevents me from using tox approach for the moment.10:51
*** icey has joined #openstack-infra10:52
*** e0ne has joined #openstack-infra10:54
*** ansmith has quit IRC11:03
*** priteau has joined #openstack-infra11:03
*** jbadiapa has quit IRC11:04
*** electrofelix has joined #openstack-infra11:04
*** dave-mccowan has joined #openstack-infra11:11
*** bhavikdbavishi has joined #openstack-infra11:14
*** ramishra_ has quit IRC11:15
*** ramishra has joined #openstack-infra11:18
*** udesale has quit IRC11:24
*** janki has quit IRC11:25
*** hashar is now known as hasharLunch11:28
*** adriancz has joined #openstack-infra11:34
*** bhavikdbavishi has quit IRC11:35
*** alexchadin has quit IRC11:36
*** bhavikdbavishi has joined #openstack-infra11:41
*** carl_cai has quit IRC11:42
*** dtantsur|afk is now known as dtantsur11:50
*** yamamoto has quit IRC11:51
*** yamamoto has joined #openstack-infra11:53
*** larainema has joined #openstack-infra11:54
*** jesusaur has quit IRC11:56
*** dabukalam has joined #openstack-infra11:57
*** jpena is now known as jpena|lunch11:57
*** yamamoto has quit IRC11:58
*** pbourke has quit IRC12:01
*** jtomasek has joined #openstack-infra12:01
*** jesusaur has joined #openstack-infra12:01
*** rh-jelabarre has joined #openstack-infra12:02
*** pbourke has joined #openstack-infra12:06
*** bhavikdbavishi has quit IRC12:07
*** pbourke has quit IRC12:09
*** pbourke has joined #openstack-infra12:09
*** jesusaur has quit IRC12:13
*** longkb has quit IRC12:14
openstackgerritTobias Henkel proposed openstack-infra/nodepool master: Revert "Cleanup down ports"  https://review.openstack.org/61419812:14
*** agopi|brb is now known as agopi12:17
*** gfidente has joined #openstack-infra12:18
*** e0ne has quit IRC12:21
*** zul has quit IRC12:27
*** wznoinsk has joined #openstack-infra12:28
*** ykarel is now known as ykarel|afk12:29
*** wznoinsk has quit IRC12:29
*** boden has joined #openstack-infra12:29
*** janki has joined #openstack-infra12:31
*** trown|outtypewww is now known as trown12:33
*** yamamoto has joined #openstack-infra12:33
*** rlandy has joined #openstack-infra12:35
*** kgiusti has joined #openstack-infra12:36
*** bhavikdbavishi has joined #openstack-infra12:39
*** jesusaur has joined #openstack-infra12:41
*** ykarel|afk has quit IRC12:44
*** emerson is now known as GDPR12:44
*** wznoinsk has joined #openstack-infra12:45
*** wznoinsk has quit IRC12:47
*** wznoinsk has joined #openstack-infra12:49
*** wznoinsk has quit IRC12:51
fungissbarnea: are you asking how to write a task which installs docker on the test node, or something else?12:57
*** hasharLunch is now known as hashar13:01
*** jbadiapa has joined #openstack-infra13:02
*** lpetrut has quit IRC13:05
*** ansmith has joined #openstack-infra13:05
*** ykarel has joined #openstack-infra13:05
*** e0ne has joined #openstack-infra13:08
*** udesale has joined #openstack-infra13:08
*** psachin has quit IRC13:09
*** lpetrut has joined #openstack-infra13:11
*** yamamoto has quit IRC13:12
*** jpena|lunch is now known as jpena13:12
*** bhavikdbavishi has quit IRC13:12
ssbarneafungi : i wonder if we don't already have nodes with docker working on them. all I know for the moment is that molecule needs docker to compile and run few containers as part of the testing, so now I need to make it work in CI.13:13
*** mriedem has joined #openstack-infra13:13
*** jtomasek_ has joined #openstack-infra13:13
ssbarneafungi: i doubt i am the first one to need this on zuul.13:14
*** tpsilva has joined #openstack-infra13:15
*** psachin has joined #openstack-infra13:15
*** jtomasek has quit IRC13:17
*** bobh has joined #openstack-infra13:20
*** auristor has quit IRC13:23
*** auristor has joined #openstack-infra13:24
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck master: Add query for (anti-)affinity nova functional test race bug 1800472  https://review.openstack.org/61420713:27
openstackbug 1800472 in OpenStack Compute (nova) "nova.tests.functional.test_server_group.ServerGroupTestV264.test_boot_servers_with_affinity_no_valid_host intermittently failing with "OpenStackApiNotFoundException: Item not found"" [Medium,Confirmed] https://launchpad.net/bugs/180047213:27
mordredssbarnea: we have a docker role13:31
mordredssbarnea: https://zuul-ci.org/docs/zuul-jobs/roles.html#role-install-docker13:32
mordredssbarnea: by default it will install docker from upstream docker, but you can set 'use_upstream_docker' to false if you want the distro-provided docker for some reason13:33
mordredssbarnea: I recommend using the role, because it also sets the docker up to point to our per-region caching mirrors13:33
mordredfungi: ^^ that's the general answer for "how does one make docker work on a test node"13:34
*** dklyle has quit IRC13:37
*** dklyle has joined #openstack-infra13:37
fungigot it13:40
fungii was thinking a package task, but that sounds much more flexible13:40
mordredyeah - needing to do special work to set up the mirrors is the bit that makes it complicated13:43
ssbarneamordred: clearly that I would want to use this install-docker role, based on http://codesearch.openstack.org/?q=install-docker&i=nope&files=&repos= i suspect there is no existing check job that is benefiting from it yet.13:43
mordredssbarnea: I totally have jobs using it ...13:43
mordredssbarnea: http://git.openstack.org/cgit/openstack/pbrx/tree/playbooks/container-images/pre.yaml13:43
mordredssbarnea: which is part of a job we run in the check pipeline for every zuul patch - as well as to publish zuul images to docker hub13:44
ssbarneamordred: thanks! now I realise what I need to do.13:44
mordredssbarnea: woot! lemme know if you have any issues - it _should_ 'just work' for you :)13:44
ssbarneain case someone wonders, i am testing use of molecule to test ansilbe roles,... so far is amazing!13:45
AJaegerfungi, could you review these two changes for infra-manual, please? https://review.openstack.org/614017 and https://review.openstack.org/60045313:45
AJaegerfungi, and also https://review.openstack.org/#/c/609731/13:46
*** weshay has joined #openstack-infra13:47
AJaegerssbarnea: do you want to update https://review.openstack.org/#/c/607151/1 - or abandon?13:47
mordredssbarnea: awesome!13:47
ssbarneaAJaeger: it should be fixed, but was far from a priority.13:48
*** sthussey has joined #openstack-infra13:49
AJaegerssbarnea: ok13:49
ssbarneaAJaeger: the problem is that i need to find a way to avoid having a section with duplicate name, "Work In Progress" in two files.13:49
ssbarneaif you can hint which names to use, it would be fixable.13:49
*** jcoufal has joined #openstack-infra13:50
ssbarneaAJaeger this is because autosectionlabel feature requires that (and is very useful feature to use).13:50
*** wznoinsk has joined #openstack-infra13:50
AJaegerssbarnea: for the single link you have, it's not worth using that feature, just use a normal label...13:51
ssbarneaAJaeger as in using a hardcoded full url instead?13:53
AJaegerssbarnea: no, as in using a label. Just place one where you need it and reference it.13:54
AJaegerssbarnea: http://git.openstack.org/cgit/openstack-infra/infra-manual/tree/doc/source/drivers.rst#n3 is a label that is used elsewhere...13:55
AJaegerthanks, fungi!13:55
AJaegerany other infra-manual reviewer around to review https://review.openstack.org/600453 , please?13:56
fungithanks for highlighting those13:56
AJaegernp14:00
*** dklyle has quit IRC14:02
*** david-lyle has joined #openstack-infra14:02
openstackgerritMerged openstack-infra/nodepool master: Revert "Cleanup down ports"  https://review.openstack.org/61419814:05
ssbarneaAJaeger : funny i fail to find the correct section i was truing to point to in that review, the one documenting creation of gerrit groups.14:05
*** d0ugal has joined #openstack-infra14:06
AJaegerssbarnea: https://docs.openstack.org/infra/manual/creators.html#id1 ?14:08
openstackgerritSorin Sbarnea proposed openstack-infra/infra-manual master: Makes documentation easier to read  https://review.openstack.org/60715114:20
*** yamamoto has joined #openstack-infra14:21
*** kiennt26 has joined #openstack-infra14:21
AJaegerssbarnea: thanks, left a comment ^14:23
*** yamamoto has quit IRC14:26
openstackgerritSorin Sbarnea proposed openstack-infra/infra-manual master: Makes documentation easier to read  https://review.openstack.org/60715114:27
*** eharney has joined #openstack-infra14:28
ssbarneaAJaeger: anything agianst adding pre-commit to infra-manual? ... so we avoid end-of-line space kind of issues?14:28
*** d0ugal has quit IRC14:28
AJaegerssbarnea: let's ask others here as well...14:29
ssbarneaindeed. better to raise question before working on a change.14:31
*** pcaruana|elisa| has quit IRC14:33
*** e0ne has quit IRC14:33
*** armax has joined #openstack-infra14:34
*** janki has quit IRC14:35
*** e0ne has joined #openstack-infra14:37
*** psachin has quit IRC14:38
fungii would want some fairly strong consensus across the team before adding any client-side hooks besides the commit hook used by gerrit, and would likely want to have it made consistent for all our team's repos14:40
fungian alternative is to use something like https://github.com/icefox/git-hooks so that you can have optional global git hooks for your dev environment14:41
fungi(it's unfortunate that git doesn't provide a solution to that problem innately)14:41
openstackgerritMerged openstack-infra/infra-manual master: Add info on needing publish-to-pypi template  https://review.openstack.org/61401714:45
openstackgerritMerged openstack-infra/infra-manual master: Remove reference to x-vrif-minus-2  https://review.openstack.org/60973114:45
*** pcaruana|elisa| has joined #openstack-infra14:45
efriedHowdy y'all. Anyone know anything about http://ci-watch.tintri.com/ ? (not loading up for me)14:46
openstackgerritBenoît Bayszczak proposed openstack-infra/zuul master: Disable Nodepool nodes lock for SKIPPED jobs  https://review.openstack.org/61326114:47
*** dtantsur is now known as dtantsur|brb14:50
*** kiennt26 has quit IRC14:51
mordredefried: nope - I don't know what that is14:53
efried:( Thanks mordred14:53
fungii skyler berg might have been running that?14:57
funginot sure if sberg is still around these days14:57
clarkbfungi: ssbarnea that is also something most editors can solve14:57
fungiclarkb: agreed. mine highlights trailing whitespace for me14:58
clarkbfungi: mine as well14:58
*** yamamoto has joined #openstack-infra14:59
mordredyeah. I actually like keeping git commit hooks to a minimum15:01
*** cfriesen has joined #openstack-infra15:02
*** dave-mccowan has quit IRC15:03
openstackgerritAlex Schultz proposed openstack-infra/project-config master: Fix instack release jobs  https://review.openstack.org/61424515:03
*** jamesmcarthur has joined #openstack-infra15:05
ssbarnea... i installed new extension to vscode for the end of line, still editor specific hack is no replacement for centralized rules.15:05
ssbarneaend-of-line is probably not the best sale case for pre-commit tool.15:06
*** ianychoi_ is now known as ianychoi15:07
ssbarneaclarkb fungi mordred : if one of you is not aware about how pre-commit (tool!) works, please read https://medium.com/@sbarnea/embracing-pre-commit-hooks-4ef1f4e7291415:08
fungiyou can also run `git diff --check` if you remember15:08
fungithat will complain about trailing whitespace15:08
ssbarneai can make a demo change if needed, it saves a lot of time by being much faster than old tox-linters approach15:08
ssbarnea"if you remember" is the key part here ;) ... me getting old...15:09
fungissbarnea: i prefer to let developers configure their own local dev environment to behave in ways they want, rather than overriding the behavior of tools on a per-repo basis15:09
ssbarneafungi: pre-commit does not installs itself, but if you enable it it will stay and work. in many repositories I run it manually without installing.15:10
ssbarneawhat I was trying to say is that the magic thing about it is not related to its name, is something different.15:11
ssbarneait does pin linter versions (like we do we python hacking) and it does reuse them at user level, as opposed to tox which will create a virtualenv for each project.15:12
ssbarneameaning that if you use flake8-1.2.3 in 50 projects, it will be installed only once in its temp directory.15:13
ssbarneathe footprint on developer machine is much much smaller15:13
clarkbssbarnea: but what if you have flake8-1.2.2 and 1.2.3 and 1.24 and 2.0.5? (this is why we use tox15:14
ssbarneaclarkb: you can have any versions of flake8 you want, this is not a problem.15:14
clarkbbut if they are globally installed, how are different versions handled15:15
mordredssbarnea: I just don't like pre-commit hooks in my dev environment - so I don't have a problem with something being setup so that people who like them can use them and collaborate on them - but for me, I do not like things running every time I run git commit15:15
ssbarneaall tools versions are pinned inside *your* repo, meaning that only identical versions will be reused.15:15
mordredbecause I run git commit A LOT15:15
mordredas essentially an autosave mechanism15:15
mordredI don't want to even run linters each time I do that15:15
ssbarneamordred: to be clear, using pre-commit, the *tool* does not mean you (or anyone else) have to install any hooks.15:16
Shrewsoh gosh no. same here. 'git commit' == 'Ctrl-S' for me15:16
mordredssbarnea: yes - totally15:16
ssbarnea... i know, the name can be bit confusing.15:16
AJaegermordred, clarkb , could either of you review this infra-manual change, please? https://review.openstack.org/60045315:16
mordredssbarnea: I'm just saying - I don't have a problem with support for it existing - we just shouldn't grow an assumption that everyone is using it15:16
clarkbAJaeger: yes!15:17
AJaegerthanks, clarkb15:17
clarkbmordred: wins15:17
AJaegerclarkb: mordred won - thanks!15:17
*** anteaya has joined #openstack-infra15:18
ssbarneamordred: ok, i like its other benefits that I described, the hook part is totally optional and i usually do not use it either. no need to worry about "magic happening on commit"15:18
*** yamamoto has quit IRC15:20
*** Tim_ok has joined #openstack-infra15:21
*** yamamoto has joined #openstack-infra15:23
*** e0ne has quit IRC15:25
rajinirhttp://ci-watch.tintri.com/ is down. Anyone know anything?15:26
AJaegerrajinir: see backscroll, we are not involved with it.15:27
*** yamamoto has quit IRC15:28
rajinirAJaeger>  Is there an alternative link where the thirdparty CI Status can be checked?15:28
AJaegerrajinir: no idea15:29
*** gyee has joined #openstack-infra15:29
*** kiennt26 has joined #openstack-infra15:31
*** agopi is now known as agopi|brb15:32
fungirajinir: maybe http://stackalytics.com/report/driverlog (which we also don't run)15:33
rajinir<fungi> thanks15:33
mmedvederajinir: efried: I have sent a message to tinrti openstack-dev contact email. Hopefully they are still willing to support the service.15:34
*** pcaruana|elisa| has quit IRC15:35
openstackgerritMerged openstack-infra/infra-manual master: Mention that docs-on-readthedocs should stay in project-config  https://review.openstack.org/60045315:35
*** e0ne has joined #openstack-infra15:35
rajinir<mmedvede> thanks for the update15:35
*** agopi|brb has quit IRC15:37
Tim_okWhere can I find working instructions on how to setup a OpenStack third party CI server using zuul3 on a CentOS7 VM?  The ones posted here:15:38
Tim_okhttps://docs.oepstack.org/infra/openstackci/third_porty_ci.html#run-masterless-puppet do not work (with CentOS7)...15:39
clarkbTim_ok: https://specs.openstack.org/openstack-infra/infra-specs/specs/zuulv3-3rd-party-ci.html is our current thoughtso n that. Basically here are the tools we expect to work and be tested (including on red hat distro), but until someone goes through the process we aren't sure ourselves15:40
clarkbTim_ok: that spec outlines where someone willing to do that work could start15:40
*** dpawlik has quit IRC15:43
*** dpawlik has joined #openstack-infra15:44
*** quiquell is now known as quiquell|off15:48
*** ccamacho has quit IRC15:48
*** dpawlik has quit IRC15:49
*** kiennt26 has quit IRC15:49
*** ykarel is now known as ykarel|away15:50
*** pcaruana|elisa| has joined #openstack-infra15:50
*** adriancz has quit IRC15:52
*** quiquell|off has quit IRC15:52
*** yamamoto has joined #openstack-infra15:57
*** pcaruana|elisa| has quit IRC16:00
*** cfriesen has left #openstack-infra16:08
openstackgerritMerged openstack-infra/elastic-recheck master: Add query for (anti-)affinity nova functional test race bug 1800472  https://review.openstack.org/61420716:09
openstackbug 1800472 in OpenStack Compute (nova) "nova.tests.functional.test_server_group.ServerGroupTestV264.test_boot_servers_with_affinity_no_valid_host intermittently failing with "OpenStackApiNotFoundException: Item not found"" [Medium,Confirmed] https://launchpad.net/bugs/180047216:09
openstackgerritClark Boylan proposed openstack-infra/zuul master: Small script to scrape Zuul job node usage  https://review.openstack.org/61367416:10
efriedmmedvede: Thanks!16:11
*** jtomasek_ has quit IRC16:12
*** lpetrut has quit IRC16:16
openstackgerritClark Boylan proposed openstack-infra/zuul master: Small script to scrape Zuul job node usage  https://review.openstack.org/61367416:18
*** dave-mccowan has joined #openstack-infra16:23
Tim_okclarkb: Thanks, I look into taking this on with approval of my mgt. team; of course.  Regards, Tim16:25
*** dtantsur|brb is now known as dtantsur16:26
clarkbTim_ok: fwiw we are happy to help, we just don't run third party ci so aren't in the greatest spot to do the work16:27
Tim_ok<clarkb>  Thanks for the offer.  Q:  where is the third party CI IRC group?  is there one?16:30
*** e0ne has quit IRC16:30
rajinirTim_ok: #openstack-third-party-ci16:30
*** trown is now known as trown|lunch16:33
Tim_okThank you, rajinir16:34
*** agopi|brb has joined #openstack-infra16:41
*** udesale has quit IRC16:43
*** ykarel has joined #openstack-infra16:46
*** erlon has joined #openstack-infra16:47
*** ykarel|away has quit IRC16:48
*** armax has quit IRC16:50
*** jrist has quit IRC16:51
*** agopi|brb is now known as agopi16:53
*** jrist has joined #openstack-infra16:54
*** hashar is now known as hasharAway16:55
*** quiquell has joined #openstack-infra16:56
*** quiquell is now known as quiquell|off16:56
*** rossella_s has quit IRC16:59
*** rossella_s has joined #openstack-infra17:01
*** rossella_s has quit IRC17:01
fungibandersnatch 3.0.0 changelog includes "Add `bandersnatch verify` subcommand to re-download + delete unneeded packages"17:02
clarkbmnaser: https://review.openstack.org/614305 cleans up some non voting josb in the puppet integration test templates17:02
fungithat could make running a bandersnatch mirror viable again17:02
clarkbmnaser: any idea who I should talk to about cleaning stuff like that in general re puppet?17:02
*** florianf is now known as florianf|afk17:03
*** rossella_s has joined #openstack-infra17:03
clarkbfungi: interesting. We'd still need to improve our blacklisting of packages to be more complete, but ya that should make it possible to clean out our existing mirror17:03
clarkbfungi: any chance you want to try doing that manually on the existing volume to see how small it becomes with the existing blacklist?17:03
fungiclarkb: it also mentions "Add whitelist + blacklist plugins" so i wonder if someone's maintaining a plugin which does what we want?17:04
clarkboh maybe17:04
clarkbwould be great if there was a precanned go back tio the way pypi was plugin :)17:04
fungipypi classic17:04
fungihttps://github.com/pypa/bandersnatch/blob/master/src/bandersnatch_filter_plugins/blacklist_name.py17:05
openstackgerritFabien Boucher proposed openstack-infra/zuul master: WIP - Pagure driver  https://review.openstack.org/60440417:05
clarkbfungi: looks like it would take our existing list then and not provide its own17:06
clarkbshould still be a good test to see how much smaller the mirror is after cleaning out those unwanted packages17:06
fungii'll see what i can figure out on mirror-update.o.o for that17:07
*** jpich has quit IRC17:08
fungilooks like we removed the crontab for updating it17:08
fungier, crontab entry i mean17:08
clarkbya17:08
clarkbbut the volume and support framework to do the updates should still be there17:08
fungihttps://git.openstack.org/cgit/openstack-infra/puppet-bandersnatch/tree/manifests/cron.pp#n2317:09
fungiyeah17:09
*** ykarel is now known as ykarel|away17:11
AJaegerfungi, https://review.openstack.org/607151 is another infra-manual change if you have time ...17:11
*** rossella_s has quit IRC17:11
fungiclarkb: currently using 1999990784 of 2000000000 quota on volume mirror.pypi17:13
clarkbya that is why we turned it off17:13
fungiseeing if this can move the needle17:13
clarkb++17:13
fungijust wanted a starting number17:13
*** rpioso is now known as rpioso|afk17:14
fungiugh, bandersnatch wants aiohttp==3.4.4 but it appears that "Requires: Python >=3.5.3"17:16
*** ykarel|away has quit IRC17:16
clarkbugh that was why we removed aiohttp from zuul too :/17:16
fungixenial has "Python 3.5.2"17:16
clarkbwhat is weird is aiohttp is a webserver I thought17:18
clarkbwhy does bandersnatch need a webserver17:18
clarkbmaybe it will serve the mirror itself now? and maybe we can disable that?17:18
mordredoh good. just what I always wanted. a web server17:18
clarkbreadme says it needs >= 3.6.117:18
mordredclarkb: what was that thing that was next to us at ansiblefest?17:19
clarkbmordred: pulp17:19
clarkbiirc17:19
mordredclarkb: they did pypi too ... should we look at whether that would be more pleasinger?17:19
fungilove detailed commit messages like https://github.com/aio-libs/aiohttp/commit/bc7bb0f17:19
fungiyeah, looks like bandersnatch 3.0.0 needs some named tuple behaviors specific to python>=3.6.117:22
fungiboo17:22
*** lpetrut has joined #openstack-infra17:23
*** rfolco|rover has joined #openstack-infra17:23
*** ccamacho has joined #openstack-infra17:23
clarkbmordred: fungi https://pulpproject.org/ is the thing that was at ansiblefest17:23
*** dmsimard has quit IRC17:24
*** yamamoto has quit IRC17:24
clarkbit possibly would solve a bunch of issues, we'd likely need to sort out if afs makes sense in its world view though17:24
clarkb(I have no idea)17:24
*** ykarel|away has joined #openstack-infra17:26
rfolco|roverany known issues in reaching docker registry ? I've noticed failures in jobs17:28
*** kopecmartin is now known as kopecmartin|off17:30
*** derekh has quit IRC17:30
clarkbrfolco|rover: not known to me, can you link to examples and do those examples have logs showing what was connected to?17:31
rfolco|roverhttp://logs.openstack.org/00/611800/2/gate/tripleo-ci-centos-7-scenario001-multinode-oooq-container/4d223f2/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz#_2018-10-30_10_28_1017:32
rfolco|roverHTTPError: 503 Server Error: Service Unavailable for url: https://registry-1.docker.io/v2/tripleomaster/centos-binary-nova-api/tags/list"]17:32
clarkbrfolco|rover: Server Error: Service Unavailable for url: https://registry-1.docker.io/v2/tripleomaster/centos-binary-nova-api/tags/list" that appears to talk directly to docker and not use our proxy caches17:32
clarkbrfolco|rover: our proxy caches are similar to http://mirror.dfw.rax.openstack.org:8080/docker (I forget exact url)17:34
rfolco|roverhmm.. cool thanks for clarifying17:34
rfolco|roverclarkb, I've seen also http://mirror.regionone.limestone.openstack.org:8080/buildlogs.centos/centos/7/virt/x86_64/container/repodata/repomd.xml: [Errno 14] HTTP Error 502 - Bad Gateway"]17:35
rfolco|roverin http://logs.openstack.org/07/612607/4/gate/tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates/f1f926a/job-output.txt.gz17:35
rfolco|roverthat's why I associated to networking issues17:35
clarkbrfolco|rover: the 502 is probably related to docker failing behind our proxy caches when the data isn't cached or needs to be refreshed17:36
clarkbrfolco|rover: well a 500 error from the http server isn't a network error, it is an application layer error17:36
clarkbbasically the caches only help if they have the data cached already when the upstream server is broken17:36
*** panda is now known as panda|off17:37
rfolco|roverclarkb, thank you17:39
clarkb(there may be network issues on the backend though)17:42
rfolco|roverweshay, ^17:44
*** jamesmcarthur has quit IRC17:48
clarkbfyi my email provider (fastmail) is currently experiencing an outage17:49
clarkbso if I don't respond to email you know why17:50
clarkboh maybe it is back again17:50
*** electrofelix has quit IRC17:50
*** trown|lunch is now known as trown17:50
fungifastfixed17:53
*** bobh has quit IRC17:53
*** jcoufal has quit IRC17:53
*** armax has joined #openstack-infra17:54
*** bobh has joined #openstack-infra17:55
*** shardy_ has quit IRC17:55
*** dtantsur is now known as dtantsur|afk17:57
*** ansmith has quit IRC17:57
*** bobh has quit IRC17:59
*** yamamoto has joined #openstack-infra18:03
*** dpawlik has joined #openstack-infra18:03
AJaegerconfig-core, please review https://review.openstack.org/614245 to help with instack retirement - and infra-manual reviewers, please look at https://review.openstack.org/60715118:05
*** chandankumar is now known as chkumar|off18:05
*** jpena is now known as jpena|off18:10
*** e0ne has joined #openstack-infra18:10
*** yamamoto has quit IRC18:12
*** ansmith has joined #openstack-infra18:12
*** rpioso|afk is now known as rpioso18:13
AJaegerthanks, clarkb !18:18
*** jamesmcarthur has joined #openstack-infra18:19
*** dpawlik has quit IRC18:20
*** jamesmcarthur has quit IRC18:22
*** diablo_rojo has joined #openstack-infra18:24
openstackgerritMerged openstack-infra/infra-manual master: Makes documentation easier to read  https://review.openstack.org/60715118:25
*** e0ne has quit IRC18:25
openstackgerritMerged openstack-infra/project-config master: Fix instack release jobs  https://review.openstack.org/61424518:29
*** bobh has joined #openstack-infra18:29
*** lpetrut has quit IRC18:29
clarkbI'm going to pop out for a bit before the meeting, but back by then18:30
*** bobh has quit IRC18:33
*** jcoufal has joined #openstack-infra18:34
*** ansmith has quit IRC18:34
*** dpawlik has joined #openstack-infra18:36
*** dpawlik has quit IRC18:40
*** bobh has joined #openstack-infra18:41
*** ansmith has joined #openstack-infra18:55
clarkbOk I'm back and we are about 4 minute away from the meeting. See you there18:57
fungigood timing18:57
*** shardy has quit IRC19:06
*** e0ne has joined #openstack-infra19:13
*** weshay is now known as weshay_doctor19:15
*** jamesmcarthur has joined #openstack-infra19:15
*** shardy has joined #openstack-infra19:18
*** dmsimard has joined #openstack-infra19:23
openstackgerritDirk Mueller proposed openstack/diskimage-builder master: Update test coverage for openSUSE-minial to 15.0  https://review.openstack.org/61043619:29
*** ykarel|away has quit IRC19:32
*** ykarel|away has joined #openstack-infra19:32
*** ykarel|away has quit IRC19:38
*** gfidente has quit IRC19:47
*** armax_ has joined #openstack-infra19:49
*** armax has quit IRC19:50
*** armax_ is now known as armax19:50
diablo_rojoSorry for missing the meeting fungi and clarkb was at the doctor getting my knee adjusted again19:50
*** gary_perkins has joined #openstack-infra19:51
fungigotta keep those knees in tune!19:52
diablo_rojofungi, more like: gotta keep that meniscus flat!19:54
diablo_rojoI hope to get an updated patchset out there for review today- assuming nothing crops up as higher prioirity.19:56
*** shardy has quit IRC20:03
mordreddiablo_rojo: I dunno - I prefer a concave meniscus myself20:03
diablo_rojomordred, I'd take concave over folded flat like the first step in most origami20:04
fungiouch20:05
*** pcaruana|elisa| has joined #openstack-infra20:12
Tim_okwhois clarkb20:12
*** shardy has joined #openstack-infra20:12
*** _ari_ has quit IRC20:13
*** rascasoft has quit IRC20:13
fungiour fearless leader20:16
*** rascasoft has joined #openstack-infra20:16
anteayaand rather a nice chap20:16
*** ralonsoh has quit IRC20:27
*** kgiusti has left #openstack-infra20:29
*** pcaruana|elisa| has quit IRC20:32
*** agopi is now known as agopi|brb20:33
*** imacdonn has quit IRC20:33
*** priteau has quit IRC20:33
*** imacdonn has joined #openstack-infra20:34
clarkbaww20:34
clarkbanyone know why stats.timers.zuul.tenant.openstack.pipeline.check.resident_time.median values are so low? do we only start the timer once the change is active?20:40
clarkbtrying to figure out if we have existing data captured on how long changes sit in the check queue as part of the backlog and ^ has data but it doesn't seem to reflect reality20:41
*** agopi|brb has quit IRC20:41
*** ansmith has quit IRC20:42
clarkboh or maybe the data needs scaling and I'm looking at raw values?20:43
*** weshay_doctor is now known as weshay20:51
clarkbya the raw value is in microseconds I think20:53
*** shardy has quit IRC20:55
clarkbsorry milliseconds20:57
*** bobh has quit IRC21:00
*** erlon has quit IRC21:02
fungiyeah, i thought resident time incremented even for inactive changes?21:03
clarkbya it does http://graphite.openstack.org/render/?from=-1day&target=scale(stats.timers.zuul.tenant.openstack.pipeline.check.resident_time.mean,%200.00001666666) shows the data as I expect21:05
clarkbnow to add some time ranges and respond to bnemec21:05
*** jcoufal has quit IRC21:06
*** boden has quit IRC21:06
*** trown is now known as trown|outtypewww21:09
*** e0ne has quit IRC21:18
*** eharney has quit IRC21:21
*** mriedem has quit IRC21:30
clarkbhttp://graphite.openstack.org/render/?height=1024&width=1280&from=20180501&until=20181031&target=scale(stats.timers.zuul.tenant.openstack.pipeline.check.resident_time.mean,%200.00001666666) is a neat graph21:43
clarkbas is http://graphite.openstack.org/render/?height=1024&width=1280&from=20180501&until=20181031&target=scale(stats.timers.zuul.tenant.openstack.pipeline.gate.resident_time.mean,%200.00001666666)21:43
clarkbshows mean time in queue for check and gate queues. Since May21:43
*** jamesmcarthur has quit IRC21:46
mordredclarkb: nice21:47
clarkbin minutes21:51
clarkbits neat how closely that correlates to "return from ptg"21:52
*** dmsimard has quit IRC21:56
clarkbskimming nodepool server contents there are some things in ~nodepool that may be interesting to people? mordred and ianw in particular they appear to have your fingerprints on them (patches and scripts related to shade)21:59
clarkbnodepool.o.o is not a server we backup because all the nodepool running data is "stateless" due to clouds storing the info21:59
clarkbmordred: ianw care to take a quick look before we delete the server?22:00
clarkbcorvus also has an open shell, but I am goingto assume we can ignore that22:01
mordredclarkb: there's a clean-ports.py in my home directory :)22:02
mordredclarkb: nope. nothing needed22:03
clarkbmordred: any sense if we should wait for corvus? Looking in that homedir its a lot of stuff related to when gearman was used more. Which I dobut we need today22:04
fungicould just tar it up and drop it on zuul.o.o or something, worst case22:06
clarkbya, though I'd prefer us to not need to do that, particular if we don't backup the server :)22:06
clarkbbut recognize this server is like 5 years old22:06
clarkbso it may have grown unexpected things22:06
fungimy best guess is that anybody who needed anything off that server has already copied it22:07
fungibut can always take a final snapshot after shutting it down, prior to deletion22:07
clarkbthats a good point22:07
fungii've done that with a number of other servers we've deleted22:07
clarkbI'll do that instaed /me goes to do that now22:07
*** david-lyle has quit IRC22:08
ianwclarkb: yeah, nothing i want in ~nodepool, /root or ~ianw thanks22:09
clarkbfungi: `openstack server image create` look like the right command to you?22:09
clarkbhelp says "Create a new server disk image from an existing server"22:09
mordredyes - that should be the right command22:10
clarkbok it is running22:10
*** yamamoto has joined #openstack-infra22:10
mordredfwiw - the sdk method is "create_image_snapshot"22:10
fungiyeah, that's what i've used in the past22:11
mordredthat method has one of my better comments in it22:11
clarkbalright when that is done I am deleting acfa8539-10a2-4bc4-aabc-e324aa855c70 nodepool.openstack.org if anyone wants to double check those values too22:12
fungiserver show says that's the one22:14
mordred'id': 'acfa8539-10a2-4bc4-aabc-e324aa855c70'22:15
mordredI agree with fungi22:15
*** yamamoto has quit IRC22:15
clarkbstatus: queued, I'm very impatient :P22:16
openstackgerritIan Wienand proposed openstack-infra/system-config master: Add yamlgroup inventory plugin  https://review.openstack.org/60238522:26
ianwinfra-root / mordred : ^ i've re-synced that, i think only the futureparser bits got out of whack.  are there more thoughts on this?22:26
ianwif we're going to bring up non-puppeted/ansible-only/docker-running hosts i think we want to do it like that22:27
clarkbianw: fixing that would make me really happy, as we do still have a several minute ansible startup time22:27
clarkbso ansible for ad hoc stuff isn't very useful currently22:27
*** dklyle has joined #openstack-infra22:28
ianwclarkb: yeah, from my pov, i want to have a "docker" matcher like https://review.openstack.org/#/c/605585/4/inventory/groups.yaml22:28
*** xek has quit IRC22:31
clarkbwe have a mysqld running on nodepool?22:31
clarkboh right from the land before time22:31
clarkbthis is what I think we should call the long long ago when nodepool and zuul coordinated over gearman22:32
fungiwith jenkins even22:33
*** agopi|brb has joined #openstack-infra22:33
clarkbthat is weird, my snapshot is no longer listable or showable. I wonder if there is a transition state?22:33
*** agopi|brb is now known as agopi22:34
clarkbI'll give it another 10-15 minutes and if it doesn't show up again try again?22:34
clarkbanyone ever seen ^ before?22:34
*** erlon has joined #openstack-infra22:35
ianwnope; i think only one i've ever snapshotted was maybe the translate servers and it worked as i expected22:36
*** dmsimard has joined #openstack-infra22:38
fungido our other snapshots show up for you?22:39
clarkbfungi: they do22:39
clarkbI'm assuming this one ran into an error of some sort22:40
clarkbunfortunately you don't get to see what that is22:40
anteayais the procrss still running?22:43
*** kjackal_v2 has quit IRC22:45
clarkbanteaya: as far as I can tell it is not22:45
clarkbI'm going to assume it was a random failure and trying again now22:45
anteayasounds reasonable to me22:45
mordredianw, clarkb: I definitely think we should get that yamlgroup plugin landed and in prod22:47
mordredI'm sad I can't get removing hosts from the inventory to work though22:47
*** eernst has joined #openstack-infra22:49
*** e0ne has joined #openstack-infra22:49
*** eernst has quit IRC22:50
*** eernst has joined #openstack-infra22:50
*** eernst has quit IRC22:54
*** tonyb has joined #openstack-infra23:03
ianwmordred: do we need a migration other than just merging it and babysitting?  emegency file maybe?23:10
clarkbok second snapshot did work. I'm going to delete the nodepool.o.o server nowish23:11
clarkbianw: before I forget there was a port deletion bug in openstacksdk that we need to be careful around23:11
clarkbianw: tobiash reverted the change in nodepool and openstacksdk is on its way to being fixed23:12
mordredyeah. the fix has landed in openstacksdk - waiting on a 0.19.0 release23:12
mordredbut since we reverted the code in nodepool, we should be _generall_ safe23:12
mordredianw: I don't think there needs to be any sort of migration - just merging and babysitting should be fine23:12
mordredianw: of course, given what it's touching, I'd likely want to babysit it with a root shell already open - just in case23:13
*** e0ne has quit IRC23:13
ianwclarkb: yeah, i noticed that and i think it broke the dib/nodepool tests ... actually i think tobiash probably had an easier time figuring it out with the mass port deletion as it was very obvious ... in the job it was manifesting just as a lack of ssh-ability :)23:15
*** rpioso is now known as rpioso|afk23:15
clarkb#status log Old nodepool.openstack.org acfa8539-10a2-4bc4-aabc-e324aa855c70 deleted as we no longer use any services on this host. Filesystem snapshot saved and called nodepool.openstack.org-20181030.223:16
openstackstatusclarkb: finished logging23:16
clarkbnow to clean up DNS23:16
clarkbif anyone is wondering the web ui for rackspcae has changed23:18
clarkbwhen you login you have to select the product in the top left toget the rax cloud stuff23:18
*** tpsilva has quit IRC23:18
mordredclarkb: awesome23:19
clarkbdns is cleaned up23:20
clarkbone less server to worry about23:20
*** harlowja has quit IRC23:22
*** yamamoto has joined #openstack-infra23:23
openstackgerritIan Wienand proposed openstack-infra/nodepool master: Logs stats for nodepool automated cleanup  https://review.openstack.org/61407423:27
openstackgerritIan Wienand proposed openstack-infra/nodepool master: Revert "Revert "Cleanup down ports""  https://review.openstack.org/61437023:27
ianwoh, probably should have waited for the release before pushing that, oh well23:28
mordredianw: nah - it's fine - just don't land the revert revert :)23:28
mordredianw: https://review.openstack.org/#/c/614039/ is the release patch, fwiw23:29
*** armax has quit IRC23:29
clarkband nl0* are disabled in emergency file until everything gets settled23:29
ianwmordred: yeah, it won't as it has the new pinned requirement; anyway we should run it through "check experimental" to do the devstack tests i think23:30
ianwdoes openstacksdk run them?23:30
ianwyeah, it seems to ... hrm, i wonder why dib gate was unhappy23:31
*** erlon has quit IRC23:35
*** sthussey has quit IRC23:35
*** harlowja has joined #openstack-infra23:36
*** rlandy has quit IRC23:46
*** erlon has joined #openstack-infra23:48
*** harlowja has quit IRC23:49
tonybmordred, ianw: Just waiting for zuul to report on the openstacksdk release23:56
tonybmordred: but adding it to constraints will get held up because of the cliff issue.23:57

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