Tuesday, 2015-07-07

*** sdake_ has joined #openstack-infra00:00
*** dannywilson has quit IRC00:00
lifelesssigmavirus24_awa: I nearly do. It will be fun to see how many people my first release of mock breaks00:00
lifelessThings I never want to see in a commit message. "    a mock function call. This will help to write better doctests.00:01
*** esmute__ has joined #openstack-infra00:01
*** dims has joined #openstack-infra00:01
*** sdake has quit IRC00:02
ianwdo we know that etherpad seems to be a bit stuck?00:03
clarkbianw: no00:03
clarkbianw: https://etherpad.openstack.org/p/neutron-partial-upgrades is working for me00:04
ianwhttps://etherpad.openstack.org/p/oom-in-rax-centos7-CI-job is being weird00:04
ianwfor me, anyway00:04
*** esmute__ has quit IRC00:05
lifelessianw: ahha. pounce! - dunno if you saw my request for devstack +2A's00:05
*** laughterwym has joined #openstack-infra00:05
jeblairianw: i think that one is working for me00:06
ianwi dunno, now it seems to work.  it was stuck loading, then it flashed some sort of error page quickly (it said something like press f5 to refresh), now it works00:06
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Raise exception for nova egress secgroup rule  https://review.openstack.org/19887100:06
ianwlifeless: is this the backports in the queue?00:06
clarkbhrm maybe it restarted00:07
jeblairianw: http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=1200:07
jeblairianw: something definitiely is/was going on00:07
*** _amrith_ is now known as amrith00:07
Shrewsmordred: greghaynes: turns out https://review.openstack.org/198871 had a genuine pep8 error (line too long). not sure where the output went00:07
jeblairclarkb: did you get a chance to test the gertty fix?00:07
clarkbjeblair: not yet let me do that now00:07
lifelessianw: yes. Needed pre-conditions to enable fixing the check-*-requirements jobs in kilo and juno00:08
*** chlong has quit IRC00:08
greghaynesShrews: psh, having fail output removes too much of the fun of debugging00:08
*** sdake has joined #openstack-infra00:09
*** wenlock has quit IRC00:09
*** Sukhdev has quit IRC00:10
*** fallenpe_ has quit IRC00:11
clarkbjeblair: I get some sql error now, but I think maybe it "corrupted" its db relative to where it expects it to be via the prvious fail00:11
jeblairclarkb: urg.  sorry.  i might be able to help you reverse that, or we could declare it fixed if you want to continue using your new db.00:12
clarkbI am happy to declare it fixed00:12
openstackgerritMerged stackforge/gertty: Fix get_repo call even more  https://review.openstack.org/19887500:13
jeblairso declared00:13
*** r-daneel has quit IRC00:13
*** sdake_ has quit IRC00:13
anteayaits fixed00:13
anteayaso glad I was here for it00:13
clarkbI can confirm that that change works with my working db00:14
*** baoli has quit IRC00:14
*** baoli has joined #openstack-infra00:15
*** yamamoto has joined #openstack-infra00:17
openstackgerritPaul Van Eck proposed openstack-infra/puppet-refstack: Add .gitreview & fix puppet-lint errors/warnings  https://review.openstack.org/19891000:17
*** rfolco has joined #openstack-infra00:18
openstackgerritPaul Van Eck proposed openstack-infra/puppet-refstack: Add .gitreview & fix puppet-lint errors/warnings  https://review.openstack.org/19891000:20
*** asettle has joined #openstack-infra00:20
*** asettle has quit IRC00:20
*** asettle has joined #openstack-infra00:21
*** yamamoto has quit IRC00:21
jeblairclarkb: i wanted to make sure that https://review.openstack.org/189316 looked good before removing the d-g jobs (even though the depends-on goes the other way just for safety)00:22
*** esmute__ has joined #openstack-infra00:22
jeblairclarkb: there was stuff happening at the time which i think is cleared up now, so i have rechecked that, and assuming it comes back clean, i'll unwip the job removal change00:22
*** esmute__ has quit IRC00:24
*** tsekiyama has quit IRC00:26
*** esmute__ has joined #openstack-infra00:27
*** HGarg has joined #openstack-infra00:27
*** HGarg has left #openstack-infra00:27
*** kgiusti has left #openstack-infra00:32
*** btully has joined #openstack-infra00:32
*** ddieterly has joined #openstack-infra00:32
jeblairianw: whatever was affecting etherpad, it seems to have been transient00:32
*** esmute__ has quit IRC00:33
*** esmute__ has joined #openstack-infra00:34
*** heyongli has quit IRC00:35
*** heyongli has joined #openstack-infra00:35
*** tqtran has quit IRC00:39
lifelessspekaing of gertty00:41
lifelessjeblair: when I review a change, it gets hidden (as I expect, I have show-reviewed-off), then it gets shown again00:42
lifelessjeblair: but the only change to the review I can see is that my comment has now been applied.00:42
lifelessjeblair: is this a known defect (do you agree that its a defect?) or would you like me to file a bug00:42
jeblairlifeless: ah, i noticed that too recently; i think it's something i overlooked in the change to add 'held changes'00:43
jeblairlifeless: i agree it's a defect; it's on my mental bug list; a formal bug would not hurt, but i'm also not likely to forget it if you don't feel like bothering00:43
lifelessthanks00:45
*** mayurig has quit IRC00:46
lifelessjeblair: https://storyboard.openstack.org/#!/story/200031300:46
*** dims has quit IRC00:50
*** hdd has joined #openstack-infra00:51
*** fawadkhaliq has joined #openstack-infra00:53
*** jyuso1 has quit IRC00:54
*** jyuso1 has joined #openstack-infra00:54
lifelesshmm00:55
lifelessI see some git yak shaving in my future00:55
*** fawadkhaliq has quit IRC00:56
*** vnod has quit IRC00:57
*** Swami has quit IRC00:57
ianwjeblair: yeah, page has been there for a while and i haven't been kicked off.  glitch in the matrix00:58
*** dims has joined #openstack-infra00:58
*** nelsnels_ has joined #openstack-infra00:59
*** nelsnels_ has quit IRC00:59
*** woodster_ has quit IRC01:01
*** nelsnelson has quit IRC01:02
*** amotoki has joined #openstack-infra01:02
*** zz_dimtruck is now known as dimtruck01:03
*** cody-somerville has quit IRC01:06
*** esmute__ has quit IRC01:07
*** erlon has quit IRC01:10
*** jasondotstar has quit IRC01:12
*** achanda has joined #openstack-infra01:13
*** gyee has quit IRC01:13
*** shashank_hegde has quit IRC01:14
*** Longgeek has joined #openstack-infra01:18
*** weshay has joined #openstack-infra01:21
*** mtanino has quit IRC01:22
*** yamamoto has joined #openstack-infra01:22
*** jasondotstar has joined #openstack-infra01:25
*** yamamoto has quit IRC01:27
*** chlong has joined #openstack-infra01:30
openstackgerritMerged openstack/diskimage-builder: Remove deprecated disk-image-get-kernel  https://review.openstack.org/19881001:37
*** patrickeast has quit IRC01:38
*** patrickeast has joined #openstack-infra01:41
*** asettle has quit IRC01:45
jheskethfungi: ping01:47
*** cody-somerville has joined #openstack-infra01:47
*** ari has joined #openstack-infra01:47
*** Longgeek has quit IRC01:51
*** tonytan4ever has joined #openstack-infra01:52
*** btully has quit IRC01:53
*** patrickeast has quit IRC01:55
openstackgerritfumihiko kakuma proposed openstack-infra/project-config: Make gate-networking-ofagent-python34 non-voting  https://review.openstack.org/19865702:00
openstackgerritMerged openstack-infra/shade: Raise exception for nova egress secgroup rule  https://review.openstack.org/19887102:04
*** patrickeast has joined #openstack-infra02:08
*** puranamr has quit IRC02:09
openstackgerritChris Hoge proposed openstack-infra/project-config: Add openstack bot to #openstack-diversity channel  https://review.openstack.org/19893602:10
*** yamahata has quit IRC02:10
openstackgerritJoshua Hesketh proposed openstack-infra/project-config: Fix ACL's on projects with puppet-system  https://review.openstack.org/19893802:12
*** PinkFreud has joined #openstack-infra02:13
hogepodgeclarkb: the patch configures accessbot. Do I need to somehow invite it to the room too?02:17
clarkbnope just need to give it perms whoch chanserv says you have done02:17
clarkbbut you also need meetbot if you want those services primarily logs02:18
*** esmute__ has joined #openstack-infra02:18
clarkball the accessbot does us control who has access02:20
clarkbmostly ops02:20
*** dboik_ has joined #openstack-infra02:24
*** spzala has quit IRC02:24
*** ddieterly has quit IRC02:25
clarkbhogepodge what irc things are you wanting?02:27
hogepodgeclarkb: meetbot and logging, and anything else I might me missing02:27
hogepodgeclarkb: there's no repository, so gerritbot not needed02:28
clarkbthe other common thibg is gerritbot ya ok02:28
clarkbso just need to add to the meetbot list. thats in system-config in eavesdrop.pp iirc02:28
openstackgerritChris Hoge proposed openstack-infra/system-config: Add eavesdrop to #openstack-diversity channel  https://review.openstack.org/19894102:29
hogepodgeclarkb: :-)02:30
*** otter768 has quit IRC02:30
*** dboik_ has quit IRC02:31
*** esmute__ has quit IRC02:32
*** yamamoto has joined #openstack-infra02:33
*** pal has joined #openstack-infra02:34
*** ari has quit IRC02:35
*** oomichi has joined #openstack-infra02:37
*** nijaba has joined #openstack-infra02:38
*** nijaba has joined #openstack-infra02:38
openstackgerritMerged openstack-infra/system-config: Rename pypi_slave class to release_slave  https://review.openstack.org/19844402:38
*** esmute__ has joined #openstack-infra02:39
openstackgerritMerged openstack-infra/system-config: Add release.slave.openstack.org  https://review.openstack.org/19844502:44
*** asselin_ has joined #openstack-infra02:45
lifelessI always forget02:51
lifelesswhats02:51
*** BharatK has joined #openstack-infra02:51
lifeless:02:51
lifelesson its own in a shell script do ?02:51
*** patrickeast_ has joined #openstack-infra02:51
*** pal has quit IRC02:51
*** patrickeast has quit IRC02:51
*** patrickeast_ is now known as patrickeast02:51
larskslifeless: it does nothing, but arguments are evaluated, unlike a comment.  It is sort of like the "true" command.02:55
larsksSo while this does nothing:02:56
larsks# ${myvar:=default}02:56
larsksThis will set a myvar to 'default' if it isn't set:02:57
larsks: ${myvar:=default}02:57
*** tonytan4ever has quit IRC02:58
lifelesslarsks: what about -just- : ?02:58
larsksThen it does nothing.02:58
lifelessk, thanks.02:59
*** weshay has quit IRC03:02
*** koolhead17 has joined #openstack-infra03:05
openstackgerritSean M. Collins proposed stackforge/gertty: Catch errno.EPERM  https://review.openstack.org/19894503:11
*** Longgeek has joined #openstack-infra03:18
*** ddieterly has joined #openstack-infra03:26
openstackgerritGuojian Shao proposed openstack-infra/project-config: test against neutron for osc gate job  https://review.openstack.org/19894803:26
*** asettle has joined #openstack-infra03:28
*** Longgeek has quit IRC03:29
*** rkukura has joined #openstack-infra03:30
*** ddieterly has quit IRC03:30
*** julim has joined #openstack-infra03:31
*** rkukura has quit IRC03:33
*** dims has quit IRC03:39
*** heyongli has quit IRC03:44
*** heyongli has joined #openstack-infra03:45
*** mayurig has joined #openstack-infra03:46
*** mayurig has quit IRC03:51
*** ianw has quit IRC03:56
*** ianw has joined #openstack-infra03:56
*** koolhead17 has quit IRC04:07
openstackgerritMerged openstack-infra/irc-meetings: Add id and url for third party meetings.  https://review.openstack.org/19872204:08
*** Krinkle_ is now known as Krinkle04:10
*** btully has joined #openstack-infra04:12
*** yamada-h has joined #openstack-infra04:14
*** pal has joined #openstack-infra04:17
*** julim has quit IRC04:19
*** xyang1 has quit IRC04:24
*** ddieterly has joined #openstack-infra04:27
*** laughterwym has quit IRC04:29
*** otter768 has joined #openstack-infra04:31
*** ddieterly has quit IRC04:31
*** yamada-h has quit IRC04:32
*** otter768 has quit IRC04:36
*** pal_ has joined #openstack-infra04:36
openstackgerritMerged openstack-infra/project-config: Add openstack bot to #openstack-diversity channel  https://review.openstack.org/19893604:38
*** Sukhdev has joined #openstack-infra04:38
*** pal has quit IRC04:39
*** dims has joined #openstack-infra04:40
*** BharatK has quit IRC04:41
*** dims_ has joined #openstack-infra04:41
*** baoli has quit IRC04:43
*** dims has quit IRC04:45
*** GonZo2000 has joined #openstack-infra04:46
*** dims_ has quit IRC04:46
GonZo2000Hi can i have multiple jenkins servers configured on JJB ?04:50
*** hdd has quit IRC04:51
*** ari has joined #openstack-infra04:55
clarkbI dont think so, we do it by runnig it on each server independenrly04:59
clarkbbut iirc the config ini file accepts just one04:59
GonZo2000thanks clarkb that was my idea from reading the docs.05:00
GonZo2000maybe i can use multiple config files05:00
*** ari has quit IRC05:00
*** yfried has quit IRC05:01
*** deepakcs has joined #openstack-infra05:04
*** asselin_ has quit IRC05:05
*** devanand_ has joined #openstack-infra05:05
*** deepakcs has quit IRC05:06
*** BharatK has joined #openstack-infra05:08
*** patrickeast has quit IRC05:13
*** gildub has joined #openstack-infra05:15
*** madong_ is now known as larainema05:20
openstackgerritYAMAMOTO Takashi proposed openstack-infra/project-config: Add python3 jobs for neutron-lbaas  https://review.openstack.org/19861105:21
*** e0ne has joined #openstack-infra05:22
*** _nadya_ has joined #openstack-infra05:23
*** achanda has quit IRC05:23
*** dannywilson has joined #openstack-infra05:23
*** ildikov has joined #openstack-infra05:23
*** deepakcs has joined #openstack-infra05:25
*** dannywilson has quit IRC05:25
*** dannywilson has joined #openstack-infra05:26
*** deepakcs has quit IRC05:26
*** Longgeek has joined #openstack-infra05:26
*** ddieterly has joined #openstack-infra05:28
openstackgerritYAMAMOTO Takashi proposed openstack-infra/project-config: Add python3-jobs for networking-midonet  https://review.openstack.org/19896005:28
ianwfar too much debugging later ... random centos failures are because "devstack-centos7-1435932846.template.openstack.org.novalocal" is too long a hostname to put in a SSL certificate05:30
rm_worklol05:30
rm_workthat's an awesome reason for failure05:30
greghaynesianw: that might go away - we just released glean which will now be setting hostname to just devstack-centos7-143593284605:32
greghaynesonce the images are rebuilt05:32
*** ddieterly has quit IRC05:32
*** flepied has quit IRC05:32
*** esmute__ has quit IRC05:33
*** gmann has quit IRC05:33
ianwgreghaynes: yeah; i mean i guess really it's a rpm (mod_ssl) bug, that's trying to set this up05:33
greghaynesturns out when we make really long hostnames we break a whole lot of things ;)05:34
ianwgreghaynes: so will glean be on all images or is it part of dib setup?05:35
greghaynesianw: all of them05:36
greghaynesif the build and upload works ;)05:36
greghaynesoh, actually, IMBW05:36
greghaynesI think it might just be dib so the hp ones05:36
*** laughterwym has joined #openstack-infra05:37
greghaynesianw: were you seeing that issue in rax too?05:38
ianwgreghaynes: i do think it's only hpcloud05:39
ianwwhat's with the novalocal on the end?05:39
greghaynesdomain?05:40
ianwgreghaynes: yeah, but after templte.openstack.org05:41
greghaynesyea, so the issue is that glean is setting a full fqdn for the hostname which should really just be a bottom level domain, so its setting devstack-centos7-1435932846.template.openstack.org05:41
greghaynesso if something else has the idea that novalocal is the domain, itll append that too05:42
*** BharatK has quit IRC05:42
*** unicell has joined #openstack-infra05:44
*** e0ne has quit IRC05:44
clarkbnovalocal is what nova metadata says is your domain05:45
ianwgreghaynes: ok, so i'll file the mod_ssl bug, and keep an eye on the images tomorrow and see what's going on05:45
*** juggler_ has joined #openstack-infra05:46
*** juggler has quit IRC05:49
*** Trozz_ is now known as Trozz05:50
*** yamahata has joined #openstack-infra05:50
*** ig0r_ has joined #openstack-infra05:52
*** Longgeek has quit IRC05:52
*** deepakcs has joined #openstack-infra05:52
*** heyongli has quit IRC05:53
*** heyongli has joined #openstack-infra05:54
*** _nadya_ has quit IRC05:54
openstackgerritBorneMace proposed openstack-infra/project-config: Added new validate-contents test for kolla  https://review.openstack.org/19896805:54
*** abregman has joined #openstack-infra05:54
*** ig0r__ has quit IRC05:55
*** Longgeek has joined #openstack-infra05:56
*** unicell has quit IRC05:56
*** yamamoto has quit IRC05:57
*** Guest20028 has joined #openstack-infra05:58
*** unicell has joined #openstack-infra05:58
*** fallenpegasus has joined #openstack-infra05:59
*** heyongli has quit IRC05:59
*** yamamoto has joined #openstack-infra06:00
*** heyongli has joined #openstack-infra06:00
*** laughterwym_ has joined #openstack-infra06:01
*** Guest20028 has quit IRC06:01
*** kiran-r has joined #openstack-infra06:02
*** kdas_ has quit IRC06:03
*** laughterwym has quit IRC06:04
*** bmace has joined #openstack-infra06:05
bmacehey folks, i tossed up a review a couple of mins ago and i am just trying to make sure i have the process / syntax right for this change.  https://review.openstack.org/#/c/198968/06:06
*** kdas_ has joined #openstack-infra06:06
bmacethe entry is already added into the tox.ini file in the kolla codebase.06:06
sdakeAJeager_ you happen to be around for that one?06:07
bmaceit isn't super clear to me if i need to define it in jenkins/jobs/kolla.yaml or not.  it doesn't need any extra args or anything.06:07
*** BharatK has joined #openstack-infra06:08
sdakebmace for questions you will probably have to try in the us timezones, for normal reviews, it will follow the normal review process - we don't ping infra unless we have some critical emergency for reviewing a busted-gate patch :)06:10
sdakeso feel free to ask in the morning - but for the pinging on the review please let the folks address the reviews using their normal worflows k?06:11
*** mrunge has joined #openstack-infra06:15
*** flepied has joined #openstack-infra06:16
*** gmann has joined #openstack-infra06:19
ianwgreghaynes: https://bugzilla.redhat.com/show_bug.cgi?id=124049506:19
openstackbugzilla.redhat.com bug 1240495 in httpd "mod_ssl install fails with a long hostname" [Unspecified,New] - Assigned to webstack-team06:19
*** bmace is now known as bmace_away06:23
openstackgerritYAMAMOTO Takashi proposed openstack-infra/project-config: Remove Midonet v2 jobs  https://review.openstack.org/19897906:27
*** _nadya__ has joined #openstack-infra06:27
*** ddieterly has joined #openstack-infra06:28
*** dtantsur|afk is now known as dtantsur06:28
*** armax has quit IRC06:29
*** pal_ has quit IRC06:30
openstackgerritMerged openstack-dev/pbr: This documents the use of comments that work between 2.6, 2.7 and 3.4  https://review.openstack.org/19392706:31
*** otter768 has joined #openstack-infra06:32
*** ddieterly has quit IRC06:33
*** Sukhdev has quit IRC06:34
*** dimtruck is now known as zz_dimtruck06:34
*** otter768 has quit IRC06:37
*** rossella_s has quit IRC06:37
*** rossella_s has joined #openstack-infra06:37
*** sshnaidm has joined #openstack-infra06:38
*** yfried has joined #openstack-infra06:38
*** yamamoto has quit IRC06:41
*** yfried_ has joined #openstack-infra06:43
*** yfried has quit IRC06:43
*** fifieldt has joined #openstack-infra06:43
*** kdas_ has quit IRC06:44
*** asettle has quit IRC06:45
*** scheuran has joined #openstack-infra06:45
clarkbbmace the validate contents job template doesnt exist06:47
*** kdas_ has joined #openstack-infra06:47
clarkbthat is why jenkins is -106:47
clarkbI dont understand us pep8 being replaced?06:48
clarkbwhy not just fix pep8? its all self testing so write a change or a stack of changes to get it where you want06:48
clarkbno new job necessary06:48
*** afazekas has joined #openstack-infra06:50
*** jtomasek has joined #openstack-infra06:50
*** Murad has quit IRC06:53
*** btully has quit IRC06:54
*** k4n0 has joined #openstack-infra06:54
larainemahi all, if someone can help, when I try to  summit patchset to gerrit, I catch this error: http://paste.openstack.org/show/351046/06:54
clarkblarainema you will need to set your contact info at the location provided06:55
openstackgerritDmitry Tantsur proposed openstack-infra/project-config: Add check-ironic-inspector-dsvm-nv to ironic experimental pipeline  https://review.openstack.org/19838106:56
larainemaclarkb, i have checked all the info, i thought it correct, but still not work06:56
clarkbhttps://ask.openstack.org/en/question/56720/cannot-store-contact-information-when-updating-info-in-openstack-gerrit/ may be of help as well06:56
*** e0ne has joined #openstack-infra06:57
*** fhubik has joined #openstack-infra06:57
*** soren has joined #openstack-infra07:00
openstackgerritSteve Kowalik proposed openstack-infra/project-config: Add a register-zanata-projects.py script and job  https://review.openstack.org/19823607:02
clarkblarainema can you  see if that ask url sorts it out?07:03
*** e0ne has quit IRC07:04
*** dannywilson has quit IRC07:04
larainemaclarkb, I can succefully connect over ssh, but git review can't work07:07
*** shashank_hegde has joined #openstack-infra07:08
clarkbyes you need to update your contact info at the url provided07:08
tchaypoupdate it to what?07:09
clarkbif you there and attempt to update it does that fix the issue? if not please see the ask url07:09
clarkbtchaypo whatevr you deem correct, we do t actaully have a way to check it07:09
tchayposo as long as https://review.openstack.org/#/settings/contact has a non-zero value, it should be fine?07:10
tchaypoor does it need to match your launchpad ID or something like that?07:10
*** mpavone has joined #openstack-infra07:10
clarkbit is fine as long as you get the bit flipped to info updated07:11
larainemaclarkb tchaypo, i can update the contact info07:11
larainemaand after updated it works07:11
*** yfried__ has joined #openstack-infra07:11
*** yfried_ has quit IRC07:11
clarkbdoing so requires a foundation member account with a primary email addr that matches the primary email addr in gerrit07:11
clarkbbut thats the only mapping/check07:12
clarkbthe ask answer should cover ^07:12
tchayposo there’s some check to ensure that the email address matches a foundation member, that only fires if you update the details in gerrit?07:13
tchayposo if you provide details to gerrit first, then later become a foundation member, you have to make a no-op update in order to trigger a re-check of the details?07:13
tchaypoor am I misunderstanding?07:14
clarkbno it fails to update without a member account07:14
clarkbso its impossible to go that order07:14
clarkbin this case I think larainema just never provided contact info as part of cla signing07:15
clarkbso it just needes doing07:15
tchaypoand yet, if I’ve understood what he’s said, larainema seems to have got into a situation where https://review.openstack.org/#/settings/contact seemed to show info that matched the foundation member details, but didn’t allow for pushing reviews…07:16
clarkbnope it was never updated07:17
larainematchaypo, I think the issue is i never update the contact info, i just thought only provide the email address07:17
tchaypoah07:18
larainemathanks clarkb tchaypo for helping07:19
*** nilasae has joined #openstack-infra07:20
*** yfried__ has quit IRC07:21
*** oomichi has quit IRC07:22
*** juggler_ is now known as juggler07:23
*** BharatK has quit IRC07:26
*** ParsectiX has joined #openstack-infra07:26
*** ddieterly has joined #openstack-infra07:29
*** dizquierdo has joined #openstack-infra07:30
*** binbincong has quit IRC07:31
tchaypoOh wait07:32
tchaypothis is the page with the in-real-life contact info, that’s stored in a way that the web app can’t decrypt07:32
clarkbyes07:32
tchaypohence every time you look at the form it’s blank, so it’s hard to tell the difference between “I gave you my data” and “I haven’t given you data yet"07:32
*** cdent has joined #openstack-infra07:33
clarkbwell it encrypts it but does not store it locally  and it uses a last updated on to tell you when it was last updated07:33
*** esmute__ has joined #openstack-infra07:34
tchaypoI’m wondering if it’s worth a note in http://docs.openstack.org/infra/manual/developers.html#sign-the-appropriate-individual-contributor-license-agreement - it currently specifically says “make sure the primary email address …"07:34
*** ddieterly has quit IRC07:34
*** arxcruz has joined #openstack-infra07:35
tchaypoperhaps a .. note:: saying that you need to provide the other details as well and won’t be able to proceed unless there’s a last-updated date here would be helpful07:35
*** fhubik is now known as fhubik_afk07:35
*** e0ne has joined #openstack-infra07:36
*** jistr has joined #openstack-infra07:37
*** larainema has quit IRC07:38
*** larainema has joined #openstack-infra07:38
*** heyongli has quit IRC07:39
*** esmute__ has quit IRC07:39
*** heyongli has joined #openstack-infra07:40
*** devanand_ has quit IRC07:41
*** dims has joined #openstack-infra07:43
*** btully has joined #openstack-infra07:43
*** binbincong has joined #openstack-infra07:45
*** laughterwym has joined #openstack-infra07:45
*** yamada-h has joined #openstack-infra07:45
*** btully has quit IRC07:48
*** dims has quit IRC07:48
*** HeOS has quit IRC07:50
*** BharatK has joined #openstack-infra07:54
*** yamada-h has quit IRC07:55
*** macjack has joined #openstack-infra07:56
*** Krinkle is now known as Krinkle_07:57
*** heyongli has quit IRC08:00
*** shashank_hegde has quit IRC08:00
*** heyongli has joined #openstack-infra08:01
*** e0ne is now known as e0ne_08:01
openstackgerritFlavio Percoco proposed openstack-infra/project-config: Move amqp1's devstack plugin under openstack  https://review.openstack.org/19900308:02
*** fhubik_afk is now known as fhubik08:02
*** e0ne_ is now known as e0ne08:03
*** hashar has joined #openstack-infra08:03
eantyshevhello! our CI on cinder has started failing to merge, with "error: Merge requires file-level merging" in merger logs. Anyone has any ideas?08:10
*** koolhead17 has joined #openstack-infra08:11
clarkbfirst check if that is a valid fail and attemlt to do the merge locally08:11
clarkbif that works then next check if the repo happy on the merger with a git fsck08:12
clarkbif that all looks good we likely need more specifics to hunt this down08:12
*** chlong has quit IRC08:13
*** ZZelle has joined #openstack-infra08:13
*** bexelbie has joined #openstack-infra08:16
*** derekh has joined #openstack-infra08:17
eantyshevclarkb: thank you, will start with that08:17
*** notnownikki has joined #openstack-infra08:17
*** ildikov has quit IRC08:17
*** shardy has joined #openstack-infra08:17
*** gildub has quit IRC08:20
*** Krinkle_ is now known as Krinkle08:20
openstackgerritMerged openstack-infra/irc-meetings: Update neutron-drivers team meeting times.  https://review.openstack.org/19880608:20
*** yfried__ has joined #openstack-infra08:21
*** larainema has quit IRC08:21
*** madong has joined #openstack-infra08:22
*** binbincong has quit IRC08:22
*** jcoufal has joined #openstack-infra08:22
*** dizquierdo has quit IRC08:23
*** madong is now known as larainema08:23
*** fhubik is now known as fhubik_afk08:23
*** larainema has quit IRC08:23
*** madong has joined #openstack-infra08:24
anteayaeantyshev: missed you at the meeting, I ended it already, see you next week perhaps08:25
anteayaclarkb: everybody up or just you?08:25
*** fhubik_afk is now known as fhubik08:25
anteayawell I'm back to bed, g'night08:26
*** mpaolino has joined #openstack-infra08:28
*** bexelbie_ has joined #openstack-infra08:29
clarkbjust waiting to feed twins the sleep hopefully08:30
*** ddieterly has joined #openstack-infra08:30
*** pblaho has joined #openstack-infra08:30
*** kdas_ has quit IRC08:30
*** ildikov has joined #openstack-infra08:32
*** otter768 has joined #openstack-infra08:33
*** dpose has joined #openstack-infra08:34
*** ddieterly has quit IRC08:35
*** binbincong has joined #openstack-infra08:36
eantyshevanteaya: I'm really sorry, I had urgent thing to do08:36
*** otter768 has quit IRC08:38
*** bexelbie has quit IRC08:39
*** pelix has joined #openstack-infra08:39
*** bexelbie has joined #openstack-infra08:40
*** sambetts has quit IRC08:44
*** sambetts has joined #openstack-infra08:45
*** laughterwym has quit IRC08:51
*** laughterwym_ has quit IRC08:51
openstackgerritIWAMOTO Toshihiro proposed openstack/requirements: Add ryu  https://review.openstack.org/15435408:52
*** Ala has joined #openstack-infra08:55
*** mfmcdonagh has quit IRC08:56
*** mcdonagm has quit IRC08:56
*** yamada-h has joined #openstack-infra08:57
*** flepied1 has joined #openstack-infra08:58
*** flepied2 has joined #openstack-infra08:58
*** yamamoto has joined #openstack-infra08:58
*** asahlin has quit IRC08:59
*** fhubik has quit IRC08:59
*** ParsectiX has quit IRC09:00
*** yamada-h has quit IRC09:00
*** dtantsur is now known as dtantsur|brb09:00
*** dizquierdo has joined #openstack-infra09:00
*** asahlin has joined #openstack-infra09:00
*** yamada-h has joined #openstack-infra09:00
*** shardy_ has joined #openstack-infra09:00
*** flepied has quit IRC09:01
*** ParsectiX has joined #openstack-infra09:02
*** shardy has quit IRC09:02
*** HeOS has joined #openstack-infra09:03
*** flepied1 has quit IRC09:03
*** yamada-h has quit IRC09:05
*** yamada-h has joined #openstack-infra09:05
*** shardy_ has quit IRC09:05
*** shardy has joined #openstack-infra09:06
*** yfried__ has quit IRC09:09
*** yamada-h has quit IRC09:09
*** yfried has joined #openstack-infra09:09
*** e0ne is now known as e0ne_09:11
*** ParsectiX has quit IRC09:12
*** ParsectiX has joined #openstack-infra09:12
*** vikram has joined #openstack-infra09:16
*** madong has quit IRC09:16
*** madong has joined #openstack-infra09:17
*** madong has quit IRC09:17
openstackgerritVictor Stinner proposed openstack-infra/project-config: Move gate-keystone-python34 to check  https://review.openstack.org/19835009:18
*** madong has joined #openstack-infra09:18
openstackgerrittakehirokaneko proposed openstack-infra/project-config: Changed the gerrit acl of stackforge/rack  https://review.openstack.org/19901809:18
*** yfried is now known as yfried|afk09:19
*** boris-42 has quit IRC09:22
*** e0ne_ is now known as e0ne09:24
*** laughterwym has joined #openstack-infra09:26
*** laughterwym_ has joined #openstack-infra09:26
openstackgerritMerged openstack/requirements: Adjust yaql version in upper-constraints  https://review.openstack.org/19869609:29
openstackgerritMerged openstack/requirements: Adding python-mistralclient to the global-requirements  https://review.openstack.org/19503509:29
*** ddieterly has joined #openstack-infra09:31
*** btully has joined #openstack-infra09:32
*** yfried|afk is now known as yfried09:34
*** ddieterly has quit IRC09:35
*** btully has quit IRC09:36
openstackgerritvikram.choudhary proposed openstack-infra/project-config: Adding networking-onos as a part of neutron project.  https://review.openstack.org/19874009:42
openstackgerritVictor Stinner proposed openstack-infra/project-config: Make gate-os-brick-python34 voting and add gate  https://review.openstack.org/19902309:42
*** dims has joined #openstack-infra09:44
*** dims_ has joined #openstack-infra09:45
*** dims__ has joined #openstack-infra09:46
openstackgerritFlavio Percoco proposed openstack-infra/project-config: Move amqp1's devstack plugin under openstack  https://review.openstack.org/19900309:46
*** HeOS_ has joined #openstack-infra09:47
*** dims has quit IRC09:48
*** ociuhandu has joined #openstack-infra09:49
*** dims_ has quit IRC09:50
*** dims__ has quit IRC09:50
*** ZZelle has quit IRC09:55
*** notnownikki has quit IRC09:57
openstackgerritMerged openstack-infra/project-config: Run neutron multinode without dvr  https://review.openstack.org/19890610:06
*** shardy_ has joined #openstack-infra10:06
*** e0ne is now known as e0ne_10:07
*** shardy has quit IRC10:08
*** shardy_ has quit IRC10:12
eantyshevjeblair, jhesketh: hello! I want to make modifications to zuul, made this https://review.openstack.org/c/19866710:12
*** shardy has joined #openstack-infra10:13
*** claudiub has joined #openstack-infra10:13
*** SotK_ has joined #openstack-infra10:14
jhesketheantyshev: oh cool that could be useful10:14
jhesketheantyshev: I'll review it now10:15
jheskethyolanda: ping10:15
yolandahi10:16
eantyshevjhesketh: I'm in doubt how that fits zuulv3 spec, so I will be glad if you point me what to do exactly with this10:16
jhesketheantyshev: yes, that's why I've pinged yolanda10:16
eantyshevyolanda: hello!10:16
jheskethyolanda: why do you think 198667 should be part of zuul v3?10:17
* jhesketh doesn't see any reason to block on the work10:17
yolandahi there, that change looked to me as it would fit there, when you talk about changes on mergers10:17
yolandai don't want to block the work, but i thought it could be enclosed on that wider spec10:17
*** ociuhandu has quit IRC10:17
yolandaas you talk there about different tenants, changes needed in zuul mergers, and gitg10:18
yolandagit10:18
*** SotK has quit IRC10:18
*** SotK_ is now known as SotK10:18
yolandai told eantyshev to read that spec, and see if that could fit there, and in that case, update the commit message to reflect that10:19
jheskethhmm it may fit within the multi tenancy10:20
jheskethwe should consider how it may affect it10:21
jheskethlet me finish reviewing :-)10:21
*** e0ne_ is now known as e0ne10:21
yolandacool, that was my point, that change seemed to fit very well in that section10:21
jhesketheantyshev: if the git url is given to the merger with the appropraite gerrit details, wouldn't that allow a single merger to be attached to 2 zuul's?10:22
eantyshevjhesketh: If together with gerrit private key then yes10:23
jhesketheantyshev: then why do you need this change?10:24
*** dtantsur|brb is now known as dtantsur10:25
KiallAny requirements core about for a review? https://review.openstack.org/#/c/196633 - dnspython3 (py3 release of dnspython)10:25
eantyshevjhesketh: at the moment only username is given10:25
openstackgerritAdam Coldrick proposed openstack-infra/storyboard: Fix entry point for storyboard-worker-daemon  https://review.openstack.org/19904210:26
eantyshevjhesketh: and merger uses private key from gerrit section of config10:26
eantyshevjhesketh: so options are either provide the key within the job, or to allow merger to use its own credentials10:27
eantyshevjhesketh: I chose the latter, since I saw no connection between credentials the merger uses and credentials used by zuul-server10:28
*** viktors|afk is now known as viktors10:30
*** ddieterly has joined #openstack-infra10:31
jhesketheantyshev: I see... let me put a few comments in the review10:32
*** otter768 has joined #openstack-infra10:34
*** ddieterly has quit IRC10:36
*** pc_m has joined #openstack-infra10:37
*** esmute__ has joined #openstack-infra10:38
*** koolhead17 has quit IRC10:38
jhesketheantyshev: I've added some comments :-)10:38
*** otter768 has quit IRC10:39
eantyshevthank you, jhesketh!10:39
*** ericksonsantos has joined #openstack-infra10:40
*** koolhead17 has joined #openstack-infra10:41
*** esmute__ has quit IRC10:42
*** radez is now known as radez_g0n310:45
*** yfried is now known as yfried|afk10:45
*** dims has joined #openstack-infra10:46
*** Krinkle has left #openstack-infra10:47
*** katyafervent_ is now known as katyafervent_awa10:48
*** katyafervent_awa is now known as katyafervent10:51
*** dims has quit IRC10:51
*** koolhead_ has joined #openstack-infra10:53
openstackgerritFathi Boudra proposed openstack-infra/jenkins-job-builder: general: add support for childCustomWorkspace  https://review.openstack.org/9710810:53
*** ihrachyshka has joined #openstack-infra10:55
openstackgerritZara proposed openstack-infra/storyboard-webclient: Bugfix: 'right click is interpreted as left click'  https://review.openstack.org/19904710:56
*** koolhead17 has quit IRC10:58
*** koolhead_ has quit IRC10:58
*** yfried|afk is now known as yfried10:58
openstackgerritFlavio Percoco proposed openstack-infra/project-config: Move amqp1's devstack plugin under openstack  https://review.openstack.org/19900310:58
*** yamada-h has joined #openstack-infra10:59
*** ociuhandu has joined #openstack-infra11:00
*** yamamoto has quit IRC11:01
*** yamamoto has joined #openstack-infra11:03
*** davideagnello has quit IRC11:04
*** yamada-h has quit IRC11:04
*** koolhead17 has joined #openstack-infra11:04
*** hashar has quit IRC11:05
*** dalgaaf has joined #openstack-infra11:08
*** Longgeek has quit IRC11:08
*** koolhead17 has quit IRC11:08
openstackgerritamrith proposed openstack-infra/project-config: Experiment with an increase time for gate-trove-functional job  https://review.openstack.org/19905011:09
*** miqui has quit IRC11:11
openstackgerritJoshua Hesketh proposed openstack-infra/os-loganalyze: Set DocumentRoot for devstack plugin  https://review.openstack.org/19905311:12
*** kashyap has joined #openstack-infra11:13
*** kashyap has left #openstack-infra11:14
openstackgerritJoshua Hesketh proposed openstack-infra/project-config: Test osloganalyze in devstack  https://review.openstack.org/19872811:14
*** yamamoto has quit IRC11:15
*** kushal has joined #openstack-infra11:17
*** claudiub has quit IRC11:23
*** thorst has joined #openstack-infra11:24
*** e0ne is now known as e0ne_11:25
openstackgerritFathi Boudra proposed openstack-infra/jenkins-job-builder: builders: add 'publish over ssh' support as a build step  https://review.openstack.org/9843711:25
*** vivcheri has joined #openstack-infra11:27
*** claudiub has joined #openstack-infra11:30
*** ddieterly has joined #openstack-infra11:32
openstackgerritSean Dague proposed openstack-dev/pbr: Add wsgi_scripts support to PBR  https://review.openstack.org/19529211:35
sdaguemordred / lifeless ok - https://review.openstack.org/195292 now has tests. They are a little janky, but they are a thing.11:36
*** ddieterly has quit IRC11:37
*** ldnunes has joined #openstack-infra11:38
*** esmute__ has joined #openstack-infra11:39
*** yfried is now known as yfried|afk11:42
*** esmute__ has quit IRC11:43
*** e0ne_ is now known as e0ne11:45
*** Longgeek has joined #openstack-infra11:45
*** prad has joined #openstack-infra11:45
*** yfried|afk is now known as yfried11:52
openstackgerritSergey Vilgelm proposed openstack/requirements: Use the 1.16.0 version of the oslo.messaging library  https://review.openstack.org/19906411:56
*** jasondotstar has quit IRC11:57
*** baoli has joined #openstack-infra11:57
openstackgerritSergey Vilgelm proposed openstack/requirements: Use the 1.16.0 version of the oslo.messaging library  https://review.openstack.org/19906411:58
*** markvoelker has quit IRC11:59
*** markvoelker has joined #openstack-infra11:59
*** baoli has quit IRC12:01
*** derekh has quit IRC12:01
*** baoli has joined #openstack-infra12:01
*** yamamoto has joined #openstack-infra12:02
*** derekh has joined #openstack-infra12:02
*** dims has joined #openstack-infra12:02
*** jaypipes has joined #openstack-infra12:02
*** gordc has joined #openstack-infra12:02
sdagueclarkb: once you get around today, let me know. I want to talk through the right place to do the handoff between multinode d-g and grenade12:03
sdaguethere are a few things floating around in my head12:03
*** bknudson has quit IRC12:03
*** weshay has joined #openstack-infra12:03
*** binbincong has quit IRC12:04
*** amrith is now known as _amrith_12:04
*** flepied2 has quit IRC12:05
clarkbI mean I am awake now but this is tail end of my day (I hope)12:05
openstackgerritDmitry Tantsur proposed openstack-infra/project-config: Add devstack check job for python-ironic-inspector-client  https://review.openstack.org/19907012:05
*** chlong has joined #openstack-infra12:06
clarkbI think we can do something very similar to what devstack did and run the grenade scripts wirh a constructed env via ansible12:06
*** kiran-r has quit IRC12:06
*** koolhead17 has joined #openstack-infra12:06
sdagueclarkb: man, crazy hours :)12:07
clarkbso its set up controller, set up compute nodes, trigger tests, then trigger upgrade on cpntroller, upgrade on cpmpues,ru  tests12:07
sdagueso... the issue is, grenade needs a lot more slicing up to get to that state12:07
*** dims has quit IRC12:07
sdaguebecause grenade handles the whole flow today12:07
sdagueso... I'm wondering if we could do this slightly differently12:08
sdagueand basically put a call out here - https://github.com/openstack-dev/grenade/blob/master/grenade.sh#L21412:08
clarkbhrm my first reaction is grenade would need to change to either be multinode aware on its own or to be less in control12:08
sdaguewhich could call the subnode script12:08
*** yfried is now known as yfried|afk12:09
*** flepied has joined #openstack-infra12:09
*** yfried|afk is now known as yfried12:09
sdagueright, we'll need changes somewhere, I'm just trying to make it so it's something that's not irreconsilable with the current model, still lets you run things locally, and doesn't become such a mess that we all give up and go goat farming12:09
clarkb*diaper farming pretty sure there is money in that business12:10
sdagueheh12:10
clarkbsdague ya since sibnode is basically just a second stack.sh run12:10
sdagueright12:10
clarkbmaybe hrm12:11
*** koolhead_ has joined #openstack-infra12:11
clarkbgrenade could ask d-g to devstack12:11
clarkband we will get nested i  realky weird eays12:11
sdagueso... if when we get to the subnode step, what if d-g just writes the ansible commands it would run to a script instead12:11
sdagueand we modify grenade to have a post-stack phase12:12
sdaguewhich runs that script12:12
clarkbthis is prpbably a good reason to properly playboom this stuff12:12
*** adduarte has quit IRC12:12
*** koolhead17 has quit IRC12:12
clarkbso that d-g and grenade can both use it12:12
*** duartea has joined #openstack-infra12:12
clarkbmordred ^ this is probably the reason we needes12:12
clarkbso they both do ansible-playbook make a devstack12:13
sdaguesure, though, honestly, I'd rather not make grenade ansible aware, just give it a script callout12:13
sdaguethough, I guess we could also make grenade take a configurable "stack" point12:13
sdaguehttps://github.com/openstack-dev/grenade/blob/master/grenade.sh#L211-L21412:13
sdaguemy concern though, is the gate code would end up so different from the base code, that it's not possible to local repro any thing12:14
clarkbthats why my change to ansible d-g used it for everything12:14
*** Ala has quit IRC12:15
*** vikram has quit IRC12:15
clarkbso multinide and single node d-g are the same12:15
clarkbthere arent even any branches really12:15
clarkbjust noop commands in single node12:15
*** Ala has joined #openstack-infra12:15
*** dprince has joined #openstack-infra12:15
*** dizquierdo has quit IRC12:16
sdaguesure, but the run mode for grenade is a bit different12:17
sdaguebecause it's also an orchestration layer to a degree12:17
Shrewsmordred: do you plan to cut a 0.9.1 shade release? or 0.10.0... whatever12:17
sdaguewe could argue that grenade should become ansible, but that's a whole other thing beast12:18
*** fabrice_ has quit IRC12:18
*** yamamoto_ has joined #openstack-infra12:18
sdagueclarkb: so I'm thinking something like this - https://review.openstack.org/19907312:19
*** vikram has joined #openstack-infra12:19
clarkbya same argument with d-g I personally find the shell with ansible as atool/cpmmand much easier to read12:19
clarkbsdague what if we gave grenade an inventory file and had it do its thing absed on that?12:20
*** binbincong has joined #openstack-infra12:20
sdagueI find the ansible stuff pretty hard to read personally, the way it's done. because all the meat is buried deep in the command string12:20
clarkband use a default local cpnnectio  only inventory12:20
*** fabrice_ has joined #openstack-infra12:20
sdagueso, today, grenade doesn't know anything about nodes12:20
*** ddieterly has joined #openstack-infra12:21
sdaguewhich is kind of ok12:21
clarkbI think it has to be aware if it is orvhestration and you want multinode tests12:21
sdagueclarkb: why?12:21
clarkbbecause orchestration across multiple nodes really wants to know the nodes exist12:22
mordredclarkb: nod12:22
mordredShrews: yes12:22
*** yamamoto has quit IRC12:22
sdagueright, but grenade doesn't really need to be aware of any of that12:22
Shrews\o/12:22
sdagueit is building a stack12:22
sdagueand upgrading a stack on the one machine12:22
sdaguethe fact that there might be more workers, is kind of irrelevant12:23
*** samuelBartel has joined #openstack-infra12:23
mordredShrews: done12:23
sdagueand tempest is running against an API12:23
mordredsdague: so you're arguing that multi-machine awareness shoudl be only in d-g and not in grenade?12:24
*** ari has joined #openstack-infra12:24
sdaguemordred: right.12:24
sdagueespecially with the partial upgrade case, where the reason for going down this path is that the workers should be left alone by grenade.12:24
*** rossella_s has quit IRC12:25
sdagueand, I don't think full multinode grenade upgrade is a thing we're going to go near any time soon. This is just about the partial upgrade case, which more projects want12:25
*** bknudson has joined #openstack-infra12:27
*** rossella_s has joined #openstack-infra12:28
*** bswartz has quit IRC12:28
*** aysyd has joined #openstack-infra12:28
*** kgiusti has joined #openstack-infra12:28
mordredsdague: awesome: "my_ip = socket.gethostbyname(socket.gethostname())\nsocket.gaierror: [Errno -2] Name or service not known"12:29
sdaguemordred: huh?12:29
mordredsdague: might be the same issue that hit multi-node that we cut a new glean for ...12:29
sdaguewhat is the issue?12:30
*** tlian has joined #openstack-infra12:30
mordredsdague: well, we started setting hostnames on the hosts, but there was a bug, so it set the hostname to a truncated version12:30
mordredhrm12:30
mordredthat can't possibly be this problem12:30
mordred(as I said it out loud ...)12:30
*** markvoelker has quit IRC12:31
*** bexelbie has quit IRC12:31
sdagueyeh, socket.gethostbyname(socket.gethostname()) not working is beyond what I expected from these environments12:31
*** bexelbie_ has quit IRC12:31
clarkbso I still dont grok the demand for partial multinode grenade but thats a different issue12:32
*** jlanoux has joined #openstack-infra12:32
clarkbwe can partial test single node pretty easily12:32
sdagueclarkb: well, lets come back to that question, because that's the primary design point12:32
clarkband will be much easier to debug12:33
sdagueclarkb: so, here is the problem with single node partial12:33
sdagueit's hyper synthetic12:33
clarkbthats fine12:33
*** bexelbie has joined #openstack-infra12:33
sdaguethe old code is only still working because it *probably* loaded all the libraries it needed before upgrade, and they are running in memory12:33
clarkbit tickles code paths, iseasier to debug and can be run lically12:34
sdagueand it adds a lot of really tricky conditionals in the grenade and devstack bits that get kind of fragile12:34
sdagueso making the barrier for upgrade be node level, makes all that a ton more clean12:34
*** otter768 has joined #openstack-infra12:35
sdagueand means that the number of combinations of different kinds of partial upgrade models people want can be pretty unlimitted12:35
sdaguebecause you just don't upgrade the worker12:35
sdaguepartial upgrade in a single node isn't easier to debug, because you can't stop the old side to add debug bits into it, because it's no longer old side code (it has different libraries)12:36
*** miqui has joined #openstack-infra12:36
clarkbuse vrnvs  or containers then12:36
clarkbthis isnt an issue with single node its an issue with how we single node12:37
mordredsure.12:37
mordredbut that would take reengineering all the thigns we do12:37
mordredwhich might be a thing to do think about12:38
mordredbut probably isn't a thing to make a ton of progress on, like, soon12:38
*** yamamoto_ has quit IRC12:38
clarkbya my poiny is either you orchrstrate and be node aware or you dont12:39
*** dguerri` is now known as dguerri12:39
clarkbfirst is use ansible second is singlw node12:39
clarkbboth work12:39
sdaguethere are other options here12:39
*** otter768 has quit IRC12:40
openstackgerritDmitry Tantsur proposed openstack-infra/project-config: Do not run ironic-inspector-dsvm job on stable/1.1 branch  https://review.openstack.org/19907912:40
*** markvoelker has joined #openstack-infra12:40
sdaguewhat is your objection to the post-stack.sh ?12:40
clarkbits hacky and doesnt make it easy to reproduce locally12:41
sdagueno one reproduces grenade locally except me12:41
sdaguein these partial cases12:41
sdagueI don't even do the partial cases locally most of the time, because there are enough differences in the gate that it's not really valid12:42
clarkbthen maybe we shpuld reduce them rather than add to them?12:42
sdaguereduce which?12:42
clarkbthe differneces with the gate12:43
clarkband insert spwcial gate script here would be a big difference12:43
clarkbits also forces us into a very simple set of upgrade tests on multi ode12:43
sdaguesure, so the first step there would be make something like devstack-vagrant to make it possible to exactly get the devstack-gate env up locally, including multinode12:44
sdaguethat's a whole other issue12:44
clarkband to do proper upgrade testing on multinode would reqyire doing it properly anyways12:44
*** fdegir has quit IRC12:44
clarkbso why not just do it especually since effort is likely the same (cost is in making it work not writing the change)12:44
sdagueI'm not sure I understand what should just be done?12:45
clarkbdo orchestration of >1 node in grenade12:46
clarkbI think it can be done simply with absible and configurable inventory file12:47
clarkbgrenade can explicitly affect subnodes properly if they are present that way12:47
clarkballowing you control over full upgrade of partial(service) upgrades12:47
clarkbwhile using the same code paths locally and in the gate12:48
sdaguebut unlike devstack-gate it doesn't have a whole system around to allocate it nodes for you already12:48
mordredthat's not how the ansible in d-g works though12:48
sdagueand adding more complexity into grenade, with basically only 1 person that understands it, is just a black swan event waiting to happen12:48
clarkbmordred it is12:48
mordredclarkb: no, I mean what sdague said12:48
clarkboh12:48
mordredclarkb: I was agreeing with you12:48
mordredsdague: the inventory file associates one or more hosts with one or more names. so the things you run are thigns like "run this on the controllers" "run this on the compute nodes" - if you have one node, you just have a simple inventory file that maps all the things to the one host12:50
*** ParsectiX has quit IRC12:50
sdagueright, except I only ever want to run 1 thing on the compute nodes, stack.sh12:50
sdagueonce12:50
sdagueafter it was run on the controller12:51
sdaguethen never talk to the compute nodes ever again12:51
*** yfried is now known as yfried|afk12:51
sdagueso rewrite all of grenade as ansible seems... a pretty big hammer to get that12:51
mordredwell, I think "rewrite all of grenade as ansible" is likely not the suggestion12:52
sdaguemordred: that's fair12:52
clarkbya just the stack run12:52
clarkbwhrre you linked12:52
clarkbI am saying make it configurable rather than $randomscripyable12:52
clarkband ansible is my sugeestion for doi g that12:53
*** fdegir has joined #openstack-infra12:53
*** yamamoto has joined #openstack-infra12:53
sdagueso that would make the communication boundary between d-g and grenade a bunch of ansible config12:54
mordredit would make it one inventory file, which it's already writing out12:54
sdagueand require also being able to understand ansible to do grenade in the single node case12:54
sdaguefor people trying to work on things12:54
sdaguewhich I'm not sure why that's better than the call out script12:54
sdaguewhich is how devstack-gate hooks work for all our various jobs12:55
*** rlandy has joined #openstack-infra12:55
*** BharatK has quit IRC12:55
clarkbya and hooks have been tons of troubel...12:55
mordred"being able to understand ansible" is going to become a requirement for touching pretty much any gate job in a few months12:55
clarkbI dunno I dislike the d-g hooks very much12:56
mordredso, not making value judgements on whether it's the right choice here - I don't think the learning curve of ansible should be a huge blocker12:56
sdaguemordred: sure, but the whole point of pushing this back into grenade was to make it possible to replicate outside the gate12:56
mordredsure12:56
*** bswartz has joined #openstack-infra12:56
clarkbthey are poorly declared interfaces so the natural progression of them has been to add more redundant unneeded ones to fit someones use casr12:56
*** yfried|afk is now known as yfried12:56
clarkbrather than grok interface and use it properly12:57
mordredbut if we wind up reimplementing ansible in shell as a result of doing that, it's worth considering just using the tool12:57
sdagueso.... https://review.openstack.org/#/c/199073/ doesn't seem like it's reimplementing ansible in shell12:57
*** bexelbie has quit IRC12:58
*** yamamoto has quit IRC12:58
mordredI agree. it does not seem like a reimplementation of ansible in shell :)12:59
openstackgerritSamuel BARTEL proposed openstack-infra/project-config: add project fuel-plugin-availability-zones  https://review.openstack.org/19908412:59
*** yamamoto has joined #openstack-infra13:00
*** yamamoto has quit IRC13:00
*** nilasae has quit IRC13:01
mordredsdague: btw - this has ALMOST nothing to do with that - but I'm spinning up a devstack using devstack-vagrant right now13:01
sdaguemordred: neat13:02
mordredand it strikes me, d-v uses puppet to do things ... which is what we use to configure the d-g nodes ... this has me wondering if it would be valuable to have a d-g-vagrant13:02
mordredthat would run literally all of the exact things we run on d-g nodes except for repo caching13:02
sdaguemordred: so... that was my end game for devstack-vagrant13:02
sdagueI just never got there13:02
mordrednod13:02
mordredso, if I wrote a patch for that at some point, you would not scream and throw your arms in the air and whatnot13:03
sdaguenope, I'd +A that like there was no tomorrow13:03
mordredcool13:03
sdagueit even can do multinode13:03
mordredI've _also_ been thinking of taking a step (after dib-nodepool is finished, of course) to publishing our dib images to vagrant boxes and docker containers13:03
sdaguethat would be pretty awesome13:04
mordredsince we already have a thing doing the work of making the images13:04
*** ddieterly has quit IRC13:04
mordredbiggest challenge is that i think most people dont' want to download the image that has the caching in it13:04
clarkb15GB vagrant images13:04
mordredyah13:04
*** hashar has joined #openstack-infra13:04
sdagueyeh, trimming the cache might be worth while.13:05
*** _nadya__ has quit IRC13:05
*** yamamoto has joined #openstack-infra13:06
*** eharney has joined #openstack-infra13:07
*** changbl has quit IRC13:08
*** sshnaidm has quit IRC13:08
*** jasondotstar has joined #openstack-infra13:09
*** jasondot_ has joined #openstack-infra13:10
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add stable ACLs for Neutron sub-projects  https://review.openstack.org/19875013:10
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add a create ACL for networking-bgpvpn  https://review.openstack.org/19875113:10
openstackgerritKyle Mestery proposed openstack-infra/project-config: Update ACLs for Neutron sub-projects  https://review.openstack.org/19874913:10
openstackgerritKyle Mestery proposed openstack-infra/project-config: Update merge commit gerrit ACL for Neutron repositories  https://review.openstack.org/19908913:10
*** _amrith_ is now known as amrith13:11
*** jsavak has joined #openstack-infra13:11
*** bexelbie has joined #openstack-infra13:12
mordredsdague: yup. vagrant just crashed my laptop13:12
mordredsdague: I _have_ used it successfully that one time though13:13
mordredso that's good13:13
*** notnownikki has joined #openstack-infra13:13
*** jasondot_ has quit IRC13:13
*** jasondotstar has quit IRC13:14
sdagueI've not had it crash my env in a long time, virtualbox up to date with your kernel?13:14
*** jasondotstar has joined #openstack-infra13:14
*** skolekonov has quit IRC13:16
*** heyongli has quit IRC13:17
yolandamordred, can you update access on puppet-os_client_config-core, and allow puppet-cores to have rights there?13:18
*** kiran-r has joined #openstack-infra13:18
yolandagood afternoon btw :)13:18
*** yamamoto has quit IRC13:18
*** heyongli has joined #openstack-infra13:19
*** dims has joined #openstack-infra13:19
*** dboik has joined #openstack-infra13:20
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode  https://review.openstack.org/19909113:21
*** dizquierdo has joined #openstack-infra13:22
sdagueanyway, I'm sure that doesn't actually work, but that's kind of what I was thinking13:22
mordredyolanda: yup13:22
mordredyolanda: done13:23
*** BharatK has joined #openstack-infra13:23
yolandacool, thx13:23
Sam-I-Ammornings13:23
*** julim has joined #openstack-infra13:24
openstackgerritZara proposed openstack-infra/storyboard-webclient: Stop 'about' and 'license' buttons overlapping others  https://review.openstack.org/19909513:25
*** heyongli has quit IRC13:26
*** mriedem_away is now known as mriedem13:27
*** ddieterly has joined #openstack-infra13:28
*** heyongli has joined #openstack-infra13:28
*** ddieterly has quit IRC13:28
*** ddieterly has joined #openstack-infra13:29
openstackgerritZara proposed openstack-infra/storyboard-webclient: Display 'about' and 'license' icons on higher zooms  https://review.openstack.org/19909713:30
*** mtanino has joined #openstack-infra13:31
*** spzala has joined #openstack-infra13:31
*** BharatK has quit IRC13:32
*** heyongli has quit IRC13:32
*** amitgandhinz has joined #openstack-infra13:33
*** amotoki has quit IRC13:33
*** boris-42 has joined #openstack-infra13:33
*** heyongli has joined #openstack-infra13:36
*** mpavone has quit IRC13:36
*** mika has quit IRC13:37
*** e0ne is now known as e0ne_13:37
*** mika has joined #openstack-infra13:37
*** bhunter71 has joined #openstack-infra13:37
*** yamamoto has joined #openstack-infra13:38
openstackgerritvikram.choudhary proposed openstack-infra/project-config: Adding networking-onos as a part of neutron project.  https://review.openstack.org/19874013:38
*** e0ne_ is now known as e0ne13:39
*** dustins has joined #openstack-infra13:39
*** jasondotstar has quit IRC13:40
*** deepakcs has quit IRC13:42
*** annegentle has joined #openstack-infra13:42
*** _nadya_ has joined #openstack-infra13:42
*** ddieterly has quit IRC13:44
*** fallenpegasus has quit IRC13:44
*** heyongli has quit IRC13:44
*** AJaeger_ is now known as AJaeger13:45
*** blaedd has joined #openstack-infra13:46
openstackgerritSean Dague proposed openstack-infra/project-config: add experimental grenade multinode jobs  https://review.openstack.org/19910313:46
*** heyongli has joined #openstack-infra13:47
AJaegeranteaya: yeah, I missed that. But what you wrote is not actionable for me. jeblair what do you want with reviews like https://review.openstack.org/#/c/196085/? anteaya said "made aware" which can mean many things. Adding you to the review? Waiting for your +2? Telling you on IRC?13:49
openstackgerritBorneMace proposed openstack-infra/project-config: Added new validate-contents test for kolla  https://review.openstack.org/19896813:49
*** smccully has joined #openstack-infra13:49
AJaegermordred: did you solve your po file issue for openstack-manuals?13:50
*** mpavone has joined #openstack-infra13:50
mordredAJaeger: I believe I learned that I did not need to solve my po file issue - which was even better13:50
AJaegermordred: yeah, po files are all autogenerated - with the exception of initial setup. So, don't touch them ;)13:51
*** claudiub has quit IRC13:52
pelixany infra core reviewers happy to review/approve changes for JJB?13:53
*** yamamoto has quit IRC13:53
pelixhttps://review.openstack.org/181541 & https://review.openstack.org/17945513:53
openstackgerritValeriy Ponomaryov proposed openstack-infra/project-config: Move experimental Manila jobs in Rally to check pipeline  https://review.openstack.org/19910813:55
*** signed8bit_ZZZzz is now known as signed8__13:55
*** r-daneel has joined #openstack-infra13:56
openstackgerritValeriy Ponomaryov proposed openstack-infra/project-config: Move experimental Manila jobs in Rally to check pipeline  https://review.openstack.org/19910813:56
*** annegentle has quit IRC13:57
openstackgerritValeriy Ponomaryov proposed openstack-infra/project-config: Move experimental Manila jobs in Rally to check pipeline  https://review.openstack.org/19910813:57
*** sigmavirus24_awa is now known as sigmavirus2413:57
mordredsdague: any chance you know what I need to go to get my devstack to have run with keystone v3?13:57
mordredsdague: ah - found it in stackrc13:59
*** zz_dimtruck is now known as dimtruck13:59
sdagueyeh, it should always be there now13:59
sdaguethere is a flag to turn off v213:59
mordredwell ... IDENTITY_API_VERSION=${IDENTITY_API_VERSION:-2.0}13:59
Sam-I-Amkeystone v3 for middleware?13:59
*** nelsnelson has joined #openstack-infra13:59
mordredis what's set in stackrc in the thing I have locally13:59
*** blaedd has quit IRC14:00
mordredso I apparently need to set that to 3 - but I can do that14:00
*** blaedd has joined #openstack-infra14:00
sdagueif [ "$ENABLE_IDENTITY_V2" == "False" ]; then14:00
sdague    IDENTITY_API_VERSION=314:00
sdaguefi14:00
sdagueok, that's pretty convoluted14:00
mordredyah. so - either disable v2 or enable v314:00
samueldmqmordred: sdague that's experimental ...14:01
mordredsdague: is there an easy way for me to pass a localrc to devstack-vagrant?14:01
samueldmqdisabling v2 requires devstack to set up resources using keystone v3 only (of course)14:01
mordredsamueldmq: I just need a keystone v3 running that I can run service and endpoint adding to - I don't actually need the rest of openstack to work with it :)14:01
*** annegentle has joined #openstack-infra14:02
samueldmqthere are some patches to make that happen (starting at https://review.openstack.org/#/c/186681/)14:02
*** jsavak has quit IRC14:02
samueldmqmordred: it is already running you you run a simple devstack without setting anything else14:02
mordredsamueldmq: really?14:02
sdaguemordred: I don't think so, normally14:02
sdaguenormally I just work with an existing vagrant if I'm hacking on that, but that shouldn't be too bad to add in14:03
*** jsavak has joined #openstack-infra14:03
Sam-I-Amiirc, keystone v3 lives at the usual /v2.0 url14:03
mordredsamueldmq: WARNING: unsupported identity-api-version 3, falling back to 2.014:03
*** Kennan2 has joined #openstack-infra14:03
mordredAuthorization Failed: The resource could not be found. (HTTP 404) (Request-ID: req-b50d736c-6796-4503-8daa-4d2e7ee368ef)14:03
mordredwait14:03
mordredwhat?14:03
samueldmqmordred: yes it is https://github.com/openstack/keystone/blob/master/etc/keystone-paste.ini#L10514:03
mordredyou put keystone v3 at the v2.0 url?14:03
mordred*headdesk*14:03
samueldmqmordred: what did you do ?14:03
samueldmqSam-I-Am: ?14:03
Sam-I-Amfor kilo thats how it works14:03
*** Kennan has quit IRC14:03
sdagueSam-I-Am: ?? in what context14:03
samueldmqkeystone v3 lives in :5000/v314:03
sdagueyeh, definitely devstack uses /v314:04
Sam-I-Amsdague: might have been a well-needed change in liberty14:04
Sam-I-Amfor the kilo install guide, nothing wanted to work right with /v3 (although it existed)14:04
*** signed8__ is now known as signed8bit_ZZZzz14:04
mordredok. I agree - there is a keystone endpoint running on :5000/v314:05
*** xyang1 has joined #openstack-infra14:05
samueldmqSam-I-Am: no, you can't access v3 API at v2.0 at all14:05
Sam-I-Ami think it was mostly client-side problems14:05
mordredhowever, WARNING: unsupported identity-api-version 3, falling back to 2.014:05
Sam-I-Amsamueldmq: interesting. how am i getting v3 auth responses with /v2.0 endpoints?14:05
samueldmqmordred: you using the clients ? what command are you trying to run ?14:05
Sam-I-Amsamueldmq: unless thats what changed14:05
mordredsamueldmq: keystone catalog14:06
Sam-I-Ammordred: you need to use the openstack client for v3, iirc14:06
*** dimtruck is now known as zz_dimtruck14:06
Sam-I-Amkeystone client was deprecated for v314:06
mordredSam-I-Am: mordred@camelot:~/src/openstack-dev/devstack-vagrant$ openstack catalog list14:06
mordredERROR: openstack 'NoneType' object does not support item assignment14:06
dstanekSam-I-Am: v3 tokens or v3 api?14:06
samueldmqSam-I-Am: v2.0 operations are equivalent in v3 using the default domain14:06
samueldmqdstanek: :-)14:06
dstaneksamueldmq: yes. yes, we do make things too confusing.14:07
Sam-I-Amsamueldmq: that makes sense. there's only one domain in this case.14:07
Sam-I-Amdstanek: tokens14:07
*** annegentle has quit IRC14:07
dstanekSam-I-Am: that's different than the v3 api14:07
samueldmqmordred: ah you setting openstack client to use v3 ?14:08
Sam-I-Amdstanek: ah.14:08
mordredsamueldmq: should be14:08
mordredsamueldmq: OS_IDENTITY_API_VERSION=314:08
Sam-I-Amdstanek: is 'catalog' a v3 api thing? because that fails on my system.14:08
mordredcatalog should work everywhere14:09
mordred"keystone catalog" -- "openstack catalog list"14:09
mordredshould be the same thing14:10
samueldmqmordred: could you try something similar to openstack endpoint list --os-url $KEYSTONE_SERVICE_URI_V3 --os-identity-api-version 314:10
*** bmace_away is now known as bmace14:10
mordredERROR: openstack 'NoneType' object does not support item assignment14:10
samueldmqmordred: I am scared ..14:10
mordredopenstack endpoint list --os-url http://192.168.1.231:5000/v3 --os-identity-api-version 314:11
samueldmqmordred: that's how jamielennox is doing at https://review.openstack.org/#/c/186681/14:11
Sam-I-Amah, i see14:11
Sam-I-Ami had a problem with my openrc14:11
mordredok. so - I used shade instead of the command line clients14:11
samueldmqmordred: let me get a vm up and see what I can do :)14:12
mordredand got this: keystoneclient.openstack.common.apiclient.exceptions.BadRequest: Expecting to find domain in project - the server could not comply with the request since it is either malformed or otherwise incorrect. The client is assumed to be in error. (HTTP 400)14:12
mordredso - I believe that's telling me "hi, you didn't specify a domain"14:12
mordredone would expect the error message elsewhere to be better than "'NoneType' object does not support item assignment"- just fwiw14:12
samueldmqmordred: ++ yes I agree14:13
Sam-I-Ammordred: here's what seems to work - http://paste.openstack.org/show/352089/14:13
Sam-I-Ammy keystone endpoints are /v2.0 though14:13
Sam-I-Ambut openstack catalog list works now14:13
samueldmqmordred: in fact, that's what shade is for, right ? fix things openstack :)14:13
mordredheh14:13
*** sdake has quit IRC14:14
mordredsamueldmq, Sam-I-Am: yes. adding OS_USER_DOMAIN_NAME=default and OS_PROJECT_DOMAIN_NAME=default makes it work. thank you both14:14
*** sdake has joined #openstack-infra14:14
Sam-I-Ammordred: os_project_domain_id ?14:15
Sam-I-Amyeah14:15
* Sam-I-Am slow14:15
mordredwell, it's 'default', so I used name not id14:15
Sam-I-Amfind moar coffee14:15
mordredalthough you are right, the id param _Does_ also accept names14:15
clarkbwouldnt ot be nice if there was a ... wait for it ...default domain?14:16
mordredbtw - for anyone interested, I tested this via ansible and shade like this:14:16
mordredansible localhost -m os_auth14:16
*** signed8bit has joined #openstack-infra14:17
mordredwhich helpfully printed out the service catalog for me14:17
*** signed8bit_ZZZzz has quit IRC14:17
mordredclarkb: right?14:18
Sam-I-Amclarkb: isnt the default domain default? ?)14:18
Sam-I-Amer, :)14:18
Sam-I-Amstupid fingers14:18
*** heyongli has quit IRC14:18
*** haypo has joined #openstack-infra14:18
mordredSam-I-Am: it is...14:18
bmacehey folks, sorry if this is a silly question.  my review failed last time so i think i fixed it and patched it appropriately, but it doesn't seem to be getting run through jenkins again?  i checked zuul and there is no job for it and some others posted more recently seem to have finished.14:18
clarkbexcept mordred had to specify it14:18
clarkbwhich is not a default14:18
mordredSam-I-Am: except yeah ^^ it didn't do a good job of being a default :)14:18
Sam-I-Amthat was the joke14:18
* mordred hands Sam-I-Am a fluffy bunny14:19
openstackgerritPaul Belanger proposed openstack-infra/project-config: Upload applytest folder gate-infra-puppet-apply-*  https://review.openstack.org/19912214:19
Sam-I-Amha14:19
clarkbbmace link to change?14:19
Sam-I-Amdoes it contain caffeine?14:19
haypohi. DevStack setup failed on "edit-constraints .../requirements/upper-constraints.txt -- keystone ..." with "pkg_resources.RequirementParseError: Missing distribution spec /opt/stack/new/oslo.utils" on my patch https://review.openstack.org/#/c/198026/14:19
*** mpaolino_ has joined #openstack-infra14:19
pabelangerany chance for a some what fast review / +2 for above?  ^ Make things easier when debugging puppet-apply jobs14:19
clarkbmordred btw if you really want to get people hand them twins14:19
samueldmqmordred: no you used the id ..14:19
mordredsamueldmq: I did what?14:20
*** mpaolino has quit IRC14:20
samueldmqmordred: the default domain has id = 'defaul' :)14:20
samueldmqdefault*14:20
haypoi'm not more sure that lifeless told me that he's working on this issue. is anyone aware of this upper-constraints issue?14:20
*** asselin_ has joined #openstack-infra14:20
bmacehrm, and of course as soon as i mention it in here, from out of nowhere, jenkins marks it as verified though i saw nothing in zuul about it :/  https://review.openstack.org/#/c/198968/14:20
mordredsamueldmq: you realize I'm going to punch someone right?14:20
*** heyongli has joined #openstack-infra14:20
*** sshnaidm has joined #openstack-infra14:20
AJaegerclarkb: good morning. Could you add Robert Clark (Security PTL) as core to security-specs ACL, please? https://review.openstack.org/#/admin/groups/976,members14:20
samueldmqmordred: :-)14:20
mordredsamueldmq: in a world where everything is UUIDs for ids and humand readable names for names .... *GAH*14:20
bmaceclarkb thanks for asking.  that must have been the magic to knock it loose :)14:20
* mordred goes back to debugging the actual thing14:20
samueldmqdstanek: is the id of default domain 'default' only in devstack ? or do we do that in our migration stuff ?14:21
clarkbAJaeger dobe14:21
clarkber done14:21
openstackgerritZara proposed openstack-infra/storyboard-webclient: Display 'about' and 'license' icons on higher zooms  https://review.openstack.org/19909714:21
clarkbhaypo I think thats the upgtafe pbr issue14:22
AJaegerclarkb: thanks!14:22
clarkbhaypo you need to track down what is downgrading pbr14:22
clarkbarg more docs.o.o https fails14:24
dstaneksamueldmq: not sure, but i assume it can be whatever14:24
openstackgerritPaul Belanger proposed openstack-infra/project-config: Upload applytest folder gate-infra-puppet-apply-*  https://review.openstack.org/19912214:24
*** btully has joined #openstack-infra14:24
pabelangernibalizer, ^14:24
samueldmqmordred: dstanek it can be configurable .. but the it's default is 'default' https://github.com/openstack/keystone/blob/master/etc/keystone.conf.sample#L79014:24
haypoclarkb: before the error, i see pbr 1.2 which looks correct14:25
samueldmqyes, this sentence is confusing including too many 'default' :-)14:25
samueldmqdstanek: thanks14:25
mordredso - there are more than one issue14:25
mordredone is that the ID is being set to default14:25
mordredthat's silly14:25
mordredthe _name_ should be set to default14:25
mordredand it shoudl have an ID that nobody should care about14:25
*** chlong has quit IRC14:25
dstaneksamueldmq: yeah, i believe you still have to create it. so there isn't any magic.14:25
mordredalso - region_name has a default value of RegionOne14:26
mordredbut if that is the default region on the cloud, you do not have to specify it as a parameter14:26
mordred_that_ is how a default value for something global like that should behave14:26
mordredputting this:14:26
mordred       user_domain_id: default14:26
mordred       project_domain_id: default14:26
mordredinto a config file for a user is just bonkers14:26
samueldmqmordred: if you don't specify anything, fall into the default thing14:27
mordrednot that I don't love everyone, of course14:27
samueldmqhehe14:27
samueldmqmordred: I think that's for v2 (where the default domain is used).. in v3 domains should be set explicitely14:28
AJaegerclarkb: was https ever working on docs.openstack.org?14:28
samueldmqmordred: I think this is from security guys vs UX guys (what you're playing here I think)14:28
mordredsamueldmq: I hear that - but I'd like to suggest that that is incorrect14:28
mordredsamueldmq: nope. there is nothing security about this14:28
mordredif the cloud has one and only one domain14:28
clarkbAJaeger no14:28
mordredit should be ABSOLUTELY fine to have it behave as a default14:29
clarkbAJaeger https everywhere was previosuly at fault14:29
mordredthis is especially important for the transition period from v2 to v314:29
mordredwhere people will have their old creds, and where their clouds will have only one domain14:29
mordredmaking everyone in the world pass the two strings "default" into the API14:29
mordredis _not_ security14:29
openstackgerritFathi Boudra proposed openstack-infra/jenkins-job-builder: builders: add 'publish over ssh' support as a build step  https://review.openstack.org/9843714:29
clarkbmordred: but making it impossible to use the api means no users which means you are most secure14:30
samueldmqmordred: you have interesting points, I'd love to see them discussed with other folks :)14:31
mordredclarkb: good point. I forgot about that14:31
Sam-I-Ammordred: this is one of the reasons i lose hair writing the install guide14:31
Sam-I-Amespecially when keystone middleware changes without more docs than a blog post :/14:31
samueldmqSam-I-Am: hehe14:32
*** annegentle has joined #openstack-infra14:32
samueldmqmordred: clarkb I am kind of learning how things work in a UX vs security approach :-)14:32
*** yamamoto has joined #openstack-infra14:32
samueldmqwhen I wonder why certain things aren't different :14:32
*** Ala has quit IRC14:34
Sam-I-Amsamueldmq: when i wonder, i just dive into neutron for a while :)14:35
Sam-I-Amthen other things dont seem so crazy14:35
samueldmqSam-I-Am: hehee I need to see more about neutron .. I am very keystone so far14:35
samueldmqSam-I-Am: but willing to get a wider knowledge by contributing to/understanding  other projects as well14:36
*** otter768 has joined #openstack-infra14:36
Sam-I-Amsamueldmq: i sort of figured out its near impossible to know everything about every project, even just the 'core' ones14:36
*** bmace has quit IRC14:36
Sam-I-Ambrain capacity limited14:36
Sam-I-Amor i keep killing brain cells14:37
*** rossella_s has quit IRC14:37
*** BharatK has joined #openstack-infra14:37
*** rossella_s has joined #openstack-infra14:37
sdagueAJaeger: can you help me figure out what's going on with this failure - https://review.openstack.org/#/c/199103/ ?14:37
AJaegersdague: let me check...14:37
samueldmqSam-I-Am: sure, it's not about knowing everything about everyproject at all14:39
*** woodster_ has joined #openstack-infra14:39
*** jsavak has quit IRC14:39
openstackgerritSean Dague proposed openstack-infra/project-config: add experimental grenade multinode jobs  https://review.openstack.org/19910314:39
sdagueAJaeger: oh, I think I figured it out14:39
samueldmqSam-I-Am: I'm working on keystone for a while, and I am far from understanding everything14:39
samueldmqSam-I-Am: that's just about having a better knowledge of openstack as a whole, if you know what I mean :)14:39
*** fallenpegasus has joined #openstack-infra14:39
Sam-I-Amsamueldmq: sure14:40
AJaegersdague: yeah, that looks fine now.14:40
Sam-I-Amwriting the install guide helped a bit14:40
*** jsavak has joined #openstack-infra14:40
*** otter768 has quit IRC14:40
openstackgerritPaul Belanger proposed openstack-infra/system-config: WIP: Migrate to puppet-httpd module  https://review.openstack.org/19881914:41
openstackgerritPaul Belanger proposed openstack-infra/system-config: Actually test for puppet errors when we run puppet-apply  https://review.openstack.org/19913214:41
*** esmute__ has joined #openstack-infra14:41
AJaegerjeblair: https://review.openstack.org/#/c/199103/ is a change that touches devstack-gate. Is that something you like to review?14:42
*** annegentle has quit IRC14:43
AJaegerdhellmann: regarding https://review.openstack.org/#/c/197672, what's wrong with that one and it's depends-on?14:44
*** otter768 has joined #openstack-infra14:44
*** austin81 has joined #openstack-infra14:44
*** austin81 has left #openstack-infra14:45
*** austin81 has joined #openstack-infra14:46
*** esmute__ has quit IRC14:46
*** BharatK has quit IRC14:46
*** yfried has quit IRC14:48
*** tonytan4ever has joined #openstack-infra14:49
dhellmannAJaeger: oh, just that I didn't notice the depends-on14:50
*** radez_g0n3 is now known as radez14:50
*** bmace has joined #openstack-infra14:50
dhellmannAJaeger: actually, renaming isn't something we could block on with just a depends-on, right? it requires manual action and gerrit downtime?14:51
*** vponomaryov has left #openstack-infra14:51
*** vponomaryov has joined #openstack-infra14:51
*** jsavak has quit IRC14:52
sdaguedhellmann: the way renames for existing projects in openstack occur is the governance change depends-on the project-config change14:52
*** jsavak has joined #openstack-infra14:52
sdagueand the project-config change is just handled delicately14:52
*** rlandy has quit IRC14:52
dhellmannsdague: ok14:52
sdagueso not approved until there is a window14:52
sdagueif it's a "new" openstack project, the depends is the other way around14:52
dhellmannI guess there's no way to avoid the small window where the names in the 2 places don't match14:52
sdagueso that the TC must ok it first14:53
sdagueright14:53
*** rlandy has joined #openstack-infra14:53
sdaguebut we've got 2 TC members in infra, it just requires someone to twizzle it again post land on the project-config side14:53
pabelangermordred, AJaeger any chance for some +2 action? https://review.openstack.org/#/c/199122/14:55
*** bradjones|away is now known as bradjones14:56
*** achanda has joined #openstack-infra14:56
*** hdd has joined #openstack-infra14:57
*** wenlock has joined #openstack-infra14:59
*** BharatK has joined #openstack-infra14:59
*** wenlock1 has joined #openstack-infra15:00
*** jsavak has quit IRC15:01
*** jsavak has joined #openstack-infra15:01
*** achanda has quit IRC15:02
*** ddieterly has joined #openstack-infra15:03
*** arxcruz has quit IRC15:03
*** cindyo has joined #openstack-infra15:03
*** nilasae has joined #openstack-infra15:03
*** annegentle has joined #openstack-infra15:04
*** laughterwym has quit IRC15:04
*** armax has joined #openstack-infra15:05
*** _kiran_ has joined #openstack-infra15:05
*** thedodd has joined #openstack-infra15:05
*** _kiran_ has quit IRC15:06
openstackgerritRussell Bryant proposed openstack/requirements: Stop adding extra newline after unparseable lines.  https://review.openstack.org/19914315:06
*** zz_dimtruck is now known as dimtruck15:06
*** puranamr has joined #openstack-infra15:07
*** annegentle has quit IRC15:13
*** e0ne is now known as e0ne_15:16
*** dannywilson has joined #openstack-infra15:16
*** shashank_hegde has joined #openstack-infra15:16
*** dims has quit IRC15:16
*** _nadya_ has quit IRC15:16
*** mpaolino has joined #openstack-infra15:18
*** mpaolino_ has quit IRC15:18
*** bexelbie has quit IRC15:20
*** dannywilson has quit IRC15:20
*** tsekiyama has joined #openstack-infra15:20
*** annegentle has joined #openstack-infra15:21
sdaguemordred / clarkb (who I know is probably asleep) can I get your look at - https://review.openstack.org/#/c/199091/ as an approach here15:22
clarkbI am not asleep yet, there is something terribly wrong here15:22
sdagueAJaeger: hmmmm ... after addressing the duplicate entry thing, now I get a not defined error on https://review.openstack.org/#/c/199103/15:24
*** flepied has quit IRC15:26
*** sabeen1 has joined #openstack-infra15:26
*** e0ne_ is now known as e0ne15:26
*** dannywilson has joined #openstack-infra15:27
*** cdent has quit IRC15:28
openstackgerritRussell Bryant proposed openstack-infra/project-config: Publish docs for networking-ovn.  https://review.openstack.org/19915315:29
*** mpaolino has quit IRC15:29
openstackgerrityolanda.robla proposed openstack-infra/puppet-gerrit: cron to clear gerrit logs  https://review.openstack.org/19915415:30
* mriedem has spent entirely too much time in the guts of pbr / sphinx / autodoc this morning15:30
*** changbl has joined #openstack-infra15:31
*** shashank_hegde has quit IRC15:32
anteayaeantyshev: oh no problem I just didn't want you to be surprised if you popped in15:32
anteayaAJaeger: ah I didn't know you would only consider my comments if they were framed as an action item15:33
*** kiran-r has quit IRC15:33
anteayaAJaeger: yes, let's hear from jeblair what he would like, as I am motivated by his wishes to be informed15:33
*** bexelbie has joined #openstack-infra15:33
AJaegeranteaya: you misunderstood me. I missed that when doing my +2A. Still, I wonder what's the proper work flow is.15:33
openstackgerritDmitry Tyzhnenko proposed openstack-infra/git-review: Add reviewers by group alias on upload  https://review.openstack.org/19504315:34
*** sabeen3 has joined #openstack-infra15:34
anteayawell I think part of my reluctance to review is the lack of acknowledgement my reviews get15:34
anteayaas I feel only a green check mark, if it is exists is acknowledged15:34
*** sabeen1 has quit IRC15:35
anteayacomments I leave tend to get glossed over, which I find tiring15:35
AJaegeranteaya: And that was the only complaint there - you gave a comment which I should have read properly and come back to IRC and ask: What shall be done?15:35
AJaegeranteaya: If I would have read it properly, I wouldn't have approved it ;( Sorry.15:36
anteayait is okay, it is a small thing15:37
anteayabut it is indicative of something i have been feeling for a while15:37
sdagueAJaeger: any thoughts on https://review.openstack.org/#/c/199103/ because now I'm more confused15:37
anteayamany small things15:37
AJaegersdague: you have "pipeline: check" in projects.yaml but zuula/layout.yaml uses "gate-..."15:37
*** soren has quit IRC15:37
sdagueah....15:37
sdagueAJaeger: thanks15:38
*** davidlenwell has quit IRC15:38
AJaegeranteaya: the problem is that when I'm giving my reviews, the input box does not show any previous review comments.15:38
AJaegeranteaya: and if I didn't read them properly before, I might act on what I see. That happens to several of us...15:39
AJaegeranteaya: so, what I do in such case: Give my comment with just saying "LGTM, I'll only +2 once this or that is done."15:39
*** shardy_ has joined #openstack-infra15:40
AJaegeranteaya: and for others where it's really critical, I even -2 like project renames: https://review.openstack.org/#/c/195341/15:41
*** shardy has quit IRC15:41
*** dims has joined #openstack-infra15:43
openstackgerritSean Dague proposed openstack-infra/project-config: add experimental grenade multinode jobs  https://review.openstack.org/19910315:43
sdaguehmmm... is there a hostname truncation issue - http://logs.openstack.org/91/199091/1/experimental/check-tempest-dsvm-multinode-full/e519323//console.html#_2015-07-07_15_11_42_302 ?15:45
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Exclude .test directory from pep8 check  https://review.openstack.org/19915915:45
clarkbsdague: yes there is, image builds from this morning should correct it15:45
*** shardy_ has quit IRC15:45
clarkbsdague: this was the other thing I was sying about multinode, it doesn't work at all currently and would be good to get working before we try to add a bunch of complexity15:45
*** spredzy is now known as spredzy|afk15:45
clarkbhopefully that is the only thing that has broken int he last month or so15:45
anteayaAJaeger: the input box does not show any previous review comments, that sounds like the problem right there15:46
*** shardy has joined #openstack-infra15:46
*** asselin_ has quit IRC15:46
*** whoops has joined #openstack-infra15:47
*** sdake_ has joined #openstack-infra15:48
*** sdake has quit IRC15:48
sdagueit looks like the key local regression is brctl isn't in the subnodes - http://logs.openstack.org/32/189632/6/check/check-tempest-dsvm-multinode-smoke/b6214f4/logs/subnode-2/devstacklog.txt.gz15:48
*** BharatK has quit IRC15:49
hogepodgehow often does the puppet scripts that manages the openstack bot run?15:49
clarkbhogepodge: every half hour to 45 minutes depending on how slow the previous puppet run is15:49
*** unicell has quit IRC15:49
sdagueclarkb: yes, I do understand that point. I also understand that 0 people besides me look at grenade race failures (that often show up more often in the partial jobs), even when they are a 40% failure rate. So making that part more complicated isn't a win either :)15:49
fungihogepodge: basically it's fired from cron every 15 minutes, but happens under a global lock to prevent concurrent runs, and currently takes around 30 minutes to complete15:50
clarkbsdague: I think the hostname -f is the failure there15:50
clarkbsdague: not sure if brctl is related to that or not, I thought that we were explicitly installing the bridge utils package prior to any bridge setup15:50
fungihogepodge: which can mean every 30-60 minutes depending on how long the previous run takes15:50
sdagueclarkb: oh, good point15:50
hogepodgefungi: hmm, I'm wondering why openstack bot hasn't joined #openstack-diversity yet. That change merged last night. https://review.openstack.org/#/c/198936/115:50
sdagueclarkb: right, we might be, that's exposing in the worlddump15:50
fungihogepodge: i'll have a look15:51
sdagueclarkb: so the hostname -f failure is because of the hostname truncation?15:51
clarkbsdague: yes, /etc/hosts has the full name but the hostname is truncated so the name lookup fails15:51
clarkbbecause foo != foobar15:51
fungihogepodge: that's the change to add access permissions for our bots and infrastructure admins, not a change make the openstack bot join the channel15:52
sdagueclarkb: gotcha15:52
hogepodgefungi ah ha.15:52
fungihogepodge: did you have a corresponding change to the openstack_project::eavesdrop class in the openstack-infra/system-config repo?15:52
*** hichihara has quit IRC15:53
hogepodgefungi: yes, but it hasn't merged yet https://review.openstack.org/#/c/198941/15:53
*** akshai has joined #openstack-infra15:53
*** davidlenwell has joined #openstack-infra15:53
openstackgerritZara proposed openstack-infra/storyboard-webclient: Display 'about' and 'license' icons on higher zooms  https://review.openstack.org/19909715:54
fungihogepodge: lgtm, but we generally avoid approving those while meetings are going on since that's also our meetbot and the eavesdrop class restarts it on config updates, which can be disruptive mid-meeting15:54
*** kushal has quit IRC15:54
*** davidlenwell is now known as daviddavid15:54
*** daviddavid is now known as davidlenwell15:54
hogepodgefungi: I understand. personally, I'd just wait for the infra meeting to restart it ;-)15:55
pelixAnyone on the infra team available to approve a change to enable some more tests in JJB? https://review.openstack.org/18154115:55
fungiif we ever find a vict^H^H^Holunteer for reimplementing our irc bots in errbot we should dig into ways to make that sort of thing more graceful15:55
*** annegentle has quit IRC15:56
*** ivar-lazzaro has joined #openstack-infra15:56
zaropelix: can do15:56
zaromorning15:56
*** ivar-lazzaro has quit IRC15:57
pelixzaro: you've already +2'ed it ;)15:57
clarkbsdague: so I think that worlddump brctl can't happen if we fail early enough to have not installed brctl15:57
*** ivar-lazzaro has joined #openstack-infra15:57
clarkbsdague: so in theory that problem goes away with the new images today15:57
*** ddieterly has quit IRC15:57
anteayazaro: morning15:57
*** hashar is now known as hasharMeeting15:58
*** davidlenwell has quit IRC15:58
*** davidlenwell has joined #openstack-infra15:58
*** salv-orlando has joined #openstack-infra15:58
openstackgerritMerged openstack-infra/jenkins-job-builder: Enable tests of builder object  https://review.openstack.org/18154115:59
zaropelix: i would have obliged for tests only changes :)15:59
fungiclarkb: are you working toward a once-every-two-days sleep schedule now? :/16:00
*** jistr has quit IRC16:00
*** BharatK has joined #openstack-infra16:00
zarothat would be impressive16:00
*** jtomasek has quit IRC16:01
clarkbfungi: I hope not, I just got caught being up last night and now its cool so time to get stuff done16:01
yolandafungi, mordred, i don't seem to have core rights to puppet-refstack , can you please check that?16:01
clarkbwill probably check out after hte meeting if I make it that far16:01
*** aswadr has joined #openstack-infra16:02
clarkbhave to take advantage of all sub 25C hours16:02
fungiyolanda: fixing, we probably missed adding our puppet core group to that once it was created16:02
anteayaclarkb: I'm for you catching a nap anytime you can get one16:02
yolandamm, if i look into the acl, i just see "inherit from all-projects", i cannot see any special puppet group there16:03
clarkbpart of my problem is I have always been bad at taking naps16:03
*** ddieterly has joined #openstack-infra16:03
clarkbso I end up awake far too long then just crash and burn and thankfully have enough help that that isn't too terrible16:03
fungiyolanda: oh! that acl never applied. something may be broken with it16:03
fungilooking now to see if i can tell what16:03
*** annegentle has joined #openstack-infra16:03
anteayaclarkb: yeah you don't strike me as a good napper16:03
yolandamaybe some typo in the acl16:04
anteayaclarkb: I can offer lessons16:04
fungiyolanda: it's better than that. there is no acl file for it in project-config16:04
fungii wonder if there was an acl in the change which added the repo16:04
*** dtantsur is now known as dtantsur|afk16:04
yolandawas that added recently?16:04
fungiyolanda: https://review.openstack.org/18243716:05
yolandaso no acl16:05
fungiyolanda: oh, it's got an acl-config specified in project.yaml though16:05
yolandamm, puppet-system is the generic one, right?16:05
sdagueclarkb: cool, when do we expect new images?16:06
*** laughterwym_ has quit IRC16:06
fungiyolanda: nope, there is no gerrit/acls/openstack-infra/puppet-system.config in the project-config repo. it's a nonexistent file i guess16:06
clarkbsdague: prior to me disappearing for forever they were showing up across the board by around noon pdt, so about 3 hours from now16:07
yolandai remember having used or seen that puppet-system sometime16:07
*** scheuran has quit IRC16:07
*** ddieterly has quit IRC16:07
*** jistr has joined #openstack-infra16:07
fungiyolanda: looks like it might have been intended to use system-config.config instead16:07
anteayait is in openstack16:07
anteayanot openstack-infra16:07
anteayathe path is wrong16:07
yolandaanteaya, thx, i remember having seen that somewhere16:08
*** davidlenwell has quit IRC16:08
*** ddieterly has joined #openstack-infra16:08
fungianteaya: i don't see one there either. can you link to it?16:08
*** ashleighfarnham has joined #openstack-infra16:08
*** jlanoux has quit IRC16:08
*** davidlenwell has joined #openstack-infra16:08
fungianteaya: are you thinking of puppet-modules.config in openstack?16:08
anteayasorry you are correct16:08
*** ashleighfarnham has quit IRC16:08
anteayayes I was thinking of puppet-mondules in openstack16:08
*** ashleighfarnham has joined #openstack-infra16:09
anteayaas that seems to be the catchall acl for puppety things16:09
fungianteaya: anyway, puppet-refstack is an openstack-infra project so likely wouldn't use the puppet-openstack team's acls16:09
anteayaah16:09
anteayaso yeah, no acl16:09
yolandafungi, is there some generic acl for openstack-infra puppet projects?16:09
*** ddieterly has quit IRC16:09
*** kushal has joined #openstack-infra16:09
sdagueclarkb: coolio16:10
fungiyolanda: no, they're separate. i think the fix is to remove the acl-config entry from projects.yaml and add a puppet-refstack.config acl16:10
*** annegentle has quit IRC16:10
yolandaok, i can do it in a moment16:10
*** k4n0 has quit IRC16:11
yolandafungi, that puppet-system is appearing in more projects16:11
yolandaopenstack-infra/beaker-localhost, beaker-nodepool..16:12
*** davidlenwell is now known as omnibus716:12
*** omnibus7 is now known as davidlenwell16:12
yolandaso at some point there was some intention to have that acl ?16:12
fungiyolanda: i wonder if someone has cargo culted someone else's typo16:12
*** davidlenwell is now known as david_enwell16:12
anteayayolanda: the intention to have a common acl has to come from puppet16:13
*** kgiusti has left #openstack-infra16:13
zaropelix: you didn't answer my question regarding folder support.16:13
yolandamm, so i wonder if that beaker projects introduced the bug16:14
fungianteaya: yolanda: found. it used to exist and was removed by https://review.openstack.org/18621716:14
fungianteaya: yolanda: i think people keep copying old reviews to add new projects16:14
anteayaah ha16:14
anteayayou are the best detective16:14
fungianteaya: i owe it all to git log -p16:15
anteaya:)16:15
pelixzaro: which patchset was that on?16:15
*** fawadkhaliq has joined #openstack-infra16:15
zaropelix: https://review.openstack.org/#/c/180185/16:15
yolandafungi, ok, so then better to remove that puppet-system from the 3 projects, and add individual acls for them, right?16:16
*** MarkAtwood has joined #openstack-infra16:17
fungiyolanda: yep, they're all broken. they should have separate acls each16:18
*** mayurig has joined #openstack-infra16:18
*** flepied has joined #openstack-infra16:18
openstackgerrityolanda.robla proposed openstack-infra/project-config: Fix puppet-refstack acl  https://review.openstack.org/19917716:18
fungialso reviewers might want to keep an eye out for new acolytes of that particular cargo cult16:18
openstackgerritVladimir Kozhukalov proposed openstack-infra/project-config: Add project fuel-agent repo  https://review.openstack.org/19917816:18
yolandaok, change for puppet refstack16:18
openstackgerritMerged openstack-infra/elastic-recheck: Remove obsolete cinder LVM no space  querie  https://review.openstack.org/18878216:19
fungiyolanda: thanks!16:19
*** david_enwell is now known as davidlenwell16:19
yolandaabout the other 2 projects, they are more related to test, doesn't seem to be puppet modules, so not sure about the right acl. i guess no release is needed16:19
*** sshnaidm has quit IRC16:20
*** mpavone has quit IRC16:20
*** davidlenwell has quit IRC16:20
*** soren has joined #openstack-infra16:20
*** dpose has quit IRC16:20
*** davidlenwell has joined #openstack-infra16:20
*** hogepodge has quit IRC16:21
anteayafungi: thank you16:21
*** BharatK has quit IRC16:21
*** austin81 has left #openstack-infra16:22
*** kdas_ has joined #openstack-infra16:23
*** bexelbie has quit IRC16:23
*** kushal has quit IRC16:23
*** unicell has joined #openstack-infra16:23
pleia2good morning16:23
*** davidlenwell has quit IRC16:23
*** davidlenwell has joined #openstack-infra16:24
yolandathe beaker ones seem to be from nibalizer, i'll check with him16:24
*** ildikov has quit IRC16:25
*** amrith is now known as _amrith_16:26
*** davidlenwell has quit IRC16:27
*** davidlenwell has joined #openstack-infra16:28
*** unicell has quit IRC16:28
anteayamorning pleia216:28
*** unicell has joined #openstack-infra16:28
*** jcoufal has quit IRC16:28
*** armax has quit IRC16:28
*** ddieterly has joined #openstack-infra16:29
clarkbgood morning16:29
*** ddieterly has quit IRC16:30
*** davidlenwell has quit IRC16:31
zaroclarkb: any interest in updating gearman-plugin with your change?16:31
*** notnownikki has quit IRC16:31
greghaynes'morning16:32
*** e0ne has quit IRC16:33
nibalizerYa I created the beaker repos16:33
*** armax has joined #openstack-infra16:34
*** jistr has quit IRC16:35
*** sdake_ is now known as sdae16:35
*** davidlenwell has joined #openstack-infra16:35
*** davidlenwell is now known as davidlenwell_16:36
*** luqas has quit IRC16:37
anteayamorning greghaynes16:38
*** davidlenwell_ is now known as davidlenwell16:38
*** abregman is now known as abregman|brb16:38
*** armax has quit IRC16:38
*** Swami has joined #openstack-infra16:39
*** kdas_ has quit IRC16:39
*** haypo has quit IRC16:39
*** sdae is now known as sdake16:39
zaroclarkb: i'm refering to https://review.openstack.org/#/c/18037116:39
*** haypo has joined #openstack-infra16:40
*** annegentle has joined #openstack-infra16:40
clarkbzaro: re your comments the block at line 249 is supposed to handle that16:41
*** haypo has left #openstack-infra16:41
clarkbat least I think that was the intent, but if we don't call register after we close in all cases that wont work16:41
*** annegentle has quit IRC16:44
*** patrickeast has joined #openstack-infra16:45
clarkbzaro: I likely wont get to that today16:45
clarkbbut feel free to push a new patchset if you like16:45
*** claudiub has joined #openstack-infra16:46
zaroclarkb: ahh, i missed that,  but it definately doesn't seem to handle all cases on reset.16:47
*** Longgeek has quit IRC16:47
*** shashank_hegde has joined #openstack-infra16:47
zaroclarkb: i'll take a look again to see if i can get that to work for all cases.16:47
*** rockyg has joined #openstack-infra16:47
*** kiran-r has joined #openstack-infra16:47
*** wenlock has quit IRC16:48
zaroclarkb: i didn't mean to try it today.  i'm just wondering if you were interested in updating the plugin at some point in the future since the plugin seems to be working without problems now.16:48
*** tonytan4ever has quit IRC16:48
*** duartea has quit IRC16:49
clarkbits a less pressing change16:49
clarkbbut would probably be good16:49
openstackgerritVipul Sabhaya proposed openstack-infra/project-config: Move Cue and related repos to OpenStack namespace  https://review.openstack.org/19626816:51
*** kdas_ has joined #openstack-infra16:51
openstackgerritVipul Sabhaya proposed openstack-infra/project-config: Move Cue and related repos to OpenStack namespace  https://review.openstack.org/19626816:52
*** claudiub has quit IRC16:52
zaroclarkb: ok, i just didn't want to put more working into it than necessary :)16:52
*** claudiub has joined #openstack-infra16:52
*** ZZelle has joined #openstack-infra16:53
*** hasharMeeting has quit IRC16:53
*** davideagnello has joined #openstack-infra16:56
AJaegeryolanda, fungi : There's already https://review.openstack.org/#/c/198938/1/gerrit/projects.yaml16:58
*** shashank_hegde has quit IRC16:59
*** jsavak has quit IRC16:59
*** ddieterly has joined #openstack-infra17:00
fungiAJaeger: good find. i guess that could be updated to add acl files17:01
*** afazekas has quit IRC17:01
*** esmute__ has joined #openstack-infra17:01
*** shashank_hegde has joined #openstack-infra17:02
*** wenlock has joined #openstack-infra17:02
AJaegerfungi: yeah, I'll approved yolanda's and will update the other now on top of it17:02
*** ociuhandu has quit IRC17:04
fungiAJaeger: thanks17:04
fungii also commented on 19893817:04
*** ihrachyshka has quit IRC17:05
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Fix ACL's on projects with puppet-system  https://review.openstack.org/19893817:06
* AJaeger updated 198938...17:06
*** wenlock has quit IRC17:06
*** aswadr has quit IRC17:07
*** ihrachyshka has joined #openstack-infra17:07
AJaegerfungi: Now there're no more acl-config lines in openstack-infra ;)17:07
fungiAJaeger: thanks!17:07
openstackgerritMerged openstack-infra/project-config: Fix puppet-refstack acl  https://review.openstack.org/19917717:07
*** cindyo has quit IRC17:07
pabelangerfungi, AJaeger: sorry for the noise, but do you mind checking out 199122 ?  It's gating me for some git and puppet stuff17:08
pabelangeruploads log files basically17:08
*** salv-orl_ has joined #openstack-infra17:08
*** ildikov has joined #openstack-infra17:08
*** e0ne has joined #openstack-infra17:09
AJaegerpabelanger: +2.17:09
*** yamahata has quit IRC17:09
*** jsavak has joined #openstack-infra17:09
pabelangerAJaeger, danke17:10
*** vahidh has joined #openstack-infra17:10
*** salv-orlando has quit IRC17:11
*** HeOS has quit IRC17:11
AJaegerspeaking about uploads, I'd like some advise on https://review.openstack.org/#/c/198802/2/jenkins/jobs/puppet-module-jobs.yaml . The change copies over  devstack-logs macro but this is not a devstack setup. So, should it use the macro - or should it use the full definition? fungi, clarkb, do you have any style comments, please?17:11
AJaegerpabelanger: gern geschehen!17:11
*** gyee has joined #openstack-infra17:12
*** derekh has quit IRC17:12
*** ddieterly has quit IRC17:12
AJaegerfungi: Do you want to add krotscheck to puppet-refstack acl? Or wait until he asks for it? ;)17:13
krotscheckAJaeger: Oh god please no.17:13
* krotscheck is FAR more productive if he doesn't have to worry about reviewing projects.17:13
AJaegerkrotscheck: ok, we leave puppet-refstack without any cores for now ;)17:14
krotscheckAJaeger: I think hogepoge is the right person to go to for that.17:14
fungiAJaeger: i leave anything beyond including the infra-core and infra-puppet-core groups up to jeblair to formally delegate17:14
rockygAJaeger, krotscheck  Yes, hogepodge and maybe davidlenwell?17:15
rockygOr pvanek?17:15
davidlenwellI'm here17:15
fungiAJaeger: we have a core review team for our puppet modules. it just needs to be added as an included group17:15
funginot sure why this devolved into a pingfest17:15
AJaegerfungi: Agreed, let's add the usual setup and let rockyg discuss with jeblair...17:16
rockygdavidlenwell, it's a discussion on whether puppet-refstack has any refstack cores.  I put forth you, chris, paul17:16
fungiinfra puppet modules have infra-core and infra-puppet-core review teams added to the initial core review group. any additional core reviewers are up to the infra ptl to formally delegate17:16
davidlenwellwe can just match the refstack-core group17:16
*** austin81 has joined #openstack-infra17:16
davidlenwellor add the refstack-core group as core17:17
*** dizquierdo is now known as dizquierdo_afk17:17
fungirockyg: davidlenwell: krotscheck: apologies for you getting pinged on this. it's not something that needs addressing right now17:17
davidlenwellcool17:18
davidlenwellthanks17:18
krotscheckkk17:18
rockygdavidlenwell, I think the question is, who from Refstack should be proposed for the puppet module maintenance/changes17:18
*** bdemers has quit IRC17:18
rockyg'nuff said17:18
fungiand "nobody" is a perfectly fine answer to that question17:18
*** e0ne is now known as e0ne_17:18
*** bdemers has joined #openstack-infra17:18
AJaegermordred: are you around? I have a question on the Meiji release name...17:19
anteayaI'm curious how it should be pronounced17:19
davidlenwellfungi: nobody it is ;)17:19
*** david-lyle has quit IRC17:19
anteayame-jee17:19
anteayamay-jee17:19
krotscheck"Tasty Chocolate Treat"?17:19
fungianteaya: mostly the latter17:19
AJaegeranteaya: I see everybody in Tokyo at the first keynote learn to pronounce it ;)17:19
anteayafungi: thanks, I'll work with that17:20
*** abregman|brb is now known as abregman17:20
anteayaAJaeger: yeah no kidding17:20
anteayaI need a brain pattern though to start to attach things too17:20
anteayaI'll go with may-jee until I learn differently17:20
*** austin81 has left #openstack-infra17:20
rockygAJaeger, yah.  Not so much.  No one in the US has been able to pronounce Huawei, even when we work with them on it :P17:20
anteayaI'm going to get a walk in before the meeting17:21
*** pradk has joined #openstack-infra17:21
*** e0ne_ is now known as e0ne17:21
AJaegerrockyg: it's not better in Germany - and pronouncing SUSE has also it's challenges. But so is pronouncing Linux ;)17:21
fungirockyg: i still don't understand why "wah-way" is hard to remember17:21
*** jtomasek has joined #openstack-infra17:22
pleia2fungi: it has none of the same letters! ;) (ok, a couple)17:22
*** pblaho_ has joined #openstack-infra17:22
rockygfungi, Mirkans.  Furrin words are hard How-way, Hoo-ah-way, haw-way17:22
*** pblaho_ has quit IRC17:22
krotscheckanteaya: That video that mordred posted has the pronunciation. The last three seconds of this movie have it as well: https://www.youtube.com/watch?v=Cuw5VK4ZvxM17:22
krotscheckanteaya: It's three syllables.17:23
*** kiran-r has quit IRC17:23
*** jklare has quit IRC17:23
*** _amrith_ is now known as amrith17:24
rockyganteaya, consider it a Hawaiian style pronounciation ;-)17:24
*** _nadya_ has joined #openstack-infra17:25
*** achanda has joined #openstack-infra17:25
*** e0ne is now known as e0ne_17:26
fungiyeah, "meh-ee-jee" is a little closer, though the first two syllables are sort of a diphthong and get slurred together a bit17:26
*** e0ne_ is now known as e0ne17:27
AJaegeranteaya, mordred : My question was whether the release name Meiji is named after the emperoror or after a location. This came up in https://review.openstack.org/#/c/199060/2/doc/glossary/glossary-terms.xml17:27
fungijapanese pronunciation gets a little interesting around distinguishing syllables. for example the n sound is treated as a distinct syllable, and is more akin to a vowel than a consonant in the way it's used in word construction17:28
*** wenlock has joined #openstack-infra17:28
fungiAJaeger: technically it's named after the location17:28
AJaegerfungi: The shrine or the park? Or either? ;)17:29
*** mrunge has quit IRC17:29
rockygfungi, and no Japanes word starts with "en"17:29
*** annegentle has joined #openstack-infra17:30
krotscheckAnyone remember off the top of their head what the appropriate bitwise operator is to check if a file is readable? I've got the mode, need to check if I can read it.17:30
clarkb44417:30
clarkbin octal17:31
AJaeger"The code name comes from the locations Meiji shrine and Meiji park which are named after the emperor Meiji."17:31
fungiAJaeger: our wiki is a little vague on which (or whether both) but since we name based on locations (kilo being a bit of an exception) it's definitely a place name rather than intended as a direct reference to a person or a historical time period17:31
clarkbthere is a defined val for all three bits you should use though17:31
clarkbI would need to read docs for that though17:31
AJaegerfungi: thanks. So, my wording above should be fine, isn't it?17:31
fungiAJaeger: looks right to me17:32
fungiAJaeger: you might extend that to "Meiji shrine and Meiji park in Tokyo" since that was the relevant selection criteria17:33
fungi(i.e. the candidate names were distilled to only locations in tokyo for the final poll)17:33
*** e0ne is now known as e0ne_17:33
smoserhey, is there something that i can just give a url to a review and have it run 'git fetch ... git branch' for me ?17:33
smoserie, i'd like:17:33
jrollsmoser: git review -d 12345617:34
jrollwill check out that review17:34
jrolls/-d/-x/ if you want to cherry-pick17:34
AJaegerthanks, fungi!17:34
smoserjroll, thats exactly what i was looking for. thanks.17:35
jrollnp :)17:35
*** annegentle has quit IRC17:35
krotscheckclarkb: I was looking for &, ^ or |, but it turns out your answer helped me figure it out. Thanks!17:35
*** shashank_hegde has quit IRC17:37
smoserthe one thing i'd like is for it to take a url and scrape the review number out of it . cause i'm too lazy, and i have the review url in my browser.17:37
smoserbut oh well.17:37
openstackgerritKhai Do proposed openstack-infra/jenkins-job-builder: Improve delete all job performance  https://review.openstack.org/19919317:38
fungismoser: i just cut/paste the review number out of the url for that, but i suppose you could whip up a wrapper that just does a regex to pull the value out of a url string17:39
*** mayurig has quit IRC17:39
smoseryeah, thats what i'md oding fungi .  you're just less lazy than me.17:39
*** e0ne_ is now known as e0ne17:39
greghaynesclarkb: so this nodepool gearman race - somehow the nodepool gear client is trying to connect to the totally wrong gearman server port17:40
*** jklare has joined #openstack-infra17:40
AJaegerpleia2: could you check https://review.openstack.org/#/c/198236/ , please? The scripts look fine but need some local testing. And I think this needs to be part of the post queue, not the gate queue.17:40
clarkbkrotscheck &17:40
greghaynesclarkb: we do the same kind of config templating trick as we do for the db name in test setup to do the gearman server... so its wierd there could be some kind of race there17:41
*** austin81 has joined #openstack-infra17:41
fungismoser: i dunno, i'm pretty darned lazy. but i'm too lazy to compete with you on who's the laziest17:41
smosertouche17:41
clarkbgreghaynes its not just open port 0 ask socket for port?17:41
pleia2AJaeger: added to my list, thanks for bumping it up in my queue17:42
greghaynesclarkb: that is basically what happens, yes, then we template that out in the nodepool config so the clients use it17:42
AJaegerthanks, pleia2 !17:43
fungigreghaynes: are the clients reusing the same generated config/same gearman server or are you creating them per-test?17:43
*** mayurig has joined #openstack-infra17:44
greghaynesfungi: it gets made per test, we copy it to a temp location17:44
*** yamahata has joined #openstack-infra17:44
*** pcrews has quit IRC17:46
jeblairsmoser: if you use gertty, you can just hit 'x' (cherry-pick) or 'c' (checkout) while looking at the change17:46
*** armax has joined #openstack-infra17:46
*** _nadya_ has quit IRC17:46
fungigreghaynes: hrm, i wonder if the temp location for the configs is maybe not unique and clients end up reading configs not meant for their test?17:47
AJaegerfungi, jeblair: Speaking about new repos: Do you have any guidance on https://review.openstack.org/#/c/198905/ - stackforge/xstatic-angular-gettext? All the other xstatics are in the stackforge namespace but should they stay there or should we encourage them moving to openstack?17:47
openstackgerritgreghaynes proposed openstack-infra/nodepool: Set gearman client_id for nodepool client  https://review.openstack.org/19920017:47
*** dguerri is now known as dguerri`17:47
greghaynesfungi: hrmmm17:48
jeblairAJaeger: i'd keep them together; better for developers, and the marginal cost for us to move one is small17:48
AJaegerjeblair: thanks, will review and approve...17:49
fungiAJaeger: i agree with jeblair. but also one of the issues that resulted in xstatic packages being in stackforge and not officially governed by horizon in the past was a question of licensing (since it's common for the xstatic package to duplicate the license of the software it's packaging)17:49
*** esmute__ has quit IRC17:49
openstackgerritgreghaynes proposed openstack-infra/nodepool: Run geard for nodepool tests  https://review.openstack.org/19492817:50
AJaegerfungi: thanks, good to know!17:50
openstackgerritgreghaynes proposed openstack-infra/nodepool: Trigger image builds using gearman  https://review.openstack.org/19517517:50
jeblairAJaeger, anteaya: i don't have to block devstack-gate job changes, sdague and other cores there can help evaluate those17:50
openstackgerritgreghaynes proposed openstack-infra/nodepool: Trigger image uploads using gearman  https://review.openstack.org/19772217:52
openstackgerritgreghaynes proposed openstack-infra/nodepool: Trigger image delete using gearman  https://review.openstack.org/19811417:52
AJaegerjeblair: so, who should review devstack-gate as well? One of sdague, jeblair, clarkb, fungi?17:53
*** Longgeek has joined #openstack-infra17:53
fungiAJaeger: yep https://review.openstack.org/#/admin/groups/267,members17:53
fungiAJaeger: specifically (at the moment) infra-core plus sdague17:54
*** yfried has joined #openstack-infra17:54
*** yfried has quit IRC17:54
AJaegerSo, in other words: Anteaya and myself shouldn't be the only +2s when it gets approved right now ;)17:54
AJaeger(that's easier for me to remember ;)17:54
fungiAJaeger: the goal being that we have some oversight against the (thankfully infrequent) trend to want to add any job that runs devstack-gate as also voting on devstack-gate changes17:54
jeblairAJaeger: well, if the change comes from sdague, probably okay ;)17:55
AJaegerjeblair: ;)17:55
jeblairbut yes, what fungi said17:55
fungithe qa team is in a similar situation with people proposing jobs for devstack and tempest just because they happen to run one or both17:56
AJaegerfungi, jeblair: I'll keep both in mind when reviewing. Thanks!17:58
openstackgerritMerged openstack-infra/jenkins-job-builder: Exclude .test directory from pep8 check  https://review.openstack.org/19915917:59
*** reed_ has quit IRC17:59
*** abregman has quit IRC17:59
*** esmute__ has joined #openstack-infra18:01
*** jsavak has quit IRC18:02
*** jsavak has joined #openstack-infra18:03
*** julim_ has joined #openstack-infra18:03
*** marzif_ has joined #openstack-infra18:03
*** Longgeek has quit IRC18:04
*** tonytan4ever has joined #openstack-infra18:04
*** coolsvap|away is now known as coolsvap18:04
*** changbl has quit IRC18:05
*** julim has quit IRC18:05
*** ihrachyshka has quit IRC18:07
*** shashank_hegde has joined #openstack-infra18:09
anteayajeblair: thank you18:09
openstackgerritMerged openstack-infra/project-config: Add Zanata support for openstack-manuals projects  https://review.openstack.org/19822118:11
openstackgerritMerged openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-gettext  https://review.openstack.org/19890518:11
openstackgerritMerged openstack-infra/project-config: midonet: Specify Q_PLUGIN=midonet in local.conf  https://review.openstack.org/19253118:11
*** hogepodge has joined #openstack-infra18:11
openstackgerritMonty Taylor proposed openstack-infra/shade: Add CRUD methods for Keystone domains  https://review.openstack.org/19039618:11
openstackgerritMonty Taylor proposed openstack-infra/shade: Support project/tenant and domain vs. None  https://review.openstack.org/19044218:11
openstackgerritMonty Taylor proposed openstack-infra/shade: Allow use of admin tokens in keystone  https://review.openstack.org/19920918:11
mordredShrews, fungi, jeblair: ^^ enjoy18:12
greghaynesoo fun18:12
*** esmute__ has quit IRC18:12
mordredgreghaynes: yah.18:12
*** vnod has joined #openstack-infra18:13
Shrewsmordred: i will enjoy so much18:13
openstackgerritYAMAMOTO Takashi proposed openstack-infra/project-config: Remove Midonet v2 jobs  https://review.openstack.org/19897918:15
*** jsavak has quit IRC18:15
*** yamamoto has quit IRC18:16
anteayakrotscheck: thanks, sounds like may-he-gee to my ears18:16
*** kgiusti has joined #openstack-infra18:16
*** soren has quit IRC18:16
*** ZZelle has quit IRC18:16
*** jsavak has joined #openstack-infra18:16
*** _nadya_ has joined #openstack-infra18:17
*** e0ne is now known as e0ne_18:17
*** hogepodge has quit IRC18:20
*** marzif_ has quit IRC18:21
smoserjeblair, i did not know of gertty. thanks for that too.18:22
*** salv-orl_ has quit IRC18:22
*** _nadya_ has quit IRC18:26
Shrewsmordred: i still would like for 190396 to keep the get/list/search methods with the others to at least keep *some* code organization in that large __init__.py file18:26
Shrewsmordred: but i'm ocd'ish that way. if i'm being overly stupid, please say so18:27
jeblairi try to aim for just the right amount of stupid in any given situation18:28
mordredShrews: oh - piddle. one sec18:28
mordredShrews: we actually don't seem to have the methods sorted that way in OperatorCloud18:29
Shrewsjeblair: it is a fine line  :)18:29
jeblairShrews: are you calling me stupid?18:29
Shrewsmordred: those aren't in OperatorCloud18:29
mordredShrews: yah they are18:30
Shrewsmordred: check again  :)18:30
Shrewsthey are inherited18:30
mordredShrews: oh - actually - list/search/get are in OpenStack - create/delete are in operator18:31
Shrewsmordred: we have had this problem often enough, i'm going to split that class out into a separate file, i think18:31
Shrewsjeblair: certainly not18:31
mordredShrews: and now I remember - I needed to rework this to use the new methods that the keystone team added into ksc for us18:31
mordredShrews: I agree18:31
jeblairShrews: ;)18:31
mordredShrews: let's do it after this keystone pile though18:31
Shrewsmordred: oh, totally18:31
mordredShrews: ok - so - there are two ways to get a domain list18:33
mordredShrews: one is via ksc.auth.domains() - which gives you the list of domains your token is scoped to18:33
mordredShrews: the other is ksc.domains.list() - which gives you all the domains and is admin only18:34
mordredShrews: I think that list_domains() should call ksc.auth.domains() for sure18:34
*** rockyg has quit IRC18:34
mordredShrews: shoudl we put an override in OperatorCloud that calls the unscoped one?18:34
*** john-davidge_ has joined #openstack-infra18:34
mordredShrews: or leave that for another day?18:34
*** pcrews has joined #openstack-infra18:34
mordredjeblair: (you may also have design thoughts there)18:35
Shrewsmordred: that sounds right to me18:35
mordredShrews: I asked an either/or?18:35
Shrewsmordred: but, it can be done later18:36
Shrewsmordred: add the override18:36
mordredcool18:36
mordredso - override, but do it later18:36
Shrewsyeah18:36
mordredossum18:36
jeblairthat seems sensible18:36
*** sdake has quit IRC18:36
*** esmute__ has joined #openstack-infra18:36
*** signed8bit is now known as signed8bit_ZZZzz18:37
*** rossella_s has quit IRC18:37
*** rossella_s has joined #openstack-infra18:37
*** esmute__ has quit IRC18:38
*** esmute__ has joined #openstack-infra18:38
*** esmute__ has quit IRC18:39
*** e0ne_ is now known as e0ne18:39
*** heyongli has quit IRC18:39
*** koolhead_ has quit IRC18:39
john-davidge_is jenkins failing on python34 currently a known issue?18:40
fungijohn-davidge_: you'll have to be a _lot_ more specific18:41
fungijohn-davidge_: an example is probably the best place to start your questions18:41
john-davidge_fungi: haha, sure. Here’s 3 examples: https://review.openstack.org/#/c/199164/ https://review.openstack.org/#/c/165023/ https://review.openstack.org/#/c/158697/18:42
john-davidge_fungi: from my quick look through reviews these failures seem to be intermittent18:43
fungijohn-davidge_: those all look like recent neutron test failures. i know there was a release of the "routes" dependency that broke python 3 compatibility in the past week-ish. let me find you a bug report and you can see if that matches what you're getting18:44
john-davidge_fungi: thanks18:44
fungior it's possible we bumped the upper cap on routes to include a broken version18:45
*** heyongli has joined #openstack-infra18:45
*** Sukhdev has joined #openstack-infra18:45
pabelangerjeblair, anything else specific you want to see for the Host trystack.o.o site spec before adding it into the spec approval section for the meeting?  review 19509818:45
fungijohn-davidge_: upstream report seems to be https://github.com/bbangert/routes/issues/5118:46
fungihaypo's not around right now or i'd ask for a shortcut to related bug reports18:47
john-davidge_fungi: Ok thanks, looks like its a legit failure then. will hold off on the rechecks for now :)18:49
*** salv-orlando has joined #openstack-infra18:49
*** heyongli has quit IRC18:50
*** pcrews has quit IRC18:50
*** e0ne has quit IRC18:50
sc68calanteaya: quick update on the linuxbridge job - we've got https://review.openstack.org/198833 in the pipe which fixes another bug we've seen in the LB CI job18:50
anteayayay another bug fix18:51
anteayasc68cal: thank you!18:51
fungijohn-davidge_: hrm... https://review.openstack.org/182746 blacklists routes 2.1 and is still failing its python34 job, so might be something new18:51
sc68calanteaya: do you have any thoughts on merging https://review.openstack.org/195653 next week or so?18:51
*** achanda has quit IRC18:51
*** sdake has joined #openstack-infra18:52
*** baoli has quit IRC18:52
anteayasc68cal: let's see where we are after 198833 gets some exercise18:52
anteayabut yes would be happy to look at it even at the end of this week18:53
anteayasc68cal: thank you18:53
john-davidge_fungi: :/18:53
*** dims has quit IRC18:53
fungirussellb: you commented on https://review.openstack.org/182746 about routes breaking python 3. can you elaborate?18:55
pelixzaro: I believe I have a way to test the JJB timeout capability, though it's currently using gevent.server.StreamServer which is probably a bit heavy for the need18:55
*** mriedem has quit IRC18:55
fungirussellb: i see that change is blacklisting routes 2.1 which has an open github issue about wanting a 2.2 release so that python 3 fixes will be available18:55
pelixbasically just creates a server in a thread to open a port but ignores all connection attempts18:56
*** baoli has joined #openstack-infra18:56
sc68calanteaya: that sounds good, thank you18:56
*** annegentle has joined #openstack-infra18:56
*** sdake_ has joined #openstack-infra18:56
jheskethMorning18:57
*** heyongli has joined #openstack-infra18:57
fungirussellb: the corresponding requirements change https://review.openstack.org/196867 suggests that routes 2.1 was blacklisted because of bug 146988318:57
openstackbug 1469883 in OpenStack Compute (nova) "Unit tests fail with Routes==2.1" [Medium,Fix committed] https://launchpad.net/bugs/1469883 - Assigned to Davanum Srinivas (DIMS) (dims-v)18:57
*** dims has joined #openstack-infra18:58
anteayajhesketh: hey18:59
anteayasc68cal: thank you18:59
*** sdake has quit IRC19:00
pleia2meeting time19:00
*** hogepodge has joined #openstack-infra19:00
fungiaha... russellb: john-davidge_: dims: seems i was previously aware of this and then drank those braincells away https://github.com/bbangert/routes/issues/3719:00
dimshaha fungi :)19:01
fungidims: did you maybe have a cached wheel?19:01
*** ildikov has quit IRC19:02
*** _nadya_ has joined #openstack-infra19:02
fungidims: wondering now if pip autobuilding and caching wheels is bringing this issue back19:02
*** harlowja_ has joined #openstack-infra19:02
*** harlowja has quit IRC19:03
*** armax has quit IRC19:03
dimsfungi: something made me believe that haypo took care of it in https://github.com/haypo/routes/commit/10400ba8a269cd3d000db4f47543b9238cc01d5519:03
dimsso will wait for next rev of Routes19:03
*** ildikov has joined #openstack-infra19:03
*** heyongli has quit IRC19:03
*** ociuhandu has joined #openstack-infra19:05
*** heyongli has joined #openstack-infra19:05
*** Rockyg has joined #openstack-infra19:06
*** mriedem has joined #openstack-infra19:06
*** jsavak has quit IRC19:08
*** kdas_ has quit IRC19:08
*** armax has joined #openstack-infra19:10
openstackgerritScott Moser proposed openstack-infra/git-review: support urls as change-id input  https://review.openstack.org/19922519:10
*** heyongli has quit IRC19:10
openstackgerritDoug Hellmann proposed openstack-infra/infra-specs: cleanups for centralize-release-tagging  https://review.openstack.org/19922619:10
*** puranamr has quit IRC19:10
*** radez is now known as radez_g0n319:10
*** soren has joined #openstack-infra19:10
*** cindyo has joined #openstack-infra19:11
*** jsavak has joined #openstack-infra19:11
*** esmute__ has joined #openstack-infra19:12
openstackgerritEmilien Macchi proposed openstack-infra/project-config: specs: fix puppet specs URL  https://review.openstack.org/19922819:13
*** zz_ja is now known as zz_zz_ja19:13
*** ayoung is now known as ayoung-afk19:13
*** signed8bit_ZZZzz is now known as signed8bit19:14
*** yamamoto has joined #openstack-infra19:16
*** heyongli has joined #openstack-infra19:17
fungidims: got it. apparently that's blocking requirements sync to neutron since neutron's gating on python 3 so was working with routes 2.1 but not with the routes 1.x that they end up with after blacklisting 2.119:19
*** pcrews has joined #openstack-infra19:20
dimsah i see. we weave a really tangled web :)19:20
fungiseriously :/19:20
*** yamamoto has quit IRC19:21
*** changbl has joined #openstack-infra19:22
*** heyongli has quit IRC19:24
*** mriedem has quit IRC19:25
*** bdemers has quit IRC19:25
pelixzaro: I'm thinking that although I have this working on JJB at the moment, that I should move it to python-jenkins19:26
dimsfungi: would something like this work for them? http://paste.openstack.org/show/352789/19:26
dims(and all of us!)19:26
pelixsince that is where we really want to test the timeout behaviour and with JJB only should care that we set the correct properties19:26
*** heyongli has joined #openstack-infra19:27
*** signed8bit is now known as signed8bit_ZZZzz19:28
*** abregman has joined #openstack-infra19:28
*** signed8bit_ZZZzz is now known as signed8bit19:28
fungidims: possibly. you could try amending 182746 to see if it passes19:29
fungisince the requirements jobs aren;t going to check unit tests for individual projects19:29
*** mriedem has joined #openstack-infra19:29
*** linkedinyou has joined #openstack-infra19:30
*** shardy_ has joined #openstack-infra19:31
*** heyongli has quit IRC19:32
*** shardy has quit IRC19:32
*** bswartz has quit IRC19:32
*** vnod has quit IRC19:32
*** bswartz has joined #openstack-infra19:32
pleia2mordred: you seem tight with the openstack ansible folks, want to tell them to get their rename patch in? they're still over in stackforge19:34
pleia2and it's been a while :)19:34
mordredpleia2: sure!19:35
pleia2would be great to see them moved over on friday19:35
*** shardy_ has quit IRC19:36
*** tonytan4ever has quit IRC19:37
*** achanda has joined #openstack-infra19:37
*** shardy has joined #openstack-infra19:37
*** luqas has joined #openstack-infra19:38
*** rlandy has quit IRC19:38
*** john-davidge_ has quit IRC19:38
*** smccully has quit IRC19:38
*** heyongli has joined #openstack-infra19:40
*** bdemers has joined #openstack-infra19:41
lifelessmorganfainberg: and yeah - loook at the content of the reply to me on the git list :/19:42
*** tonytan4ever has joined #openstack-infra19:42
lifelessmorganfainberg: 99% 'fit into our community norms', 1% actual review19:42
*** luqas has quit IRC19:42
*** austin81 has left #openstack-infra19:45
*** heyongli has quit IRC19:45
*** austin81 has joined #openstack-infra19:45
morganfainberglifeless: will it male19:46
morganfainbergMe19:46
morganfainbergMake me* sad?19:46
* morganfainberg throws phone smudgy little screen across the room.19:47
Clintit will mail you sad19:47
openstackgerritSean Dague proposed openstack-dev/pbr: Add wsgi_scripts support to PBR  https://review.openstack.org/19529219:47
openstackgerritEmilien Macchi proposed openstack-infra/project-config: puppet & tripleo: use devstack-logs publisher for logs  https://review.openstack.org/19880219:47
EmilienMAJaeger: ^19:47
*** dims has quit IRC19:48
zaropelix: cool and agreed!19:48
*** rcarrillocruz has joined #openstack-infra19:49
rcarrillocruzasselin: heya, it's a shame tomorrow i'll be prepping my vaca (leaving on Thu), won't be attending the sprint19:49
rcarrillocruzand i really wanted to be there19:49
rcarrillocruz:/19:49
*** _nadya_ has quit IRC19:49
*** maishsk has joined #openstack-infra19:49
pleia2asselin: care to send out a reminder about the puppet virtual sprint tomorrow+thursday?19:50
asselinrcarrillocruz, np19:50
asselinpleia2, sure will do19:50
pleia2thanks :)19:50
sdaguelifeless: so I'm not going to say I'm proud of the tests added to https://review.openstack.org/#/c/195292/, but there are some now. And they do test the critical functionality. But it took a lot of jiggery pokery to come up with a pattern that didn't just deadlock python.19:50
asselinrcarrillocruz, enjoy your vacation19:50
morganfainbergClint: i see what you did there.19:50
*** david-lyle has joined #openstack-infra19:50
*** esmute__ has quit IRC19:52
*** ayoung-afk is now known as ayoung19:52
*** heyongli has joined #openstack-infra19:52
rcarrillocruzthx sir19:53
openstackgerritScott Moser proposed openstack-infra/git-review: support urls as change-id input  https://review.openstack.org/19922519:53
*** e0ne has joined #openstack-infra19:54
*** esmute__ has joined #openstack-infra19:55
*** heyongli has quit IRC19:57
*** heyongli has joined #openstack-infra19:59
*** austin81 has quit IRC20:00
*** jtomasek has quit IRC20:00
pleia2not in time for the meeting, but worth mentioning, we're at the midpoint with our Outreachy intern, taron has done an admirable job getting up to speed and we've discussed & reported such to the outreachy community for the midterm update, eyes on hound changes continue to be appreciated :)20:00
*** wshao has joined #openstack-infra20:00
mordredpleia2: w00t20:00
crinkle\o/20:00
persiajeblair: What's the best way to send you the names/email of the folk in question?  Email?  IRC?20:00
*** Longgeek has joined #openstack-infra20:01
taronshould be more patches to review later today, just need to do some reshuffling so I have an ubuntu test vm again20:01
thingeefungi: going to be doing a tag release soon for cinderclient with the revert of version discovery. putting together the release notes, should be quiet otherwise.20:01
pleia2taron: yay!20:01
jeblairpersia: either is good20:01
fungithingee: thanks for the heads up20:01
tarongo has the strangest, ugliest way of compiling things I have ever seen20:01
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Allow JJB to set connection timeout to jenkins server  https://review.openstack.org/15147220:01
fungitaron: that's excellent news!20:02
fungiwell, i mean not the ugliness of go compilation. that's not news at all20:02
taronhahaha20:02
nibalizertaron: \o/20:02
pleia2hehe20:02
fungijust sort of a fact of life20:02
taronvcsrepo { '/opt/hound/src/github.com/etsy/hound':20:02
tarondied a little inside while writing that20:02
pelixzaro: I've modified the timeout patch to abstract from socket and confirmed it works as expected, also updated to support floats which are supported.20:03
nibalizerpleia2: fungi could I have core on beaker-localhost and beaker-nodepool20:03
anteayataron pleia2 fungi is there a gerrit topic selected for hound changes?20:03
Clint/opt is pretty bad20:03
pelixzaro: be tomorrow or the next day when I get around to adding the tests to python-jenkins for timeout behaviour20:03
nibalizerassuming jeblair doesn't want to pop those out since it did go without his +120:03
pleia2anteaya: no, but the work is happening in puppet-hound (eventually system-config for our specific stuff)20:03
pelixzaro: just need to convert from using gevent to something more lightweight20:03
pleia2puppet-hound repo, I mean20:04
yolandafungi, jeblair, i was wondering if you could cut some release for gerritlib. I have a patch for manage-projects, to add descriptions of projects, blocked for very long time, and just depends on some code that landed in gerritlib20:04
taronClint: that's about the one part of that path I can modify if you suggest it being placed somewhere else20:04
anteayapleia2: oh so just anything to the puppet-hound repo20:04
taronsuppose I could do it in a service user's homedir or something instead20:04
Clinttaron: you should wait several months before baiting me to express an opinion on that20:04
*** puranamr has joined #openstack-infra20:04
pleia2anteaya: yep!20:04
anteayapleia2: okay thanks20:04
taronheh, k20:04
pleia2anteaya: the boilerplate landed yesterday20:05
anteayaokay20:05
anteayagood20:05
*** pelix has quit IRC20:05
lifelesssdague: thanks, I'll look soon20:05
nibalizeranteaya: taron 'code-search' is the topic the spec states should be our topic for the hound stuff20:05
pleia2nibalizer: oh, thanks20:05
*** chlong has joined #openstack-infra20:06
*** jtomasek has joined #openstack-infra20:07
*** marzif_ has joined #openstack-infra20:07
anteayanibalizer: ah thank you20:07
*** armax has quit IRC20:08
lifelessmorganfainberg: dunno, does http://www.spinics.net/lists/git/msg255543.html make you sad ?20:08
*** reazem_ has quit IRC20:09
funginibalizer: those are the projects yolanda was talking about having been originally configured to point to a nonexistent acl20:09
*** jtomasek has quit IRC20:09
*** heyongli has quit IRC20:09
*** jtomasek has joined #openstack-infra20:10
morganfainberglifeless: yes. Can haz gerrit. Or at least something with pull requests barring a nice tool for bigger communities? :(20:10
funginibalizer: https://review.openstack.org/198938 needs to be approved to fix them20:10
*** amrith is now known as _amrith_20:11
*** austin81 has joined #openstack-infra20:11
pleia2on our way20:11
*** heyongli has joined #openstack-infra20:14
*** fawadkhaliq has quit IRC20:15
*** puranamr has quit IRC20:15
*** marzif__ has joined #openstack-infra20:16
*** armax has joined #openstack-infra20:17
*** marzif_ has quit IRC20:18
*** ivar-laz_ has joined #openstack-infra20:18
*** jtomasek has quit IRC20:19
*** puranamr has joined #openstack-infra20:19
openstackgerritMerged openstack-infra/project-config: Fix ACL's on projects with puppet-system  https://review.openstack.org/19893820:19
*** spzala has quit IRC20:19
*** salv-orlando has quit IRC20:20
*** kgiusti has left #openstack-infra20:21
*** ivar-lazzaro has quit IRC20:21
*** Longgeek has quit IRC20:24
*** marzif__ has quit IRC20:26
*** akraminsky has joined #openstack-infra20:28
akraminskyI'm looking for a file called multinode_setup.sh which is used as the ready-script for multinode projects. I can't seem to locate it, does anyone know off the top of their head where I can find it?20:28
fungiakraminsky: http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/scripts/multinode_setup.sh20:29
akraminskyfungi, thanks!20:30
fungiakraminsky: you're welcome20:31
mgagneso nodepool thinks it's a good idea to delete my node while a job is still running on it. Is there a known bug or broken python-jenkins/jenkins versions in the wild?20:33
*** dprince has quit IRC20:34
fungimgagne: do you have info from the nodepool debug log? it should normally base that decision on receipt of a zeromq message from the jenkins master stating that the job has completed20:35
fungimgagne: is it possible the worker is not being put into single-use mode and has started running a new job before nodepool processes the deletion?20:36
mgagnefungi: I don't know how to check that info. How can I check?20:36
akraminskymgagne: you can tail -f /var/log/nodepool.debug.log while running the job20:37
fungimgagne: see if a running job has an OFFLINE_NODE_WHEN_COMPLETE parameter being set20:38
akraminskymgagne: err sorry /var/log/nodepool/debug.log20:38
mgagneforgot about that file, checking now20:38
mgagneOFFLINE_NODE_WHEN_COMPLETE is not found in the file20:38
fungimgagne: the parameter value should be 1 to tell jenkins to set the node offline so that it won't pick up a second job while it awaits deletion20:38
fungimgagne: i mean in the job parameters in jenkins20:38
*** coolsvap is now known as coolsvap|away20:39
fungimgagne: like https://jenkins01.openstack.org/job/gate-puppet-keystone-puppet-unit-4.0/80/parameters/20:39
mgagne:O wth is this parameter :O20:39
*** heyongli has quit IRC20:39
fungimgagne: we use a zuul parameter function to set it for our jobs that run on single-use workers20:40
mgagnesure but I never read anything about it in nodepool docs if you see what I mean20:40
fungimgagne: it's a jenkins+zuul detail not anything specific to nodepool20:40
mgagnehmmm, I'm not using zuul or gearman20:41
mgagneonly zeromq plugin20:41
*** jamielennox is now known as jamielennox|away20:41
fungimgagne: aha, the triggering on that parameter is done within the jenkins-gearman plugin so that might be a coupling concern we didn't foresee20:41
*** madong_ has joined #openstack-infra20:42
mgagnenode is marked "USED" as soon as the job starts20:42
fungimgagne: yep, and then marked as delete when the job completes. but if it takes a few seconds for nodepool to remove it from jenkins then it might start running a second job20:43
mgagneyep, that's what happened20:43
fungiand then get deleted in the beginning of that second job20:43
*** madong has quit IRC20:43
*** e0ne has quit IRC20:44
*** heyongli has joined #openstack-infra20:44
fungijeblair: ianw: clarkb: zaro: ^ any ideas on how/if that can be worked around, short of running zuul? like set the parameter in job configuration and install jenkins-gearman but point it at a dummy gearman on localhost or something?20:44
mgagnefungi: could it be related to that bug? https://review.openstack.org/#/c/193600/20:45
fungiwondering if there could be a stripped-down jenkins-gearman plugin which just does the OFFLINE_NODE_WHEN_COMPLETE bit20:45
*** smccully has joined #openstack-infra20:45
*** Sukhdev has quit IRC20:45
*** erlon has joined #openstack-infra20:46
fungimgagne: maybe that could decrease the incidence, but the underlying issue (and reason we have OFFLINE_NODE_WHEN_COMPLETE in the jenkins-gearman plugin) is so that jenkins itself can avoid a race condition between job completion and node deletion20:47
mgagnetrue20:47
mgagneso unless I use gearman plugin or write my own to offline the node, there isn't much I can do for now20:48
fungimgagne: yeah, i mean the routines and framework are there in the jenkins-gearman plugin source already, so it might not be hard to strip out the bulk of the things you're not using to just have an offline-when-complete plugin for jenkins20:48
mgagneor I move the cleanup cron to a time where jobs aren't running and accept the fact that a node could be used multiple times before it's cleaned up20:50
fungithough i'm not well versed in java, so consider that a guess on my part20:50
*** jasondotstar has joined #openstack-infra20:50
fungimgagne: that's a possible workaround too, i think. we have a bunch of jobs where rerunning them on the same worker more than once will just outright fail, so that's not really something we've tried20:51
*** jamielennox|away is now known as jamielennox20:52
mgagneour jobs aren't that destructive (yet) so it's a compromise I can accept for now20:52
*** rtom has joined #openstack-infra20:54
*** bswartz has quit IRC20:54
*** heyongli has quit IRC20:55
*** ivar-laz_ has quit IRC20:57
*** ivar-lazzaro has joined #openstack-infra20:57
*** HeOS has joined #openstack-infra20:58
*** marzif__ has joined #openstack-infra20:58
*** puranamr has quit IRC21:01
*** puranamr has joined #openstack-infra21:03
jeblairmgagne: you want to run nodepool without zuul?21:03
mgagneyea21:03
jeblairwhy?21:03
*** wshao has quit IRC21:03
*** thorst has quit IRC21:03
mgagnejeblair: yea, I am already and it's working fine (except when a race occurs)21:03
mgagnejeblair: because I don't have time to setup all that stuff21:03
mgagnejeblair: I'm working on it on borrowed time21:04
*** julim_ has quit IRC21:04
jeblairmgagne: well, it's not designed for that, as you have found.  you would need to add a new plugin to jenkins to do the locking that the gearman plugin does.  it's not easy, which is why it's implemented the way it is.21:05
mgagnejeblair: well, nodepool docs doesn't mention zuul much and let you understand that it's possible to run it without zuul21:05
mgagnejeblair: which I'm currently doing21:06
mgagnehttp://docs.openstack.org/infra/nodepool/installation.html#zuul21:06
jeblairmgagne: i'd recommend either using zuul (puppet-zuul and puppet-openstackci are intended to make it easier, but might not be there yet), or using something like jclouds which is designed for jenkins generally21:06
openstackgerritCody Herriges proposed openstack-infra/project-config: Move "bundle install" for puppet-modules-jobs.  https://review.openstack.org/19929121:06
jeblairthat might be very old :/21:07
*** maishsk has quit IRC21:07
*** aysyd has quit IRC21:07
*** dustins has quit IRC21:07
mgagnenodepool is working fine except for that race which I "fixed" by moving the cleanup cron to 4am when there is no job running21:08
*** heyongli has joined #openstack-infra21:08
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add stable ACLs for Neutron sub-projects  https://review.openstack.org/19875021:08
openstackgerritKyle Mestery proposed openstack-infra/project-config: Add a create ACL for networking-bgpvpn  https://review.openstack.org/19875121:08
openstackgerritKyle Mestery proposed openstack-infra/project-config: Update merge commit gerrit ACL for Neutron repositories  https://review.openstack.org/19908921:08
openstackgerritKyle Mestery proposed openstack-infra/project-config: Update ACLs for Neutron sub-projects  https://review.openstack.org/19874921:08
mgagnezuul might come later, not now21:08
asselinpleia2, comments on reminder e-mail? https://etherpad.openstack.org/p/common-ci-sprint-reminder21:08
jeblairmgagne: it's not deleting nodes after jobs run?21:08
mgagnejeblair: it's something I can accept with my kind of jobs21:08
jeblairmgagne: oh i see21:09
mgagnejeblair: we already have permanent slaves21:09
openstackgerritScott Moser proposed openstack-infra/git-review: support urls as change-id input  https://review.openstack.org/19922521:09
jeblairso you're not trying to use it to make single-use slaves?21:09
mgagnejeblair: just trying to move to nodepool to automate slaves update and move away from puppet21:09
*** rtom has quit IRC21:09
openstackgerritCody Herriges proposed openstack-infra/project-config: Move "bundle install" for puppet-modules-jobs.  https://review.openstack.org/19929121:09
mgagnejeblair: if it works, yes. otherwise I don't mind much for now21:09
jeblairi think i get the picture21:09
pleia2asselin: reviewing, and looking up ref re: timezones now21:10
pleia2(I think it was at a recent meeting)21:10
openstackgerritCody Herriges proposed openstack-infra/project-config: Move "bundle install" for puppet-modules-jobs.  https://review.openstack.org/19929121:10
mgagneif I'm *really* annoyed by garbage on slaves and need *real* single-use, I'll check zuul or just write a plugin for jenkins21:10
asselinpleia2, yeah...not sure how to search for that (easily)21:10
pleia2asselin: found and adding21:11
*** abregman has quit IRC21:11
asselinpleia2, did you login and grep the file systm?21:12
*** dims has joined #openstack-infra21:13
*** austin81 has quit IRC21:13
pleia2asselin: no, I just opened up recent meeting logs until I found it21:13
*** ociuhandu has quit IRC21:13
taronmordred: how're you going about generating the config file for hound.o.o? trying to come up with a plan of attack for the module21:13
*** wshao has joined #openstack-infra21:13
pleia2used the standard logs for the link, since that allows me to link to a timestamp <321:13
asselinpleia2, thanks21:13
taronit builds which is all well and good until you try to run it and it dies from lack of config.json21:13
pleia2asselin: email looks good to me now21:14
jeblairfungi, mgagne: i don't think the suggestion from fungi at 20:44 will work -- i believe that parameter is only considered by jenkins gearman plugin if it comes with a gearman job start request21:14
mordredtaron: I wrote a patch to jeepyb ... one sec21:14
asselinpleia2, ok...need to go to a meetgin...will send it off when I get back (about 30 mins)21:14
fungijeblair: ahh, yep that seems likely. so it really would need a separate plugin or an actual zuul21:14
mgagnejeblair: yes. However I'm suggesting writing a new plugin replicating this behavior21:14
EmilienMhello infra, I hope sun is shinning for you, can we have a review on https://review.openstack.org/#/c/199291/ to fix our upgrade job please ?21:15
mordredtaron: https://review.openstack.org/#/c/178474/21:15
jeblairmgagne: yes, that may work.  however, i believe that will be more work than setting up zuul.21:15
taronmordred: much obliged21:15
mordredtaron: idea is that we have a projects.yaml already - so we can just tranform that information21:16
*** jasondotstar has quit IRC21:16
*** maishsk has joined #openstack-infra21:16
jeblair(but with a payoff for the community, and the gearman plugin could have a cleaner design; so not a bad idea)21:16
*** ldnunes has quit IRC21:16
*** heyongli has quit IRC21:17
*** armax has quit IRC21:17
taronmordred: yeah, that's how I assumed it would work for our purposes, generalizing it for the module may take a bit more thinking21:17
taroneither that or it could just not manage config.json, but that seems a bit odd21:17
*** annegentle has quit IRC21:18
taronsomething toggleable, perhaps21:18
fungitaron: also you can crib from where we install jeepyb in http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/git_backend.pp to run create-cgitrepos21:18
fungitaron: and also from that same puppet class you can see how to get projects.yaml on the system21:18
taronoh cool21:19
* taron will do some reading21:19
fungitaron: though a lot of the same glue can be found in the hound class from monty's poc change https://review.openstack.org/#/c/178488/5/modules/openstack_project/manifests/hound.pp (which is linked in the code-search spec)21:19
*** dizquierdo_afk is now known as dizquierdo21:20
fungiso some viable examples to work from21:20
*** sdake_ is now known as sdake21:20
* taron needs to reread the spec, evidently21:20
taronindeed21:20
fungitaron: for reference, mentioned in work item #2 at http://specs.openstack.org/openstack-infra/infra-specs/specs/code-search.html#work-items21:21
*** MarkAtwood has quit IRC21:21
*** cindyo has quit IRC21:21
*** bmace has quit IRC21:21
*** salv-orlando has joined #openstack-infra21:22
*** fallenpegasus has quit IRC21:22
*** heyongli has joined #openstack-infra21:24
*** dboik has quit IRC21:25
*** marzif_ has joined #openstack-infra21:26
*** marzif__ has quit IRC21:26
*** austin81 has joined #openstack-infra21:26
*** miqui has quit IRC21:27
*** heyongli has quit IRC21:28
*** heyongli has joined #openstack-infra21:29
*** eharney has quit IRC21:29
*** salv-orlando has quit IRC21:29
*** salv-orlando has joined #openstack-infra21:30
*** dims has quit IRC21:30
*** jsavak has quit IRC21:30
*** esmute__ has quit IRC21:32
*** marzif_ has quit IRC21:32
*** pradk has quit IRC21:32
*** esmute__ has joined #openstack-infra21:33
*** heyongli has quit IRC21:33
*** bmace has joined #openstack-infra21:33
*** heyongli has joined #openstack-infra21:35
anteayapleia2: ah just reading the email you sent out about the renames21:35
*** dims has joined #openstack-infra21:35
greghaynesclarkb: so I think I fugured out the race21:35
anteayapleia2: the compass ones21:35
*** bhunter71 has quit IRC21:35
*** dims has quit IRC21:35
*** Guest7393 has joined #openstack-infra21:35
anteayapleia2: compass was discussed at the tc meeting today and I belief the agreement was they have some things to work on yet21:35
greghaynesclarkb: I wasnt naming the nodepool builder thread, so it was getting a default name of Thread- which is in out wait_for_thread whitelist21:36
pleia2anteaya: I just copied from https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Upcoming_Project_Renames21:36
anteayas/belief/believe21:36
anteayapleia2: yup I understand21:36
pleia2anteaya: so we can update that if needed :)21:36
anteayaI'm thinking that entry bears some further scrutiny21:36
krotscheckI remember conversations about migrating to a different IRC bot. Which one did we want to migrate to?21:36
pleia2the list is in flux, so we'll just want to make sure it's up to date on Thursday when I do prep21:36
jeblairkrotscheck: undecided, potentially errbot, but needs some more thought and a plan21:37
pleia2krotscheck: I think we were looking at eir21:37
*** mriedem is now known as mriedem_away21:37
pleia2well, something with an e and an r anyway!21:38
krotscheckjeblair, pleia2: Thanks. Was going to put together an ansible script to create an info bot for a private team channel, because OMG organizational knowledge. Wanted to make sure whatever I ended up using was the same thing in use in infra.21:38
krotscheckHrm. errbot is GPLv3 :/21:38
*** heyongli has quit IRC21:39
mordredkrotscheck: yup21:39
funginothing complex about running a gplv3 tool unless you need to modify its to get it to do what you want it to do21:39
mordredfungi: nothing complex about that either21:40
krotscheckfungi: Yeah, but if we want to write a plugin, that's considered derivative according to the apache foundation.21:40
mordredfungi: the only thing that's complex about it is if you want to ship a proprietary product based on that code21:40
fungimordred: other than if you need your employer to okay you contributing changes to a gplv3 project on work time21:40
krotscheckthe GPLv3 people consider even linking to it a derivative work.21:40
jeblairkrotscheck: when you say "private team channel", i'm assuming your talking about something extra-curricular to openstack, yeah?21:40
mordredkrotscheck: the openstack foundation's legal council has ok'd the infra project having and producing GPLv3 code21:40
jeblairand yes, if we were to use it, i would expect that if we wrote a plugin, it would be gplv3 for that reason21:41
krotscheckmordred: Oh good.21:41
*** _amrith_ is now known as amrith21:41
*** amitgandhinz has quit IRC21:41
mordredkrotscheck: so - yeah - I think errbot is our best theory so far - but we do not yet have concrete plans21:41
fungiyep, i'm not _personally_ concerned about working on gplv3 projects. just noting that some people may not be allowed by their employers to contribute as part of their jobs. though the company i'm thinking of has at least a couple of people here that likely would mention if that were really a problem ;)21:42
Clintthat is also true of Apache-2.021:42
fungiindeed it is21:43
mordredfungi: are any of those companies companies that I work for?21:43
*** heyongli has joined #openstack-infra21:43
fungimordred: possibly, yes21:43
mordredfungi: well, theoretically, it's not a problem21:43
mordredhypothetically of course21:43
*** rlandy has joined #openstack-infra21:44
fungithough i do think we would need to _not_ turn on cla enforcement for any gplv3 project we're directly maintaining in our gerrit, because that would just be silly21:44
mordredfungi: funny story ...21:44
Clintpretty sure the company i work for distributes gplv3 code21:44
fungidco would likely be fine. our icla is pretty apache-specific though21:44
* fungi makes ianal handsign to ward off evil eye21:45
mordredfungi: I believe the terms of the apache license are such that code contributed under it could be added to a gplv3 final work21:45
jeblairfungi: when i dig deep -- really really deep -- for a reason for openstack to have a cla, the _only_ thing i can come up with is "so that the foundation can, in the future, decide to release openstack under gplv3".21:45
openstackgerritsebastian marcet proposed openstack-infra/openstackid: [smarcet]  https://review.openstack.org/19845821:46
jeblair(i mean, after it changes its bylaws)21:46
Clintit could do that without a cla, no?21:46
mordredfungi: I'm not saying I think this is necessary, of course, but I dont' think having it on breaks anything more than it breaks other things21:46
mordredClint: it can do literally everything without a CLA that it thinks it needs a CLA to do21:46
fungidual-license apl2/gpl3 would be an interesting combo21:46
jeblairmordred: er, it could not relicense code written by others21:46
*** fallenpegasus has joined #openstack-infra21:47
mordredjeblair: no need - it could release a gplv3 licensed derivative work21:47
openstackgerritsebastian marcet proposed openstack-infra/openstackid: OIDC - OpenId Connect Implementation  https://review.openstack.org/19845821:47
jeblair(but of course, apache is liberal so that's pretty eassy to do in effect anyway)21:47
mordredyah. that's my point21:47
jeblairmordred: agreed, the outcome would be effectively the same21:47
mordrednow - the foundation being allowed to make that decision would involved a fascnatingly large amount of bylaws changes I believe21:48
krotscheckFor those of us not on the private channel where I just answered jeblair's question: It's a channel of HP people who do a lot of infra work, and we need infobot things for all the crazy links and forms and hoops. I'd rather just use the thing we're likely going to use in public.21:48
mordredthere are SO MANY CRAZY LINKS AND FORMS AND HOOPS21:48
fungisounds like lots of fun i don't want to know about ;)21:48
mordredwait. no there aren't... you shold all come work for us21:48
*** sdake_ has joined #openstack-infra21:48
zaromgagne: sorry i was afk, did you figure out a workaround?21:48
jeblairkrotscheck: anyway, back to your question; we currently run supybot, so investment in using a supybot infobot thingy might not be a waste (in that we'd be converting eventually anyway)21:49
mgagnezaro: yea, just delete the used node at night instead of every minutes and let jobs reuse them (which I'm fine with)21:49
krotscheckjeblair: Aight. I can do that.21:49
EmilienMfungi, mordred: can I get a final +A on https://review.openstack.org/#/c/199291/ please ?21:50
jeblairkrotscheck: i'm not actually sure which would be easier/better.  hth.21:50
mordredjeblair: based on a thing Clint said - I want to start geotagging all of my git commits ...21:51
krotscheckjeblair: ¯\_(ツ)_/¯21:51
*** heyongli has quit IRC21:51
*** wshao has quit IRC21:51
jeblairmordred: and add that to stackalytics?21:51
jeblairkrotscheck: pretty much :)21:51
mordredyup21:51
*** Longgeek has joined #openstack-infra21:51
mordredjeblair: preferrably with some time data to show paths of patches for a person through a cycle21:51
*** sdake has quit IRC21:52
*** tonytan4ever has quit IRC21:52
*** nilasae has quit IRC21:53
*** whoops has quit IRC21:54
openstackgerritMerged openstack-infra/release-tools: Close milestones for post-version releases  https://review.openstack.org/19513021:54
mordredjeblair: ZOMG. THIS IS ACTUALLY A THING!!!!!21:54
mordredjeblair: pip install geocommit21:54
openstackgerritMerged openstack-infra/release-tools: Handle more errors when updating bugs  https://review.openstack.org/19524221:54
*** pc_m has quit IRC21:55
*** heyongli has joined #openstack-infra21:55
openstackgerritMerged openstack-infra/release-tools: Catch errors checking for zuul status  https://review.openstack.org/19531921:56
*** tonytan4ever has joined #openstack-infra21:56
fungineat. it just stores the location data in git notes21:56
openstackgerritgreghaynes proposed openstack-infra/nodepool: Trigger image builds using gearman  https://review.openstack.org/19517521:57
*** annegentle has joined #openstack-infra21:57
*** Longgeek has quit IRC21:57
greghayneshow does that work with patchsets though - do you just get the last location you worked added?21:57
greghaynesreally you want a list of locations21:57
mordredgreghaynes: that's an excellent point21:58
greghaynesso maybe git-review should support geocommit21:59
fungi"all the pubs^H^H^H^Hplaces where i worked on this commit"21:59
greghaynesoh yes! we can also add in a BAC feature21:59
mordredgreghaynes: clearly this is what we shoudl all drop everything to work on21:59
*** changbl has quit IRC21:59
*** heyongli has quit IRC21:59
greghaynesmordred: I thought thats why we have priority efforts - for when stuff like this comes along22:00
mordredgreghaynes: there is also a thing called "doko" which is a command line python thing for getting you your GPS coordinates22:01
openstackgerritMerged openstack-infra/project-config: Move "bundle install" for puppet-modules-jobs.  https://review.openstack.org/19929122:01
openstackgerritMerged openstack-infra/release-tools: Fix --stable flag  https://review.openstack.org/19671222:01
*** heyongli has joined #openstack-infra22:01
greghaynesmordred: heh, I may be a nerd about that and have a beaglebone hooked up to a SPI gps that parses that out using pynmea22:02
mordredgreghaynes: I may want one of those22:03
*** bmace has left #openstack-infra22:03
greghayneshrm, I bet you could find a serial gps, then just usb -> serial and do it on your laptop22:03
fungior bt it from your phone22:04
greghaynesoh right, we do all have gps things on us nowdays dont we ;)22:04
fungii see several smartphone apps for different platforms that let you query your gps data over bluetooth from a computer22:05
*** wshao has joined #openstack-infra22:05
fungino idea whether there are free software client implementations or if you'd need to reverse-engineer a hackish and unversioned high-level protocol22:05
*** heyongli has quit IRC22:06
openstackgerritSpencer Krum proposed openstack-infra/system-config: Test the apply test  https://review.openstack.org/19932122:06
*** heyongli has joined #openstack-infra22:07
*** thorst has joined #openstack-infra22:07
*** dimtruck is now known as zz_dimtruck22:09
*** unicell has quit IRC22:10
*** unicell has joined #openstack-infra22:10
*** heyongli has quit IRC22:12
*** heyongli has joined #openstack-infra22:13
*** ari has quit IRC22:13
*** esmute__ has quit IRC22:15
mordrednibalizer: best commit message eer22:16
*** esmute__ has joined #openstack-infra22:16
*** prad has quit IRC22:16
nibalizerheh22:16
nibalizerI want to make sure it fails when it should be failing, pabelanger has me all worried22:17
nibalizerlooks like it does22:17
mordrednibalizer: he'll do that22:17
mordrednibalizer: wily canadians22:17
*** thorst has quit IRC22:18
mordredlifeless: you think we do not need the oscc patch?22:19
*** austin81 has quit IRC22:19
*** heyongli has quit IRC22:19
*** esmute__ has quit IRC22:19
lifelessmordred: I think we do not need it. Its not solving a problem that I can see, and it creates inconsistency and surprises for developers.22:20
*** wshao has quit IRC22:20
*** esmute__ has joined #openstack-infra22:20
lifelessas well as likely throwing crap into stdout during the test suite since I very much doubt it was developed entirely cleanly so far [in that you've had the belts-and-braces, which stops you seeing cruft]22:20
mordredlifeless: you are correct on this22:21
mordredlifeless: I'm happy to abandon the change in oscc ... I just mainly want to make sure the oscc-based fixture is in oslotest so that everyone can use it in their functional tests22:21
*** heyongli has joined #openstack-infra22:22
*** fawadkhaliq has joined #openstack-infra22:22
lifelessso a fixture that uses oscc makes sense to me22:22
lifelessis it in a dedicated file?22:22
lifelessI'd make the dependency from oscc optional I think22:22
dhellmannif it's a fixture, why not just put it in oscc?22:22
lifelesse.g. in setup.cfg22:22
lifeless[extras]22:23
lifelessos-client-config =22:23
dhellmannwe're doing similar things in the other oslo libs, for similar reasons22:23
lifeless  os-client-config22:23
lifelessor as dhellmann says just have the fixture inside os-client-config22:23
lifelessfolk that need it have to know that they're using os-client-config one way or another22:23
dhellmannright22:23
*** fallenpegasus has quit IRC22:24
mordredno - they don't22:24
mordredthat's the entire point22:24
*** linkedinyou has quit IRC22:24
mordredis that they do NOT need to know that they're using oscc, and are, in fact, unlikely to know much about it per-se22:25
mordredit's used in a general setup fixture for getting cloud credentials in a devstack-based functional test22:25
*** fallenpegasus has joined #openstack-infra22:26
mordredwhat it winds up returning is a python dictionary with expected auth info in it22:27
lifelessok22:27
*** austin81 has joined #openstack-infra22:27
dhellmannok. Why don't we want them to know where that comes from? Just to make it easier?22:27
*** puranamr has quit IRC22:28
mordredwell, I mean, they could - but they already depend on oslotest, so it's an easy way to just get them the functionality now that devstack exports it22:28
*** linkedinyou has joined #openstack-infra22:28
*** armax has joined #openstack-infra22:28
mordredI mean, I could also just go patch-bomb all the projects ... my patch count _is_ getting low :)22:28
*** annegentle has quit IRC22:29
*** puranamr has joined #openstack-infra22:29
dhellmannI'm just trying to make sure I understand. We get a lot of "but there are so many extra dependencies" about some of these libs that folks start not wanting to use them. If this is a thing they need, then making that dependency explicit elsewhere may be educational.22:29
mordreddhellmann: I can see taht22:30
dhellmann*I* want oslotest to have stuff like this, I'm just not sure its users do.22:30
dhellmannalso, with this circular dependency issue...22:30
lifelesswell22:30
*** heyongli has quit IRC22:30
lifelessits not a circular dependency22:30
dhellmannalthough I guess it's a test dep, so -- yeah22:30
lifelessneither setup_requires the other22:31
*** rlandy has quit IRC22:31
lifelessso you can build them both sanely to binary form.22:31
*** bknudson has quit IRC22:31
dhellmannoslotest does require oscc, no?22:31
lifelessYou can run the tests for the other with the prior version22:31
dhellmannoh, I see22:31
dhellmanninstall_requires22:31
lifelessand once you've done that you can do the other, and then install them both22:31
lifelessno races22:31
* dhellmann doesn't know how those things apply to the way the debs are being packaged22:31
lifelessno binary-blob-from-deity-to-get-started22:32
lifelessdhellmann: Debian is right now evolving its tooling to better support such nuance.22:32
lifelessdhellmann: but since they're all already in the archive,there should be no drama22:32
* dhellmann hates it when we all spend so much time on things that shouldn't be an issue22:33
*** dizquierdo has quit IRC22:33
dhellmannso should we just leave everything alone? because even making oscc an optional dependency of oslotest means we have to tell users they need it to use some of the oslotest features, which I thought was what mordred wanted to avoid22:33
*** esmute__ has quit IRC22:33
openstackgerritKhai Do proposed openstack-infra/puppet-openstackci: Add Jenkins Job Builder manifest  https://review.openstack.org/18491922:34
*** esmute__ has joined #openstack-infra22:34
*** Guest15874 has joined #openstack-infra22:34
lifelessdhellmann: I think we should just leave well enough alone22:35
lifelessdhellmann: until we've really understood zigo's issue22:35
dhellmannlifeless: ++22:35
lifelessif its what I think it is, I think its a non-issue22:35
lifelessif its something else22:35
zigoWhat issue?22:35
lifelesslets work through it and not knee-jerk22:36
lifelesszigo: see the last 40 or so lines of backscroll22:36
* dhellmann heads off to dinner22:36
zigoOh, about the build-dependency loop?22:36
*** gordc has quit IRC22:36
zigoI think it's not too bad, as removing os-client-config from the build-depends is ok.22:36
zigoIs it really used?22:37
zigoEither it's not, or the unit tests aren't using it at least.22:37
zigoBut really, if it was in use in the unit tests, it would be.22:39
zigoI just hope it never will.22:40
zigoWhich is why I sent a mail to the list.22:40
*** esmute__ has quit IRC22:42
lifelesszigo: I don't understand why its a build-dependency loop22:42
lifelesszigo: its not upstream22:42
openstackgerritKhai Do proposed openstack-infra/system-config: Use third party ci manifest to setup jenkins job builder.  https://review.openstack.org/19933522:42
zigolifeless: oslotest has os-client-config in requirements.txt, so I've put it in Build-Depends-Indep. But in fact, it doesn't use it during tests, it seems, so I finally removed it. Then os-client-config uses oslotest during its unit tests...22:45
*** Sukhdev has joined #openstack-infra22:45
zigoSo, it's a non-issue if oslotest *never will* use os-client-config during its unit tests.22:45
*** chlong has quit IRC22:46
lifelesszigo: requirements.txt is for install requirements, not build requirements.22:46
lifelesszigo: all openstack packages have one and only one build requirement: pbr.22:47
*** sabeen3 has quit IRC22:47
lifeless(unless you have a different definition of build than I'm expecting)22:47
zigolifeless: In most cases, all stuff in requirements.txt are also build dependencies required to run unit tests.22:47
lifelessyou can run unit tests with the prior version of runtime dependencies22:47
zigoI do run unit tests on build, always...22:47
lifelesswe know this because we sequence all our commits in one at a time22:47
*** tonytan4ever has quit IRC22:48
*** _nadya_ has joined #openstack-infra22:50
openstackgerritgreghaynes proposed openstack-infra/nodepool: Trigger image uploads using gearman  https://review.openstack.org/19772222:51
asselinzaro, thanks for the updates....will take a look22:52
*** esmute__ has joined #openstack-infra22:54
*** _nadya_ has quit IRC22:54
*** Rockyg has quit IRC22:57
hogepodgejeblair: fungi: fits-list is nothing but a big spam dump now, and is replaced by defcore-committee. What can I do to help shut that list down?22:57
*** esmute__ has quit IRC22:57
*** r-daneel has quit IRC23:00
*** fawadkhaliq has quit IRC23:02
*** Guest15874 has left #openstack-infra23:02
fungihogepodge: i don't think we've (yet) closed down a mailing list on our mailman instance. it's been a while since i've done that. i suspect the options there (you seem to be an admin for it) are to remove all the subscribers and adjust the moderation mode so that all new posts will be rejected with a message saying the list is retired (or just black-hole them)23:03
*** esmute__ has joined #openstack-infra23:04
*** soren has quit IRC23:04
pleia2we shut down and deleted the non-announce third party one23:04
jeblairhogepodge, fungi: yeah, i think either something like that, or we could see about adding something to puppet to tell exim to ignore some lists (and therefore, msgs addressed there would be rejected)23:04
mordred++23:04
mtreinishfungi: we were going to close down the openstack-qa list, but until the fabled dashboard exists we still need it for periodic job notifications23:05
jeblairpleia2: yeah, i think there were mitigating circumstances for that one that led us to think we could delete it (such as, no substantial messages were sent to it)23:05
pleia2yeah, we didn't need to keep anything, it was just deleted23:05
*** esmute__ has quit IRC23:06
*** heyongli has joined #openstack-infra23:08
lifelessfungi: so23:09
*** puranamr has quit IRC23:09
*** ashleighfarnham has quit IRC23:10
lifelessfungi: http://logs.openstack.org/periodic/propose-requirements-constraints-master/50fc7d4/console.html#_2015-07-07_06_05_57_92623:10
fungilifeless: i think we're hoping for additional reviews on https://review.openstack.org/19892623:10
*** esmute__ has joined #openstack-infra23:11
*** sridhar_ram has joined #openstack-infra23:11
hogepodgefungi: jeblair: pleia2: I'd be happy with just removing the subscribers and rejecting all new posts. I can set it up, and set a message out saying the list is about to be shut down. Then after some point I can ping you again to delete it23:11
*** thedodd has quit IRC23:11
lifelessthere's a -1 there already23:12
lifelessmordred: https://review.openstack.org/#/c/198926/ <- can you please either +2A or -1 ?23:12
*** annegentle has joined #openstack-infra23:12
*** heyongli has quit IRC23:13
*** esmute__ has quit IRC23:13
*** wenlock has quit IRC23:13
fungijhesketh: or if you're up to revisit your current -123:13
*** esmute__ has joined #openstack-infra23:14
fungiunclear whether you -1'd for the lack of a cleanup patch (which i've tested and should not need to be set as a dependent patch from what i can tell) or for the choice of requirements file location, or both23:15
jheskethfungi: yep, let me look23:15
mordredI +A'd it. I think the cleanups are thingks we can do with a followup23:15
fungii actually think that adding the cleanup patch after will be safer anyway23:15
fungibecause this lets us be absolutely certain it's working before we potentially rip it off our nodepool images and wind up with a fire drill scramble to delete new images and recheck a bunch of failed changes23:16
jheskethfungi: ah it was because I wasn't a fan of putting a reqs file in etc. In fact I don't think we need one at all and we could just pip install into the venv23:17
*** hichihara has joined #openstack-infra23:17
fungionce it ends up applied to the proposal/pypi job workers i'll propose the cleanup for our prep scripts/dib elements23:17
*** wenlock1 has quit IRC23:17
*** akshai has quit IRC23:18
fungijhesketh: perfectly fair, however we already have this patch which should work and it's blocking getting requirements constraints proposed to many, many repos. we can always improve the mechanism we use to install it later23:18
openstackgerritSridhar Ramaswamy proposed openstack-infra/project-config: Add tacker-horizon to gerritt IRC notifications  https://review.openstack.org/19934123:18
jheskethSure, it's certainly pretty harmless as is so I'm fine with that :-)23:18
fungithat will allow us to take our time on a more aesthetically pleasing modification to use a pip exec rather than a requirements file23:19
*** bswartz has joined #openstack-infra23:19
*** nelsnelson has quit IRC23:20
openstackgerritMerged openstack-infra/system-config: Install zuul-env venv with puppet  https://review.openstack.org/19892623:20
jheskethSounds good to me23:20
*** bhunter71 has joined #openstack-infra23:22
*** linkedinyou has quit IRC23:23
*** esmute__ has quit IRC23:23
*** esmute__ has joined #openstack-infra23:23
*** yamada-h has joined #openstack-infra23:26
*** gildub has joined #openstack-infra23:26
*** salv-orlando has quit IRC23:26
*** krtaylor has quit IRC23:29
*** heyongli has joined #openstack-infra23:30
lifelessnuts23:34
lifelesshttp://logs.openstack.org/28/184328/5/check/gate-oslo.db-requirements/55e3703/console.html#_2015-07-07_22_41_43_22723:34
*** annegentle has quit IRC23:34
lifelessI need to iterate again the project-config script there23:34
*** heyongli has quit IRC23:34
lifelessI wish more of it was in the requirement repo and unit tested23:34
*** austin81 has left #openstack-infra23:35
*** heyongli has joined #openstack-infra23:37
*** krtaylor has joined #openstack-infra23:41
*** Longgeek has joined #openstack-infra23:44
*** heyongli has quit IRC23:44
*** heyongli has joined #openstack-infra23:45
*** yamada-h has quit IRC23:46
mordredlifeless: ++23:48
*** Longgeek has quit IRC23:49
*** heyongli has quit IRC23:50
*** erlon has quit IRC23:50
*** jasondotstar has joined #openstack-infra23:51
*** gyee has quit IRC23:51
*** heyongli has joined #openstack-infra23:52
greghaynesNg: In hp2 theres some nodes in the machine-information doc that say youre reserving them, im guessing thats a thing of the past and no longer relavent?23:52
*** baoli has quit IRC23:53
*** heyongli has quit IRC23:56
*** camunoz__ has joined #openstack-infra23:57
*** camunoz__ is now known as camunoz_mtg23:57
*** laughterwym has joined #openstack-infra23:58
mordredgreghaynes: I hear he's making a nice lemon cream sauce with them23:58
*** dalgaaf has quit IRC23:58
greghaynesoh, I should definitely leave them be23:59
*** camunoz has quit IRC23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!