Friday, 2015-05-15

*** davideagnello has quit IRC00:01
*** dannywilson has quit IRC00:02
*** mahito has joined #openstack-infra00:02
*** salv-orlando has quit IRC00:03
*** fawadkhaliq has quit IRC00:04
*** kmartin has quit IRC00:04
*** Rockyg has quit IRC00:05
*** sputnik13 has quit IRC00:06
*** davideagnello has joined #openstack-infra00:07
*** shashankhegde has quit IRC00:07
*** mahito has quit IRC00:08
openstackgerritMerged openstack-infra/puppet-user: Replace ci.o.o links with docs.o.o/infra
*** mahito has joined #openstack-infra00:09
ociuhanduhi all, we’re experiencing some issues with our CI at the moment, devstack fails while stacking, trying to fetch
*** Longgeek has quit IRC00:13
ociuhandudoes anyone have similar problems?00:13
*** Longgeek has joined #openstack-infra00:13
ociuhanduwe’ve tried the url from several locations and any time we get the same “503 Service Unavailable” reply00:13
fungiociuhandu: yeah, i started seeing the same thing a few minutes ago00:14
*** ayoung has joined #openstack-infra00:14
asselinociuhandu, I get the same error00:14
*** ildikov has joined #openstack-infra00:14
fungidstufft: ^ fun with pypa.org00:14
fungier, .io00:14
*** mestery has quit IRC00:14
fungimy money's on it's cloud-hosted and in the process of being restarted as fallout from venom patching00:15
*** dims has quit IRC00:15
dstufftserver got rebooted00:15
dstufftdidn't come back up00:15
fungithose are my favorite00:15
ociuhandufungi asselin: whew, thanks for the confirmation, we just managed to recover after another issue we had, been online for like an hour and now this :) Was hoping it’s not related to our CI :)00:16
dstufftfixed now00:17
fungidstufft: thanks!00:17
fungiwe're your network-based monitoring and alerting system00:17
dstufftfungi: +100:19
*** sarob has quit IRC00:19
ociuhandufungi: I tried the url from several locations with same results but after the latest events today I’m becoming a bit paranoid :)00:19
ociuhandufungi: maybe also being like 3:20 AM here helps a bit ;)00:20
*** mriedem has joined #openstack-infra00:20
dstufftfungi: had any problems with your bandersnatch mirrors today?00:21
vipulmordred: ping00:21
fungidstufft: no idea... not that any of _our_ meat-based monitoring and alerting systems have reported anyway ;)00:21
openstackgerritMerged openstack-infra/puppet-zuul: Replace ci.o.o links with docs.o.o/infra
vipulmordred: which shade repo is the one that adds openstack modules for ansible. or
dstufftfungi: okay, cool00:22
vipulthe two are quite different00:22
dstufftfungi: PyPI is a bit unhappy, wasn't sure how much it was actually affecting people though00:22
fungidstufft: not enough to disrupt our testing, at any rate00:22
fungiso... great job!00:22
*** dims has joined #openstack-infra00:23
*** banix has quit IRC00:24
openstackgerritMatt Riedemann proposed openstack/requirements: Raise the cap on oslo.config to <1.7.0
*** marun has joined #openstack-infra00:27
*** armax has quit IRC00:28
asselinI'm running into an issue with my nodepool nodes no longer being marked 'used'....00:28
asselinmy configurations are looking ok...been working find until just recently00:29
fungiasselin: that usually happens when nodepool loses contact with the jenkins zeromq-publisher plugin00:29
asselinyeah...was looking at that...they're on the same host.00:30
fungioh, right, you're running nodepool on your jenkins master00:30
*** banix has joined #openstack-infra00:31
*** markvoelker has quit IRC00:32
*** ociuhandu has quit IRC00:33
openstackgerritMerged openstack-infra/pypi-mirror: Replace ci.o.o links with docs.o.o/infra
*** yamamoto has joined #openstack-infra00:37
openstackgerritMerged openstack-infra/project-config: kolla: add sudo for
openstackgerritMerged openstack-infra/system-config: Use jenkins default file from puppet-jenkins
*** D1337d has quit IRC00:41
openstackgerritMerged openstack-infra/puppet-elasticsearch: Cleanup old ES logs
openstackgerritMerged openstack-infra/zuul: Replace ci.o.o links with docs.o.o/infra
*** otter768 has joined #openstack-infra00:42
*** otter768 has quit IRC00:47
*** mtanino has quit IRC00:49
mordredvipul: hey vipul!00:51
mordredvipul: I should delete shade-ansible - it's evil00:51
mordredvipul: is the shade repo which has the library00:52
vipulmordred: and then i found this that uses the shade library :|00:52
vipulso which one to use in a playbook..00:53
mordredvipul: you use that00:53
mordredvipul: there are more pull requests coming to add more and more things00:54
mordredvipul: like, we shoudl add one for cue :)00:54
*** Krinkle|detached is now known as Krinkle00:54
vipulmordred: got it.. it seemed partial00:54
mordredvipul: (also, sorry for being terse earlier - I was on my phone - harder to type)00:54
vipulmordred: no worries ;) -- ok it's starting to make sense now00:55
*** changbl has joined #openstack-infra00:55
vipulwell that would be make it a lot more complete00:55
mordredvipul: basically it hangs together like this - shade is a library which knows how to deal with all the differences in logic and stuff for where openstack is weird00:55
*** markvoelker has joined #openstack-infra00:56
mordredvipul: we're moving to using shade in nodepool, since a lot of the logic for failure handling and whatnot actually comes from nodepool in the first place (10k servers a day, you learn some things :) )00:56
mordredvipul: then we're also using shade as the basis for all the new ansible modules00:56
mordredvipul: the old ones did not support keystone v3 or auth plugins and a whole host of other things were mostly wrong00:56
*** tqtran has quit IRC00:57
vipulmordred: Ok that makes sense.. why not use existing python clients?00:57
mordredvipul: so - you should use those for sure - and if you want to add stuff, step 1 is often to add the capability to shade in the first place, then make the ansible module a thin wrapper around it which understands ansible things00:57
mordredvipul: they're too hard00:57
*** patrickeast_ has quit IRC00:57
mordredvipul: the size of the ansible modules would be GIANT00:58
mordredvipul: and most of that would have to be duplicated inside of nodepool too00:58
mordredand other things00:58
mordredvipul: because te clientlibs are libraries that expose the REST api - which is great - until you have to deal with config differences and the things deployers can choose to do00:58
vipuli see, the clients are too granular, and don't actually represent scenarios you woul want to do00:58
*** harlowja has quit IRC00:58
mordredshade is more like a business-logic resource library00:59
*** yamahata has joined #openstack-infra00:59
mordredvipul: if you wat to cry, start reading through with this method:
openstackgerritJames Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras and env markers
mordredand notice that it keeps going through to
mordredvipul: all just to be able to upload an image01:00
vipulYep.. it's painful.. starting to see why this needed to be written01:01
vipulyou're doing a bunch of orchestration in client library.. without actually mainitng any state01:01
*** tiswanso has joined #openstack-infra01:02
vipulis there concern of things failing .. causing leaks, etc01:02
vipulunless Task is something that actually maintains some sort of state01:02
*** zhiwei has quit IRC01:02
mordredit is not01:02
mordredit's mainly for enabling rate limiting01:02
mordred(which we have to do in nodepool)01:02
mordredthere are some transactions where we handle compound things well - and some where we don't01:03
*** sdake has quit IRC01:03
mordredultimately some of those need to get better - although we're also pushing back on the projects to provide us ways to accomplish things where the client doesn't leak or race01:03
mordredvipul: ask anyone in nova or neutron about clarkb and I yelling about how floating ips work :)01:04
*** tiswanso has quit IRC01:04
vipul:) well projects actually building out apis that are use-case driven might be a while ;)01:04
*** tiswanso has joined #openstack-infra01:05
*** marun has quit IRC01:05
mordredI'd actually love to see shade eventually drive some thinking about some use-case driven api things01:05
*** sdake has joined #openstack-infra01:05
mordredI know we've got folks thinking about it in both nova and neutron01:05
mordredneutron is looking for a "make me a darned network" api that would do all of the various resources needed to make a network01:06
*** mahito has quit IRC01:06
mordredand similar with nova and "please get me a server _With_ an IP"01:06
*** markvoelker has quit IRC01:06
vipuli don't have ot learn about routers, subnets, all that crap?01:06
mordredvipul: right?01:06
mordredvipul: that's what I said :)01:06
*** asrangne__ has quit IRC01:07
vipulwell good to see there is progress01:07
mordredvipul: there is a simple use case, which is "get me the basic standard network with no crazy options"01:07
*** achanda has quit IRC01:07
vipulwhich 90% of users do anyway :P01:07
mordredso, it's aWESOME that we support the complex case01:07
mordredbut we need to help the simple use cases too01:07
vipulwell thanks for the background.. it's makes more sense now01:08
mordredvipul: let me know if you have any issues with either shade or the ansible modules01:08
*** ddieterly has joined #openstack-infra01:08
vipulmordred: yup will do!01:08
mordredvipul: I'm servicing my use cases currently, but always happy to make sure I'm not being tunnelvisioned01:08
vipulWe're going to use these to deploy Cue.. so maybe this will either validate those or we'll bring up some new ones01:09
*** ddieterl_ has joined #openstack-infra01:09
*** heyongli has joined #openstack-infra01:12
clarkbya devstack shouldnt have to set up tenant networking for me, neutron should just work :/01:12
clarkbalso devstack totally does that for you btw01:12
clarkbwhich probably hides the problem for many devs01:13
*** ddieterly has quit IRC01:13
*** zz_dimtruck is now known as dimtruck01:13
vipuldevstack sets up all sorts of routes as well .. which you only realize are missing if your devstack host restart :|01:14
*** gyee has quit IRC01:14
vipuland you can no longer get into VMs booted in that devstac01:14
*** puranamr has quit IRC01:15
clarkbI have learned so much about running openstack recently01:15
vipuloh a tangent.. why do we not use virtualenvs in devstack?01:15
clarkbvipul its slower and all of our services should be coinstallable01:16
vipultry keeping a devstack vm longer than a few days.. pull new code.. and try to restack01:16
clarkbnope dont do that01:16
vipul90% of the time.. it aint going to work01:16
clarkbalso venvs wouldnt help that01:16
fungiyou get it to work 10% of the time? lucky!01:16
vipulwell a lot of times its just pip dependencies that get in the way01:17
vipuli'd rather blow away virtualenv, have devstack create new ones01:17
clarkbya but pip is broken regardless01:17
clarkbalso you can do that01:17
clarkbdevstack does support venvs its just slooow01:17
*** zhiwei has joined #openstack-infra01:18
vipulhmm i didn't realize there was actual support.. i'll try it to prove my theory01:18
*** wznoinsk has quit IRC01:20
*** wznoinsk has joined #openstack-infra01:20
*** ddieterl_ has quit IRC01:21
*** ddieterly has joined #openstack-infra01:21
*** heyongli has quit IRC01:22
openstackgerritMerged openstack-infra/puppet-jenkins: Adds jenkins default file
openstackgerritMerged openstack-infra/puppet-logstash: Logrotate the indexer.log logfile.
*** claudiub has quit IRC01:26
*** ashleighfarnham has joined #openstack-infra01:30
mriedemfungi: clarkb: is this still a thing today?
mriedemlogstash shows it spike and drop off01:32
mriedemoh damn nevermind01:33
mriedemi was looking at the wrong logs01:33
fungimriedem: there was something going on with unusual load/connectivity/something for our git server farm from about 1400-1600 utc today01:33
mriedemfixed itself?01:33
fungi<spoor>machines don't fix themselves</spoor>01:34
fungibut yeah, whatever was going on stopped going on. it will almost certainly start going on again01:35
fungiwe were just at a bit of a loss to identify a root cause. maybe next time01:36
mriedemso have you seen....this?
*** Longgeek has quit IRC01:36
fungiyep! dstufft just fixed that a little bit ago01:37
dansmithwhat was it?01:37 rebooted and didn't come back up without coaxing, apparently01:37
*** Longgeek has joined #openstack-infra01:37
dansmithso should we be rechecking now or wait?01:37
jogowe need to start a log for all the things that break us and put it on the web, so it would be a blog01:37
mriedemwas there a bug reported?01:37
fungianything that failed on that should be fine now01:38
*** banix has quit IRC01:38
fungijogo: until it went down01:38
fungijogo: then we'd have nowhere to record why it was offline ;)01:38
dansmithjust return "YES" from that page01:38
dansmithshould be easy to keep that up01:38
fungigood point01:39
dansmithif you don't get yes, then it's broken01:39
dansmithso it's always right01:39
fungieven when it's not broken, it's broken01:39
dansmithwell, even when that service is up, something is definitely broken for sure01:40
dansmithlike "it's always beer o clock somewhere"01:40
openstackgerritMerged openstack-infra/system-config: Replace ci.o.o links with docs.o.o/infra
fungii suspect a very strong correlation between those two statements, in fact01:41
lifelessfungi: so, hp cloud is broken was the result?01:42
fungilifeless: not sure. i think "hpcloud is broken" may simply be preventing me from figuring out right now why we didn't build snapshot images there for the past few days01:43
fungii'm about to try it yet again01:43
fungifrom what people in the know were saying, hpcloud was in the midst of its "reboot all the things" for the venom vulnerability01:44
lifelessfungi: and there's another glitch happening right now01:46
lifelessfungi: (yay) but - worth trying01:46
fungiwell, i've taken the opportunity to clean up leaked instances, images, floating ip addresses, ports, keypairs, et cetera01:47
jogolifeless: in *theory* OpenStack should be able to handle rolling reboots of all the things like this right?01:48
lifelessjogo: the operational issues I'm aware of are not due to openstack01:48
lifelessjogo: and are things that openstack isn't currently designed or architected to handle01:48
*** Sukhdev has joined #openstack-infra01:49
lifelessjogo: (should it be?perhaps. There's a cost to such things.)01:50
*** sdake_ has joined #openstack-infra01:51
jogolifeless: well baby steps, finalizing the multinode testing is still first01:51
fungifor example, live migration would have been an option for dealing with the qemu restarts for the venom fix. pause/resume with disk-backed memory too. rackspace at least couldn't do those. needed to hard reboot your instances instead01:51
fungibut the thread on the operators ml indicated that at least some environments which had designed around those features were doing it that way with success01:52
lifeless<3 live migration01:54
jogoneat, live migration is working pretty well in the multi-node test too01:55
*** sdake has quit IRC01:55
*** hdd has quit IRC01:58
openstackgerritJames Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras and env markers
*** Kennan2 has joined #openstack-infra02:02
*** mahito_ has joined #openstack-infra02:02
*** Kennan has quit IRC02:03
*** Kennan2 is now known as Kennan02:04
openstackgerritJeremy Stanley proposed openstack-infra/shade: Replace ci.o.o links with docs.o.o/infra
fungimordred: your wish ^ has been granted02:08
*** btully has quit IRC02:09
*** mtanino has joined #openstack-infra02:09
*** signed8bit has quit IRC02:10
greghaynesfungi: commented, commit msg is 100% unrelated now, not sure if we care02:10
*** Sukhdev has quit IRC02:10
mordredfungi: :)02:11
pleia2I dunno, it's still mostly related, just slightly adjusted to point to docs too ;)02:11
pleia2s/docs/new shade docs02:11
greghaynesyea, I dont care either way, just thought id point it out02:11
*** shashankhegde has joined #openstack-infra02:12
*** sigmavirus24_awa is now known as sigmavirus2402:12
*** sarob has joined #openstack-infra02:13
*** sigmavirus24 is now known as sigmavirus24_awa02:15
openstackgerritMerged openstack-infra/devstack-gate: Replace ci.o.o links with docs.o.o/infra
openstackgerritMerged openstack-dev/pbr: Bump integration test timeouts.
fungiindeed. the old url went to ci.o.o, which is deprecated, so the commit message is right on point. _where_ on docs.o.o/infra it's been changed to is a little less important ;)02:18
lifelessfungi: ok so we're trying again... any result?02:23
fungilifeless: i'm waiting for my delete loop over alien images to complete02:24
fungishould be almost done02:24
lifelesssee thats sad02:24
lifelessthat you have to wiat02:24
fungireminds me of watching text come over a 300bps connection02:25
*** ZZelle has joined #openstack-infra02:28
fungiimage update is underway now02:28
lifelessdefragging hard disks?02:30
*** unicell has quit IRC02:33
fungiflying toasters02:36
mordredlifeless: I agree, btw, that waiting for deletes is bong02:37
mordredlifeless: it seems, btw, to be largely related to the table lock on the nova quota code02:38
mordredlifeless: because the API calls to request a delete (not perform the delete, mind you) needs to touch the quota checking code02:38
mordredlifeless: let's ignore fora  sec that the quota code should not do a table lock ... enqueinng a delete most certainly should not trigger the lock02:39
*** rkukura has quit IRC02:39
mordredlifeless: in case you can't tell, I'm not thrilled about this :)02:39
*** baoli has joined #openstack-infra02:40
*** sigmavirus24_awa is now known as sigmavirus2402:40
*** otter768 has joined #openstack-infra02:43
openstackgerritMonty Taylor proposed openstack-infra/puppet-nodepool: Consume puppet-diskimage_builder
lifelessmordred: I think you're ecstatic02:47
*** Krinkle is now known as Krinkle|detached02:47
mordredlifeless: well, I'm certainly flying high02:47
*** otter768 has quit IRC02:47
lifelessmordred: flying | high02:48
lifeless(bitwise OR)02:48
mordredlifeless: :)02:48
*** julim has quit IRC02:49
*** yamahata has quit IRC02:50
*** sigmavirus24 is now known as sigmavirus24_awa02:52
mordredlifeless: also - in case you didn't know - "username" and "password" are _different_02:55
lifelessmordred: so not userword and passname ?02:55
mordredlifeless: which makes lines like passwd = config['username'] not as effective as you might think02:56
mordredI mean, I'm sure there are some times when that will do the right thing02:56
lifelessmordred: 'everything is awful'02:56
lifelessmordred: (Yes, I'm pushing that ear worm)02:56
openstackgerritMerged openstack-infra/shade: Add floating IP pool resource methods
*** ajmiller has joined #openstack-infra02:59
*** BharatK has joined #openstack-infra03:00
*** mtanino has quit IRC03:00
*** comstud has quit IRC03:02
*** rkukura has joined #openstack-infra03:05
*** ddieterly has quit IRC03:05
*** krtaylor has joined #openstack-infra03:05
*** mahito_ has quit IRC03:05
openstackgerritJoe Gordon proposed openstack-infra/infra-manual: Add section on how to run python unit tests.
*** hughsaunders has quit IRC03:11
*** dims has quit IRC03:12
*** hughsaunders has joined #openstack-infra03:12
*** panda has quit IRC03:13
*** panda has joined #openstack-infra03:14
*** hdd has joined #openstack-infra03:18
fungilifeless: so... my image update completed successfully03:21
lifelessfungi: \o/03:21
lifelessfungi: now we find out?03:21
*** achanda has joined #openstack-infra03:21
fungiwell, that was just one of many. i'm going ahead and updating the bare-trusty images in the remaining 4 networks. the scheduled builds don't try again until 14:14 utc03:23
openstackgerritMonty Taylor proposed openstack/requirements: Bump oslotest to 1.6.1 for functional tests
*** achanda has quit IRC03:27
*** baoli has quit IRC03:29
*** baoli has joined #openstack-infra03:29
*** sdake has joined #openstack-infra03:33
*** hdd has quit IRC03:33
openstackgerritMerged openstack/requirements: Bump python-openstacksdk version to 0.4.1
*** alexpilotti has quit IRC03:35
*** sdake_ has quit IRC03:36
*** baoli has quit IRC03:37
*** sdake_ has joined #openstack-infra03:38
mordredfungi: ooh! yay03:39
*** tiswanso has quit IRC03:40
*** sdake has quit IRC03:41
*** shashankhegde has quit IRC03:41
*** armax has joined #openstack-infra03:42
*** yamahata has joined #openstack-infra03:43
*** armax has quit IRC03:48
*** HenryG has quit IRC03:51
*** Sukhdev has joined #openstack-infra03:55
*** sarob has quit IRC03:56
fungilifeless: okay, all bare-trusty images have been updated successfully, and any nodes booted from previous images are already being used, so any new jobs should get nodes booted from the updated images03:59
fungiand with that, i'm out for the night04:00
lifelessfungi: wicked, thanks04:01
openstackgerritMerged openstack-dev/pbr: Finish removing invocations of pip.
*** stevemar has joined #openstack-infra04:01
*** tnovacik has joined #openstack-infra04:04
*** ddieterly has joined #openstack-infra04:06
lifeless\o/ time for a pbr release, 1600 friday, WGPGW04:08
lifeless[in case anyone missed it, that was a joke]04:08
mordredfungi: neat!04:09
openstackgerritIan Wienand proposed openstack-infra/project-config: [WIP] Use neutron for centos7 tests
*** ddieterly has quit IRC04:10
*** coolsvap|afk is now known as coolsvap04:12
openstackgerritIan Wienand proposed openstack-infra/devstack-gate: [WIP] run centos7 with neutron
*** BharatK has quit IRC04:15
*** achanda has joined #openstack-infra04:22
*** BharatK has joined #openstack-infra04:28
SukhdevDear infra folks, there seems to be connectivity issue with the etherpad - any information about this?04:30
clarkbSukhdev: its up for me, can you provide more info?04:31
clarkbSukhdev: is it a specific etherpad?04:32
openstackgerritMerged openstack-infra/shade: Replace ci.o.o links with docs.o.o/infra
clarkbSukhdev: it loads for me and says you are there04:32
clarkbin the yellow04:32
Sukhdevclarkb: Yes, when I try to type anything, it disconnects04:33
Sukhdevclarkb: please try to write something to it -04:33
*** dannywilson has joined #openstack-infra04:33
clarkbSukhdev: you probably have it open in two tabs, close one04:33
Sukhdevclarkb: while I was tying it got disconnected04:34
clarkbSukhdev: yes that is a known issue if you have multiple tabs open to that etherpad04:34
*** enikanorov2 has joined #openstack-infra04:34
Sukhdevclarkb: I do not have this one opened in multiple tabs, but, I have several ether pads open - can that do it?04:34
clarkbno it should be per etherpad, can you double check that o udon't have that one open elsewhere?04:35
Sukhdevclarkb: let me check04:35
*** sdake_ has quit IRC04:35
Sukhdevclarkb: I did not have this ether pad in multiple tabs - but, had several other ether pads open. I have closed them all04:39
Sukhdevclarkb: Now this one seems to be behaving OK04:40
*** otter768 has joined #openstack-infra04:44
*** stevemar has quit IRC04:46
*** otter768 has quit IRC04:49
*** dannywilson has quit IRC04:52
*** deepakcs has joined #openstack-infra04:58
*** btully has joined #openstack-infra04:58
*** yfried|afk is now known as yfried05:01
*** dimtruck is now known as zz_dimtruck05:03
*** sks has joined #openstack-infra05:04
*** erikmwilson has quit IRC05:04
*** ildikov has quit IRC05:05
*** bhunter71 has quit IRC05:05
*** bhunter71 has joined #openstack-infra05:06
*** ddieterly has joined #openstack-infra05:07
*** k4n0 has joined #openstack-infra05:07
*** markvoelker has joined #openstack-infra05:11
*** ddieterly has quit IRC05:11
*** armax has joined #openstack-infra05:13
openstackgerritMerged openstack-infra/openstackid: Replace ci.o.o links with docs.o.o/infra
*** armax has quit IRC05:15
*** ildikov has joined #openstack-infra05:18
*** ZZelle has quit IRC05:20
*** maishsk_afk has joined #openstack-infra05:21
*** gokrokve has joined #openstack-infra05:31
*** yfried is now known as yfried|afk05:32
*** tnovacik has quit IRC05:32
*** macjack has joined #openstack-infra05:35
*** bhunter71 has quit IRC05:36
*** HenryG has joined #openstack-infra05:38
*** armax has joined #openstack-infra05:42
*** mestery has joined #openstack-infra05:44
*** Swami_ has joined #openstack-infra05:45
*** afazekas has joined #openstack-infra05:47
*** markvoelker has quit IRC05:47
*** Swami has quit IRC05:48
*** sdake has joined #openstack-infra05:52
*** mestery has quit IRC05:53
*** mestery has joined #openstack-infra05:53
*** sdake has quit IRC05:53
*** sdake has joined #openstack-infra05:54
*** unicell has joined #openstack-infra05:54
*** Sukhdev has quit IRC05:55
*** gokrokve has quit IRC05:56
*** gokrokve_ has joined #openstack-infra05:56
*** jtomasek has joined #openstack-infra05:56
*** sdake has quit IRC05:58
*** ddutta has joined #openstack-infra05:59
*** maishsk_afk has quit IRC05:59
*** msuriar has quit IRC06:00
dduttaAjaegar: thx for the review .... I had forgotten to change the role06:01
*** gokrokve_ has quit IRC06:01
dduttanow its fixed06:01
lifelesshah, we're not installable again06:01
*** shashankhegde has joined #openstack-infra06:01
dduttaAjaegar: have I addressed everything else correctly now06:03
openstackgerritOpenStack Proposal Bot proposed openstack-infra/project-config: Normalize projects.yaml
*** ddieterly has joined #openstack-infra06:07
*** pal has joined #openstack-infra06:08
*** ddieterly has quit IRC06:12
*** HeOS has quit IRC06:12
*** mestery has quit IRC06:13
*** sdake has joined #openstack-infra06:16
openstackgerritDebo~ Dutta proposed openstack-infra/project-config: Add python-cognitiveclient project for Cognitive
dduttaAJaegar: fixed the other issue too ... thx a lot06:20
*** ajo has joined #openstack-infra06:21
lifelesscould I get someone to remove sphinxcontrib-docbookrestapi-0.1 from pypi ? its broken - setuptools won't install it because of the missing namespace declaration in __init__06:22
lifeless(how should this be filed?)06:22
lifelessmordred: ^06:22
*** achanda has quit IRC06:23
openstackgerritDebo~ Dutta proposed openstack-infra/project-config: Add python-cognitiveclient project for Cognitive
dduttaAJaegar: I will submit the cloudpulse fix separately06:26
*** yfried|afk is now known as yfried06:28
*** yfried_ has joined #openstack-infra06:35
*** k4n0 has quit IRC06:35
*** yfried has quit IRC06:35
*** k4n0 has joined #openstack-infra06:36
*** Swami_ has quit IRC06:38
openstackgerritMerged openstack-infra/project-config: Normalize projects.yaml
*** Kennan2 has joined #openstack-infra06:41
*** Kennan has quit IRC06:41
*** unicell1 has joined #openstack-infra06:43
*** armax has quit IRC06:43
*** otter768 has joined #openstack-infra06:45
*** unicell has quit IRC06:45
*** Sukhdev has joined #openstack-infra06:49
*** otter768 has quit IRC06:50
*** naggappan has quit IRC06:51
*** ashleighfarnham has quit IRC06:52
*** k4n0 has quit IRC06:54
*** k4n0 has joined #openstack-infra06:55
openstackgerritClint Adams proposed openstack-infra/system-config: Add #kolla to the list of statusbot channels
*** arxcruz has joined #openstack-infra07:01
*** jlanoux has joined #openstack-infra07:03
*** arxcruz has quit IRC07:03
*** arxcruz has joined #openstack-infra07:03
*** ildikov has quit IRC07:04
*** sdake_ has joined #openstack-infra07:05
*** rlandy has joined #openstack-infra07:06
openstackgerritJames Polley proposed openstack-dev/pbr: Teach pbr to read extras and env markers
*** ibiris_away is now known as ibiris07:07
*** shashankhegde has quit IRC07:07
*** ddieterly has joined #openstack-infra07:08
*** ZZelle_ has quit IRC07:08
*** sdake has quit IRC07:08
*** ZZelle_ has joined #openstack-infra07:09
*** aluria has joined #openstack-infra07:12
*** ddieterly has quit IRC07:12
*** panda has quit IRC07:13
*** panda has joined #openstack-infra07:14
*** ildikov has joined #openstack-infra07:17
openstackgerritJames Polley proposed openstack-dev/pbr: Teach pbr to read extras and env markers
*** ivar-laz_ has joined #openstack-infra07:18
*** rcarrillocruz has joined #openstack-infra07:19
*** Guest19895 has quit IRC07:21
*** freerunner has joined #openstack-infra07:21
*** ivar-lazzaro has quit IRC07:21
*** ivar-laz_ has quit IRC07:23
*** abregman has joined #openstack-infra07:24
*** luqas has joined #openstack-infra07:25
*** ihrachyshka has joined #openstack-infra07:26
*** gilliard is now known as gilllliard07:26
*** jcoufal has joined #openstack-infra07:27
*** freerunner has quit IRC07:29
*** freerunner has joined #openstack-infra07:30
*** chlong has quit IRC07:32
*** jistr has joined #openstack-infra07:33
*** nfedotov has joined #openstack-infra07:35
openstackgerritMerged openstack-infra/zuul: Simplify cloner required parameters
*** macjack has left #openstack-infra07:44
openstackgerritFabien Boucher proposed openstack-infra/puppet-zuul: Set manage_log_conf parameter to true by default
*** mrunge has joined #openstack-infra07:45
*** HeOS has joined #openstack-infra07:51
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Fix python-cloudpulseclient ACLs pasto
*** freerunner has quit IRC07:59
*** ociuhandu has joined #openstack-infra08:00
*** mpaolino has joined #openstack-infra08:02
openstackgerritFabien Boucher proposed openstack-infra/system-config: Do not set the manage_log_conf feature flag as default is true
*** Kennan2 is now known as Kennan08:04
*** ddieterly has joined #openstack-infra08:09
*** woodster_ has quit IRC08:10
*** btully has quit IRC08:11
*** derekh has joined #openstack-infra08:13
*** ddieterly has quit IRC08:13
*** Sukhdev has quit IRC08:15
*** yfried_ is now known as yfried|afk08:22
*** tnovacik has joined #openstack-infra08:25
*** dtantsur|afk is now known as dtantsur08:27
*** yamahata has quit IRC08:33
*** abregman is now known as abregman|afk08:36
*** arxcruz has quit IRC08:37
*** smccully has quit IRC08:38
*** yfried|afk is now known as yfried_08:38
*** arxcruz has joined #openstack-infra08:41
*** otter768 has joined #openstack-infra08:46
*** otter768 has quit IRC08:50
*** smccully has joined #openstack-infra08:52
*** afazekas has quit IRC08:52
*** Trozz has joined #openstack-infra08:53
*** rguillebert has joined #openstack-infra08:56
*** yfried_ is now known as yfried|afk08:58
*** niska has quit IRC08:58
*** ricky1 has joined #openstack-infra08:58
openstackgerritSteven Dake proposed openstack-infra/project-config: Set the Kolla build timeout to 3 hours
sdake_ajaegeri your still awake can you ahve a quick look at that and tell me if its correct syntax?09:03
sdake_ajaeger that is :)09:03
*** ildikov has quit IRC09:07
*** ddieterly has joined #openstack-infra09:10
*** niska has joined #openstack-infra09:12
*** katyafervent_ is now known as katyafervent_awa09:13
*** ddieterly has quit IRC09:14
*** afazekas has joined #openstack-infra09:15
*** ildikov has joined #openstack-infra09:16
*** ildikov is now known as ildikov_afk09:16
*** teran_ has quit IRC09:17
*** katyafervent_awa is now known as katyafervent09:20
*** rguillebert has quit IRC09:21
*** yfried|afk is now known as yfried_09:24
*** ildikov_afk has quit IRC09:25
*** deepakcs has quit IRC09:30
*** katyafervent is now known as katyafervent_awa09:33
*** katyafervent_awa is now known as katyafervent09:35
*** rguillebert has joined #openstack-infra09:36
*** katyafervent is now known as katyafervent_awa09:37
*** katyafervent_awa is now known as katyafervent09:37
*** ildikov_afk has joined #openstack-infra09:38
*** shardy has joined #openstack-infra09:39
*** ddutta has quit IRC09:39
*** dims has joined #openstack-infra09:45
*** rguillebert has quit IRC09:45
openstackgerritVictor Stinner proposed openstack/requirements: Drop suds dependency
*** alexpilotti has joined #openstack-infra09:46
*** yfried_ is now known as yfried|afk09:47
*** e0ne has joined #openstack-infra09:55
*** e0ne is now known as e0ne_09:56
openstackgerritRichard Pijnenburg proposed openstack-infra/jenkins-job-builder: Add logstash reporter
electricalzaro: left some comments in patch set 1 and made mods you asked in
*** e0ne_ is now known as e0ne10:01
*** rguillebert has joined #openstack-infra10:04
openstackgerritBob Ball proposed openstack-infra/devstack-gate: Add a hook for job-dependent log collection.
*** teran has joined #openstack-infra10:10
*** ddieterly has joined #openstack-infra10:10
*** pc_m has joined #openstack-infra10:13
*** pc_m_ has joined #openstack-infra10:13
*** pcm_ has joined #openstack-infra10:13
*** pc_m__ has joined #openstack-infra10:15
*** pc_m__ has quit IRC10:15
*** ddieterly has quit IRC10:15
*** teran has quit IRC10:15
*** pcm_ has quit IRC10:15
ekarlsowhy doesnt' anyone make into a elasticsearch thing instead so it's searchable ?10:16
*** yfried|afk is now known as yfried_10:17
*** pc_m has quit IRC10:17
openstackgerritMerged openstack-infra/tripleo-ci: Allow puppet jobs to also run on F21
*** pcm_ has joined #openstack-infra10:17
*** pc_m_ has quit IRC10:17
*** zhiwei has quit IRC10:18
*** teran has joined #openstack-infra10:23
*** abregman|afk is now known as abregman10:23
*** msuriar has joined #openstack-infra10:28
*** yfried_ is now known as yfried|afk10:28
*** e0ne is now known as e0ne_10:31
*** ajo has quit IRC10:33
*** e0ne_ is now known as e0ne10:33
*** yfried|afk is now known as yfried_10:34
*** ociuhandu has quit IRC10:39
openstackgerritGiulio Fidente proposed openstack-infra/tripleo-ci: Use environments file for Ceph and HA jobs
*** yamamoto has quit IRC10:46
*** otter768 has joined #openstack-infra10:47
sdagueekarlso: no one has gotten to it, I'm sure patches would be welcomed10:51
*** otter768 has quit IRC10:51
openstackgerritGiulio Fidente proposed openstack-infra/project-config: Add TripleO/Puppet/Ceph job to check-tripleo-puppet-jobs
*** weshay has joined #openstack-infra10:55
openstackgerritRichard Pijnenburg proposed stackforge/python-jenkins: Add decode to utf8 for python 3.x
electricalzaro: solved the python 3.x support for get_version.10:58
*** pal has quit IRC10:59
*** mpaolino has quit IRC10:59
*** ricky2 has joined #openstack-infra11:05
*** ricky1 has quit IRC11:07
*** ociuhandu has joined #openstack-infra11:11
*** ddieterly has joined #openstack-infra11:11
*** Longgeek has quit IRC11:12
openstackgerritRichard Pijnenburg proposed openstack-infra/jenkins-job-builder: Add logstash publisher
*** panda has quit IRC11:14
electricalzaro: made a complete mistake with this one. moved it from reporter to publisher where it belongs.11:14
*** panda has joined #openstack-infra11:14
*** ddieterly has quit IRC11:15
*** ociuhandu_ has joined #openstack-infra11:19
*** ociuhandu has quit IRC11:19
*** ociuhandu_ is now known as ociuhandu11:19
*** dguerri is now known as _dguerri11:19
openstackgerritDarragh Bailey proposed stackforge/python-jenkins: Modify tests to exercise Request on python3
*** _dguerri is now known as dguerri11:21
*** yfried_ is now known as yfried|afk11:25
openstackgerritDarragh Bailey proposed stackforge/python-jenkins: Add decode to utf8 for python 3.x
openstackgerritDarragh Bailey proposed stackforge/python-jenkins: Modify tests to exercise Request on python3
*** e0ne is now known as e0ne_11:26
pelixelectrical: just rebased your change to be able to add remaining python 3 fixes required ;)11:26
electricalpelix: ah okay, was a bit confused for a moment :-)11:27
*** bcornec1 has quit IRC11:27
electricalmust have forgot to do a rebase11:28
electricalpelix: <-- logstash publisher. i made a stupid assumption that was so wrong. i initially thought it was a reporter but its a publisher so moved it. only fails now on something i haven't touched.11:29
*** yamamoto has joined #openstack-infra11:32
*** mika has quit IRC11:33
*** BharatK has quit IRC11:34
openstackgerritJames Polley proposed openstack-dev/pbr: Teach pbr to read extras and env markers
pelixhmm, something has changed around the python 3.4 environment11:40
*** alexpilotti has quit IRC11:42
*** Longgeek has joined #openstack-infra11:45
*** yfried|afk is now known as yfried_11:45
*** baoli has joined #openstack-infra11:45
*** e0ne_ is now known as e0ne11:48
pelixelectrical: looks like the python 3 slaves no longer has a default encoding of UTF-8, so need to ensure the files are opened with the correct encoding.11:48
pelixhave a fix coming up11:48
openstackgerritGiulio Fidente proposed openstack-infra/project-config: Add TripleO/Puppet/Ceph job to check-tripleo-puppet-jobs
*** marcusvrn1 has joined #openstack-infra11:50
linuxmaniacpelix: thanks!11:50
*** baoli_ has joined #openstack-infra11:51
*** marcusvrn has quit IRC11:52
*** ldnunes has joined #openstack-infra11:52
pelixlinuxmaniac: did it actually get merged? cause I believe we just hit a bug around encoding on systems not defaulting to UTF-8?11:53
*** baoli has quit IRC11:54
*** ddieterly has joined #openstack-infra11:54
linuxmaniacpelix: No it wasn't :-(11:54
electricalpelix: ah okay. cool. thanks!11:56
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Replace open() with
pelixlinuxmaniac electrical: I believe that should allow changes to start landing again once it's approve ^^11:57
*** marcusvrn1 has quit IRC11:57
*** marcusvrn has joined #openstack-infra11:59
linuxmaniacI hope it will not take three months to be approved12:01
*** marcusvrn1 has joined #openstack-infra12:02
*** dprince has joined #openstack-infra12:02
*** marcusvrn has quit IRC12:03
*** marcusvrn has joined #openstack-infra12:08
*** marcusvrn1 has quit IRC12:08
*** freerunner has joined #openstack-infra12:08
*** gilllliard is now known as gilliard_en_rout12:09
*** ajo has joined #openstack-infra12:09
*** BharatK has joined #openstack-infra12:09
*** banix has joined #openstack-infra12:12
*** banix has quit IRC12:12
*** links has joined #openstack-infra12:13
*** ajo has quit IRC12:13
*** marcusvrn has quit IRC12:16
*** yfried_ is now known as yfried|afk12:18
*** marcusvrn has joined #openstack-infra12:18
*** e0ne is now known as e0ne_12:22
*** yfried|afk is now known as yfried_12:23
*** coolsvap is now known as coolsvap|afk12:23
*** afazekas has quit IRC12:23
*** rfolco has joined #openstack-infra12:24
*** gordc has joined #openstack-infra12:26
*** e0ne_ is now known as e0ne12:26
*** kgiusti has joined #openstack-infra12:27
*** tnovacik has quit IRC12:27
*** freerunner has quit IRC12:30
*** marcusvrn has quit IRC12:32
*** mwagner_lap has joined #openstack-infra12:32
electricallinuxmaniac: hope not :-)12:35
*** marcusvrn has joined #openstack-infra12:36
*** jtomasek has quit IRC12:38
*** Daviey has quit IRC12:39
*** Daviey has joined #openstack-infra12:39
*** ddieterly has quit IRC12:39
*** bhunter71 has joined #openstack-infra12:39
*** aysyanne has joined #openstack-infra12:42
*** dboik has joined #openstack-infra12:42
*** mpaolino has joined #openstack-infra12:43
*** yolanda has quit IRC12:43
*** yolanda has joined #openstack-infra12:44
*** doug-fish has joined #openstack-infra12:44
openstackgerritDavide Guerri proposed openstack-infra/shade: Add Neutron/Nova Floating IP list/search/get
*** dboik has quit IRC12:47
*** otter768 has joined #openstack-infra12:48
electricalhope i'll be able to get some new plugins in quickly for jjb so i can start using it at work :-)12:48
*** annegentle has joined #openstack-infra12:48
*** derekh_ has joined #openstack-infra12:49
*** claudiub has joined #openstack-infra12:51
*** bswartz has quit IRC12:52
*** otter768 has quit IRC12:52
*** derekh has quit IRC12:52
*** e0ne is now known as e0ne_12:54
*** barnaby has quit IRC12:55
*** wenlock has joined #openstack-infra12:56
*** e0ne_ is now known as e0ne12:56
*** barnaby has joined #openstack-infra12:56
*** _nadya_ has joined #openstack-infra12:56
*** links has quit IRC12:57
*** pblaho has quit IRC12:58
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Catch client exceptions during list ops
*** sdake has joined #openstack-infra12:59
openstackgerritSamuel de Medeiros Queiroz proposed openstack-infra/project-config: Experimental Devstack job for Identity v3 API only
*** sdake_ has quit IRC13:02
*** mattfarina has joined #openstack-infra13:03
*** sdake_ has joined #openstack-infra13:03
*** pblaho has joined #openstack-infra13:04
*** erikmwilson has joined #openstack-infra13:04
*** erikmwilson has quit IRC13:04
*** yfried_ is now known as yfried|afk13:05
*** sdake_ has quit IRC13:07
*** sdake has quit IRC13:07
*** yfried|afk is now known as yfried_13:08
*** freerunner has joined #openstack-infra13:09
linuxmaniacelectrical: we use it at work and in kamailio/sems projects :-)13:09
electricallinuxmaniac: nice :-) I'm hoping to use at Elastic for our job management ( assuming we both are talking about jjb :p )13:10
*** sdake has joined #openstack-infra13:11
linuxmaniacelectrical: indeed13:11
electricalI'm tired of managing all the jobs manually13:12
*** freerunner has quit IRC13:12
electricalespecially when small things change but change for a lot of jobs13:12
linuxmaniacI would like to see how people are using jjb with tons of almost same jobs13:12
*** sks has quit IRC13:12
openstackgerritMerged openstack-dev/pbr: Teach pbr to read extras and env markers
linuxmaniacelectrical: We had improved a lot since we implement jjb configs13:13
electricalI got about 25 jobs my self to manage which is only a small amount of what we have in total.13:13
electricalfor our Logstash project we got about 170 same type of jobs, only the name and repo are different :-)13:14
sdakemorning folks13:14
electricalmorning sdake13:14
*** ricky2 has left #openstack-infra13:14
sdakethe kolla gate recently introduced a more thorough ci job, nad we broke sthings badly13:14
sdakenow our gate is cmpletely blocked13:14
*** dkranz_ is now known as dkranz13:14
linuxmaniacelectrical: INFO:jenkins_jobs.builder:Number of jobs generated:  58913:15
electricallinuxmaniac: very nice!13:15
sdakecan fokls with appropriate permissions take a look at
sdakewhich shouldget us operational13:15
electricali think we got close to 1000 jobs in total now across 3 jenkins masters13:15
sdakenote to self - new ci jobs shouldbe nonvoting ;)13:15
linuxmaniacelectrical: and this is one main file. We have 313:15
electricalhehe okay :p13:15
linuxmaniacwith less jobs13:15
*** dboik has joined #openstack-infra13:16
*** erikmwilson has joined #openstack-infra13:16
linuxmaniacwe didn't migrate all the jobs yet13:16
*** dboik_ has joined #openstack-infra13:17
electricalhehe okay. I'm hoping to setup a POC soon with my own jobs and show how useful it is for us13:17
linuxmaniacI'm not quite satisfied with out actual config schema13:18
electricalHow do you mean?13:18
linuxmaniacthis config is for controlling jobs for making debian packages13:19
linuxmaniacjobs are almost the same13:19
linuxmaniacbut if I add a unit-test job for a project, I have to add a new file with the config13:20
linuxmaniacIs quite difficult for me to explain13:20
*** gordc has quit IRC13:21
*** dboik has quit IRC13:21
linuxmaniacbut essentially I think we are doing something really wrong. Because this is just growing13:21
electricalI have sort of same kind of problem. I have like 10 jobs for the same puppet module. would love to have a single file that simply inherits all the jobs required for it.  nicest thing would be to have a single file, define some puppet module names and let it handle a loop or something with all the different job configs13:22
*** erlon has joined #openstack-infra13:22
*** pblaho has quit IRC13:22
*** bknudson has joined #openstack-infra13:23
*** freerunner has joined #openstack-infra13:23
*** sdake has quit IRC13:23
*** peristeri has joined #openstack-infra13:24
*** mase_x200 has joined #openstack-infra13:26
openstackgerritDerek Higgins proposed openstack-infra/project-config: Switch all tripleo Fedora jobs to F21
*** ddieterly has joined #openstack-infra13:29
*** bswartz has joined #openstack-infra13:29
*** samueldmq has quit IRC13:33
*** dustins has joined #openstack-infra13:33
*** claudiub has quit IRC13:34
openstackgerritGiulio Fidente proposed openstack-infra/project-config: Add TripleO/Puppet/Ceph job to check-tripleo-puppet-jobs
*** sdake has joined #openstack-infra13:34
*** k4n0 has quit IRC13:36
*** amitgandhinz has joined #openstack-infra13:40
*** woodster_ has joined #openstack-infra13:41
*** sdake has quit IRC13:42
*** marcusvrn has quit IRC13:43
*** marcusvrn has joined #openstack-infra13:44
openstackgerritJeff Peeler proposed openstack-infra/project-config: make kolla check non-voting
*** wenlock has quit IRC13:45
*** banix has joined #openstack-infra13:47
*** tiswanso has joined #openstack-infra13:48
BobBallSo I'm using the skip-if definition for a job (see so it doesn't run on docs jobs etc; but zuul still votes +1 on changes where we don't run any tests (e.g. - it seems that pipelines don't have a way to skip based on whether the jobs are going to run or not; Any suggestions for how to stop voting if no jobs13:48
sdagueBobBall: I don't know if that's implemented yet13:49
*** mika has joined #openstack-infra13:50
jpeelerAJaeger: as you see above, updated i'll never make the mistake of adding a non-voting job again. that review should allow me to not bother any of you and i'm sorry about all this.13:51
*** tnovacik has joined #openstack-infra13:51
jpeeler*voting grr, see my problem? can't even keep it straight in my head13:51
*** yfried_ is now known as yfried|afk13:53
openstackgerritRicardo Carrillo Cruz proposed openstack-infra/shade: Add initial designate read-only operations
*** jlanoux has quit IRC13:54
*** dansmith is now known as superdan13:56
*** pabelanger has joined #openstack-infra13:56
*** erikmwilson has quit IRC13:58
*** annegentle has quit IRC13:59
*** gordc has joined #openstack-infra14:04
*** banix has quit IRC14:05
*** erikmwilson has joined #openstack-infra14:08
*** mpaolino has quit IRC14:08
*** sdake has joined #openstack-infra14:08
*** salv-orlando has joined #openstack-infra14:08
openstackgerritRicardo Carrillo Cruz proposed openstack-infra/shade: Add initial designate read-only operations
*** stevemar has joined #openstack-infra14:11
*** signed8bit has joined #openstack-infra14:11
fungisdake: i'm a little worried about jobs that are expected to tie up a worker for 3 hours... is there a faster/simpler version you can gate on and then run the extended job as a periodic check of a branch tip instead?14:11
*** abregman has quit IRC14:11
sdakefungi my personal machine takes 45 minutes to run the job14:12
sdakeits the bare minimum we can do14:12
sdakebut I expect vm to be slower14:12
sdakeI really don't know how lon it will take to run14:12
sdakehow about this, we try 3 hours, if its greater the 1.5, we do period?14:12
sdakeand I'll submit a change as soon asI have data?14:12
sdakeit may only be 1 hour14:13
sdakeor it could be 1.514:13
*** yfried|afk is now known as yfried_14:13
sdakeit ran for 30 minutes before timing out but our tools don't print the progress atm14:13
sdake(needs more work)14:13
sdakeso I couldn't tell how far along it was14:13
fungisdake: you should probably also consider implementing your own timeout within the guts of the job similar to how devstack-gate does, so that you get logs even if it runs up to your intended timeout threshhold14:14
electricalfungi: morning14:14
sdakefungi cool I will look into that14:14
sdakewe get logs now tho I think, atleatt they were going to jenkins14:14
sdakeit just that our test case is python and doesn't print to stdout which it needs to do14:15
*** banix has joined #openstack-infra14:15
sdakelets try 90 minutes14:15
jpeelersdake: there's this too -
*** dims is now known as dimsum__14:16
*** sdake has quit IRC14:16
*** amotoki has joined #openstack-infra14:16
openstackgerritSteven Dake proposed openstack-infra/project-config: Set the Kolla build timeout to 3 hours
fungigood afternoon electrical14:17
openstackgerritSteven Dake proposed openstack-infra/project-config: Set the Kolla build timeout to 90 minutes
*** btully has joined #openstack-infra14:18
*** eharney has joined #openstack-infra14:18
*** tnovacik has quit IRC14:19
*** jcoufal has quit IRC14:21
*** freerunner has quit IRC14:21
electricalfungi: im trying to build my first jjb job now. see if i can make it work :p14:21
electricaland then hopefully abstract everything away in templates and groups, etc.14:22
*** yfried_ is now known as yfried|afk14:23
*** dangers_away is now known as dangers14:23
fungizuul is surprisingly busy for the friday before a summit14:24
*** sigmavirus24_awa is now known as sigmavirus2414:24
*** mpaolino has joined #openstack-infra14:27
*** sdake has joined #openstack-infra14:28
*** freerunner has joined #openstack-infra14:30
*** zz_dimtruck is now known as dimtruck14:30
*** annegentle has joined #openstack-infra14:31
*** ZZelle has joined #openstack-infra14:31
openstackgerritWayne Warren proposed openstack-infra/infra-specs: Jenkins Job Builder 2.0.0 API Changes/Rewrite
*** breton has joined #openstack-infra14:34
bretonwhere is source code for openstack proposal bot?14:34
electricalfungi: yay built my first job :p lol14:36
*** achanda has joined #openstack-infra14:36
*** mpaolino has quit IRC14:37
fungibreton: "openstack proposal bot" is just an account name in our gerrit used by a number of different jenkins jobs which propose changes on certain trigger events14:37
*** erikmwilson has quit IRC14:37
*** nfedotov has left #openstack-infra14:37
fungibreton: the scripts it runs are included in our job scripts directory but vary depending on the job14:38
fungibreton: any particular change proposal type you were interested in? i can point you to the corresponding job definition so you can see what script it's running and in what manner14:39
openstackgerritWayne Warren proposed openstack-infra/infra-specs: Develop gitub-integration spec.
openstackgerritWayne Warren proposed openstack-infra/infra-specs: Develop github-integration spec.
*** nelsnelson has joined #openstack-infra14:40
*** sabeen1 has joined #openstack-infra14:43
*** achanda has quit IRC14:43
*** Krinkle|detached is now known as Krinkle14:44
bretonfungi: awesome, thank you. We were discussing docs and sample config generating in #openstack-keystone, I wanted to have a look at how it's implemented now14:45
*** salv-orlando has quit IRC14:46
*** otter768 has joined #openstack-infra14:49
*** armax has joined #openstack-infra14:49
*** alexpilotti has joined #openstack-infra14:49
*** erikmwilson has joined #openstack-infra14:50
*** arielb has joined #openstack-infra14:51
*** arielb has joined #openstack-infra14:51
*** hdd has joined #openstack-infra14:51
*** otter768 has quit IRC14:53
fungibreton: sounds like fun. once you have more specific questions, i'm happy to answer those too14:56
*** mpaolino has joined #openstack-infra14:56
*** ZZelle has quit IRC14:57
*** mikal has joined #openstack-infra14:57
*** ZZelle has joined #openstack-infra14:57
*** banix has quit IRC14:57
electricalfungi: if i want some random block of text in yaml for a key its key:|  right?14:59
openstackgerritDavide Guerri proposed openstack-infra/shade: Add Neutron/Nova Floating IP create (allocate to project)
fungielectrical: yep. for example we use that for shell builders very frequently:
electricalAh. i must had the indentation wrong. was failing for me initially :-)15:00
electricalthought i was going mad15:00
fungielectrical: the "shell" keys you see in there are basically passing in raw shell scripts as blockquote text15:00
fungielectrical: i highly recommend an editor with a yaml formatting mode15:01
electricali should yeah.15:01
electricalvi is not doing a great job at it atm15:01
fungias with python, whitespace in yaml is syntactically significant to provide context15:01
electricalyeah i noticed :p15:02
*** claudiub has joined #openstack-infra15:02
*** radez_g0n3 is now known as radez15:02
fungithe up-side, once again much like python, is that makes it much more readable15:02
*** erikmwilson has quit IRC15:02
fungirestructuredtext is also that way, for similar reasons15:03
*** sdake_ has joined #openstack-infra15:03
*** nelsnels_ has joined #openstack-infra15:04
*** nelsnelson has quit IRC15:04
*** dimtruck is now known as zz_dimtruck15:05
*** rguillebert has quit IRC15:05
*** erikmwilson has joined #openstack-infra15:05
*** armax has quit IRC15:05
*** rlandy has quit IRC15:06
*** sabeen2 has joined #openstack-infra15:07
*** amitgandhinz has quit IRC15:07
*** sdake has quit IRC15:08
*** gus has quit IRC15:08
*** gus has joined #openstack-infra15:09
*** sabeen1 has quit IRC15:10
*** erikmwilson_ has joined #openstack-infra15:10
*** erikmwilson has quit IRC15:11
*** erikmwilson_ is now known as erikmwilson15:12
*** kmartin has joined #openstack-infra15:12
*** sarob has joined #openstack-infra15:12
*** jistr is now known as jistr|mtgh15:15
*** jistr|mtgh is now known as jistr|mtg15:15
*** _nadya_ has quit IRC15:15
*** sks has joined #openstack-infra15:15
*** nelsnels_ has quit IRC15:16
*** nelsnelson has joined #openstack-infra15:17
*** claudiub has quit IRC15:20
*** freerunner has quit IRC15:22
*** yaguang has joined #openstack-infra15:22
*** smarcet has joined #openstack-infra15:24
mordredelectrical: vi's yaml mode sucks. however, there is an ansible-yaml mode you can download that is much better - even at dealing with non-ansible ymal15:26
*** wenlock has joined #openstack-infra15:27
*** wenlock1 has joined #openstack-infra15:28
*** mase_x200 has quit IRC15:28
mordredelectrical: then I put that in ~/.vim/filetype.vim to make it use ansible mode for all my yaml files15:28
*** jistr|mtg is now known as jistr15:28
mordredmy life got way better15:28
*** mtanino has joined #openstack-infra15:29
*** radez is now known as radez_g0n315:29
*** yaguang has quit IRC15:29
*** weshay has quit IRC15:30
*** abregman has joined #openstack-infra15:30
*** yamahata has joined #openstack-infra15:31
electricalmordred: will give that a try15:31
*** Krinkle is now known as Krinkle|detached15:31
openstackgerritYanis Guenane proposed openstack-infra/tripleo-ci: Provide a longer timeout for HA scenario
*** freerunner has joined #openstack-infra15:32
*** mrmartin has joined #openstack-infra15:35
*** tonytan4ever has joined #openstack-infra15:37
*** gyee has joined #openstack-infra15:37
*** ibiris is now known as ibiris_away15:38
electricalDoes the yaml stuff for jjb allow for capitalizing letters? for example if i define a project name like 'elasticsearch'  i would like to show it up in my list as 'Elasticsearch'15:39
electricalor should i get used to everything lowercase if i do that?15:39
*** krtaylor has quit IRC15:39
*** amitgandhinz has joined #openstack-infra15:39
waynryou mean the job name itself?15:40
*** u_glide has joined #openstack-infra15:41
waynrjjb doesn't automatically captalize anything as far as i know, although there -may- be plugin modules that do15:41
waynrbut if you define a "job" or "job-template" with all lower case letters then the job name that shows up in the web url will be all lowercase15:42
*** puranamr has joined #openstack-infra15:42
*** banix has joined #openstack-infra15:42
u_glidehello guys! Could anyone explain why os-loganalyze shows 'Display level:' filter not for all logs at
*** weshay has joined #openstack-infra15:43
electricalwaynr: sorry, i mean in the 'display-name' part.15:44
waynrah, JJB doesn't perform any special transformation on the display-name15:45
*** Swami has joined #openstack-infra15:45
clarkbu_glide they dont match a level format that is recognized15:45
u_glideclarkb: How to fix that for example for m-api logs?
electricalwaynr: ah okay. i assume there is no way of template parts that i can tell it to capitalize the first letter of a variable like {name} ?15:47
waynrelectrical: here's an example display name on one of my template jobs:     display-name: '{name} ({scm_branch}){dn-sort-index} {dn-qualifier}Periodic Kickoff'15:47
*** hemnafk is now known as hemna15:47
clarkbu_glide you need to patch the match regex that os-loganalyze uses to apply to that file15:47
waynryou could just capitalize the name of the project15:47
electricalwaynr: ah okay.15:47
electricaltrue. but i would use the same name in other parts where it would need to be lowercase, like the github url :-)15:48
waynrwell in that case wouldn't it be more intuitive to have the actual name of the project show up in the display name for the job?15:48
waynr(that's what i do also, by the way)15:48
electricalwaynr: true.15:49
electricalwas just something i was thinking about as i currently have things with a capital letter :-)15:50
electricalchanging it to all lowecase isn't a big problem15:50
waynrelectrical: here are some examples of jobs i manage with JJB, although to be honest the are a little messy right now:
u_glideclarkb: Here ? :)15:50
*** krtaylor has joined #openstack-infra15:50
electricalwaynr: ah i see yeah ( btw, didn't realize you were from Puppet :p )15:51
waynryeah when we moved to JJB last year we just decided to deal with renaming our jobs using the template display names, it's really helped improved the consistency between ci pipelines15:52
fungithis channel is full of people from surprising places. we mostly just try collectively not to think about that ;)15:52
*** banix has quit IRC15:52
*** marcusvrn has quit IRC15:52
electricalwaynr: yeah indeed very true.15:52
electricalfungi: haha :p15:53
fungimakes it a lot easier to focus on the work at hand15:53
electricalI know so many Puppetlabs people i could almost be working there :p lol15:53
*** arxcruz has quit IRC15:53
waynrelectrical: yeah i've seen you in the puppet irc channels, and seen your nickname in github prs15:54
electricalwaynr: hehe yeah. especially beaker + docker related stuff15:55
*** BobBall is now known as BobBall_AWOL15:56
*** fedexo has joined #openstack-infra15:57
openstackgerritIgor Malinovskiy proposed openstack-infra/os-loganalyze: Add Manila logs to filter regex
*** banix has joined #openstack-infra15:57
*** tonytan4ever has quit IRC15:57
*** banix has quit IRC15:58
*** kmartin has quit IRC15:58
*** banix has joined #openstack-infra15:58
*** marcusvrn has joined #openstack-infra15:58
*** zz_dimtruck is now known as dimtruck15:59
*** sdake_ has quit IRC16:00
SpamapSzigo: starting to see what you meant about the patch getting large. I'm at 105 changed lines in 23 files for pysaml216:01
*** harlowja has joined #openstack-infra16:01
SpamapSI'll have to break it up a bit once I get all the tests passing16:01
u_glideclarkb: thanks!16:02
zigosabari: And you're far from finished!16:02
zigoI meant SpamapS16:03
zigoSpamapS: I just noticed that Keystone in Debian is broken because of using an older pysaml2 which tries to "import urllib.parse" ... :(16:03
*** harlowja has quit IRC16:03
SpamapSzigo: yeah I'd say I'm halfway done16:03
zigoSpamapS: keystone's requirements is obviously wrong.16:03
SpamapSzigo: lower bounds are usually unreliable because of the pip crack monkies always pulling highest. :)16:04
zigoAnd I discover them too late, and I'm always told: "no, we're frozen" ...16:04
SpamapSzigo: having unit tests run with the lower bounds has been discussed before16:04
*** annegentle has quit IRC16:04
zigoWhich I never undersood.16:04
zigoI mean, if it's broken, why leave it this way?16:04
SpamapSzigo: well don't you understand, the distros hate when reqs change late.. oh.. wait16:05
zigoI just hate discovering *after the facts* that it's wrong.16:05
zigoThat's worse.16:05
* zigo # dput python-pysaml2_2.4.0-1_amd64.changes16:06
fungitesting lower-bounds would effectively need a "special" feature added to pip (and we are probably the only projects who would use it, at least initially)16:06
SpamapSfungi: find the place where we sort, reverse the sort if special feature is on. :)16:07
fungiyep, and add a toddle for it16:07
fungier, toggle16:07
*** abregman has quit IRC16:07
SpamapSif we just do unit tests with it, we'd get plenty of "oh we don't work with that old version anymore" hits. A devstack job would be good too, but ENODES16:08
fungilike i say, feature addition to pip. didn't say it's unlikely to be accepted, but since others may not see a point in our use case there's always that danger16:08
*** kmartin has joined #openstack-infra16:08
*** xyang1 has quit IRC16:08
SpamapSfungi: Yes I agree most users would not understand that we want to smoke _OLD_ crack.16:08
*** xyang1 has joined #openstack-infra16:09
SpamapSjust to see if it still pops16:09
fungigreat analogy16:09
*** kmartin has quit IRC16:10
AJaegergood evening, I'd like to have some cores review, please - so that reed can use the slides for the upstream training16:10
*** yfried|afk is now known as yfried_16:10
electricalwaynr: guess i won't avoid having multiple projects defining the same set of jobs but having a different project name.16:10
*** u_glide has quit IRC16:12
*** armax has joined #openstack-infra16:13
zigofungi: If they don't see the point, it means we didn't explain well enough.16:13
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Reformat xUnit publisher help
*** ibiris_away is now known as ibiris16:14
*** yfried_ has quit IRC16:14
fungizigo: well, openstack's use of package ranges is already unusual in the python community. most projects test and release with a pip freeze of a bunch of == (if they even declare their dependencies programmatically at all) and leave it up to you to figure out what other versions might work16:14
*** gordc has quit IRC16:15
AJaegerfungi, thanks for the review.16:15
*** e0ne is now known as e0ne_16:15
*** gordc has joined #openstack-infra16:15
*** unicell1 has quit IRC16:16
*** mriedem is now known as mriedem_lunch16:18
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Support copyartifact build selector param
*** claudiub has joined #openstack-infra16:18
*** BharatK has quit IRC16:19
jpeelerAJaeger: thank you for the review as well16:19
*** e0ne_ is now known as e0ne16:20
*** john316 has joined #openstack-infra16:23
pleia2good morning16:23
*** john316 has quit IRC16:24
pleia2I'll be kind of in and out today as I have some pre-travel errands and things to do16:24
*** claudiub has quit IRC16:24
*** dtantsur is now known as dtantsur|afk16:25
*** derekh_ has quit IRC16:25
*** dguitarbite has quit IRC16:25
*** mrmartin has quit IRC16:27
fungiyeah, same for me16:27
*** patrickeast has joined #openstack-infra16:28
jpeelerfungi: could you take a quick look at ?16:28
openstackgerritAkihiro Motoki proposed openstack-dev/pbr: Make --help-commands work without testrepository
jeblairi've managed to find a way to spend 36 hours traveling to a summit in my own timezone :)16:28
pleia2jeblair: haha16:28
*** freerunner has quit IRC16:28
*** tonytan4ever has joined #openstack-infra16:28
*** claudiub has joined #openstack-infra16:29
*** marcusvrn has quit IRC16:29
jeblair(which, incidentally, started about 3 hours ago)16:29
fungiare you shipping yourself via fedex? ;)16:29
sdaguejeblair: is that the train time up?16:29
jeblairfungi: sounds better than united economy (boom-tish)16:30
*** marcusvrn has joined #openstack-infra16:30
*** abregman has joined #openstack-infra16:30
jeblairsdague: plane to pdx today to onboard some ppl and say hi to the pdx crowd; then train up tomorrow afternoon16:31
jeblairi actually could not take the train from brk anyway; it arrives in pdx about 1 hour too late for the hack train16:32
pleia2I thought about doing plane and train because fun, but 36 hours of travel in my own time zone made me rethink ;)16:32
fungijpeeler: one small problem with that change. see comment16:32
fungijeblair: i've glibly dubbed this trip "hack on the track"16:33
jeblairpleia2: it was only going to be like 14 hours when i was planning on flying to pdx tomorrow morning16:33
pleia2jeblair: yeah, even so, I'll take advantage of my short travel time for once16:33
jeblairfungi: we'll work on opentrack16:33
*** sparkycollier has joined #openstack-infra16:33
*** tonytan4ever has quit IRC16:34
jpeelerfungi: updating... thanks16:35
jeblairplus, train travel time only counts at a rate of 25% because it's so much better than airlines :)16:35
openstackgerritJeff Peeler proposed openstack-infra/project-config: make kolla check non-voting
*** ZZelle has quit IRC16:36
pleia2I do enjoy trains16:36
fungifor example, no invasive cavity searches16:36
*** sputnik13 has joined #openstack-infra16:37
JayFI wanted to take the train from SF->Vancouver, but it would've been hundreds of dollars more than the flight, not even counting the extra time :/16:37
*** e0ne has quit IRC16:37
JayFthat's why I always have trouble taking the train, more comfy, but more expensive for 4x+ as slow16:37
jeblairpleia2: the premier security line at sfot3 was _so much_ faster than precheck today -- the precheck line was super long16:37
fungiJayF: interesting. train from portland to vancouver was surprisingly cheap for the lowest-tier coach seat16:37
pleia2jeblair: doesn't surprise me, pre at sfo has been getting longer16:38
jeblairJayF: yeah, it rarely works out in the usa, unfortunately :/16:38
clarkbwe subsidize in the NW16:38
*** yamamoto has quit IRC16:38
clarkbnot sure if california plays nice16:38
*** yamamoto has joined #openstack-infra16:38
JayFclarkb: California has "Caltrain" which actually makes money but only goes from SF->Gilroy (south of San Jose) -- that all used to be amtrak16:39
jeblairclarkb: oh.  there are a lot of trains; that splains it16:39
JayFclarkb: idk anything about how we treat the actual amtrak stuff16:39
pleia2JayF: yeah, I'm in SF too, the cost was also a factor16:39
jeblairthere's a decent amount of amtrak running through the east bay, it's pretty easy to get around sac-oak-san jose16:40
jeblairbut i think only the coast starlight up to pdx?16:41
*** fedexo has quit IRC16:41
pleia2jeblair: yeah, I think so16:41
openstackgerritDarragh Bailey proposed openstack-infra/git-review: Switch to requests to support proxying of 'https'
*** spzala has joined #openstack-infra16:41
*** unicell has joined #openstack-infra16:41
pleia2the only route would have been leaving today on the starlight, then picking up the train in pdx tomorrow16:41
*** ddutta has joined #openstack-infra16:42
pleia2long journey all trains16:42
pleia2and about $150 per segment each wayh16:42
*** smarcet has quit IRC16:42
dduttahi ... this needs one more review
*** yamamoto has quit IRC16:42
*** unicell has quit IRC16:42
*** unicell has joined #openstack-infra16:43
jeblairddutta: approved.  just out of curiosity, are you interested in becomming a big-tent openstack project?  if so, why not start there?16:44
sdaguewe need a summit on the east coast that I can train to :)16:44
dduttajeblair: eventually! chatted w Steven Dake and he suggested we build a community in stack forge16:45
jeblairddutta: ok, thanks.  that's helpful :)16:45
dduttajeblair: what would be your suggestion on this.16:45
*** freerunner has joined #openstack-infra16:45
mtreinishsdague: I think you could have taken a train to atlanta16:45
*** dguerri is now known as _dguerri16:45
jeblairddutta: i want to make it so that we can start new projects directly in openstack easier.  but we're in a transition phase right now, so i think either way is fine.16:46
dduttajeblair: cool. We would like to do this as a proper openstack project whenever the transition is over then16:47
sdaguemtreinish: I would do DC by train16:47
*** yamahata has quit IRC16:47
sdagueatl is a bit much16:47
*** subscope has joined #openstack-infra16:47
*** ildikov_afk has quit IRC16:48
jpeelersomething wrong with zuul-cloner? not sure if that's going to cause problems or not16:48
*** nelsnelson has quit IRC16:48
jeblairjpeeler: details?16:48
*** otter768 has joined #openstack-infra16:49
clarkbhrm that seems likely related to my change?16:51
fungioof, we likely need to revert/hotfix that and rebuild images?16:51
*** mpaolino has quit IRC16:51
*** pabelanger has quit IRC16:51
clarkbmaybe, I did test it but clearly not well enough16:51
*** armax has quit IRC16:52
*** nelsnelson has joined #openstack-infra16:53
fungithat looks like maybe a negative check error... seems it dislikes that all the expected zuul parameters are set, and throws an error about needing to set an empty list of parameters16:53
asselinwhich zmq is infra using?
clarkbfungi the intent was to require all 3 if any one was set but otherwise allow empty16:54
clarkbasselin latest one16:54
*** nicodemos has joined #openstack-infra16:54
asselinclarkb, ?16:54
*** sparkycollier has quit IRC16:55
*** otter768 has quit IRC16:55
*** radez_g0n3 is now known as radez16:55
pelixwaynr: read your suggestions on the JJB 2.0 api, hopefully will have time to comment on them sometime over the weekend16:55
*** pabelanger has joined #openstack-infra16:55
*** radez is now known as radez_g0n316:56
*** pelix has quit IRC16:56
*** radez_g0n3 is now known as radez16:56
*** pabelanger has quit IRC16:56
*** pabelanger has joined #openstack-infra16:56
*** sparkycollier has joined #openstack-infra16:57
*** pabelanger has quit IRC16:57
*** radez is now known as radez_g0n316:57
*** pabelanger has joined #openstack-infra16:58
*** armax has joined #openstack-infra16:58
*** Sukhdev has joined #openstack-infra16:58
clarkbfungi: we can delete this mornings image builds16:58
clarkbI am still staring at the check and not immediately understanding why it would be breaking here16:58
clarkboh derp16:59
*** pabelanger has quit IRC16:59
*** pabelanger has joined #openstack-infra16:59
clarkbfungi: it needs to be if len(zuul_missing) and len(zuul_missing) < len(ZUUL_ENV_SUFFIXES)16:59
*** thedodd has joined #openstack-infra17:00
timrcpabelanger, Hi hi.  I've been super heads down with stuff inside HP-land, but I just noticed this: -- did you look into this?17:00
*** ildikov_afk has joined #openstack-infra17:01
pabelangertimrc, I did! In fact, I have something working already. It works very similar to JJB17:01
pabelangertimrc, is the code17:02
pabelangerSo far you can create a dashboard, with rows and text or dashlist panels17:02
jpeelerclarkb: is there a review that i can monitor for that change?17:02
*** ibiris is now known as ibiris_away17:02
*** jogo is now known as flashgordon17:02
pabelangertimrc, for example:
openstackgerritClark Boylan proposed openstack-infra/zuul: Allow all zuul args to be set in zuul-cloner
clarkbjpeeler: ^17:03
*** markmcclain has quit IRC17:03
pabelangerI plan to hack more on it this coming week at the summit17:03
jpeelerperfect - assuming that'll get in fairly quickly17:04
*** dimsum__ has quit IRC17:04
*** mpaolino has joined #openstack-infra17:04
sdagueclarkb / fungi - if anyone wants to review this - that will let the optimized hideci zuul support land. It will still be disabled behind a feature flag, but will be something that can be poked at.17:05
fungiclarkb: good eye17:05
*** dimsum__ has joined #openstack-infra17:05
*** annegentle has joined #openstack-infra17:05
*** shashankhegde has joined #openstack-infra17:05
clarkbfungi: should I go ahead and delete bare-* images from today?17:05
fungiclarkb: are you deleting the old images or should i? also note that i rebuilt all the bare-trusty images in hpcloud late last night because they hadn't been updated for something like 4 days... so we might not be able to roll back there17:05
*** marcusvrn1 has joined #openstack-infra17:06
clarkbfungi: the change to zuul to add this bug merged after those image builds so we should be ok17:07
clarkbfungi: I am deleting images now17:07
*** thedodd has quit IRC17:07
*** banix has quit IRC17:07
fungiokay, perfect. thanks!17:08
* fungi needs to stop e-mailing and get back to packing17:08
greghaynesoh, packing17:08
greghayneshadnt even thought of that17:08
clarkbgreghaynes: yes packing17:08
fungigreghaynes: don't for get your wooly viking hat. we're venturing into the great white north17:09
fungisnow train to the frozen wastes of vancouver17:09
*** marcusvrn has quit IRC17:09
timrcpabelanger, I'll also be at the summit.  We should meet up some time.17:09
greghaynesyes. itll be a real life snowpiercer17:09
AJaegerjeblair, pleia2, clarkb: If you have a chance to review for publishing training-guide, it would be great. Thanks!17:10
*** annegentle has quit IRC17:10
pabelangertimrc, indeed. Even though I didn't note anything on the -infra schedule, I wanted to atleast discuss it with people.  Eg yourself and jeblair17:10
*** yamahata has joined #openstack-infra17:10
waynrso my pointy-haired boss denied funding my trip to Openstack Summit in Vancouver17:10
pabelangerI don't see it being to hard to add in all the grafana functionality for dashboards, just a matter of getting the coverage17:10
*** markmcclain has joined #openstack-infra17:11
*** mpaolino has quit IRC17:11
waynri was looking forward to meeting more of y'all17:11
*** freerunner has quit IRC17:12
*** jistr has quit IRC17:12
clarkbjpeeler: fungi images deleted so we should be good until tomorrow morning and hopefully by then the fix can merge17:12
pleia2AJaeger: I might be confused here, but I don't see gate-training-guides-tox-doc-publish-checkbuild being used..?17:14
AJaegerpleia2: let me double check...17:14
fungipleia2: it's referenced via a project-template if memory serves17:14
*** ashleighfarnham has joined #openstack-infra17:14
jpeelerclarkb: is there an override process? kolla gate is broken dependent on a infra broken gate that can't be fixed until tomorrow?17:14
pleia2fungi: aha! name: ^gate-.*-tox-doc-publish-checkbuild$17:14
clarkbjpeeler: no I am saying its fixed right now17:14
AJaegerpleia2: openstack-doc-gate is the template that gets it in17:14
pleia2AJaeger: thanks17:15
fungipleia2: look for a gate-{name}-tox-doc-publish-checkbuild in a project-template in the layout.yaml17:15
jpeelerclarkb: ok sorry to misunderstand, will recheck now17:15
*** e0ne has joined #openstack-infra17:15
*** reed has quit IRC17:15
clarkbjpeeler: there may be a handful of straggler nodes on the old image but they wil get used up17:15
AJaegerpleia2: thanks for double checking and reviewing!17:15
*** miqui has joined #openstack-infra17:15
pleia2fungi: yep, thanks17:15
fungiright, it was openstack-doc-gate17:16
fungii should have remembered that... i was just looking at it when i reviewed earlier17:16
clarkbin other news didn't we just add f21 to kolla's gate?17:16
*** marcusvrn1 has quit IRC17:17
fungiclarkb: it's been moved to check-only nonvoting17:17
fungiapparently it takes at least 90 minutes to run, maybe longer17:17
*** Guest77307 has joined #openstack-infra17:17
fungithis was an unforeseen complication17:17
clarkbI wonder if firewalld is back17:17
*** HeOS has quit IRC17:18
asselinclarkb, not sure what happened, but zmq doesn't seem to open the port anymore. I don't see any log messages either in /var/log/jenkins.log when changing the port, even to invalid/already used.17:18
clarkbasselin: it won't open until a job runs17:18
clarkbasselin: and yo uhave to enable it per job or globall17:18
jpeelerfungi: would there be a way to have persistent data in between runs?17:18
*** panda is now known as panda|gone17:19
fungijpeeler: some... it would need to get generated or retrieved, then cached during our nodepool image builds17:19
jpeelera project for another day, but there's really no reason for us to rebuild every container every time17:20
fungijpeeler: for example you could have one job which periodically generates an artifact and publishes it to, then we could retrieve that during the image builds and cache it for use by your check jobs17:20
*** Guest77307 is now known as redrobot17:20
jpeeleryeah that seems workable17:21
fungias an example, we generate subunit timing info and then use that to pre-seed optimizations which can be reused by tempest jobs17:21
clarkbya there are other issues though like the containers not having our other cached data17:21
clarkbI think in general we will need to think about how we expose the carefully curated awesome in the host machine to containers becuse right now container based jobs are pretty sad and get none of that17:22
*** _dguerri is now known as dguerri17:22
*** marcusvrn has joined #openstack-infra17:22
*** mrmartin has joined #openstack-infra17:22
openstackgerritPaul Belanger proposed openstack-infra/system-config: Change out SSH key for pabelanger
*** banix has joined #openstack-infra17:24
*** redrobot has left #openstack-infra17:26
*** redrobot has joined #openstack-infra17:26
*** bhuvan has joined #openstack-infra17:26
*** reed has joined #openstack-infra17:26
*** mestery has joined #openstack-infra17:26
*** achanda has joined #openstack-infra17:27
*** redrobot has quit IRC17:29
*** kmartin has joined #openstack-infra17:29
*** sparkycollier has quit IRC17:30
*** Krinkle|detached is now known as Krinkle17:32
*** ryanpetrello has quit IRC17:33
*** notnownikki has quit IRC17:33
*** amotoki has quit IRC17:33
*** gordc has quit IRC17:34
jpeeleri fear the centos job is stuck -
*** annegentle has joined #openstack-infra17:35
jpeelerand of course i'm wrong17:35
*** redrobot has joined #openstack-infra17:36
*** redrobot is now known as Guest6319317:36
fungisometimes git on centos6 gets stuck on remote operations. sometimes it gets unstuck, sometimes it doesn't17:36
fungii've been unable to reproduce it under a debugger as it's infrequent and nondeterministic as far as i can tell17:37
fungiso i'm mainly just looking forward to the near future when we can stop caring about centos 6.x17:37
*** markmcclain has quit IRC17:37
*** luqas has quit IRC17:38
jpeelerit was a good five minute hang there17:38
*** Guest63193 is now known as redrobot17:39
fungiyeah, we saw that problem start right after the glibc update for the ghost vulnerability (like later that same day). i can't help thinking it's related17:39
jpeelerthat's disturbing17:40
*** annegentle has quit IRC17:40
jpeelerwhen you say not care about centos6, you mean because of upgrading to 7?17:41
*** dannywilson has joined #openstack-infra17:41
*** mriedem_lunch is now known as mriedem17:41
pleia2I think most of it was that we needed centos6 for py26 testing17:42
*** abregman has quit IRC17:44
*** subscope has quit IRC17:44
openstackgerritMerged openstack-infra/project-config: make kolla check non-voting
*** dannywilson has quit IRC17:45
*** barnaby has quit IRC17:46
fungiright. aside from a couple of servers we (infra) run centos6 for because of packages not being as great on ubuntu (cgit, asterisk) the only other reason we keep it around is as a testing platform for python 2.6 based jobs17:46
*** marcusvrn has quit IRC17:48
*** ibiris_away is now known as ibiris17:48
*** ddutta has quit IRC17:48
*** subscope has joined #openstack-infra17:49
*** wenlock has quit IRC17:50
*** marcusvrn has joined #openstack-infra17:50
*** tonytan4ever has joined #openstack-infra17:51
clarkbthough in theory we can upgrade both cgit and asterisk to centos7 machines now17:53
clarkbmight be good to replace one backend and one frontend for the git mirros with centos717:53
pleia2yeah, that's a good idea17:53
clarkbthen we can see how it does hopefulyl without anyone noticing the change17:53
* pleia2 nods17:53
*** sks has quit IRC17:54
asselinclarkb, fungi fyi I was able to verify zmq is publishing events using jeblair's nodepool/tools script. very nice:  python
pleia216.04 will have cgit, so if centos7 doesn't work out we could head back over to ubuntu next year17:54
*** wenlock has joined #openstack-infra17:55
jpeelerthere's got to be some testing advantage to having a slight mix in the gate17:56
clarkbjpeeler: we find lots of bugs in all the platforms :)17:57
*** ibiris is now known as ibiris_away17:57
pabelangerclarkb, not sure which packages are in centos7 for asterisk17:57
pabelangerbut should be easy to check17:58
pabelangerI think we are using the digium repos for asterisk17:58
clarkbsysv compat with upstart on ubuntu is completely broken, fedora's firewalld makes libvirt really slow, etc etc17:58
pleia2and I'm sure systemd on ubuntu will bring us all kinds of fun next year17:59
clarkboh yum install x y z will return 0 if any of x y or z install successfully18:00
pleia2hah, I didn't know about that one18:00
fungijpeeler: yeah, the balance is between "more platforms means we find more bugs in our software" and "more platforms means we spend more time working around bugs in those platforms themselves"18:03
*** freerunner has joined #openstack-infra18:03
openstackgerritMerged openstack-infra/project-config: Publish traning-guide files from master again
openstackgerritJoe Gordon proposed openstack-infra/infra-manual: Add section on how to run python unit tests.
fungiclarkb: except if yum can't resolve dependencies for at least one requested package18:04
jpeeleri already appreicated what you all do, but i never really thought about that aspect before18:04
fungie.g. if you've uninstalled python-setuptools on centos 6.x, which yum can't find to be able to install again if something depends on it18:05
* fungi can't grok that there are packages preinstalled on the os that can't be reinstalled with the package manager18:05
*** rbradfor has joined #openstack-infra18:05
flashgordonAJaeger: thanks for the review on
fungimore specifically, that you can have packages in a yum channel which depend on packages not present in any available channel (the base set of packages seems not to actually be obtainable via yum)18:07
fungithat would flat out never be allowed in debian18:07
openstackgerritMerged openstack-infra/project-config: Set the Kolla build timeout to 90 minutes
fungithere is a "base" channel, which we have enabled, but alas no python-setuptools in it18:08
openstackgerritMerged openstack-infra/project-config: Add python-cognitiveclient project for Cognitive
AJaegerflashgordon: thanks for writing it. LGTM now!18:09
*** Krinkle is now known as Krinkle|detached18:10
*** mrunge has quit IRC18:11
flashgordonAJaeger: cool, I figure over time we can add more sections in there about standard python practices in openstack18:11
*** annegentle has joined #openstack-infra18:11
SpamapSugh.. I forgot what its like to work on projects where people can break the build18:12
SpamapSpysaml2 doesn't gate.. tests may, or may not, work. :-P18:13
pleia2flashgordon: this is really great18:13
fungiSpamapS: it's scary out there18:13
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Fix python-cloudpulseclient ACLs pasto
*** mriedem has quit IRC18:13
flashgordonpleia2: thanks, this should deduplicate a lot of docs18:13
AJaegerpleia2, fungi: The patch above should enable python-cloudpulseclient to merge changes (seems they didn't noticed it - and we missed this during review). Would be nice to get this in for them18:14
AJaegerflashgordon: yeah, having a common place is great18:14
*** _nadya_ has joined #openstack-infra18:15
tim_oThe issue we were having yesterday was 1) A known bug: all is well now with the 'work around' in place, just an FYI/update18:16
openstackLaunchpad bug 1455028 in Cinder "cinder cannot start because of: oslo_config.cfg.DuplicateOptError: duplicate option: fatal_exception_format_errors" [Medium,Triaged]18:16
openstackgerritDavide Guerri proposed openstack-infra/shade: Add Neutron/Nova Floating IP delete (de-allocate from project)
*** ArxCruz has joined #openstack-infra18:16
pleia2AJaeger: oops, lgtm18:16
openstackgerritMerged openstack-infra/system-config: Change out SSH key for pabelanger
*** _nadya_ has quit IRC18:20
AJaegerfungi, good point on flashgordon's infra-manual patch. I was also thinking whether placing it closer to the sandbox or not.18:20
*** dannywilson has joined #openstack-infra18:20
fungiAJaeger: yeah, reorging the index and document title also won't invalidate links, so i think it's fine to merge as-is and then improve later18:21
*** ryanpetrello has joined #openstack-infra18:21
*** openstackgerrit has quit IRC18:22
*** openstackgerrit has joined #openstack-infra18:22
*** mriedem has joined #openstack-infra18:24
*** freerunner has quit IRC18:27
*** ayoung has quit IRC18:27
*** yamamoto has joined #openstack-infra18:31
*** Krinkle|detached is now known as Krinkle18:31
*** dguerri is now known as _dguerri18:31
*** ivar-lazzaro has joined #openstack-infra18:35
openstackgerritMerged openstack-infra/project-config: Fix python-cloudpulseclient ACLs pasto
openstackgerritMerged openstack-infra/puppet-nodepool: Consume puppet-diskimage_builder
openstackgerritMerged openstack-infra/puppet-bandersnatch: Initial import of puppet-bandersnatch module
*** yamamoto has quit IRC18:36
*** marcusvrn has quit IRC18:40
*** tim_o has quit IRC18:42
*** vponomaryov has quit IRC18:42
electricalfungi: i've had some good success with jjb today. managed to create some jenkins jobs with it using the job-template and job-group parts18:43
*** marcusvrn has joined #openstack-infra18:44
electricalGot a template per type of job and i use those in the job group. and that job group i use in every job18:44
*** rguillebert has joined #openstack-infra18:47
sdagueclarkb: yeh, that yum success on partial fail thing caused all manner of fun in devstack18:47
*** mrunge has joined #openstack-infra18:48
*** mrunge has quit IRC18:48
*** mrunge has joined #openstack-infra18:48
*** marcusvrn has quit IRC18:48
rguillebertlifeless, looks like we fixed pypy :)18:49
*** tonytan4ever has quit IRC18:49
*** stevemar has quit IRC18:50
lifelessrguillebert: \o/18:50
*** otter768 has joined #openstack-infra18:50
rguillebertthat's what I'm supposed to look at right?18:51
openstackgerritMerged openstack-infra/subunit2sql: Add graph command to show aggregate counts for tests
mtreinish^^^ time to push a new release, what could possible go wrong18:52
openstackgerritMerged openstack-infra/infra-manual: Add section on how to run python unit tests.
*** reed has quit IRC18:53
mtreinishor possibly (not an omen)18:53
*** markvoelker has joined #openstack-infra18:53
*** doug-fish has left #openstack-infra18:54
*** otter768 has quit IRC18:55
*** enikanorov2 has quit IRC18:55
*** mrunge has quit IRC18:55
*** tonytan4ever has joined #openstack-infra18:55
*** Krinkle is now known as Krinkle|detached18:56
*** marcusvrn has joined #openstack-infra18:57
*** ibiris_away is now known as ibiris18:57
*** Sukhdev has quit IRC18:58
electricalmtreinish: Sods law hit me more times then i care to remember, so i'm fairly cautious with those kind of remarks :p lol18:59
*** stevemar has joined #openstack-infra18:59
sdaguefungi / clarkb did you see the elasticsearch cve?19:00
electricalsdague: which one? the dynamic scripting ? :-)19:01
sdagueoh, that was a couple months ago, eh?19:02
sdagueI should read dates on posts better19:02
fungispeaking of...19:02
electricalsdague: hehe :p19:02
fungimrmartin: do you have any information about
clarkbsdague ours has been old enough to not have the exploitable scripting stuff they added19:02
clarkbsdague pretty sure we have avoided terribleness by being old19:03
electricalbut yeah, please disable it if you guys haven't done that yet sdague. i know of 1 customer that got hacked via that.19:03
clarkbwe are 0.90.9 no scripting19:03
fungi(just other security vulnerabilities which got fixed without anyone realizing they were there)19:03
electricalEeek. its like EOL you know :-)19:03
clarkbbut it works you know19:04
clarkband upgrading sucks19:04
electricalSome major improvements have been done along the way :-) memory wise as well19:04
electricalhehe i know :p19:04
electricalstill a bad reason though ;-)19:05
clarkbif you all fix the logstash + es upgrade story I might upgrade19:05
electricalES is easy. and LS works with any ES version19:05
clarkbbut as it stands I have almost no motivation because newer logstash never works and older logstash doesnt work against newer es19:05
*** pjenvey has joined #openstack-infra19:05
electricaldo you know we just released LS 1.5.0 ? :-)19:06
clarkb(sorry been burned too much in the past)19:06
electricali know :-)19:06
*** enikanorov2 has joined #openstack-infra19:06
electricalits something on the list to do to provide better upgrade paths19:06
electricalwhich LS version are you on now?19:06
clarkb1.3.something iirc19:06
electricalokay. upgrading to 1.5 will be a bit of pain yeah. conditional system and stuff. but performance has been very much improved and plugins have been separated ( my work )19:07
clarkbok will have to look at it again19:08
clarkbdoes es have auth yet? iirc that was going to happen and I am willing to upgrade for that19:08
electricalwe do have it but its a commercial plugin :-(19:08
clarkbwow ok19:08
clarkbfungi I didnt delete the devstack-* images, on a phone so cant do it but zuul cloner bug hitting dib jobs there19:09
clarkbpleia2 ^19:09
electricali have to say is extremely flexible, can grant access up to field level. but yeah. only comes with a support contract.19:09
fungilifeless: i love your anthropomorphization of pypi19:09
greghaynes for the broken jerb19:09
clarkbany chance one of you can clean those up?19:09
fungiclarkb: i'll knock those out19:09
lifelessfungi: :)19:10
clarkbI am sitting in the room I first met mordred in and he is here too19:10
fungiclarkb: that's got to be eerie19:11
*** ashleighfarnham has quit IRC19:12
*** panda|gone has quit IRC19:14
fungiclarkb: greghaynes: okay, devstack images are rolled back now too19:14
mrmartinfungi: I just see that, sounds fantastic.19:14
mrmartinfungi: need to ask Evgeny19:14
greghaynesfungi: ty19:14
*** panda|gone has joined #openstack-infra19:14
fungimrmartin: that would be awesome. even more awesome would be if he would respond to the oss-security mailing list about it19:15
*** ajo has joined #openstack-infra19:15
mrmartinfungi: interesting to see that not i'm the only people in the world who is complaining due release mgmt of that project. anyway it is based in django, but 1.5, not the 1.8 horizon changed to19:16
*** ashleighfarnham has joined #openstack-infra19:16
openstackgerritMatthew Treinish proposed openstack-infra/subunit2sql: Add release notes for 0.5.0
*** ajo has quit IRC19:20
*** shashankhegde has quit IRC19:21
adam_ghaving an issue pushing  a cherry-pick to a stable branch. gerrit rejects it saying the (original) change is closed. any ideas? these are the first cherry-picks this repo has seen19:24
clarkbadam_g does there exist a change on that branch with the same changeid somehow?19:24
openstackgerritFabien Boucher proposed openstack-infra/puppet-zuul: Set zuul server and merger services with ensure running
adam_gclarkb, nope19:25
*** teran has quit IRC19:26
*** teran has joined #openstack-infra19:26
clarkbadam_g can you paste the git review output?19:27
*** HeOS has joined #openstack-infra19:28
adam_gseems to happen on other repos too19:29
*** isviridov_away is now known as isviridov19:29
adam_gi see19:29
*** ihrachyshka has quit IRC19:29
adam_g.gitreview is missing its branch defaultbranch19:30
*** teran has quit IRC19:30
clarkb is the one it is confused with19:31
clarkbis that due to the .gitreview19:31
fungiwell, 182119 not 18219919:32
*** annegentle has quit IRC19:33
zarowhy is sched so horrible?19:36
*** dustins has quit IRC19:37
*** soren_ has joined #openstack-infra19:37
*** sigmavirus24 is now known as sigmavirus24_awa19:38
*** markmcclain has joined #openstack-infra19:38
*** _dguerri is now known as dguerri19:38
fungimaybe it runs on openstack ;)19:38
zaroactually i selected the wrong schedule, there's one for regular conference and one for design sumit.  the regular one doesn't have a filter for 'design summit'19:39
fungiright, that's intentional, to try and cut down on the number of people who wander into design summit sessions thinking they're conference sessions19:40
fungiso the conference only schedule is the one that's being made generally visible on websites and whatnot19:41
zarobut still no sync with mobile app and the mobile app keeps dying on my phone.19:41
*** hashar has joined #openstack-infra19:43
openstackgerritMerged openstack-infra/zuul: Allow all zuul args to be set in zuul-cloner
hasharoh zuul cloner love!19:47
*** ashleighfarnham has quit IRC19:48
*** dimtruck is now known as zz_dimtruck19:51
*** tnovacik has joined #openstack-infra19:53
*** subscope has quit IRC19:54
rcarrillocruzyeah, i was wondering also what's up with mobile app, i was expecting it to have vancouver summit agenda already19:57
openstackgerritgreghaynes proposed openstack-infra/project-config: Gate on diskimage-builder jobs in DIB
*** marcusvrn has quit IRC20:02
*** zz_dimtruck is now known as dimtruck20:03
fillerhey, did the vip ips roll recently?  say in the last 24-48 hours?20:04
*** banix has quit IRC20:04
*** Sukhdev has joined #openstack-infra20:05
greghaynesfiller: everything got rebooted recently20:05
greghaynesfiller: so that would probaly have that effect...20:05
clarkbbut the IPs did not change20:05
greghaynesyea, unclear what 'vips roll' means20:05
fillersorry.  i resolve to two routable ipv4 addresses now.20:05
clarkband if your client respects DNS round robins you should have not seen any blips20:05
fillerdid those addresses change at all.20:05
*** shashankhegde has joined #openstack-infra20:06
*** ashleighfarnham has joined #openstack-infra20:06
*** banix has joined #openstack-infra20:07
*** nelsnelson has quit IRC20:09
fillerok.  thanks clarkb and greghaynes!20:11
*** tiswanso has quit IRC20:12
*** ibiris is now known as ibiris_away20:14
*** ivar-laz_ has joined #openstack-infra20:14
*** sparkycollier has joined #openstack-infra20:15
*** sparkycollier has quit IRC20:16
*** ivar-lazzaro has quit IRC20:16
*** alexpilotti has quit IRC20:17
flashgordonso landed but I don't see the link up on yet20:18
*** achanda has quit IRC20:18
mtreinishflashgordon: you have to wait for the post job iirc20:25
mtreinishthat does the doc publishing20:25
*** mrmartin has quit IRC20:26
flashgordonmtreinish: I didn't see a post job running for the patch in zuul20:26
openstackgerritDavide Guerri proposed openstack-infra/shade: Add Neutron/Nova Floating IP attach/detach
mtreinishflashgordon: I see one in there it uses the sha1 instead of the review number20:28
mtreinishflashgordon: look for f3f688920:28
flashgordonahhh that must be it20:28
flashgordonhow is that still queued 1.5 hours later20:29
mtreinishlow prio?20:29
flashgordonare we somehow at capacity? at 200 nodes?20:29
*** baoli_ has quit IRC20:31
*** baoli has joined #openstack-infra20:31
*** tonytan4ever has quit IRC20:32
notmynamecan someone please remind me where the gate jobs are configured?20:32
mtreinishnotmyname: job definitions or the actually queue configurations?20:32
mtreinisheither way both are in project-config20:33
notmynamethe job definitions. I want to change which jobs are run on a specific branch for swift20:33
*** reed has joined #openstack-infra20:33
*** reed has quit IRC20:33
*** reed has joined #openstack-infra20:33
mtreinishnotmyname: well that's what I meant by queue configuration. :) Should be in here:
notmynamethanks :-)20:34
*** ajo has joined #openstack-infra20:34
*** ashleighfarnham has quit IRC20:35
notmynamehmm...I see where the "swift" jobs are set. is there an example of setting a different set of jobs for a particular branch?20:35
flashgordonmtreinish: the post queue is backed up by 3 hours20:35
flashgordonand nothing really idle in check20:36
flashgordonnotmyname: there are regex at the top by branch20:36
flashgordonnotmyname: search for 'stable/'20:36
*** Swami has quit IRC20:37
notmynamethanks. looking20:37
*** achanda has joined #openstack-infra20:37
*** mwagner_lap has quit IRC20:40
*** ivar-laz_ has quit IRC20:40
*** ivar-lazzaro has joined #openstack-infra20:40
flashgordonclarkb: is the post queue having issues?20:41
openstackgerritMatthew Treinish proposed openstack-infra/subunit2sql: WIP: Add doc about using subunit2sql-graph
*** ivar-lazzaro has quit IRC20:42
*** bswartz has quit IRC20:42
*** ivar-lazzaro has joined #openstack-infra20:43
*** annegentle has joined #openstack-infra20:44
openstackgerritDavide Guerri proposed openstack-infra/shade: Add Neutron/Nova Floating IP tests
clarkbflashgordon I dont think so its just a lower gearman priority20:48
*** peristeri has quit IRC20:49
flashgordonclarkb: I can't seem to find any other signs of us running out of nodes, or is it just the specific node type20:51
*** otter768 has joined #openstack-infra20:51
*** tnovacik has quit IRC20:54
*** aysyanne has quit IRC20:55
*** otter768 has quit IRC20:56
*** ashleighfarnham has joined #openstack-infra20:56
openstackgerritGeorge Peristerakis proposed openstack-infra/system-config: Move the server class settings out of the ask class
*** ashleighfarnham has quit IRC21:00
*** davideagnello has quit IRC21:01
*** sputnik13 has quit IRC21:01
*** sputnik13 has joined #openstack-infra21:02
*** armax has quit IRC21:03
*** _david_ has joined #openstack-infra21:03
*** sputnik13 has quit IRC21:04
*** hashar has quit IRC21:04
*** davideagnello has joined #openstack-infra21:04
*** mattfarina has quit IRC21:04
*** amitgandhinz has quit IRC21:05
*** ashleighfarnham has joined #openstack-infra21:06
openstackgerritMorgan Fainberg proposed openstack-infra/project-config: Add keystoneauth-core group
*** hashar has joined #openstack-infra21:06
morganfainbergclarkb, mordred, fungi, ^ important so we can get keystoneauth core group to include people such as dtroyer. This should help ensure what is developed for keystoneauth can avoid the client/server-hell issues (avoid breaking things in silly ways)21:07
morganfainberganteaya cc ^21:08
*** ldnunes has quit IRC21:15
*** stevemar has quit IRC21:17
*** rainya has joined #openstack-infra21:19
*** rguillebert has quit IRC21:21
*** kgiusti has quit IRC21:22
*** AL34N1X has joined #openstack-infra21:24
*** AL34N1X has quit IRC21:27
*** ashleighfarnham has quit IRC21:28
*** freerunner has joined #openstack-infra21:30
*** ociuhandu has quit IRC21:31
*** prad has quit IRC21:33
openstackgerritDavanum Srinivas (dims) proposed openstack/requirements: Switch on requirements check for nova-docker
*** freerunner has quit IRC21:36
*** eharney has quit IRC21:36
*** freerunner has joined #openstack-infra21:36
*** dprince has quit IRC21:37
fungisorry, was between internets. am now on car internet, catching up21:37
_david_fungi, thx for the input. I have nova.git now, Will see if  could reproduce that failure21:39
notmynameflashgordon: the regexes at the top are for jobs, not for projects. so can I add a "branch" to the swift project section (with a proper regex)? or do I do something in the jobs?21:40
notmynameflashgordon: my goal is to turn off the python jobs of the feature branch where we aren't using python21:41
flashgordonnotmyname: for jobs yes, but I *think* you can limit to project as well21:41
*** erlon has quit IRC21:41
notmynameactually, for now I want only the noop job to run on feature/hummingbird21:41
notmynameuntil we get the golang tests set up21:41
flashgordonnotmyname: not sure how to do that, perhaps  someone else knows21:42
*** dimtruck is now known as zz_dimtruck21:42
notmynameno rush, I'll pick it up later. I've got a presentation to write now21:43
flashgordonnotmyname: so does this mean you are going forward with the go? or is this still just an experiment?21:43
notmynameflashgordon: still an expirament21:43
flashgordonnotmyname: ahh21:43
notmynameflashgordon: but there's no reason to run all the tests on an R&D branch, especially when it's a different language21:43
fungi_david_: you probably want an identical copy of the bare nova repository from the gerrit server rather than cloning it. for troubleshooting purposes we put a tarball of that at
flashgordonnotmyname: any idea of how big of a perf difference you get with it?21:43
notmynamenope, not yet21:43
flashgordonnotmyname: makes sense to me, I was wondering why you were running python tests on that branch21:44
_david_fungi, Sure, i downloaded nova.git form this link21:44
notmynameflashgordon: only because i hadn't figured out to turn them off yet21:44
_david_fungi, I'm looking at this pack file: git verify-pack -v pack-93ad57004de887eb835b2bd4df2d7c3f6a5c394b.pack21:45
funginotmyname: see the second example at
*** rkukura_ has joined #openstack-infra21:46
_david_fungi, interestingly, can be reverted without conflicts21:46
flashgordonnotmyname: maybe the skip-if logic would work?21:46
funginotmyname: basically just add the merge-check and noop-jobs templates to your project entry for that hummingbird repo21:46
funginotmyname: assuming it's a separate git repo21:47
flashgordonfungi: its a branch in swift repo21:47
flashgordonbecause that is what branches are good for21:47
_david_fungi, And all tests are passing, ... I think it would worth trying to try JGit with this change reverted21:47
*** hashar has quit IRC21:47
*** rkukura has quit IRC21:48
fungiflashgordon: notmyname: yeah that's going to get complicated. you'd need to explicitly list all the jobs which run on your normal branches and set them to a regex that excludes the hummingbird branch21:48
*** annegentle has quit IRC21:49
fungi_david_: yeah, clarkb even compiled the same jgit version from our 2.10 war and tried inspecting the repo with the jgit cli, no errors21:49
*** rkukura has joined #openstack-infra21:50
fungi_david_: it seems like there may be a performance-related component to this issue, where it only emerges under heavy activity21:51
*** annegentle has joined #openstack-infra21:51
morganfainbergyay i got a pleia2 +2 :)21:51
morganfainbergpleia2: thnx! :)21:51
_david_fungi, define "heavy activity"?21:52
*** rkukura_ has quit IRC21:52
morganfainbergfungi: jgit - it's like going into a funhouse with those mirrors...except with git repos...and bad message pack files.21:52
pleia2morganfainberg: sure, easy one21:52
*** sabeen2 has quit IRC21:53
fungimorganfainberg: if my gertty syncs soon over this modem, you'll have a +321:53
morganfainbergfungi: yay!21:53
morganfainbergfungi: then i'll bug you to put me in said new group so i can do $things21:54
pleia2I can do it too if fungi is car-ing21:54
fungi_david_: well, normal activity for us... maybe 10 events a second, a hundred or so established connections to the ssh api, ssh connection turnover of a few a second21:54
morganfainbergpleia2: yep!21:55
_david_fungi, Are you using slaves for CI pulls?21:55
morganfainbergpleia2: <3 the infra peoples! you're all great!21:55
fungi_david_: and not sure how many requests a second to jetty. i'd have to run some stats21:55
morganfainberg(and make my life better in openstack land)21:56
fungi_david_: yeah, our ci doesn't clone from gerrit, and we have an intermediary layer to fetch change refs and rehost them to serve to the workers21:56
_david_fungi, OK.21:57
fungi_david_: so most of the load is from end users and stream-events listeners21:58
*** Swami has joined #openstack-infra21:58
*** davideagnello has quit IRC21:58
*** teran has joined #openstack-infra22:00
openstackgerritMerged openstack-infra/project-config: Add keystoneauth-core group
*** tnovacik has joined #openstack-infra22:04
*** Sukhdev has quit IRC22:05
*** ddieterly has quit IRC22:09
*** sputnik13 has joined #openstack-infra22:09
*** davideagnello has joined #openstack-infra22:09
*** bswartz has joined #openstack-infra22:09
*** isviridov is now known as isviridov_away22:10
*** bknudson has quit IRC22:10
*** hdd has quit IRC22:11
*** hdd has joined #openstack-infra22:14
*** _david_ has quit IRC22:14
*** claudiub has quit IRC22:17
*** claudiub has joined #openstack-infra22:17
*** ociuhandu has joined #openstack-infra22:22
*** tnovacik has quit IRC22:25
*** Krinkle|detached is now known as Krinkle22:26
*** banix has quit IRC22:28
*** rfolco has quit IRC22:34
*** ashleighfarnham has joined #openstack-infra22:35
*** spzala has quit IRC22:40
*** sabeen1 has joined #openstack-infra22:43
*** arielb has quit IRC22:43
*** yamahata has quit IRC22:48
*** jamielennox|away is now known as jamielennox22:48
*** freerunner has quit IRC22:48
*** enikanorov2 has quit IRC22:50
*** rainya has quit IRC22:50
*** annegentle has quit IRC22:51
*** mwagner_lap has joined #openstack-infra22:52
*** otter768 has joined #openstack-infra22:52
morganfainbergpleia2: ping - mind adding me to the keystoneauth-core and release groups?22:53
*** hdd has quit IRC22:53
pleia2morganfainberg: it's still in the post queue 7d4332322:53
morganfainbergjamielennox was looking at the group in gerrit22:54
pleia2oh, hm22:54
morganfainbergi hadn't checked the actual review yet22:54
pleia2aha, there it is22:54
morganfainbergpleia2 :)22:54
pleia2morganfainberg: done22:54
morganfainbergpleia2: thanks!!22:54
morganfainbergpleia2: will also need to be added to keystoneauth-release22:55
morganfainbergpleia2 if you don't mind.22:55
morganfainbergpleia2 :)22:55
pleia2morganfainberg: there you go :)22:56
morganfainbergpleia2: I think i owe ya a beer... cause you're awesome.22:56
pleia2beer \o/22:56
* morganfainberg probably owes all of -infra beers22:56
fungipleia2: yeah, that has nothing to do with post queue jobs. once it's merged, the next puppet run to start will pick it up (cron job on the puppetmaster)22:56
*** otter768 has quit IRC22:56
pleia2fungi: aha :)22:56
*** rainya has joined #openstack-infra22:57
fungipleia2: our rube-goldberg machine where cron fires a shell script that updates git and runs ansible which logs into machines and runs puppet agent which runs scripts which pulls more things from git which connects to gerrit which...22:58
fungithat mess22:58
pleia2fungi: yeah, wasn't quite sure where gerrit group creation landed22:58
*** dangers is now known as dangers_away22:58
*** signed8bit is now known as signed8bit_ZZZzz22:58
fungipleia2: manage-projects does it22:58
*** achanda_ has joined #openstack-infra23:01
jeblairfungi: i can't wait to see your architecture diagram!23:02
fungijeblair: i want to build a representative physical model rube goldberg machine out of junk and label each of the components with the names of our systems23:03
fungiit would make a great art installation for the dev lounge23:03
pleia2this old boot represents gerrit23:03
*** ZZelle has joined #openstack-infra23:03
jeblairnow i know why we should have a midcycle23:04
*** e0ne has quit IRC23:04
*** sputnik13 has quit IRC23:05
*** achanda has quit IRC23:05
*** dims_ has joined #openstack-infra23:05
*** ociuhandu has quit IRC23:06
*** achanda_ has quit IRC23:06
*** dimsum__ has quit IRC23:07
*** dboik_ has quit IRC23:09
*** dboik has joined #openstack-infra23:10
*** claudiub has quit IRC23:12
*** ArxCruz has quit IRC23:14
*** ArxCruz has joined #openstack-infra23:15
*** puranamr has quit IRC23:16
*** wznoinsk_ has joined #openstack-infra23:17
openstackgerritMerged stackforge/python-jenkins: Add decode to utf8 for python 3.x
openstackgerritGerman Eichberger proposed openstack/requirements: Adding responses libray
*** wznoinsk has quit IRC23:19
clarkbmordred ^ yet another23:19
*** ajo has quit IRC23:19
*** bhuvan has quit IRC23:20
*** soren_ has quit IRC23:21
*** ashleighfarnham has quit IRC23:24
*** ArxCruz has quit IRC23:24
*** hemna is now known as hemnafk23:27
*** shashankhegde has quit IRC23:27
*** ashleighfarnham has joined #openstack-infra23:28
*** ashleighfarnham has quit IRC23:28
clarkb(there was conversatiom about how we have a ton of different http response mock tools)23:29
*** enikanorov2 has joined #openstack-infra23:33
openstackgerritGerman Eichberger proposed openstack/requirements: Adding responses libray
fungithere can be only one!23:36
sdagueclarkb: heh. In nova we just start it for reals and be done with it. Bringing up our wsgi stack really doesn't take very long.23:36
*** ashleighfarnham has joined #openstack-infra23:36
*** dboik has quit IRC23:37
*** hdd has joined #openstack-infra23:38
*** dannywilson has quit IRC23:40
*** sabeen1 has quit IRC23:40
*** ivar-lazzaro has quit IRC23:44
*** ivar-lazzaro has joined #openstack-infra23:45
*** ivar-lazzaro has quit IRC23:45
*** ivar-lazzaro has joined #openstack-infra23:46
jamielennoxfungi: httpretty needs to go away, i've done it for a couple of projects, the upstream is just dodgy23:46
jamielennoxi think that just leaves requests-mock23:46
jamielennoxassuming we don't add responses23:46
*** Sukhdev has joined #openstack-infra23:49
mtreinishheh, looks like the new release didn't install correctly:
mtreinishhmm, it doesn't like matplotlib...23:50
*** annegentle has joined #openstack-infra23:51
clarkbwebtest is another23:51
clarkbwerkzeug maybe?23:53
*** baoli has quit IRC23:53
clarkbmtreinish missing C dev headers?23:54
patrickeasti’m seeing problems with matplotlib on my 3rd party ci too while trying to build new nodepool images
mtreinishclarkb: yeah that's what I was thinking. But, the installed version requires pandas which I thought already had a dep on matplotlib23:55
sdaguejamielennox: what projects still use httpretty?23:55
patrickeasterr wait
patrickeastwrong paste23:55
mtreinishpatrickeast: heh, that's probably my fault too, from the new subunit2sql release23:55
jamielennoxsdague: i don't know off the top of my head, i can propose to remove it from reqs and see23:55
*** pelix has joined #openstack-infra23:56
mtreinishpatrickeast: thanks that paste probably explains why it doesn't work on subunit-worker01, it's libpng23:56
*** rainya has quit IRC23:56
mtreinishI'll push out a a 0.5.1 which removes matplotlib from requirements as a temporary measure and just put something in the readme about needing it for the graph command23:57
sdaguesomething did, because people were scrambling for the pin23:57
*** annegentle has quit IRC23:57
* mtreinish can't wait for extras in pbr to be a thing23:57
clarkbironic does23:57
*** yamamoto has joined #openstack-infra23:57
*** pcm_ has quit IRC23:59
*** vahidh has joined #openstack-infra23:59

Generated by 2.14.0 by Marius Gedminas - find it at!