Tuesday, 2015-06-09

ianwis anyone around who can hold me a centos7 vm so i can try and diagnose the centos7+neutron failures?00:04
*** dims_ has joined #openstack-infra00:05
jeblairianw: can do00:05
*** sputnik13 has joined #openstack-infra00:05
jeblairmordred: can you look at https://review.openstack.org/184559 particularly my review comment?00:05
*** marun has joined #openstack-infra00:06
jeblairianw: can you trigger a centos7+neutron failure and point me at a node number?00:06
*** unicell has quit IRC00:06
*** unicell has joined #openstack-infra00:06
*** sputnik13 has quit IRC00:07
*** dims has quit IRC00:08
*** ociuhandu has quit IRC00:08
*** mtanino has quit IRC00:09
lifelessfungi: pop (context switched back)00:09
ianwjeblair: devstack-centos7-rax-iad-293258100:11
jeblairianw: ssh key?00:12
ianwjeblair: https://github.com/ianw.keys <- the one ending in "UcMr"00:13
*** yamamoto has quit IRC00:15
jeblairianw: ssh jenkins@
ianwjeblair: thankyou.00:16
*** shuoy has joined #openstack-infra00:17
openstackgerritJeremy Stanley proposed openstack-infra/project-config: Add periodic propose-requirements-constraints job  https://review.openstack.org/18951100:17
fungilifeless: ^00:17
*** freerunner has quit IRC00:17
fungithough i'm going to -1 it for a bit while we work out security concerns00:17
*** ho has joined #openstack-infra00:19
*** koolhead17 has joined #openstack-infra00:19
*** tonytan4ever has joined #openstack-infra00:21
tristanCasselin: now I don't understand why jenkins succeed https://review.openstack.org/#/c/177373 because it still use the old variable name prv_ssl00:21
* asselin looks00:21
tristanCasselin: so I'll fix that and propose a follow up to clean the snakeoil cert usage00:22
tristanCunless ruby doesn't care if we use undefined variable...00:23
asselintristanC, I think you need to do a recheck00:23
ianwjeblair: any record of when jenkins reruns the jobs if the vm drops offline?00:23
ianwhttp://logs.openstack.org/89/179689/30/check/check-tempest-dsvm-centos7/45ab41d/ < 06:1000:23
asselinthat patch you didn't update since may 2600:23
*** freerunner has joined #openstack-infra00:24
lifelessfungi: any thoughts on the security angle - like, is that a month long -1, or a few days ? Can I do something to help ?00:24
ianwhttp://logs.openstack.org/89/179689/30/check/check-tempest-dsvm-postgres-full/47e3fa4/ <- 05:34 , i'm sure the centos7 one is a rerun00:24
*** puranamr has quit IRC00:25
asselintristanC, sorry...still not seeing the issue?00:25
*** zz_dimtruck is now known as dimtruck00:25
jeblairianw: i might be able to dig it out of the log; is that important?00:25
fungilifeless: i mostly want some weigh-in from other cores. the credentials aren't as high-value as, say, the ones on the pypi slave. also this is running code which has already been approved/merged to projects, not arbitrary gerrit changes00:25
*** bhuvan has joined #openstack-infra00:25
*** cathy_ has joined #openstack-infra00:26
*** shuoy has quit IRC00:26
tristanCasselin: https://review.openstack.org/#/c/177373/6/modules/openstack_project/manifests/jenkins.pp line 38, it still use $prv_ssl_cert_file which is unset00:26
fungilifeless: the related issue here is that the only worker with access to those credentials, currently, is not a nodepool node but a persistent jenkins slave, so we need to worry about whether that job will generate a lot of stale data and need some maintenance attention00:27
openstackgerritTristan Cacqueray proposed openstack-infra/system-config: Use common jenkins master implementation  https://review.openstack.org/17737300:27
ianwjeblair: ok, not at this point; i'm monitoring that job and sending telemetry out ATM ... "hopefully" it fails00:27
ianwit seems like OOM issues similar to glusterfs that make the machine look dead00:27
*** koolhead17 has quit IRC00:27
ianwbut this is something that started recently, just last week00:28
lifelessfungi: so there is a vector here00:28
asselinjhesketh, are you around?00:28
lifelessfungi: the generate.py code isn't what I'm worried about00:28
*** yamamoto has joined #openstack-infra00:28
lifelessfungi: we're pip installing in a venv00:28
*** shuoy has joined #openstack-infra00:28
asselintristanC, yeah..I see that00:28
tristanCasselin: so this is fixed in patchset 700:28
mordredlifeless: I love your tweet about egg_info00:29
lifelessfungi: it would leave a trail a mile wide, but someone could upload a cffi that escapes the virtualenv, scrovles through the fs, emails stuff out00:29
*** sdake has joined #openstack-infra00:29
*** baoli has joined #openstack-infra00:29
lifelessfungi: it would have to be a package that we've approved into global-requirements of course00:29
lifelessfungi: so I'm not very concerned.00:29
asselintristanC, I'm also seeing now that the prv_ssl_cert_file should stay in system-config, and not migrate at all to puppet-openstack ci......00:29
fungilifeless: yeah, when i say "running code which has already been approved/merged to a repo" i mean either for one of our many participating projects or released deps of theirs00:29
mordredtristanC: are you a good person to ask questions about cauth? or would that be a different human?00:30
asselintristanC, what do you think?00:30
lifelessmordred: you'll want to come to my talk @ pyconau about pip horrors00:30
mordredlifeless: done00:30
mordredjeblair: looking00:30
asselintristanC, sorry...wrapping my head around this and trying to keep it simple.00:31
tristanCasselin: I think it's fine to do the refactor in two step: first move the code out of project-config, then shift that key_file logic around00:31
tristanCmordred: I can help00:31
lifelessmordred: can you review this https://review.openstack.org/#/c/189491/ and https://review.openstack.org/#/c/189476/100:31
mordredjeblair: yes, I agree00:31
*** gyee_ has quit IRC00:32
fungilifeless: if we decide the risks are not acceptable, we can consider a multi-stage job which generates the constraints files on a nodepool node and then a dependent job which gets the generated data files and proposes the diffs to gerrit00:32
lifelessfungi: mordred: https://review.openstack.org/#/c/189249/1 needs review love too00:32
lifelessfungi: sure. I will follow the leader here.00:32
*** sdake_ has quit IRC00:32
*** kutija|away has quit IRC00:33
mordredtristanC: cool. I may bug you with details tomorrow, but my current theory is that adding openid support to cauth and using it in front of phabricstor would be the best SSO option for it00:33
asselintristanC, but then you'll have 2 more patches: 1 to refactor puppet-openstackci to use snakeoil as defaults, and then another to add back the ${vhost_name} options into system-config00:33
mordredlifeless: on my phone ... bad time for reviews00:33
lifelessfungi: I'd hope a certain amount of pragmatism can apply, since folk do run 'sudo pip install' in places - and we're not even that bad :)00:33
lifelessmordred: ack.00:33
fungilifeless: yep, i am thinking it's probably not a bad solution as is00:33
fungilifeless: from a functionality front, we'll want to consider how to expand this to additional branches besides master at some point, but perhaps not until liberty releases?00:34
lifelessfungi: shouldn't it handle that now, as-is ?00:34
lifelessfungi: the existing branch stuff in the script, and it reads from the global-requirements.txt in the branch being proposed to.00:34
lifelessfungi: we'll run the latest generator code against the branch in question.00:35
fungilifeless: well, i think i'll need to set the branch override in the job definition actually, and have additional jobs per branch00:35
lifelessfungi: oh sure. Glue. :)00:35
tristanCmordred: sure thing, mhu is the main dev these days, I'll let him know00:35
fungilifeless: is it assumed to work presently on any stable branches? if so, which ones? all of them?00:36
*** cathy_ has quit IRC00:36
lifelessfungi: from the session we were going to do kilo only.00:36
lifelessfungi: I need to put a similar set of patches together to the ones in my requirements stack that pin to 2.7 only the non-3.4 deps00:37
fungilifeless: okay, easier that way at least. i'll fix up my review to have master and stable/kilo jobs00:37
lifelessfungi: but, until we smoke out all the places that env markers tickle, we can't sensibly do that. having the kilo job is good though, as it will give diagnostics.00:37
jheskethasselin: yep00:38
*** freerunner has quit IRC00:38
lifelessdstufft has merged my -c patch to pip; now I think the next thing is to patch up devstack to use that, which for now will need pip-from-git. Yay.00:38
asselinjhesketh, hi, ok, tristanC's around too...and I'd like us to get in sync on jenkins patches00:38
fungilifeless: oh, another snag... proposal.slave.o.o is running ubuntu precise, so we can't do python 3.4 there. might need a newer/upgraded worker to run this00:38
*** shuoy has quit IRC00:39
*** sdake_ has joined #openstack-infra00:39
lifelessfungi: eek. Funny, I asked about that yesterday ;)00:39
asselinjhesketh, tristanC this is the currently system-config change https://review.openstack.org/#/c/177373/7/modules/openstack_project/manifests/jenkins.pp00:39
openstackgerritTristan Cacqueray proposed openstack-infra/puppet-openstackci: Use snakeoil default certificate  https://review.openstack.org/18951500:39
fungii was probably not around, sorry :/00:39
openstackgerritTristan Cacqueray proposed openstack-infra/system-config: Use common ssl path logic from openstackci  https://review.openstack.org/18951600:39
lifelessfungi: its ok, I blame mordred and pleia2 :)00:39
*** Swami has quit IRC00:39
fungiwell, they _are_ easy targets00:39
asselinjhesketh, tristanC I think we need to add back in the vhost_name part..which is infra specific00:40
fungibut yeah... Welcome to Ubuntu 12.04.5 LTS (GNU/Linux 3.2.0-83-generic x86_64)00:40
*** sputnik13 has joined #openstack-infra00:40
*** rkukura_ has joined #openstack-infra00:40
pleia2lifeless: we have servers that support what you need, proposals.o.o just isn't it :)00:40
fungilifeless: i don't anticipate replacing that worker with one running trusty to be at all negatively impacting00:40
tristanCasselin: jhesketh: so https://review.openstack.org/#/c/189515/ is the follow-up patch to use snakeoil certificates00:40
asselinjhesketh, tristanC then in the puppet-openstackci, we can use the default snakeoil.pem, and not do any special confs00:40
pleia2fungi: ++00:41
fungiand it needs to happen sooner or later anyway00:41
lifelesswe could run each python version in a container00:41
lifelesswould address security00:41
lifelessand give us more flex in future00:41
lifelessYAGNI ?00:41
*** yamada-h has joined #openstack-infra00:42
*** sputnik13 has quit IRC00:42
*** hdd has joined #openstack-infra00:42
*** rkukura has quit IRC00:42
*** rkukura_ is now known as rkukura00:42
fungilifeless: any guesses as to how time-intensive this job would end up being? wondering if having a separate worker for it would make sense from a not-tying-up-the-proposal-slave perspective00:42
*** sdake has quit IRC00:42
jheskethtristanC, asselin: I would switch the dependencies around.. so system-config gets updated first, then puppet-openstackci00:42
jheskethtristanC, asselin: what you have is correct, but I think we should still use the vhost_name variable (and hence the if block) just in case we rename jenkins or want somebody else to consume etc00:43
*** fallenpegasus has joined #openstack-infra00:43
asselinjhesketh, please clarify where you expect the vhost_name block to exist, system-config (infra/3rd party ci specify) or puppet-openstackci (common) ?00:44
tristanCjhesketh: the idea (i guess) is to have openstackci generic, so path could have usefull default and system-config can still use specifics path00:44
asselin(i'm assuming we agree it should only be in one of the two)00:44
*** yamamoto has quit IRC00:45
lifelessfungi: time .... > 2m14s00:45
jheskethasselin, tristanC: yeah only in one place. It probably doesn't really matter, but I think it'd be neater/better in system-config00:45
lifelessfungi: SSDs, primed pip 7 wheel cache.00:45
*** yamada-h has quit IRC00:46
asselinjhesketh, +1 I agree that's a better location. tristanC ?00:46
lifelessfungi: is once a day, so even if it was longer I wouldn't be too worried; linux is good at scheduling :)00:46
fungilifeless: probably not bad on a static worker with a persistent pip wheel cache and a warm disk cache too00:46
*** sputnik13 has joined #openstack-infra00:46
*** sputnik13 has quit IRC00:46
fungiwhich the current proposal slave would be, just need to update the distro release so it can also have python 3.400:47
lifelessor install a random build from a ppa00:47
lifelesslets update :)00:47
tristanCjhesketh: asselin: why not... but I can't find more than two usecase for openstack_project::server and the ssl path confusion could easily get solved00:47
tristanCjhesketh: asselin: this change seems like removing the need of this logic: https://review.openstack.org/#/c/189516/00:48
jheskethtristanC: are you saying it should be in puppet-openstackci?00:48
*** bhuvan has quit IRC00:48
jheskethtristanC: I think we're proposing that 189516 uses vhost_name if the ssl_cert file is empty00:49
*** chuckC_ has quit IRC00:49
asselintristanC, jhesketh I think it's easier to leave it in 17737300:51
asselini.e. go back to patch set 5: https://review.openstack.org/#/c/177373/5..7/modules/openstack_project/manifests/jenkins.pp00:52
jheskethasselin: either way is fine to me... the end result is the same, but reviewing might be easier in two steps since we've done it now00:52
asselinjhesketh, I would prefer not moving the vhost_name part back and forth which seems what's we're doing.00:53
tristanCasselin: jhesketh: hum I confused now... either way is fine to me, as long as it satisfy everyone :-)00:53
asselinhere's what I think now: system-config using patch set 5 of 17737300:54
asselinand puppet-openstackci using patch set 4 17557100:55
asselinthis way puppet-openstack ci is clean and simple. the vhost complexity remains in system-config where it belongs00:56
jheskethasselin: sounds good to me :-)00:57
jhesketh(sorry I missed your comments in the thread, we could have avoided this confusion)00:57
tristanCok deal :) I'll update these review first thing tomorrow if that's ok with you00:58
openstackgerritJeremy Stanley proposed openstack-infra/project-config: Add periodic propose-requirements-constraints jobs  https://review.openstack.org/18951100:58
asselinjhesketh, tristanC thanks...glad we can all get on the same page. ;)00:58
*** otter768 has joined #openstack-infra00:58
asselintristanC, tomorrow's fine.00:59
*** shuoy has joined #openstack-infra01:01
*** shuoy has quit IRC01:01
*** dimtruck is now known as zz_dimtruck01:02
jlvillalIs there a good reason that 'infra-core' includes 'infra-core'?  https://review.openstack.org/#/admin/groups/306,members01:03
*** otter768 has quit IRC01:03
fungijlvillal: we love circular dependencies?01:04
fungijlvillal: i blame an overzealous script which added infra-core to the core groups of all infra repos in gerrit01:04
jlvillalfungi: Yeah....  /me not so fond of it :)01:05
jlvillalfungi: Thanks for the possible why.01:05
fungijeblair: ^ what i assume was unanticipated fallout from your group update script. i'll fix that group no01:05
ianwjeblair: thanks, somewhat disappointingly the job passed -> http://logs.openstack.org/89/179689/30/check/check-tempest-dsvm-centos7/5521332//console.html01:06
*** sigmavirus24 is now known as sigmavirus24_awa01:06
fungijlvillal: fixed now01:06
jeblairfungi: perhaps?  maybe the earlier version; i think the current version shouldn't do that.01:06
jlvillalfungi: Thanks.01:06
*** dguerri is now known as dguerri`01:07
*** patrickeast has quit IRC01:07
fungijlvillal: i take it you're running something to map out the group interrelationships on review.openstack.org01:07
ianwjeblair: possibly https://review.openstack.org/#/c/188664/ (set swapines in sysctl.conf, so it doesn't get overridden if we do sysctl -p later on) fixed it01:08
jlvillalfungi: Good guess! :)01:08
fungijlvillal: fwiw the logic in the who-approves.py script in system-config didn't seem to get confused by it01:08
fungithough i think i was expressly considering the potential for circular relationships when i wrote that01:09
*** rguillebert has quit IRC01:09
jlvillalfungi: I added a check to my code now.  Thanks.01:09
*** shuoy has joined #openstack-infra01:10
*** tqtran_ has quit IRC01:11
fungijlvillal: though also note that if you just want to find out direct and indirect group members, the gerrit rest api's groups members method has a recursive option too01:11
jlvillalfungi: Ah, cool.01:12
fungiwhich is much easier than trying to track down group inclusions01:12
* jlvillal thinks things will be easier...01:12
*** shuoy has quit IRC01:13
*** NobodyCam has quit IRC01:13
*** NobodyCam has joined #openstack-infra01:13
*** shuoy has joined #openstack-infra01:13
*** sarob has quit IRC01:14
*** kutija has joined #openstack-infra01:16
*** hdd has quit IRC01:18
*** Sukhdev has quit IRC01:19
*** shuoy has quit IRC01:20
*** sigmavirus24_awa is now known as sigmavirus2401:20
*** asselin_ has joined #openstack-infra01:25
*** rkukura has quit IRC01:26
*** mriedem_away has quit IRC01:27
*** jerryz has quit IRC01:29
*** jamielennox|away is now known as jamielennox01:32
*** rkukura has joined #openstack-infra01:36
*** ashleighfarnham has quit IRC01:40
*** spzala has quit IRC01:42
*** dims_ has quit IRC01:43
*** yamamoto has joined #openstack-infra01:45
*** tiswanso has joined #openstack-infra01:48
*** yamamoto has quit IRC01:50
*** tiswanso has quit IRC01:51
*** tiswanso has joined #openstack-infra01:51
*** sdake has joined #openstack-infra01:58
StevenKpleia2: The languages thing camunoz I talked about most at the summit was exposing them via the REST API. I'm not certain if there is a bug for that.01:59
*** unicell has quit IRC02:00
*** baoli has quit IRC02:00
pleia2StevenK: hm, didn't you talk about language deletion too?02:00
*** sarob has joined #openstack-infra02:00
*** baoli has joined #openstack-infra02:00
StevenKpleia2: I mentioned it -- let me get my thoughts in order and I'll push myself to create another bugzilla account02:01
pleia2StevenK: thank you :)02:01
*** sdake_ has quit IRC02:01
pleia2they do seem to want to put more work in their api, but I don't think there's a specific bug for making languages available via the api, just https://bugzilla.redhat.com/show_bug.cgi?id=1190027 (which covers a fair number of things)02:03
openstackbugzilla.redhat.com bug 1190027 in Component-API "RFE: Project Version contributor API" [High,New] - Assigned to camunoz02:03
pleia2I should follow up with daisy to see what her stackalytics-related ones were too02:03
StevenKpleia2: camunoz had scribbled notes about exporting languages via the API, at least02:04
lifelesshow are folk feeling about a pbr release?02:04
pleia2StevenK: nods02:05
*** hdd has joined #openstack-infra02:05
*** Sukhdev has joined #openstack-infra02:06
*** hdd has quit IRC02:07
*** sarob has quit IRC02:11
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632602:14
*** Hunner has quit IRC02:16
*** taron has quit IRC02:16
*** bradjones has quit IRC02:17
*** bradjones has joined #openstack-infra02:18
*** bradjones has quit IRC02:18
*** bradjones has joined #openstack-infra02:18
*** yamada-h has joined #openstack-infra02:18
openstackgerritDi Xu proposed openstack-infra/jenkins-job-builder: Support deep merge of defaults  https://review.openstack.org/15073002:19
ianwfungi / jeblair: devstack-centos7-rax-iad-2932979  just failed ... any chance of seeing anything on the console?02:19
*** Sukhdev has quit IRC02:19
*** Hunner has joined #openstack-infra02:20
*** yamahata has quit IRC02:21
*** amotoki has joined #openstack-infra02:24
*** taron has joined #openstack-infra02:29
*** amotoki has quit IRC02:31
*** koolhead17 has joined #openstack-infra02:33
*** davideagnello has quit IRC02:35
*** rkukura_ has joined #openstack-infra02:38
*** amotoki has joined #openstack-infra02:38
*** unicell has joined #openstack-infra02:38
*** amotoki_ has joined #openstack-infra02:40
*** rkukura has quit IRC02:40
*** rkukura_ is now known as rkukura02:40
*** amotoki_ has quit IRC02:40
*** davideagnello has joined #openstack-infra02:41
*** unicell1 has joined #openstack-infra02:42
*** unicell has quit IRC02:43
greghaynesooo https://mail.python.org/pipermail/cryptography-dev/2015-June/000468.html02:43
*** yamada-h has quit IRC02:43
lifelessthere's a bug open in pip related to that02:43
*** dims has joined #openstack-infra02:43
lifelessslow sdist timings02:43
lifelessdims: I didn't get an answer from my ping to infra about releasing pbr02:44
lifelessdims: so I think you'll need to (or I can my tomorrow am)02:44
*** yamamoto has joined #openstack-infra02:45
*** dguerri` is now known as dguerri02:46
*** dims has quit IRC02:50
*** alivigni has joined #openstack-infra02:51
*** praveenkumar_ is now known as praveenkumar02:54
*** dguerri is now known as dguerri`02:56
*** rossella_s has quit IRC02:58
*** rossella_s has joined #openstack-infra02:58
*** markvoelker has joined #openstack-infra02:59
*** otter768 has joined #openstack-infra02:59
*** otter768 has quit IRC03:04
*** Sukhdev has joined #openstack-infra03:04
*** markvoelker has quit IRC03:04
mordredlifeless: you make a ping?03:05
*** Sukhdev_ has joined #openstack-infra03:05
lifelessI did03:05
lifeless14:04 < lifeless> how are folk feeling about a pbr release?03:05
mordredlifeless: I feel great about it03:05
mordredlifeless: is it going to break things?03:06
*** ameade has quit IRC03:06
*** briancurtin has quit IRC03:06
*** zhiyan has quit IRC03:06
*** simonmcc has quit IRC03:06
*** jraim has quit IRC03:06
*** nzeer has quit IRC03:06
*** thingee has quit IRC03:06
*** ctracey has quit IRC03:06
lifelessno more chance than usual, and less than often03:06
*** j^2 has quit IRC03:07
*** kun_huang has quit IRC03:07
*** Ng has quit IRC03:07
*** rackertom has quit IRC03:07
lifelessautodoc exclusions03:08
lifelesssphonx 1.303:08
lifelesstesting stuff03:08
*** jeblair has quit IRC03:08
*** Sukhdev has quit IRC03:09
mordredsounds great03:09
mordredgo for it03:09
dougwiglifeless: it's always a good time to crack open a pbr.03:09
mordreddougwig: ++03:10
mordreddougwig: or, in my case, a caipirinha03:11
* mordred should start naming software after brazilian alcohol03:11
*** sbos99 has joined #openstack-infra03:12
*** yamada-h has joined #openstack-infra03:14
*** yamada-h has quit IRC03:20
*** tonytan4ever has quit IRC03:21
*** harlowja has quit IRC03:22
*** harlowja_ has joined #openstack-infra03:22
* purp can see it now ...03:22
purpgit clone .../caphirenia.git ... git clone .../caipirhnina.git ... @#%&! git clone .../jackandcoke.git03:23
*** gordc has joined #openstack-infra03:24
*** tiswanso has quit IRC03:27
*** yamamoto has quit IRC03:28
*** sdake has quit IRC03:31
*** sdake has joined #openstack-infra03:32
*** sdake has quit IRC03:34
openstackgerritMerged openstack-infra/shade: Return new secgroup object  https://review.openstack.org/18941403:34
*** yamamoto has joined #openstack-infra03:35
*** ivar-laz_ has joined #openstack-infra03:35
*** Sukhdev_ has quit IRC03:36
*** shuoy has joined #openstack-infra03:38
*** ivar-lazzaro has quit IRC03:39
*** ivar-laz_ has quit IRC03:40
*** sigmavirus24 is now known as sigmavirus24_awa03:40
*** shuoy has quit IRC03:42
lifelesswtf is the ubuntu package python-pip-whl03:43
lifelessoh wow, just wow.03:44
*** rkukura_ has joined #openstack-infra03:45
*** rkukura has quit IRC03:46
*** rkukura_ is now known as rkukura03:46
*** patrickeast has joined #openstack-infra03:46
*** oomichi has joined #openstack-infra03:49
*** puranamr has joined #openstack-infra03:49
*** zz_dimtruck is now known as dimtruck03:51
mordredlifeless: tell me things03:57
mordredpurp: ++03:57
lifelessyou'll go blind03:57
mordredlifeless: I may already be03:58
lifelessin utopic, for instance03:58
lifelessit has the pip 1.5.6 wheel that is normally vendored within virtualenv03:58
mordredlifeless: I really want to stop asking questions don't I?03:58
lifelessdpkg -L python-pip-whl03:58
lifelessmordred: only if you value your sanity03:59
lifelessmordred: when I go 'wow, just wow' that might be a time to not ask.03:59
mordredmordred@camelot:~$ dpkg -L python-pip-whl03:59
mordreddpkg-query: package 'python-pip-whl' is not installed03:59
mordredUse dpkg --info (= dpkg-deb --info) to examine archive files,03:59
mordredand dpkg --contents (= dpkg-deb --contents) to list their contents.03:59
lifelessrighto, thats devstack using constraints files. Polish and push, then next.03:59
mordredlifeless: have I mentioned that I don't believe in installing pip from distros because BATSHIT?04:00
lifelessmordred: oh btw04:00
*** coolsvap|afk is now known as coolsvap04:00
lifelessmordred: thursday C is having tonsils out04:00
mordredlifeless: oh my!04:00
lifelessmordred: I will be fragmented due to hospital stuff04:00
mordredlifeless: well, she will not enjoy that04:00
lifelessmordred: better than the obstructive sleep apnea she currently does not enjoy04:00
lifelesshers aren't very pitted, but they are HUGE04:01
*** dimtruck is now known as zz_dimtruck04:01
lifelessOMG how can she swallow food huge04:01
*** dboik has joined #openstack-infra04:02
*** shuoy has joined #openstack-infra04:03
*** yamada-h has joined #openstack-infra04:05
*** baoli has quit IRC04:05
*** dboik has quit IRC04:07
*** sputnik13 has joined #openstack-infra04:14
*** sputnik13 has quit IRC04:16
*** koolhead_ has joined #openstack-infra04:18
*** koolhead17 has quit IRC04:19
*** yamamoto has quit IRC04:19
*** morgan has joined #openstack-infra04:21
*** yfried has quit IRC04:27
*** dpaterson has quit IRC04:31
*** sputnik13 has joined #openstack-infra04:34
*** dguerri` is now known as dguerri04:35
*** aswadr has joined #openstack-infra04:40
*** morgan has quit IRC04:40
*** ayoung has quit IRC04:41
*** puranamr has quit IRC04:43
*** dguerri is now known as dguerri`04:45
*** maishsk_afk has quit IRC04:46
*** dims has joined #openstack-infra04:46
*** rkukura_ has joined #openstack-infra04:47
gordclifeless, i've had tonsils taken out before. all the ice cream/popsicles you want... it's pretty sweet.04:48
*** markvoelker has joined #openstack-infra04:48
*** morgan has joined #openstack-infra04:49
*** rkukura has quit IRC04:50
*** rkukura_ is now known as rkukura04:50
*** sdake has joined #openstack-infra04:51
*** dims has quit IRC04:51
*** sputnik13 has quit IRC04:51
*** sputnik13 has joined #openstack-infra04:52
*** sbos99 has quit IRC04:53
*** markvoelker has quit IRC04:53
*** sarob has joined #openstack-infra04:58
*** sasi has joined #openstack-infra04:59
*** otter768 has joined #openstack-infra05:00
*** yamamoto has joined #openstack-infra05:00
*** boris-42 has quit IRC05:04
*** otter768 has quit IRC05:05
*** yamahata has joined #openstack-infra05:05
*** shuoy has quit IRC05:07
*** juggler_ has joined #openstack-infra05:12
*** sabeen has joined #openstack-infra05:12
*** alexpilotti has quit IRC05:14
*** juggler has quit IRC05:15
*** marun has quit IRC05:17
*** fallenpegasus has quit IRC05:20
*** patrickeast has quit IRC05:22
*** sputnik13 has quit IRC05:22
lifelessgordc: both her parents have too; but its changed from our days.05:23
lifelessgordc: now they want you to eat rough edged foods, apparently knocks back rough bits forming around the suture05:23
gordclifeless, that sounds terrible... doctors are sadists05:26
*** sasi1 has joined #openstack-infra05:26
*** yfried has joined #openstack-infra05:27
*** achanda has joined #openstack-infra05:27
*** shuoy has joined #openstack-infra05:27
*** sasi has quit IRC05:29
*** e0ne has joined #openstack-infra05:29
*** abregman has joined #openstack-infra05:30
*** sarob has quit IRC05:30
*** yamahata has quit IRC05:32
*** achanda has quit IRC05:36
*** bkero has quit IRC05:38
*** sputnik13 has joined #openstack-infra05:39
*** tqtran has joined #openstack-infra05:39
*** bkero has joined #openstack-infra05:39
*** armax has quit IRC05:45
*** sputnik13 has quit IRC05:45
*** e0ne has quit IRC05:47
*** yamada-h has quit IRC05:48
*** mpopow has quit IRC05:50
*** BharatK has joined #openstack-infra05:52
*** deepakcs has joined #openstack-infra05:53
*** k4n0 has joined #openstack-infra05:55
sasi1Hi, I am building a CI system and devstack is faling due to "sudo: no tty present and no askpass program specified" when run from jenkins. can any one encountered this error ? any thoughts05:59
*** rkukura_ has joined #openstack-infra06:00
*** jerryz has joined #openstack-infra06:01
*** rkukura has quit IRC06:02
*** rkukura_ is now known as rkukura06:02
*** sputnik13 has joined #openstack-infra06:02
openstackgerritMark McLoughlin proposed openstack-infra/project-config: OpenStack Foundation Transparency Policy project  https://review.openstack.org/17964806:02
sasi1 error log: http://paste.openstack.org/show/276617/06:03
*** devvesa has joined #openstack-infra06:06
*** rbradfor has quit IRC06:10
*** nzeer has joined #openstack-infra06:13
*** achanda has joined #openstack-infra06:16
*** sdake has quit IRC06:16
*** jraim has joined #openstack-infra06:17
*** ihrachyshka has joined #openstack-infra06:19
*** maishsk_afk has joined #openstack-infra06:20
*** shuoy has quit IRC06:20
*** rbradfor has joined #openstack-infra06:20
*** dguerri` is now known as dguerri06:24
openstackgerritLimor Stotland proposed openstack/requirements: Adding apscheduler to global requirement  https://review.openstack.org/18956406:25
openstackgerritSergey Kraynev proposed openstack-infra/project-config: Make heatclient-dsvm-functional check for heatcli  https://review.openstack.org/18828406:26
openstackgerritSergey Kraynev proposed openstack-infra/project-config: Make heatclient-dsvm-functional gating for heatcli  https://review.openstack.org/18920406:26
*** sbos99 has joined #openstack-infra06:27
*** j^2 has joined #openstack-infra06:27
*** simonmcc has joined #openstack-infra06:27
*** Nikkau has joined #openstack-infra06:27
openstackgerritMarc Koderer proposed openstack-infra/project-config: Make Manila scenario testing voting  https://review.openstack.org/18923006:29
*** ctracey has joined #openstack-infra06:30
*** scheuran has joined #openstack-infra06:30
*** dougwig has quit IRC06:33
*** electrical has quit IRC06:33
*** dguerri is now known as dguerri`06:34
*** mgkwill has quit IRC06:34
*** edwarnicke has quit IRC06:34
*** sweston has quit IRC06:34
*** achanda has quit IRC06:36
*** gsagie has joined #openstack-infra06:38
gsagieHello, if anyone can please review the following patch: https://review.openstack.org/#/c/187042/   , we want to be able to create a stable/kilo branch for Dragonflow and its currently blocked on this06:39
gsagieand we cant continue working on the main branch without potentially breaking that.. :)06:40
AJaegergsagie: already reviewed - please ping at US time again for other reviewers to see it.06:40
gsagieThanks! will do06:40
*** ruagair has quit IRC06:40
*** ameade has joined #openstack-infra06:41
*** ruagair has joined #openstack-infra06:44
*** sasi1 has left #openstack-infra06:44
*** dtantsur|afk is now known as dtantsur06:45
*** ruagair has quit IRC06:45
*** ruagair has joined #openstack-infra06:45
*** yamada-h has joined #openstack-infra06:45
*** zhiyan has joined #openstack-infra06:46
*** Ng has joined #openstack-infra06:49
*** tchaypo has quit IRC06:53
*** woodster_ has quit IRC06:53
*** DuncanT has quit IRC06:53
*** ramishra has quit IRC06:53
*** tchaypo has joined #openstack-infra06:55
*** tchaypo is now known as Guest4845806:55
*** Guest48458 is now known as tchorizo06:56
*** sbos99 has quit IRC06:57
*** samuelBartel has joined #openstack-infra06:57
*** rossella_s has quit IRC06:57
*** rossella_s has joined #openstack-infra06:58
*** mrunge has joined #openstack-infra06:59
*** tchorizo is now known as tchaypo07:00
*** otter768 has joined #openstack-infra07:01
*** akscram has quit IRC07:03
*** sbos99 has joined #openstack-infra07:04
StevenKAJaeger: Thanks for the review. The zanata.ini file contains the URL, username and API key to talk to the server.07:04
*** rkukura_ has joined #openstack-infra07:05
AJaegerStevenK: Ah... So, we need to get these onto the server with some other patch. Not sure how it's done for transifex07:05
StevenKAJaeger: Yup, I'll be looking at that tomorrow07:06
StevenKNot sure how to get a jenkins user or something on zanata07:06
*** otter768 has quit IRC07:06
*** rkukura has quit IRC07:07
*** rkukura_ is now known as rkukura07:07
*** akscram has joined #openstack-infra07:09
*** briancurtin has joined #openstack-infra07:09
*** dguerri` is now known as dguerri07:10
*** sputnik13 has quit IRC07:10
*** jyuso1 has quit IRC07:12
*** arxcruz has joined #openstack-infra07:14
*** tqtran has quit IRC07:15
*** rackertom has joined #openstack-infra07:16
*** HeOS has quit IRC07:16
*** ihrachyshka has quit IRC07:20
*** Guest75296 has quit IRC07:21
*** electrical has joined #openstack-infra07:23
*** jtomasek has quit IRC07:23
*** thingee has joined #openstack-infra07:24
*** Hal has joined #openstack-infra07:25
*** Hal is now known as Guest9072707:25
*** yfried is now known as yfried|afk07:26
*** edwarnicke has joined #openstack-infra07:28
*** yfried|afk is now known as yfried07:28
*** achanda has joined #openstack-infra07:28
*** kun_huang has joined #openstack-infra07:28
*** sdake has joined #openstack-infra07:29
*** dabos has joined #openstack-infra07:32
*** yamamoto has quit IRC07:33
*** woodster_ has joined #openstack-infra07:34
*** dougwig has joined #openstack-infra07:34
*** sbos99 has quit IRC07:36
*** dguerri is now known as dguerri`07:36
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820707:36
*** sweston has joined #openstack-infra07:37
*** jlanoux has joined #openstack-infra07:37
*** markvoelker has joined #openstack-infra07:37
*** sdake_ has joined #openstack-infra07:39
*** Ala has joined #openstack-infra07:40
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820707:40
*** sdake has quit IRC07:43
*** markvoelker has quit IRC07:43
*** sdake has joined #openstack-infra07:44
*** tchaypo_ has joined #openstack-infra07:44
*** ramishra has joined #openstack-infra07:45
*** DuncanT has joined #openstack-infra07:45
*** markus_z has joined #openstack-infra07:46
*** sdake_ has quit IRC07:47
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820707:52
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820707:52
*** mpaolino has joined #openstack-infra07:52
*** jyuso1 has joined #openstack-infra07:53
*** mpaolino has quit IRC07:53
*** cdent has joined #openstack-infra07:54
*** sabeen has quit IRC07:55
flip214AJaeger: good morning!07:57
flip214please give me one more hint.07:57
AJaegermorning, flip214 !07:57
AJaegerflip214: please give me a hint on what ;)07:57
flip214https://review.openstack.org/#/c/187568/ wants to test https://review.openstack.org/#/c/178573/ which isn't merged yet; and I can't merge the devstack plugin as long as its test fails.07:58
*** tchaypo has quit IRC07:58
*** tchaypo_ is now known as tchaypo07:58
*** tchaypo1 has joined #openstack-infra07:58
AJaegerflip214: there's no way to do what you want.07:59
AJaeger573 needs to merge first07:59
*** tchaypo1 has quit IRC07:59
AJaegerTry the following:07:59
AJaegerAdd to https://review.openstack.org/#/c/187568/08:00
*** yamamoto has joined #openstack-infra08:00
AJaegerDepends-On: Icb7347d55dc5b22720269405b6d2b4053c2ad81908:00
AJaegerin the commit message08:00
flip214Okay, thanks.08:00
*** ZZelle has quit IRC08:01
AJaegerThen the two patches will be moved together through the gate.08:01
*** ZZelle has joined #openstack-infra08:01
*** dguerri` is now known as dguerri08:02
flip214Newline before "Change-Id", or both in same paragraph?08:02
AJaegerNot sure whether this works in this case but if 568 needs a cinder tree with 573 included, then this will work08:02
AJaegerflipline: Just on a line of its own08:02
*** mgkwill has joined #openstack-infra08:02
AJaegerflipline? Sorry08:02
flip214Never mind ;)08:02
*** marun has joined #openstack-infra08:02
flip214thanks, uploaded. I'll wait for the result!08:03
*** dizquierdo has joined #openstack-infra08:03
AJaegerflip214: tell me whether it worked ;)08:03
*** jklare has quit IRC08:05
flip214I will, you can rest assured of that.08:06
*** Guest90727 has quit IRC08:06
*** rkukura_ has joined #openstack-infra08:07
*** marun has quit IRC08:08
*** rkukura has quit IRC08:09
*** rkukura_ is now known as rkukura08:09
*** jerryz has quit IRC08:09
*** ihrachyshka has joined #openstack-infra08:10
*** yamamoto has quit IRC08:11
*** jklare has joined #openstack-infra08:11
*** fhubik has joined #openstack-infra08:12
*** achanda has quit IRC08:13
*** Hal has joined #openstack-infra08:14
*** ParsectiX has joined #openstack-infra08:14
*** Hal is now known as Guest8910208:14
openstackgerritJens Rosenboom proposed openstack-infra/system-config: Make puppet work with Ubuntu Vivid  https://review.openstack.org/18914208:15
*** Somay has joined #openstack-infra08:16
*** ihrachyshka_ has joined #openstack-infra08:18
*** sdake_ has joined #openstack-infra08:22
*** gordc has quit IRC08:22
*** jklare has quit IRC08:23
*** jklare has joined #openstack-infra08:24
*** sdake has quit IRC08:25
openstackgerritSomay Jain proposed openstack-infra/jenkins-job-builder: Support for more conditions type in conditions-buildstep plugin  https://review.openstack.org/18959008:25
*** freerunner has joined #openstack-infra08:25
openstackgerritSpencer Krum proposed openstack-infra/puppet-dashboard: Deprecate Module  https://review.openstack.org/18959108:26
*** mfmcdonagh has joined #openstack-infra08:27
*** mcdonagm has joined #openstack-infra08:27
*** yamamoto has joined #openstack-infra08:27
*** jyuso1 has quit IRC08:29
*** jyuso1 has joined #openstack-infra08:29
*** freerunner has quit IRC08:30
*** freerunner has joined #openstack-infra08:31
*** jyuso1 has quit IRC08:32
*** jyuso has joined #openstack-infra08:32
*** koolhead_ has quit IRC08:32
*** sdake_ has quit IRC08:34
*** Ala has quit IRC08:35
*** HeOS has joined #openstack-infra08:35
*** sdake has joined #openstack-infra08:36
*** achanda has joined #openstack-infra08:37
*** Ala has joined #openstack-infra08:37
*** sasi1 has joined #openstack-infra08:37
*** sdake_ has joined #openstack-infra08:39
*** pelix has joined #openstack-infra08:41
*** sdake_ has quit IRC08:41
*** leopoldj has joined #openstack-infra08:42
*** sdake has quit IRC08:42
*** jyuso has quit IRC08:43
*** jyuso1 has joined #openstack-infra08:43
*** isviridov_away is now known as isviridov08:44
openstackgerritLimor Stotland proposed openstack/requirements: Adding 'Advanced python scheduler' (apscheduler) to global requirement  https://review.openstack.org/18956408:45
*** e0ne has joined #openstack-infra08:46
openstackgerritDerek Higgins proposed openstack-infra/project-config: [WIP] Import rdo packaging into openstack namespace  https://review.openstack.org/18949708:47
*** sasi1 has left #openstack-infra08:47
*** derekh has joined #openstack-infra08:50
*** chlong has quit IRC08:52
*** soren has joined #openstack-infra08:57
*** maishsk_afk has quit IRC08:57
*** soren is now known as soren_08:59
*** soren_ is now known as soren08:59
*** jyuso1 has quit IRC09:01
*** otter768 has joined #openstack-infra09:02
*** maishsk_afk has joined #openstack-infra09:03
flip214AJaeger: https://review.openstack.org/#/c/187568/ check is still not being run...09:03
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel VMware DVS plugin  https://review.openstack.org/18959909:04
*** gsagie has left #openstack-infra09:05
*** amotoki_ has joined #openstack-infra09:05
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel VMware DVS plugin  https://review.openstack.org/18959909:06
*** otter768 has quit IRC09:06
*** freerunner has quit IRC09:07
*** fhubik is now known as fhubik_afk09:08
*** amotoki_ has quit IRC09:09
*** freerunner has joined #openstack-infra09:13
*** rkukura has quit IRC09:13
*** rkukura has joined #openstack-infra09:13
*** amotoki_ has joined #openstack-infra09:14
AJaegerflip214: it is running, see http://status.openstack.org/zuul09:15
*** koolhead17 has joined #openstack-infra09:16
AJaegerflip214: finished but seems the depends-on  did not help in setting up the repos properly09:19
*** afazekas has joined #openstack-infra09:19
AJaegerflip214: So, let's get the other job merged first, you have a chicken-and-egg problem09:19
AJaegerflip214: Run the test locally and post the results09:19
openstackgerritSomay Jain proposed openstack-infra/jenkins-job-builder: Support for more conditions type in conditions-buildstep plugin  https://review.openstack.org/18959009:20
openstackgerritDerek Higgins proposed openstack-infra/project-config: [WIP] Import rdo packaging into openstack namespace  https://review.openstack.org/18949709:22
*** Somay has quit IRC09:25
*** Hal has joined #openstack-infra09:26
*** Guest89102 has quit IRC09:26
*** freerunner has quit IRC09:26
*** Hal is now known as Guest9118009:27
*** markvoelker has joined #openstack-infra09:27
*** freerunner has joined #openstack-infra09:28
*** jyuso1 has joined #openstack-infra09:28
*** nicodemos has quit IRC09:28
*** ericksonsantos has quit IRC09:28
*** markvoelker has quit IRC09:32
*** freerunner has quit IRC09:33
*** dabos has quit IRC09:34
*** yamamoto has quit IRC09:34
*** freerunner has joined #openstack-infra09:34
*** woodster_ has quit IRC09:41
*** yamamoto has joined #openstack-infra09:41
*** fhubik_afk is now known as fhubik09:45
*** lxsli_away is now known as lxsli09:47
*** yfried is now known as yfried|afk09:48
*** dims has joined #openstack-infra09:48
*** haypo has joined #openstack-infra09:50
*** jtomasek has joined #openstack-infra09:50
haypohi. FYI all swift tests are failing, but there is a fix: https://review.openstack.org/#/c/189472/ -- can someone please take a look?09:50
haypothe change adds libjerasure-dev dependency09:51
openstackgerritDavide Guerri proposed openstack-infra/shade: Add secgroup update API  https://review.openstack.org/18949009:52
openstackgerritDavide Guerri proposed openstack-infra/shade: poc changeset, please disregard  https://review.openstack.org/18962109:52
*** redixin has joined #openstack-infra09:53
*** dims has quit IRC09:54
*** achanda has quit IRC09:56
*** yamada-h has quit IRC09:56
*** koolhead17 has quit IRC09:57
*** achanda has joined #openstack-infra09:58
*** freerunner has quit IRC09:59
flip214AJaeger: hmmm, locally I get failed tests - even before anything touches my driver...10:00
flip214the cinder-only tests run okay, though.10:00
AJaegerflip214: what fun ;(10:00
*** e0ne is now known as e0ne_10:02
flip214AJaeger: yeah, right.10:02
*** yfried|afk is now known as yfried10:03
*** freerunner has joined #openstack-infra10:03
*** isviridov is now known as isviridov_away10:03
openstackgerritMerged openstack-infra/project-config: Make Manila scenario testing voting  https://review.openstack.org/18923010:03
*** achanda has quit IRC10:03
electricalAJaeger: perhaps you know. With JJB, is there a way to set the jenkins login and host as env variables? i couldn't find anything about it.10:07
*** fhubik is now known as fhubik_afk10:08
*** dims has joined #openstack-infra10:09
*** freerunner has quit IRC10:10
*** freerunner has joined #openstack-infra10:11
pelixelectrical: no support for that, though it would be useful. Since there is a refactoring planned around the API which includes checking how we handle config, it would be something that could be looked at10:11
electricalpelix: ah cool thanks.10:11
electricalWe want to save our JJB stuff in a public repo but not the config file with the jenkins login details :-)10:11
*** derekh has quit IRC10:11
electricalbut we still would need to manage it since we have a sub dir per project.10:11
openstackgerritMerged openstack-infra/project-config: Make heatclient-dsvm-functional check for heatcli  https://review.openstack.org/18828410:12
*** e0ne_ has quit IRC10:12
*** achanda has joined #openstack-infra10:12
*** ams__ has joined #openstack-infra10:13
pelixwhy not create a job inside your jenkins instance that can write out the credentials to a config file?10:14
*** yfried is now known as yfried|afk10:14
*** alexpilotti has joined #openstack-infra10:14
pelixsince jenkins is aware of it's own credentials, it can retrieve them itself10:14
electricalhuh? how do you mean? ( not enough tea yet :p )10:15
pelixYou can create a job in jenkins to execute groovy code in a system context which in turn is able to use the Jenkins API internally to extract the necessary credentials. It's what we use here for a locked down system.10:16
electricalhmm, interesting. and do you manage the paths in there as well? since for every project we need to add 2 paths ( one for the job templates and one for the scripts )10:17
pelixWe stand up a Jenkins using chef with 2 jobs prewritten out to disk to be loaded automatically. One of them populates the ini file with the needed auth conditions and the other is the upload job that uses it to upload from using the jenkins master.10:17
flip214AJaeger: is there no way to get https://review.openstack.org/#/c/187568/ merged, to break the loop?10:18
*** freerunner has quit IRC10:18
flip214can I override the "Verified" field in some way?10:19
pelixone we have doesn't need to be that complex, but that's just a case of taking the basic idea and expanding on it10:19
*** Ala has quit IRC10:19
redixinhi all. is it possible to run gate-project-config-layout locally? I run 'tox -v -e zuul' but it works (not works on gates)10:19
electricalpelix: are you able to share that groovy script?10:19
*** e0ne has joined #openstack-infra10:21
*** rkukura_ has joined #openstack-infra10:22
ams__I'm trying to have a variable number of builders in my JJB job-template, but I can't see a way to do this. Is there any way to achieve this? I basically want to give builders a list10:23
*** teran has quit IRC10:23
*** derekh has joined #openstack-infra10:24
electricalams__: can you elaborate a bit more?10:24
*** rkukura has quit IRC10:25
*** rkukura_ is now known as rkukura10:25
electricalyou can have multiple type of builders in the builders key in a job-template ( i only have use for 1 atm10:25
*** teran has joined #openstack-infra10:25
ams__I have a job-template which defines 1 builder, but I would like the ability to have give this job template a list (or a dict) of parameters and give each project the ability to vary the number of builders10:25
electricalah okay. so you have 1 default builder and you want to add more depending on what type of project it is ?10:26
ams__my use case is a job that builds some python code and then builds a docker image from it, i always do one build of the python code but sometimes there are multiple docker images (all in the same single job).10:26
pelixelectrical: looking10:26
electricalpelix: thanks :-)10:26
*** redixin has quit IRC10:27
electricalams__: how compare the docker image builder tasks to each other? are there many things the same?10:28
*** BobBall_AWOL is now known as BobBall10:28
*** markvoelker has joined #openstack-infra10:28
pelixams__: yes it's possible but a little fiddly, you end up needing to define a dedicated macro to be passed in and be expanded with the list of jobs, one sec and I'll try to knock together a gist10:29
electricalthat's what i thought about as well10:29
*** redixin has joined #openstack-infra10:29
ams__pelix: great, that would be really helpful, thankyou10:30
*** jlanoux_ has joined #openstack-infra10:30
*** teran has quit IRC10:30
*** ociuhandu has joined #openstack-infra10:31
*** fhubik_afk is now known as fhubik10:32
*** jlanoux has quit IRC10:33
*** markvoelker has quit IRC10:33
AJaegerflip214: you would need to disable the tests to have it merged. And it won't help - it does not report on cinder, or does it?10:34
*** yfried|afk is now known as yfried10:34
AJaegerredixin: which change is failing?10:34
redixinAJaeger, https://review.openstack.org/#/c/187997/10:35
pelixams__: something like https://gist.github.com/electrofelix/ab7a1da9168eebdadf28 should work, I have something similar locally, so I've just transcribed it to make it more generic10:36
AJaegerelectrical: I can't help you, sorry.10:38
electricalAJaeger: no worries bud :-)10:38
*** yamada-h has joined #openstack-infra10:38
pelixams__: we actually have a bug in template expansion really in that for lists passed through as variables we attempt to expand into multiple jobs even when the variable could not affect the name of the job. Really should be passed through directly in such cases10:39
*** rkukura has quit IRC10:40
flip214AJaeger: I've got a 590kB text file (compressed 68K) showing that the cinder tests all run.10:41
flip214but at this size I can't paste anymore (pastebin, paste.debian.org, paste.lisp.org).10:41
flip214shall I email it?10:41
openstackgerritMerged openstack-infra/system-config: Add libuuid devel packages  https://review.openstack.org/18928710:41
AJaegerflip214: please talk with the cinder team about that...10:41
*** yamada-h has quit IRC10:41
AJaegerredixin: this is strange indeed, it works locally for me as well...10:42
*** ParsectiX has quit IRC10:42
openstackgerritMerged openstack-infra/project-config: Disable py26 jos for glance_store  https://review.openstack.org/18601110:43
AJaegerredixin: rebase your change to master and run again ;)10:43
*** yamada-h has joined #openstack-infra10:43
AJaegerThere was a recent change to add gate-rally-dsvm-rally-{service} and that one is not patched by your change10:43
openstackgerritSergey Skripnick proposed openstack-infra/project-config: Add Ironic services to Rally experimental pipeline  https://review.openstack.org/18799710:44
*** samueldmq has joined #openstack-infra10:45
*** redixin has quit IRC10:46
openstackgerritAndreas Scheuring proposed openstack-infra/project-config: Add networking-macvtap to StackForge  https://review.openstack.org/18964410:46
*** redixin has joined #openstack-infra10:47
openstackgerritDmitry Tantsur proposed openstack-infra/project-config: Project rename stackforge/ironic-discoverd -> openstack/ironic-inspector  https://review.openstack.org/18803010:48
*** soren has quit IRC10:48
ams__pelix: thanks! taking a look now10:48
*** jlanoux has joined #openstack-infra10:50
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Fall back to git clone if zuul-cloner not available  https://review.openstack.org/18964710:51
*** ildikov_ has joined #openstack-infra10:51
*** amotoki_ has quit IRC10:53
*** boris-42 has joined #openstack-infra10:53
*** jlanoux_ has quit IRC10:54
ams__pelix: So I think that's close to what I want, but I want to give custom-builders a list of builders (actually, for my case all the same builder) but with different parameters for each. Is that feasible do you think?10:55
pelixjust add additional steps to the macro10:56
pelixyou can have as many builders as you want, I've updated the gist to make that clear10:58
*** claudiub has joined #openstack-infra10:58
*** dabos has joined #openstack-infra10:58
openstackgerritSean Dague proposed openstack-infra/devstack-gate: create GRENADE_PLUGINRC to pass through grenade external plugins  https://review.openstack.org/18965210:59
openstackgerritSean Dague proposed openstack-infra/devstack-gate: Add work around to enable heat grenade config  https://review.openstack.org/18965310:59
*** teran has joined #openstack-infra11:00
*** ParsectiX has joined #openstack-infra11:01
*** soren has joined #openstack-infra11:03
*** otter768 has joined #openstack-infra11:03
*** marcusvrn has joined #openstack-infra11:04
*** maishsk_afk has quit IRC11:04
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832511:04
*** yamamoto has quit IRC11:05
*** marcusvrn has quit IRC11:05
*** hdd has joined #openstack-infra11:06
*** yamamoto has joined #openstack-infra11:06
*** freerunner has joined #openstack-infra11:06
*** teran has quit IRC11:06
*** maishsk_afk has joined #openstack-infra11:07
*** otter768 has quit IRC11:08
*** marcusvrn has joined #openstack-infra11:08
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832511:08
*** teran has joined #openstack-infra11:09
*** spredzy is now known as spredzy|afk11:09
*** pc_m has joined #openstack-infra11:11
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832511:12
*** coolsvap is now known as coolsvap|afk11:12
*** che-arne has joined #openstack-infra11:15
*** ociuhandu has quit IRC11:15
ams__pelix: I see, but my problem is I want to vary them dynamically, so a project may specify custom-builders: a b c and I'll have 3 builders, but another may specify customer-builders: a and only have 1 builder11:15
ams__I suppose (as the number doesn't vary that much) I could just have builder-docker-1, builder-docker-2, builder-docker-311:16
*** ociuhandu has joined #openstack-infra11:16
pelixyou can set the value for 'custom-builders' to point to a different macro under the job part as well11:18
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Support deep merge of defaults  https://review.openstack.org/15073011:19
pelixyolanda: did you grok my explanation for https://review.openstack.org/183939 ? essentially there was a bug around how file objects would be handled by JJB so in wrapping stdin/stdout for codec support I had to fix that as well to be able to create a unit test.11:23
yolandahi pelix, i didn't check recently, sorry11:23
*** coolsvap|afk is now known as coolsvap11:23
TheJuliajamielennox: So morded mentioned a few days ago the thought of creating a proper noauth plugin.  Seems like it would be a good idea11:23
yolandalet me take a look11:23
ams__pelix: ok, so my best option is to have (for my case) 3 build macros, x-build1, x-build2, x-build3 (defining 1 builder, 2 builders and 3 builders respectively) and then pass those depending on my template11:24
*** wznoinsk_ has joined #openstack-infra11:25
*** wznoinsk_ has quit IRC11:25
*** arxcruz has quit IRC11:25
*** achanda has quit IRC11:26
*** coolsvap is now known as coolsvap|afk11:27
*** markvoelker has joined #openstack-infra11:29
*** yfried is now known as yfried|afk11:29
dstaneka proposal bot failure from 6 months ago: https://review.openstack.org/#/c/128930/  can that just be abandoned?11:30
*** yamamoto has quit IRC11:30
*** fhubik is now known as fhubik_afk11:33
*** markvoelker has quit IRC11:34
*** soren has quit IRC11:34
*** e0ne is now known as e0ne_11:34
*** deepakcs has quit IRC11:35
*** e0ne_ has quit IRC11:40
*** arxcruz has joined #openstack-infra11:40
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: Run worldump if tempest fails  https://review.openstack.org/18873311:40
openstackgerritJoe Gordon proposed openstack-infra/devstack-gate: DNM: Making sure worlddump is works here  https://review.openstack.org/18901011:40
*** aswadr has quit IRC11:41
*** yfried|afk is now known as yfried11:42
*** aswadr has joined #openstack-infra11:43
*** yamamoto has joined #openstack-infra11:47
*** soren has joined #openstack-infra11:48
*** woodster_ has joined #openstack-infra11:48
*** soren has quit IRC11:48
*** soren has joined #openstack-infra11:48
*** BharatK has quit IRC11:53
*** koolhead17 has joined #openstack-infra11:56
*** freerunner has quit IRC11:57
*** weshay has joined #openstack-infra11:59
*** hdd has quit IRC11:59
sdaguedstanek: merging the 2014.2 tag back in is kind of important iirc12:00
*** jaypipes has joined #openstack-infra12:02
*** fhubik_afk is now known as fhubik12:02
*** Guest84699 has quit IRC12:03
*** sshnaidm has joined #openstack-infra12:04
openstackgerritMerged openstack-infra/jenkins-job-builder: Add support for Clang Scan-Build plugin version 1.5  https://review.openstack.org/18485412:04
openstackgerritDmitry Tantsur proposed openstack-infra/project-config: New project python-ironic-inspector-client  https://review.openstack.org/18879812:05
*** dprince has joined #openstack-infra12:05
*** achanda has joined #openstack-infra12:05
*** bdemers has quit IRC12:05
*** samueldmq has quit IRC12:08
*** Ala has joined #openstack-infra12:10
*** achanda has quit IRC12:13
*** yfried is now known as yfried|afk12:13
*** prad_ has joined #openstack-infra12:14
*** e0ne has joined #openstack-infra12:14
*** prad has quit IRC12:14
*** prad_ is now known as prad12:14
openstackgerritDoug Hellmann proposed openstack-infra/irc-meetings: add agenda_url for cross-project meeting  https://review.openstack.org/18967612:14
*** soren has quit IRC12:15
*** baoli has joined #openstack-infra12:15
*** zz_dimtruck is now known as dimtruck12:15
*** aysyanne has joined #openstack-infra12:16
openstackgerritSirushti Murugesan proposed openstack-infra/project-config: Add flag DEVSTACK_GATE_NEUTRON to heat grenade job  https://review.openstack.org/18772512:18
*** yfried|afk is now known as yfried12:19
*** dan_ has joined #openstack-infra12:19
*** dan_ is now known as Guest548412:19
*** e0ne is now known as e0ne_12:21
openstackgerritSean Dague proposed openstack-infra/devstack-gate: Add work around to enable heat grenade config  https://review.openstack.org/18965312:21
*** e0ne_ is now known as e0ne12:21
*** samueldmq has joined #openstack-infra12:22
electricalArgh. Hipchat plugin for jenkins 0.1.9 changed completely :-(12:22
*** freerunner has joined #openstack-infra12:22
electricalyolanda: pelix any idea how to handle this? they created as a publisher now and nothing in the properties anymore.12:22
electricalshould i create a publisher thing for it and do a version check if the user is running 0.1.9 ?12:23
openstackgerritSirushti Murugesan proposed openstack-infra/project-config: Add flag DEVSTACK_GATE_NEUTRON and enable heat grenade plugin  https://review.openstack.org/18772512:23
openstackgerritJoe Gordon proposed openstack-dev/hacking: Expand origin section of README  https://review.openstack.org/18525112:26
openstackgerritJoe Gordon proposed openstack-dev/hacking: Add example to versioning section of README  https://review.openstack.org/18525212:26
openstackgerritJoe Gordon proposed openstack-dev/hacking: Revise adding additional checks section of README  https://review.openstack.org/18525312:26
openstackgerritJoe Gordon proposed openstack-dev/hacking: Take advantage of RST formatting for links  https://review.openstack.org/18525712:26
*** achanda has joined #openstack-infra12:27
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632612:27
*** marun has joined #openstack-infra12:28
*** ociuhandu has quit IRC12:28
*** hdd has joined #openstack-infra12:28
*** soren has joined #openstack-infra12:28
*** e0ne is now known as e0ne_12:30
electricalAh. there is already a patch. https://review.openstack.org/#/c/174036/412:31
*** freerunner has quit IRC12:32
openstackgerritRichard Pijnenburg proposed openstack-infra/jenkins-job-builder: Adding hipchat publisher  https://review.openstack.org/17403612:34
*** mwagner_afk has quit IRC12:35
openstackgerritMerged openstack-dev/hacking: Expand origin section of README  https://review.openstack.org/18525112:35
*** e0ne_ has quit IRC12:35
openstackgerritMerged openstack-dev/hacking: Add example to versioning section of README  https://review.openstack.org/18525212:37
*** kgiusti has joined #openstack-infra12:37
openstackgerritMerged openstack-dev/hacking: Revise adding additional checks section of README  https://review.openstack.org/18525312:37
*** chlong has joined #openstack-infra12:37
openstackgerritMonty Taylor proposed openstack-infra/shade: Add very initial support for passing in occ object  https://review.openstack.org/18820112:38
*** rlandy has joined #openstack-infra12:38
*** doug-fish has joined #openstack-infra12:39
*** mpaolino has joined #openstack-infra12:40
*** unicell has joined #openstack-infra12:40
ams__pelix: sorry for all the questions, but is there any way to have my project pass a variable to may customer builder? So from your example, have "builder-bundler-1.5.2" contain "shell: "gem uninstall {uninstall}""12:41
*** sigmavirus24_awa is now known as sigmavirus2412:41
*** bknudson has quit IRC12:41
*** unicell1 has quit IRC12:41
openstackgerritRoman Vasilets proposed openstack-infra/project-config: Add Ironic services to Rally experimental pipeline  https://review.openstack.org/18799712:43
openstackgerritMerged openstack-dev/hacking: Take advantage of RST formatting for links  https://review.openstack.org/18525712:43
*** wznoinsk has quit IRC12:44
*** yamamoto has quit IRC12:48
*** yamamoto has joined #openstack-infra12:49
ams__ah, nm, I can have - '{customer-builders}': var1: '{var1}'12:50
*** koolhead17 has quit IRC12:51
*** peristeri has joined #openstack-infra12:51
fungidstanek: yeah, that's one of those changes that gets proposed just after we tag each new major release on a branch which isn't master (that was merging the tag from proposed/juno back into master)12:52
fungidstanek: ironic has a similar one. i think the release managers didn't know to be on top of the core reviewers to get those changes merged immediately following release12:53
fungii just need to test a rebase/recreation of that to make sure that merging it after the change which merged the 2015.1.0 tag won't cause any versioning issues in master12:53
fungii'll do that now before i get sidetracked12:54
*** baoli has quit IRC12:54
*** _nadya_ has joined #openstack-infra12:54
openstackgerritPh. Marek proposed openstack-infra/project-config: Cherry-pick the DRBD cinder driver before running tests  https://review.openstack.org/18969412:54
*** e0ne has joined #openstack-infra12:55
flip214asselin: ^^12:55
*** baoli has joined #openstack-infra12:55
flip214fungi: would you like to help me with that as well? ^^12:55
*** _nadya_ has quit IRC12:56
*** hdd has quit IRC12:56
fungiflip214: i suppose the drbd-devstack job doesn't run against cinder? because if it does, that will totally not test the things you expect to be tested on cinder12:57
*** _nadya_ has joined #openstack-infra12:57
fungiit will just test the 178573,12 patchset over and over no matter what cinder change the job is running for12:57
fungiflip214: oh, actually i guess it's running after the cinder change is checked out. i get it now12:58
*** ildikov_ has quit IRC12:59
openstackgerritJerome Hourquebie proposed openstack-infra/jenkins-job-builder: Add cloudbees folder support  https://review.openstack.org/18969712:59
asselin_flip214, hmmm...sorry not sure that's correct12:59
*** ildikov_ has joined #openstack-infra12:59
fungiflip214: so... the only remaining concern i have is that if 178573,12 ceases to be able to merge cleanly against the tip of cinder master, that job will fail until you both update 178573 with a working patchset _and_ update this job template for the new patchset number12:59
asselin_fungi, flip214 sorry...your depends-on should do the right thing12:59
asselin_flip214, http://logs.openstack.org/68/187568/5/check/check-tempest-dsvm-full-drbd-devstack/b0be830/logs/devstack-gate-setup-workspace-new.txt.gz#_2015-06-09_11_08_41_46513:01
*** samuelBartel has quit IRC13:01
pelixelectrical: think there is either a review up or one that was accepted into the publishers for the new plugin.13:03
electricalpelix: found an existing review.13:03
*** otter768 has joined #openstack-infra13:04
asselin_flip214, have you looked at these errors? http://logs.openstack.org/68/187568/5/check/check-tempest-dsvm-full-drbd-devstack/b0be830/logs/screen-c-vol.txt.gz?level=TRACE13:04
*** hdd has joined #openstack-infra13:04
asselin_flip214, I think your job setup is currently testing the right thing...failures are due to drbd specific issues13:05
pelixams__: I'd have to fiddle a bit with it to see13:05
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632613:05
*** amitgandhinz has joined #openstack-infra13:05
ams__pelix: I got it working, but i have to pass all the variables in I need. It'll work for my purposes, it's just not the neatest13:05
*** changbl_ has quit IRC13:06
*** julim has joined #openstack-infra13:07
*** otter768 has quit IRC13:09
*** spredzy|afk is now known as spredzy13:14
*** _nadya_ has quit IRC13:14
*** ociuhandu has joined #openstack-infra13:14
*** nikil89_ has joined #openstack-infra13:15
*** nikil89_ is now known as naggappan13:15
jamielennoxTheJulia: sure - what would it have that token_endpoint with no token doesn't?13:15
*** ayoung has joined #openstack-infra13:16
*** ociuhandu has quit IRC13:16
*** sdake has joined #openstack-infra13:17
*** ParsectiX has quit IRC13:18
*** bhunter71 has joined #openstack-infra13:19
*** dpaterson has joined #openstack-infra13:19
marunhi! is there anyone deploying multi-node configurations with devstack?  I'm a bit confused as to how rabbit configuration is supposed to work on secondary nodes13:20
openstackgerritMonty Taylor proposed openstack-infra/infra-specs: Add maniphest bug tracking spec  https://review.openstack.org/18819613:20
*** soren has quit IRC13:21
*** samuelBartel has joined #openstack-infra13:22
*** che-arne has quit IRC13:22
fungiugh, this is worse than i thought... https://review.openstack.org/#/q/status:open+topic:merge/release-tag,n,z13:22
mordredfungi: what's the problem?13:23
fungittx: dhellmann: ideas for how we can convince projects to merge those in a timely manner? they're pretty much all merge-conflicting now and have to be regenerated13:23
fungibarbican has -2'd theirs13:23
*** dkranz has joined #openstack-infra13:23
*** jamielennox is now known as jamielennox|away13:23
mordredthey what?13:24
fungiironic approved theirs, but not before it reached merge conflict13:24
*** shardy_ has joined #openstack-infra13:24
*** tiswanso has joined #openstack-infra13:25
fungii'll write up a change to include an explanation of why these are safe for the commit messages when i regenerate them all13:25
mordredfungi: I just responded to that code review13:25
*** shardy has quit IRC13:25
mordredfungi: I'm going to write a quick email to the mailing list13:26
*** alivigni has joined #openstack-infra13:26
fungimordred: awesome. it looks like it's been a long-standing issue as we've got multiple projects who didn't merge the juno release tag and many, many who didn't merge the kilo release tag13:26
*** dustins has joined #openstack-infra13:27
sdagueoh, hey zuul ui gets really slow when there are 1555 merge check changes outstanding13:27
ttxyeah, if i could come up with a decent explanation of why they need to merge those, I would have sent that email already13:27
openstackgerritRussell Bryant proposed openstack-infra/devstack-gate: Add partial-ncpu-neutron-dvr grenade mode  https://review.openstack.org/18971513:28
sdagueit's also weird for me that we aren't building more nodes, we've got pretty deep queues right now, but the ramp up rate seems lower than I'd have expected.13:29
*** shardy_ has quit IRC13:29
*** yamada-h has joined #openstack-infra13:29
dkranzfungi: THe zuul fix seems to have gone through and the periodic tempest jobs last night are back to the previous failure: "/opt/stack/new/devstack-gate/devstack-vm-gate.sh: line 457: cd: /opt/stack/new/devstack: No such file or directory"13:29
*** shardy has joined #openstack-infra13:30
mordredfungi: ok. I'm a littlebit considering that we request that the TC give us permission to just go in and merge release tags for people13:31
*** radez_g0n3 is now known as radez13:31
mordredttx: the reason they should merge them is that the release team has asked them to merge tham as part of the release process13:31
sdaguemordred: at least make the release-managers group able to merge them13:32
sdagueI think that's a very reasonable approach13:32
mordredsdague: yah - well, that would involve giving the release team ability to approve patches in arbitrary repos13:32
*** hdd has quit IRC13:33
mordredsdague: I kind of wish that we had reached the point of being ok with zuul just landing some patches13:33
mordredsdague: and rather than proposing these just landed these13:33
mordredbut I don't think we're there yet13:33
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632613:33
openstackgerritEmilien Macchi proposed openstack-infra/project-config: jenkins/jobs/tripleo/puppet: drop puppet-openstack  https://review.openstack.org/18971913:35
*** samuelBartel has quit IRC13:35
mordredemail sent13:36
mordredhopefully it didn't come across too finger-wagging13:36
*** ociuhandu has joined #openstack-infra13:36
*** yamada-h has quit IRC13:36
mordredfungi: perhaps we should add a post-script to openstack proposal bot changes that says "if you have questions or concerns about this patch, please come to #openstack-infra and ask" ... or something13:37
*** yamada-h has joined #openstack-infra13:37
sdaguemordred: well, the release team has the ability to land arbitrary patches on stable13:37
ttxmordred: I for one ca ncertanly live without those merge commits. But some people seem to care.13:37
mordredttx: oh - well, I told everyone they exist because you want them13:37
ttxmordred: and I usually redirects those people to you for an explanation. For we are even13:37
mordredI also could live without them13:38
ttxWhat I type doesn't make sense13:38
mordredbut - if they are a thing we do - we should do them everywhere13:38
ttxI blame the jetlag vapors emanating from Morgan13:38
mordredor we should stop doing them anywhere13:38
*** bknudson has joined #openstack-infra13:38
mordredttx: I don't want to know any more about vapors emanating from Morgan13:38
*** tvanderwiel has joined #openstack-infra13:38
openstackgerritMerged openstack-infra/irc-meetings: add agenda_url for cross-project meeting  https://review.openstack.org/18967613:39
*** achanda has quit IRC13:39
*** yamada-h_ has joined #openstack-infra13:40
*** mriedem has joined #openstack-infra13:41
*** ociuhandu has quit IRC13:41
*** yamada-h has quit IRC13:41
*** ociuhandu has joined #openstack-infra13:41
TheJuliajamielennox|away: I think it would have to have sufficent information to tell the caller where to go connect.  Along the lines of the user being able to submit it's own configuration to return for service URLs.  And it would have to attempt no validation aside from the presence of the input.13:42
fungisdague: looks like the spike in the check pipeline was followed by a spike more or less up to our nodepool aggregate instance quota13:42
fungisdague: just with some lag13:42
thervemordred, The thing with the release tag patches is that they look awful in gerrit13:42
therveNot sure why13:42
sdaguefungi: oh, are we only at 400 nodes?13:42
fungitherve: because gerrit displays merge commits by showing the merge conflict resolution rather than a diff between the parents13:43
fungitherve: implementation choice upstream in gerrit13:43
thervefungi, Fair enough13:43
mordredtherve: it's pretty aweful13:43
fungisdague: the test nodes graph shows ~80013:43
*** yamada-h_ has quit IRC13:43
*** zz_jgrimm is now known as jgrimm13:43
mordredlike, its not even close to being the thing you want13:43
sdagueoh, it's there now13:43
*** jeblair has joined #openstack-infra13:43
*** eharney has quit IRC13:44
*** Zhongjun has joined #openstack-infra13:44
electricalfungi: pelix mordred https://review.openstack.org/#/c/174036/  required for latest hipchat plugin in jenkins :-)13:45
*** dimtruck is now known as zz_dimtruck13:48
jgriffithmordred: BTW, one of the reasons that never merged on Cinder side: https://review.openstack.org/#/c/179287/13:49
fungimordred: keystone's master branch is an excellent example of a project which has never approved one of the tag merge changes http://paste.openstack.org/show/277769/13:49
jgriffithmordred: noboby bothered to look an recheck it :(13:50
jgriffithincluding me13:50
mordredjgriffith: yah. we probably need a better story around that13:50
*** dkranz has quit IRC13:50
jgriffithmordred: anyway, thanks for the note, I'll pay coser attention to them now13:50
openstackgerritMerged openstack-infra/yaml2ical: Add Calendar Name and Description to output ical  https://review.openstack.org/18836913:50
mordredjgriffith: thanks - and sorry for requesting that you do a manual non-interesting task - I wish we had a better story on this13:50
jgriffithmordred: have you seen the code in Cinder lately?  This is a treat :)13:51
mordredjgriffith: hahahaha13:51
*** dkranz has joined #openstack-infra13:51
mordredjgriffith: consider yourself lucky that I've been focusing on neutron/nova/floating-ips and haven't yet started to yell at you about the attachments api13:51
jgriffithmordred: That's been on my list for like 2 years :(13:52
openstackgerritTristan Cacqueray proposed openstack-infra/puppet-openstackci: Create a jenkins server like http://jenkins.openstack.org  https://review.openstack.org/17557113:52
mordredjgriffith: although hoestly, the fact that you can say "nova boot oh and also give me a volume" is awesome13:52
jgriffithmordred: maybe today is the day13:52
openstackgerritTristan Cacqueray proposed openstack-infra/system-config: Use common jenkins master implementation  https://review.openstack.org/17737313:52
mordredjgriffith: I've been asking for that for floaters for ages13:52
jgriffithmordred: yeah, that's always annoyed the crap out of me13:52
fungimordred: looks like some of those have also been abandoned, so i'll restore and update them as well https://review.openstack.org/#/q/status:abandoned+topic:merge/release-tag,n,z13:53
jgriffithmordred: but instead of "fixing it" I just ignored it and changed my workflow13:53
mordredjgriffith: :)13:53
* jgriffith is part of the problem13:53
mordredjgriffith: have you seen shade?13:53
jgriffithmordred: oh yeah!13:53
mordredjgriffith: I've written an entire library based on ignoring the problem and working around it13:53
jgriffithmordred: and I tried to move to it a while back when you and I chatted13:53
* mordred is part of the problem too13:53
jgriffithmordred: note that I did the same sort of thing in my ansible playbooks13:54
*** dtantsur is now known as dtantsur|brb13:54
mordredjgriffith: btw - the new ansible modules have started landed upstream13:54
jgriffithmordred: I saw that13:54
jgriffithmordred: I had some problems cracking the config stuff13:54
jgriffithmordred: but I'll take another look shortly13:54
mordredstill working through them, but hope to get this stack done by 2.013:54
mordredjgriffith: the config stuff may have improved -we discovered a few bugs13:55
jgriffithmordred: I'll check it out, I think I promised you help at least in the form of feedback at one point13:55
mordredit _should_ just work if you have env vars set or if you pass parameters and dont' use config files13:55
jgriffithmordred: so I still owe you at least that13:55
mordredjgriffith: woot13:55
*** koolhead17 has joined #openstack-infra13:55
jeblairfungi: the last three seem to be correctly abandoned13:55
mordredlook, it's a jeblair13:55
jeblairfungi: the other ones were at least abandoned with comments indicating the authors do not understand the situation13:56
*** sdake has quit IRC13:56
fungijeblair: yep, i'm hoping to fix that with better commit messages shortly13:56
jeblairone got a -1 for whitespace13:57
* morganfainberg wavers at jeblair13:57
yolandajeblair, i'm construction the nodepool.conf path for nodepool. I was wondering if including the jenkins url in the nodepool.conf file or not13:57
* morganfainberg falls over after wavering13:57
mordredmorganfainberg: haha13:57
yolandaas it seems that it will be the only missing bit in jenkins target13:57
mordredmorganfainberg: blast it man - you're going to be jetlag recovered by the time I get there13:57
yolandataking ...13:57
yolanda      jenkins:13:57
yolanda        url: https://jenkins1.example.org/13:57
yolanda        user: username13:57
yolanda        apikey: key13:57
yolanda        credentials-id: id13:57
morganfainbergmordred: yeaaaah, it's much better today already13:58
yolandai'd prefer to completely remove that jennkins target section13:58
morganfainbergmordred: was bad by about 2200 last night13:58
* jeblair picks up morganfainberg and leans him against the bar13:58
mordredjeblair: there's a bar? where's my drink?13:58
morganfainbergmordred: we might have forgotten to restock it :P13:58
jeblairyolanda: i think in my sketch i suggested '[jenkins "jenkins1.example.org"]' as the conf section title13:59
* mordred throws an ineffective used mountain goat at morganfainberg13:59
ttxwe definitely are in a bar13:59
yolandajeblair, i was assuming that to be the target name14:00
yolandanot url14:00
*** UsedMountainGoat has joined #openstack-infra14:00
*** bdemers has joined #openstack-infra14:00
*** hashar has joined #openstack-infra14:00
* UsedMountainGoat is ineffective.14:00
jeblairyolanda: not sure what you mean?14:01
morganfainbergmordred: ^^ see what you did? :P14:01
mordredmorganfainberg: wow. that happened14:01
openstackgerritRussell Bryant proposed openstack-infra/project-config: Add check-grenade-dsvm-partial-ncpu-neutron  https://review.openstack.org/18942614:01
openstackgerritRussell Bryant proposed openstack-infra/project-config: Add check-grenade-dsvm-partial-ncpu-neutron-dvr  https://review.openstack.org/18972714:01
*** shardy_ has joined #openstack-infra14:01
yolandajeblair, something like14:01
yolanda[jenkins "target_name"]14:01
*** UsedMountainGoat has quit IRC14:01
yolandaso we are able to match targets in nodepool.yaml for that target name14:02
yolandaand url is an independent setting14:02
jeblairyolanda: ah, okay, yeah that looks good14:02
*** zz_dimtruck is now known as dimtruck14:02
*** miqui has joined #openstack-infra14:03
*** shardy has quit IRC14:03
fungimordred: any guesses where merge proposals like https://review.openstack.org/152011 are coming from? i see them happen from time to time on projects with feature branches concurrent with the dates of the changes14:03
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832514:03
fungimordred: but not sure how they're getting triggered without an actual tag getting pushed14:04
*** dboik has joined #openstack-infra14:04
fungioh, i bet i can unwind it back to a proposal job run14:04
yolandajeblair, i also see test-job is present, why is that under jenkins section , and not under the target ?14:04
yolandathis bit...                 t.jenkins_test_job = jenkins.get('test-job')14:05
yolandamakes sense to be there?14:05
mescanefSergeyLukjanov: hello. :-) please review https://review.openstack.org/#/c/189599/14:05
*** dguerri is now known as dguerri`14:06
*** shardy_ has quit IRC14:07
jeblairyolanda: because the job-launching mechanism is jenkins-specific14:07
yolandamm, then it needs to go to nodepool.conf as well14:07
*** shardy has joined #openstack-infra14:07
yolandaif we want to bring all jenkins target by nodepool14:07
jeblairyolanda: yeah, i think it's fine for now.  we could move it later, but i don't think it's important.14:08
yolandais that really used?14:08
jeblairwe don't14:08
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832514:09
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832514:10
*** eharney has joined #openstack-infra14:12
*** hdd has joined #openstack-infra14:12
openstackgerritDavide Guerri proposed openstack-infra/shade: Add get_server_external_ipv4()  https://review.openstack.org/18903014:12
*** asrangne has joined #openstack-infra14:13
*** xyang1 has joined #openstack-infra14:13
*** amotoki_ has joined #openstack-infra14:16
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832514:16
*** aswadr has quit IRC14:16
*** samuelBartel has joined #openstack-infra14:17
flaper87is there a way I can check if when/if the new images creation succeeded? What time does the CI images re-creation happens?14:20
*** hdd has quit IRC14:20
jeblairflaper87: logs are here: http://nodepool.openstack.org/14:21
fungiflaper87: http://nodepool.openstack.org/ for the logs, and 14:14 utc is when it starts, usually takes 1.5-2 hours to complete14:21
*** mtanino has joined #openstack-infra14:21
flaper87jeblair: fungi awesome, thanks :) That way I won't have to poll for updates14:22
flaper87well, not -infra, at leat14:22
*** k4n0 has quit IRC14:22
jeblairwe had a power outage last night, which lasted longer than my ups battery, in that it lasted longer than 30 seconds.14:24
jeblairso apparently, there's a bit of a margin of error in the battery runtime report, which was reporting a runtime of 45 minutes right before the outage.14:25
*** mpaolino has quit IRC14:25
mordredjeblair: hrm. that does seem like somewhat of a margin of error14:25
*** mpaolino has joined #openstack-infra14:26
jeblairyah.  now i get to figure out where to buy a new battery14:26
*** deepakcs has joined #openstack-infra14:26
mordredjeblair: batteries-r-us?14:26
mordredjeblair: I mean, there's got to be a burningman supply store in Oakland with UPS batteries, right14:27
mordredthey might also shoot fire sometimes ... but that's a feature14:27
*** dimtruck is now known as zz_dimtruck14:27
jeblaironly for 30 seconds though14:27
mordredjeblair: the power goes off, the ups starts shooting fire14:27
mordredI mean ... you'd NOTICE that the power was off14:27
mordredand you'd had light14:28
*** shuoy has joined #openstack-infra14:28
jeblairtrue, i had to read by candlelight last night, which isn't as bright or generally awesome as reading by flamethrower14:28
sdaguejeblair: you need to invest in some oil lamps14:28
jeblairsdague: i used to, but apparenty they can be bad news in earthquake country14:29
*** julim has quit IRC14:29
sdagueoh, right, well it would solve your light problem.14:29
*** spzala has joined #openstack-infra14:30
jeblair(i mean, candles aren't great either as far as that goes, but they're not as dangerous "off" as oil lamps are)14:30
sdaguejust expose other problems14:30
*** zz_dimtruck is now known as dimtruck14:30
*** julim has joined #openstack-infra14:30
pabelangerpleia2, nibalizer: I care.  Well, I should care.  I know we don't run much centos in -infa but would be nice to add coverage.  What can I do to help14:31
shuoyjeblair and fungi: just say thanks to the openstack-infra team for the help :-)14:32
openstackgerritMonty Taylor proposed openstack-infra/nodepool: Start passing cloud_config object to shade  https://review.openstack.org/18974414:32
mordredthat will not pass tests yet ^^14:32
mordreduntil we cut a new shade with the occ patch in it14:33
mordredbut I wanted to go ahead and put it up14:33
*** haypo has left #openstack-infra14:33
*** achanda has joined #openstack-infra14:33
*** hdd has joined #openstack-infra14:34
fungishuoy: were you able to work out a suitable merge commit into master from your dev/experimental branch?14:35
jeblairnow i just need to find a replacement battery for my since-discontinued ups from a since-aquired company14:36
mordredjeblair: it sounds like maybe you should learn about making your own batteries14:36
mordredjeblair: I mean, you already have lead type in the family ... why not make batteries too?14:37
*** fallenpegasus has joined #openstack-infra14:37
jeblairmordred: right; they were making batteries in the 1700s.  how hard could it be?14:38
*** kmartin has quit IRC14:39
shuoyfungi: yes, basically my requriement was like this: “1. use dev/experimental to replace the master (I know it is not a good practice — the reason was that we were in dev/experimental since we started some major surgery of our code; 2 we want to keep the git logs that happened on the dev/experimental branch even after we merge(replace) it back to master;”14:40
openstackgerritDan Prince proposed openstack-infra/tripleo-ci: Cherry pick Fedora python-ipaddress fix  https://review.openstack.org/18975614:41
shuoyfungi: though the workflow on the relase page was not entirely same as our case, after you described the principle behind it (and I did some googling), I figured out the way to achieve my goal.14:42
sdaguejeblair: also, clearly, not in any way dangerous to be playing with large chunks of lithium in earthquake country14:42
*** amotoki_ has quit IRC14:44
shuoyfungi: what I want to say though, having openstack-infra team’s online support gave me the mental assurance that I can make it  with this level of help (I did not know about how Gerrit /github work together before this). Many thanks!14:45
mordredshuoy: yay! glad you were suggessful!14:46
*** yfried is now known as yfried|afk14:48
*** yfried|afk is now known as yfried14:49
openstackgerritMerged openstack-infra/shade: Add very initial support for passing in occ object  https://review.openstack.org/18820114:49
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Add secgroup update API  https://review.openstack.org/18949014:49
openstackgerritMerged openstack-infra/shade: Add design for an object interface  https://review.openstack.org/16450614:49
mordredShrews: ooh - does the update stuff work now?14:50
Shrewsmordred: yup14:50
*** leopoldj has quit IRC14:50
mordredShrews: woot!14:51
*** salv-orl_ has joined #openstack-infra14:51
*** shuoy has quit IRC14:51
*** armax has joined #openstack-infra14:51
openstackgerrityolanda.robla proposed openstack-infra/nodepool: Update nodepool to use nodepool.conf file  https://review.openstack.org/18976214:52
*** salv-orlando has quit IRC14:52
Shrewsmordred: manually tested with both neutron and nova, so pretty confident with it14:52
mordredlooks great14:52
mordredjeblair, fungi: could I bother you for a review on https://review.openstack.org/189490 ? I'd like to get it landed so we can cut a release14:53
openstackgerrityolanda.robla proposed openstack-infra/puppet-nodepool: Add creation of nodepool.conf file  https://review.openstack.org/18832514:54
*** abregman has quit IRC14:55
yolandafungi, we see gerrit 2.8 slower than prev 2.4 version. Have you done any tweaked in terms of caching, apache config.. that is not reflected on the manifests?14:55
mordredyolanda: nope. but I think we do have a bigger/faster machine we run it on, so we may not have noticed the differences?14:56
*** dabos has quit IRC14:56
*** jaypipes has quit IRC14:57
*** dkranz has quit IRC14:57
*** tiswanso has quit IRC14:57
*** ihrachyshka has quit IRC14:57
*** tiswanso has joined #openstack-infra14:58
fungiyolanda: we run it on a 30gb performance hvpvm instance in rackspace14:58
*** ihrachyshka_ has quit IRC14:58
*** dkranz has joined #openstack-infra14:58
*** amotoki_ has joined #openstack-infra14:58
fungiyolanda: with a separate database server (a trove instance)14:58
rcarrillocruzour also is a 32GB VM, but it's on ESXi14:58
*** shuoy has joined #openstack-infra14:58
rcarrillocruzi think we will have to tweak and play cache things14:59
*** achanda_ has joined #openstack-infra15:00
*** kmartin has joined #openstack-infra15:00
*** yfried has quit IRC15:00
*** maishsk_afk has quit IRC15:00
*** koolhead17 has quit IRC15:00
mordredrcarrillocruz: fwiw, ours has 8 VCPUs as well15:00
mordredrcarrillocruz: that at least pretend to be Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz15:01
rcarrillocruzmordred: otoh, we also use ldap, something you don't in upstream... i'll check it out with jstack what threads are blocking15:01
*** achanda has quit IRC15:01
mordredrcarrillocruz: this is a good point15:01
mordredrcarrillocruz: so - you may want to look at the cauth program that enovance wrote for software factory15:02
shuoymordred: let’s make openstack successful, together :-)15:02
*** fhubik is now known as fhubik_afk15:02
rcarrillocruzmordred: googling15:02
mordredrcarrillocruz: it has ldap support in it - it basically handles the auth at the apache layer15:02
mordredrcarrillocruz: and passes remote_user to the thing on the backend15:02
mordredrcarrillocruz: if you are blocking on LDAP queries on the backend ... it might be a way to decouple the two things15:03
*** shuoy has left #openstack-infra15:03
mordredrcarrillocruz: but I have not yet used it15:03
*** shuoy has joined #openstack-infra15:03
*** Sukhdev has joined #openstack-infra15:04
*** ddieterly has joined #openstack-infra15:04
*** bdemers has quit IRC15:04
*** mpaolino has quit IRC15:04
*** achanda_ has quit IRC15:04
*** otter768 has joined #openstack-infra15:05
*** sdake has joined #openstack-infra15:05
*** deepakcs has quit IRC15:06
cschwedemordred: I’m currently refactoring cauth, and adding documentation for it15:06
*** afazekas has quit IRC15:06
*** achanda has joined #openstack-infra15:06
*** sdake has quit IRC15:06
*** dustins has quit IRC15:06
yolandamordred, that looks interesting15:06
*** salv-orl_ has quit IRC15:06
yolandaactually, running gerrit show-caches it throws an average of 7seconds per ldap request15:07
*** sdake has joined #openstack-infra15:07
*** markvoelker has joined #openstack-infra15:07
*** JamesK1 has joined #openstack-infra15:07
*** e0ne is now known as e0ne_15:08
*** salv-orlando has joined #openstack-infra15:08
*** chlong is now known as chlong-zzz15:10
*** otter768 has quit IRC15:10
*** naggappan has quit IRC15:10
*** baoli has quit IRC15:10
*** achanda has quit IRC15:11
*** markvoelker has quit IRC15:12
*** baoli has joined #openstack-infra15:13
*** vahidh has joined #openstack-infra15:15
*** amotoki_ has quit IRC15:16
*** annegentle has joined #openstack-infra15:16
*** changbl_ has joined #openstack-infra15:17
*** annegentle has quit IRC15:18
*** e0ne_ has quit IRC15:18
*** shuoy has quit IRC15:19
*** dims has quit IRC15:19
*** dtantsur|brb is now known as dtantsur15:19
*** amotoki_ has joined #openstack-infra15:19
*** soren has joined #openstack-infra15:21
openstackgerritPaul Belanger proposed openstack-infra/grafyaml: Initial unit tests for CLI  https://review.openstack.org/18977015:21
*** dabos has joined #openstack-infra15:21
*** Ala has quit IRC15:21
leseb_AJaeger: ping15:22
leseb_AJaeger: did you get a chance to look at my last comment, just wanted your opinion: https://review.openstack.org/#/c/188768/15:22
*** arxcruz has quit IRC15:23
*** hashar is now known as hasharMeeting15:23
*** e0ne has joined #openstack-infra15:23
*** Sukhdev has quit IRC15:25
openstackgerritPaul Belanger proposed openstack-infra/project-config: Add grafyaml-coverage tests  https://review.openstack.org/18977315:26
*** fhubik_afk is now known as fhubik15:27
openstackgerritDoug Hellmann proposed openstack/requirements: Raise minimums for oslo libs without namespace packages  https://review.openstack.org/18977515:27
*** hasharMeeting is now known as hashar15:27
*** shuoy has joined #openstack-infra15:27
*** jgrimm is now known as jgrimm_out15:28
*** sdake has quit IRC15:30
*** prad_ has joined #openstack-infra15:30
*** ianw has quit IRC15:30
*** ianw has joined #openstack-infra15:31
*** dabos has quit IRC15:31
dstaneksdague: fungi: good stuff15:31
*** dizquierdo has quit IRC15:33
*** sambetts has quit IRC15:33
*** gyee_ has joined #openstack-infra15:34
*** amotoki_ has quit IRC15:35
*** achanda has joined #openstack-infra15:37
*** rkukura has joined #openstack-infra15:37
*** dimtruck is now known as zz_dimtruck15:38
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel SolidFire Cinder  https://review.openstack.org/18977915:38
*** ddieterl_ has joined #openstack-infra15:39
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel SolidFire Cinder  https://review.openstack.org/18977915:39
*** zz_dimtruck is now known as dimtruck15:40
*** ddieterly has quit IRC15:40
krotscheckIs the mechanism by which global-configuration is synchronized amongst projects something that can be reused for other things?15:41
greghaynesyou mean global-requirements?15:41
krotscheckRight, that.15:41
krotscheckJavaScript doesn't have a thing like PEP8, so I'd like to find a way to keep linting configurations synchronized between projects.15:42
*** marun has quit IRC15:42
*** ajmiller has joined #openstack-infra15:43
greghaynesah, that sounds like a test job that is the same for most projects15:43
*** pradk has joined #openstack-infra15:44
*** isviridov_away is now known as isviridov15:44
*** sarob has joined #openstack-infra15:45
*** pradk has quit IRC15:45
*** sputnik13 has joined #openstack-infra15:46
krotscheckgreghaynes: Yep.15:47
krotscheckgreghaynes: Except that we'd also like to keep the projects atomic.15:47
krotscheckI mean separable.15:47
krotscheckSo dependencies, rules, and tools used to execute those things should be project specific.15:47
greghaynesSounds a lot like hacking15:48
greghaynesmaybe it could be extended to do what you want, or they might have ideas on how to do it15:49
*** hemnafk is now known as hemna15:49
*** eantysh9 has joined #openstack-infra15:50
greghayneswhich also gets you http://docs.openstack.org/developer/hacking/ :p15:50
mordredkrotscheck: we do have a somewhat generalize ability to automatically propose changes to projects15:51
mordredkrotscheck: it's used not just for requirements at the moment, but also for translations. we also use it to propose cleanups to the projects.yaml file15:51
*** ajmiller_ has joined #openstack-infra15:51
mordredkrotscheck: and the merge-release-tag job15:51
*** ajmiller has quit IRC15:52
krotscheckmordred: Hrm. I feel like the best way to do this would be to start something similar up for javascript global-requirements, and then to write a linting plugin for either eslint or jscs that enforces our rules.15:52
mordredkrotscheck: so, yeah, totally possible to make a thing that keeps a central copy of a thing and copies it around - that said, also to greghaynes' point, it might be neat to make something like hacking but for javascript15:52
krotscheckThat way, we only sync one thing.15:52
*** marcusvrn1 has joined #openstack-infra15:52
mordredI agree15:52
*** eantyshev has quit IRC15:52
*** Sukhdev has joined #openstack-infra15:53
*** ajmiller_ is now known as ajmiller15:53
*** BharatK has joined #openstack-infra15:53
mordredkrotscheck: I mean, we could also just put a javascript deps file in global-requirements and extend update.py to figure out if the target repo needs python info, javascript info, or both15:53
krotscheckmordred: Naah. That makes too much sense.15:53
mordredkrotscheck: that way it's the same mechanism and we don't need yet-another-sync thing - one could imagine horizon getting patches that have python and javascript requirements changes15:54
mordredkrotscheck: :)15:54
greghaynesIMO tests failing so that we only merge stuff that fits the style guides is better than an async cleanup commit later if possible15:54
*** marcusvrn has quit IRC15:54
krotscheckgreghaynes: Hrm. Good point. Chicken/egg15:54
*** unicell has quit IRC15:54
krotscheckWell, but we have that already.15:55
krotscheckIf someone puts in a dependency that breaks tests, that is.15:55
*** rcrit has quit IRC15:56
*** dtantsur is now known as dtantsur|afk15:56
greghaynesyes, dependencies and linting are different - dependencies in js it might be a fine idea for, it really depends on how that works15:56
*** BharatK_ has joined #openstack-infra15:57
greghaynesmordred: in CI, what is the sane way to install docker15:58
mordredgreghaynes: sudo apt-get install docker15:58
*** rossella_s has quit IRC15:58
*** che-arne has joined #openstack-infra15:58
greghaynesoh, its a system package? their docs tell me to do the lovely wget | sh15:58
*** rossella_s has joined #openstack-infra15:58
mordredgreghaynes: nah. it's a system package15:58
mordreddon't listen to the hipsters15:58
greghayneshaha, ok15:58
mordredVersion: 1.5-115:59
mordredin vivid - lemme check trusty15:59
*** yfried has joined #openstack-infra15:59
*** Kiall_ is now known as Kiall15:59
greghaynes"System tray for KDE3/GNOME2 docklet applications"15:59
*** asselin_ has quit IRC15:59
mordredsame version in trusty15:59
mordredoh - wait15:59
greghaynesdocker.io is the package15:59
*** achanda has quit IRC15:59
mordred1.0.1~dfsg1-0ubuntu1~ubuntu0.14.04.1 in trusty15:59
greghaynesI was wondering why it needed to depend on X1116:00
openstackgerritPaul Belanger proposed openstack-infra/grafyaml: Fix --version CLI output  https://review.openstack.org/18978816:00
mordred1.5.0~dfsg1-1ubuntu2 in vivid16:00
*** achanda has joined #openstack-infra16:00
*** adduarte has quit IRC16:01
*** cdent has quit IRC16:01
openstackgerritJeremy Stanley proposed openstack-infra/project-config: Add an explanation to release tag merges  https://review.openstack.org/18979116:01
nibalizerpabelanger: well basically there are these 'roles' (I hate that word) being written for openstackci16:02
fungimordred: jeblair: sdague: ttx: dstanek: morganfainberg: jgriffith: ^ see if that's sufficient, if not i can make it even more verbose16:02
nibalizer'zuul', 'graphite', 'logsevrer'16:02
*** BharatK_ has quit IRC16:02
* morganfainberg looks.16:02
nibalizerand right now they're written for ubuntu and would need code modificiatons to run successfully on centos16:03
SergeyLukjanovmordred, fungi, pabelanger, hey folks, I have a news from the mirantis marketing team about stackalytics (sorry about big latency, it takes a long time...)16:03
SergeyLukjanovmordred, fungi, pabelanger, the response is "We told the foundation that we are willing to do that if they drop Bitergia and make stackalytics the one and only main tool. They don't appear to be OK with that. "16:03
pabelangernibalizer, Right, I would expect that.  Since we don't use CentOS too much upstream16:03
*** e0ne is now known as e0ne_16:03
*** juggler_ is now known as juggler16:03
mordredSergeyLukjanov: sigh16:03
fungiSergeyLukjanov: it's too bad nobody from stackalytics was present for the analytics tools discussion at the summit, because that was the decision16:04
nibalizerpabelanger: where this hits rubber and road is here https://review.openstack.org/#/c/184891/16/spec/acceptance/basic_spec.rb16:04
*** Nikkau has quit IRC16:04
nibalizerso if we get centos tests working we can remove that if statement16:04
morganfainbergfungi: lgtm16:04
*** dpaterson has quit IRC16:04
mordredSergeyLukjanov: yah - the decision out of the summit was to drop all the tools except for stackalytics16:04
SergeyLukjanovfungi, I unfortunately was unable to attend, are there any outputs written?16:04
*** jlanoux has quit IRC16:04
fungiSergeyLukjanov: specifically we want to replace bitgeria's stuff and our other analytics tools with stackalytics, we just need a transition plan to add the missing features into stackalytics16:04
*** samuelBartel has quit IRC16:05
mordredSergeyLukjanov: although I think it may take a cycle or so to make sure that we can migrate features16:05
greghaynesThere were some shortcomings with stackalytics relative to russelb's thing that need to get addressed first though16:05
pabelangernibalizer, Because I don't know, is the infra for CentOS 7 to run the specs up and running?16:05
pabelangeror would that need to be done too16:05
SergeyLukjanovmordred, fungi, so it sounds like the things our marketing folks wants16:05
mordredgreghaynes: yah - I think the idea was "make stackalytics official to make it clear that the differences should get fixed, then remove the other things as their features are in stackalytics"16:06
mordredSergeyLukjanov: yah16:06
mordredSergeyLukjanov: so there may just be a miscommunication somewhere16:06
SergeyLukjanovmordred, fungi, are there any proves for the decision to drop other tools that I could share with them?16:06
SergeyLukjanovmordred, I have a feeling that this answer is based on the very outdated info from ~ prev. summit16:06
pleia2good morning16:06
jeblairSergeyLukjanov: https://etherpad.openstack.org/p/YVR-infra-activity-tracking16:07
jeblairSergeyLukjanov: line 5016:07
fungiSergeyLukjanov: it was a discussion involving reed and ttx, who did have the authority to decide those things with us16:07
*** dkranz has quit IRC16:07
nibalizerpabelanger: uh 'yes'16:07
crinklepabelanger: the puppet group runs tests on centos716:08
nibalizerin as much as it should be but I haven't personally had to poke it so there may be some gotcha16:08
SergeyLukjanovjeblair, ack, thx16:08
nibalizerbut actually openstack/puppet-* use it pretty heaviliy so yes it works16:08
SergeyLukjanovfungi, great16:08
mordredjeblair, SergeyLukjanov, fungi: should we propose something to the TC for the TC to bless?16:08
jeblairSergeyLukjanov: also line 51 suggests "replace activity with stackalytics".  so i don't know if the exact words are there, but i think that captures the intent16:08
fungiSergeyLukjanov: though i'm mildly concerned that the mirantis marketing team has any vested interest in community-maintained contributor analytics16:08
SergeyLukjanovso, I will try to return back to them with some links and news16:09
*** e0ne_ is now known as e0ne16:09
pabelangernibalizer, crinkle okay, so if I understand you'd just need to write the cento7 coverage with the spec tests.  Which, since we don't use it upstream seem to be work for no -infra benefit16:09
*** dkranz has joined #openstack-infra16:09
jeblairmordred: that might be a good idea, just to make absolutely sure everyone is on the same page16:09
fungiSergeyLukjanov: because one thing we don't want our community statistics system to be is a marketing platform16:09
SergeyLukjanovfungi, it has a footer with mirantis ref :)16:09
*** amit213 has joined #openstack-infra16:10
*** gyee_ has quit IRC16:10
*** redixin has quit IRC16:11
jeblairSergeyLukjanov: right; i think that's the sort of thing that mirantis should be okay with people changing if it becomes an official project.16:11
SergeyLukjanovfungi, I hope we'll be able to agree with them, I think this consensus to move from other tools to stackalytics will be a good change16:11
pabelangerfungi, Brings up a good point, since Mirantis had branding on stackalytics, guess the spec should reference that16:11
nibalizerpabelanger: well I think you'd have to modify the modules to work as well. I started out without the 'Debian' fence and it was failing centos16:12
nibalizerbute yes16:12
SergeyLukjanovyeah, it should be removed for sure and marketing folks understand this, I think that's the main reason for anticipation of the move16:12
jeblairi hope that we would replace mirantis branding with openstack project branding, and also more non-mirantis people would contribute to maintaining it16:13
jgriffithfungi: yeah, that's great16:13
pabelangernibalizer, right, I would be infavor of making them non-voting, if possible.  Then, allow the people interested to convert them to centos7.16:13
mordredSergeyLukjanov: yah16:13
pabelangerFWIW: I'd likely be submitting patches for them, since I plan to start spinning up something local here at work16:13
jeblairSergeyLukjanov: cool, i do also understand their position -- they might not want to move the domain name and lose the branding unless they are sure that the tool will be the "main" tool.16:14
mordredyah. that makes sense16:14
SergeyLukjanovyup, make sense16:14
jeblairSergeyLukjanov: so yes, i think it's good to get this all written down and agreed :)16:14
nibalizerpabelanger: well they can be made 'voting' trivially by removing that if statement in the patch that adds centos support16:14
*** MarkAtwood has joined #openstack-infra16:15
*** tonytan4ever has joined #openstack-infra16:15
pabelangernibalizer, right. And that makes sense to me.16:16
*** sasi has joined #openstack-infra16:17
*** amit213 has quit IRC16:17
pabelangerI wouldn't want centos7 support to drag down the effort to move forward. But, having non-voting jobs help people who want to add support for it too16:17
*** hdd has quit IRC16:17
*** hdd has joined #openstack-infra16:18
*** achanda has quit IRC16:18
*** yamamoto has quit IRC16:18
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Add secgroup update API  https://review.openstack.org/18949016:20
*** shuoy has quit IRC16:20
Shrewsmordred: i had missed the returned nova object normalization ^^^^16:21
*** soren has quit IRC16:21
*** derekh has quit IRC16:23
*** scheuran has quit IRC16:23
*** yfried has quit IRC16:23
openstackgerritMerged openstack-infra/project-config: Add Ironic services to Rally experimental pipeline  https://review.openstack.org/18799716:24
*** annegentle has joined #openstack-infra16:25
*** shuoy has joined #openstack-infra16:25
*** dimtruck is now known as zz_dimtruck16:27
*** shuoy has quit IRC16:27
*** unicell has joined #openstack-infra16:27
*** shuoy has joined #openstack-infra16:28
*** Zhongjun has quit IRC16:28
*** Swami has joined #openstack-infra16:29
*** baoli has quit IRC16:30
*** jtomasek has quit IRC16:30
*** shuoy has quit IRC16:30
*** ihrachyshka_ has joined #openstack-infra16:31
*** baoli has joined #openstack-infra16:31
*** zz_dimtruck is now known as dimtruck16:31
*** HeOS has quit IRC16:32
*** dannywilson has joined #openstack-infra16:32
*** bkero has quit IRC16:32
*** bkero has joined #openstack-infra16:32
*** tqtran_ has joined #openstack-infra16:33
*** yamahata has joined #openstack-infra16:33
*** ihrachyshka has joined #openstack-infra16:33
*** salv-orlando has quit IRC16:33
*** alexsyip has joined #openstack-infra16:34
*** dannywilson has quit IRC16:34
*** dannywilson has joined #openstack-infra16:35
openstackgerritSirushti Murugesan proposed openstack-infra/devstack-gate: Add work around to enable heat grenade config  https://review.openstack.org/18965316:36
*** markus_z has quit IRC16:37
*** e0ne has quit IRC16:38
*** tonytan4ever has quit IRC16:39
*** tonytan4ever has joined #openstack-infra16:40
*** tonytan4ever has quit IRC16:40
*** tiswanso has quit IRC16:40
*** ildikov_ has quit IRC16:40
*** dguerri` is now known as dguerri16:40
*** patrickeast has joined #openstack-infra16:41
*** abregman has joined #openstack-infra16:42
jrollfungi: you need that ironic 2014.2 tag merged asap?16:42
*** BobBall is now known as BobBall_AWOL16:43
*** tiswanso_ has joined #openstack-infra16:44
SergeyLukjanovjeblair, yeah, it should help16:44
*** jerryz has joined #openstack-infra16:46
fungijroll: it's not an emergency, but i'm updating all the outstanding tag merge changes to no longer conflict with master branches if needed, and adding good commit messages to all of them (based on https://review.openstack.org/189791 which should change them for us in automation for future releases as well)16:47
*** BharatK has quit IRC16:47
jrollfungi: ok, cool, I'll make sure we hit that today16:47
fungisure thing16:48
fungii've got all the 2014.2 tag merges unabandoned if necessary and updated, working on 2015.1.0 now16:49
*** ashleighfarnham has joined #openstack-infra16:50
Clintfungi: could i impose upon you to do some gerrit account futzing at some point today?16:50
openstackgerritMerged openstack-infra/jenkins-job-builder: Support for more configurable options in Plots Plugin  https://review.openstack.org/18889716:50
fungiClint: sure, what's up?16:50
*** amit213 has joined #openstack-infra16:51
Clintfungi: i want my identity changed to https://login.launchpad.net/+id/Bye7h6r (and the dup account nuked or whatever)16:51
jeblairClint: if you're just trying to change to the other clint, it's not going to work.  ;)16:51
fungiwe're onto you16:51
fungiClint: you want to keep the new account or switch that id onto the old account?16:52
Clintfungi: switch, please16:52
fungii seriously wish gerrit's openid-based workflow were better thought out16:52
Clintjeblair: i feel bad enough that i broke clint@ubuntu.com for him16:52
*** dustins has joined #openstack-infra16:52
openstackgerritJames E. Blair proposed openstack-infra/project-config: Expand explanation of auto tag merge  https://review.openstack.org/18981216:54
*** amit213 has quit IRC16:54
jeblairfungi: ^16:55
*** sayalilunkad has joined #openstack-infra16:55
*** markvoelker has joined #openstack-infra16:56
*** ivar-lazzaro has joined #openstack-infra16:56
fungijeblair: thanks! i'll update the commit message i'm using for the 2015.1.0 tags to that16:58
thingeefungi: pushing 1.2.2 of python-cinderclient now.16:58
*** thedodd has joined #openstack-infra16:58
fungithingee: awesome. was it glance_store that was breaking pending a new cinderclient release?16:59
*** fhubik has quit IRC16:59
thingeefungi: it was something related with glance16:59
*** rkukura_ has joined #openstack-infra17:00
*** ujuc has joined #openstack-infra17:00
*** freerunner has joined #openstack-infra17:00
*** marun has joined #openstack-infra17:01
*** markvoelker has quit IRC17:01
openstackgerritMerged openstack-infra/tripleo-ci: Cherry pick Fedora python-ipaddress fix  https://review.openstack.org/18975617:02
*** weshay has quit IRC17:02
*** jaypipes has joined #openstack-infra17:02
*** rkukura has quit IRC17:03
*** hdd has quit IRC17:03
*** hdd has joined #openstack-infra17:03
*** puranamr has joined #openstack-infra17:04
*** rkukura_ has quit IRC17:05
fungiClint: so... 16272 is your original account from may. 16798 was the one associated with your new openid. i've moved the new openid back to 16272 and set the 16798 account inactive now17:05
*** rlandy has quit IRC17:05
*** rlandy has joined #openstack-infra17:05
*** otter768 has joined #openstack-infra17:05
*** rkukura has joined #openstack-infra17:07
Clintfungi: looks good, thanks!17:08
*** salv-orlando has joined #openstack-infra17:09
*** otter768 has quit IRC17:10
fungiClint: sure thing, let me know if you run into any problems17:11
openstackgerritSean Dague proposed openstack-infra/devstack-gate: create GRENADE_PLUGINRC to pass through grenade external plugins  https://review.openstack.org/18965217:11
openstackgerritSean Dague proposed openstack-infra/devstack-gate: Add work around to enable heat grenade config  https://review.openstack.org/18965317:11
*** wznoinsk has joined #openstack-infra17:12
Clintwill do17:12
*** shuoy has joined #openstack-infra17:13
*** shuoy has quit IRC17:14
*** ajmiller has quit IRC17:14
wznoinskhi all, could someone from infra team have a look at https://review.openstack.org/#/c/186839/, we need 2 groups created for this new stackforge project, it was originally requested by nova cores to have our unit tests for our NFV CI public (more details in comments on that change)17:15
pleia2wznoinsk: you mean the -core and -release teams?17:16
wznoinskthanks in advance, let me know if something's not clear17:16
wznoinskpleia2: yes17:16
pleia2wznoinsk: they exist, we just need to know the gerrit email address of the initial member (they can add other members)17:16
*** rkukura has quit IRC17:17
pleia2at least, they should, hm17:17
wznoinskcan't see it on https://review.openstack.org/#/admin/groups/?filter=nfv but maybe it's not showing groups with no members, please add the author of the project-config change first: sean.k.mooney@intel.com, we'll take it from there then17:18
wznoinsk,_ pleia217:18
openstackgerritMerged openstack-infra/project-config: Add an explanation to release tag merges  https://review.openstack.org/18979117:19
anteayawznoinsk: did you ever get your personal gerrit account sorted out?17:19
*** yamamoto has joined #openstack-infra17:19
*** ericksonsantos has joined #openstack-infra17:19
anteayawznoinsk: the email thread didn't seem to reach a conclusion on the mailing list17:19
wznoinskanteaya: hi, I was setting up my irc client properly to have a chat with Jeremy Stanley and how we can test it, and to unlock my account too17:20
sdaguefungi: https://review.openstack.org/#/c/189808/ is needed to fix the grenade jobs, there was an un anticipated compat breakage with the oslo.middleware release17:20
fungiwznoinsk: yep, that's me17:20
fungisdague: looking17:20
anteayawznoinsk: are you ready to have that discussion now?17:21
sdagueonce we get test results back that it's passing the critical bit, we should hop it to the top of the gate, because everything with grenade jobs is faceplant right now17:21
wznoinskfungi: good to catch you, before I even started to look for you ;-)17:21
*** ianw has quit IRC17:21
sdaguebut it's still about 20 minutes from confirming the answer that it fixes things17:21
openstackgerritMerged openstack-infra/shade: Add secgroup update API  https://review.openstack.org/18949017:21
jkthi there, so I'm using Zuul for doing post-push CI in KDE in projects without gating, and I'd like to create some pretty dashboard about how these jobs are doing17:21
*** ianw has joined #openstack-infra17:21
fungiwznoinsk: i'm usually around, though i tend to be disturbingly unconscious between 0400-1200 utc17:22
jktbecause the projects in question are *not* using Gerrit, I cannot really reuse that work which you guys put into the 3rdparty CI dashboards etc17:22
*** pelix has quit IRC17:22
pleia2fungi: I thought these groups created themselves, but maybe I'm wrong17:22
wznoinskfungi: do you have a moment to try to figure out what the problem is maybe? it would be great if I could use some test server on your side to test it17:22
pleia2(though now I worry we're conflating two different problems)17:22
jktI'm browsing through zuul's source now, and I originally wanted to use reportes by simply adding another class and storing JSON-formatted stuff about which jobs in my ref-updated pipeline17:23
fungipleia2: the groups create themselves but initially empty. if they haven't it's either that the change adding them to the acls has merged so recently that our automation hasn't applied them yet, or something was/went wrong17:23
pleia2I'm still talking about adding people to the initial groups re: 18683917:23
sdaguejkt: so... actually, we were discussing just such a thing at summit17:23
openstackgerritMike Perez proposed openstack/requirements: Bumping CinderClient to 1.2.2  https://review.openstack.org/18982517:23
pleia2fungi: ok good, it merged yesterday afternoon so it should have done its thing by now17:23
fungiwznoinsk: sure, we have review-dev.openstack.org17:23
jktsdague: jhesketh pointed me to https://review.openstack.org/#/c/135170/ yesterday, is there something newer?17:23
thingeekragniz: ^17:23
*** yamamoto has quit IRC17:24
pleia2fungi: so should we manually create intel-nfv-ci-tests-release && -core or..?17:24
*** e0ne has joined #openstack-infra17:24
fungiwznoinsk: if stream-events gets stuck on review-dev, restarting the server is non-impacting for production17:24
wznoinsklet me know when I can test it on review-dev.openstack.org17:24
sdaguejkt: https://etherpad.openstack.org/p/YVR-QA-testing-beyond-the-gate17:24
wznoinsk<- fungi17:24
*** viglesias has left #openstack-infra17:24
sdaguejkt: it's concept phase atm17:24
anteayawznoinsk: you can do so now17:24
pleia2ah, the project isn't in gerrit either17:25
fungipleia2: are those groups mentioned in an acl? if so we shouldn't need to create them by hand17:25
anteayawznoinsk: you should be able to sign into review-dev.openstack.org with your same launchpad credientials17:25
pleia2fungi: yeah, they're in the acl17:25
anteayawznoinsk: ensure your account has your public ssh key and away you go17:25
kragnizthingee: looks good to me17:25
fungipleia2: sounds like there's either something wrong with the change that we didn't catch in review, or manage-projects broke on us17:25
*** teran has quit IRC17:26
pleia2the patch seems ok, inclined to think manage-projects is to blame17:26
fungisdague: 189808 is failing all its jobs so far17:27
*** hashar has quit IRC17:27
sdaguefungi: the only job reporting back so far is heat, right?17:27
jktsdague: thanks, according to that page, it looks similar in (some of) the goals to what I need17:27
fungisdague: also partial-ncpu now17:27
sdaguefungi: ok17:27
fungiand that one's voting17:27
jktsdague: but it seems that zuul is not mentioned in there, and that the thing will probably rely on log processing via logstash17:28
openstackgerritMerged openstack-infra/project-config: Expand explanation of auto tag merge  https://review.openstack.org/18981217:28
sdaguejkt: it's mostly going to rely on the zuul graphite meters actually17:29
jktsdague: which I sort of understand because you guys are probably most interested in 3rd-party CI which will likely not use your central zuul, would it?17:29
fungisdague: http://logs.openstack.org/08/189808/2/check/check-grenade-dsvm-partial-ncpu/9b944c2/logs/grenade.sh.txt.gz#_2015-06-09_17_21_07_511 "/opt/stack/new/devstack/lib/nova: line 753: apache_site_config_for: command not found"17:30
sdaguejkt: actually, this is mostly not about 3rd party ci17:30
jktsdague: the statsd reporting about events?17:30
sdaguefungi: yeh, it's not that, it's nova-api17:30
openstackgerritDavide Guerri proposed openstack-infra/shade: Add get_server_external_ipv4() to meta  https://review.openstack.org/18903017:30
*** asrangne has quit IRC17:30
fungisdague: ahh, i love those sorts of cryptic errors17:30
*** shardy has quit IRC17:30
sdagueyeh, oslo.middleware is sinking multiple services17:31
jktsdague: what I had in mind is dumping machine-readable info gathered upon reporting time into couchdb, and having some client-side JS reading from there and showing status matrices17:32
jktsdague: would something like that provide what you're looking for, too? Would the implementation fit into the way you usually solve these things?17:33
sdaguejkt: well, zuul already puts out data to graphite, I think creating another datastore is something we wanted to avoid17:33
jktsdague: is that documented somewhere?17:34
jktsdague: http://docs.openstack.org/infra/zuul/statsd.html I suppose, right?17:34
sdagueyeh, sounds right17:35
*** amitgandhinz has quit IRC17:35
wznoinskanteaya: I get Pmierssions denied (publickey) from review-dev.openstack.org (same as openstack.org), could you reenable my account if that's the problem here?17:36
*** ivar-lazzaro has quit IRC17:36
jktsdague: well, we're using a shared pipeline with a non-unique job name for many projects at once, so it won't work for me right now17:36
jktsdague: it would require me to encode project name into the job name at least17:36
*** ivar-lazzaro has joined #openstack-infra17:36
fungiwznoinsk: you need to go into the webui at https://review-dev.openstack.org/ and login via launchpad openid, then add your ssh public key17:36
fungiwznoinsk: accounts aren't copied over from review.openstack.org to review-dev.openstack.org, they're separate17:37
*** ociuhandu has quit IRC17:37
wznoinskfungi: will do, thought they were17:37
*** ivar-lazzaro has quit IRC17:37
Clintis there some significance to topics of "(detached"17:37
openstackgerritsebastian marcet proposed openstack-infra/openstackid: Migration of resource server  https://review.openstack.org/17885417:38
jktsdague: http://quickgit.kde.org/?p=sysadmin%2Fgerrit-project-config.git&a=blob&f=zuul%2Flayout%2Flayout.yaml is how my layout currently looks like17:38
*** ivar-lazzaro has joined #openstack-infra17:38
fungiwznoinsk: also thanks for helping us figure this out. i'm more strongly suspecting gerrit's event stream workers block writing on half-dead sockets and never timeout17:38
*** mpaolino has joined #openstack-infra17:38
fungiClint: if you git review from a local detached head and don't specify a topic, the first word git review grabs as the "branch" name is (detached17:39
openstackgerritMerged openstack-infra/project-config: Added new SF repository for Fuel VMware DVS plugin  https://review.openstack.org/18959917:39
*** mpaolino has quit IRC17:39
wznoinskfungi: I think I'm not without a blame (leaving jobs in the background that might have been still alive is not a good thing), to me it should be either a 'normal' stream event listening or a problem with the connection so gerrit should terminate, let see what it could be17:39
pleia2fungi: when you have a moment, could you help me figure out how to retrigger manage-projects (or otherwise debug) so we can get stackforge/intel-nfv-ci-tests on their way?17:40
pleia2busy morning :)17:40
wznoinskfungi: i'm subscribed with one connection to event stream now17:40
fungiwznoinsk: honestly, i think any ability of a user to block the stream-events interface for all other users counts as a gerrit bug, and a pretty bad one17:40
wznoinskfungi: I agree17:41
wznoinskfungi: please monitor my connection, I'll make the job 'stop' (which will not close the connection my side)17:41
Clintfungi: aha17:41
fungipleia2: lemme get through these last three tag merge updates and send a quick reply to the ml thread about them, and then i'll see if i can figure out what happened with manage-projects17:42
mordredClint: it's not the best feature of git-review and we could probably make that better17:42
jktsdague: but I see that unlike what the documentation says, project name actually is encoded in there, nice17:42
pleia2fungi: thanks17:42
fungipleia2: and yes, busy morning17:43
sdaguejkt: yeh, there is actually a ton more data in the statsd feed17:43
sdagueand I expect we'll extend it more for this17:43
Clintmordred: by detecting that case and substituting random words?17:43
sdaguethe docs are just lagging17:43
pabelangerfungi, pleia2: I don't mind offering to help, assuming I'm able too17:43
sdagueit's why we thought we could get quite far just with that17:43
anteayaClint: I favour the string "Clint was here"17:44
jktsdague: correction, the proj name is only included in some aggregates, not when reporting job results, but as you say, that's easily fixable17:44
*** dguerri is now known as dguerri`17:44
fungipabelanger: it's not publicly logged (puppet can sometimes leak sensitive info in logs), so one of the root admins has to check up on it17:44
Clintanteaya: that takes me back to a horrible episode from the '90s17:45
jktsdague: one won't get access to stuff like being able to query job results for a given commit, but I don't think that's a critical feature for you17:45
anteayaClint: ew sorry for the bad memory17:45
anteayaClint: that wasn't my goal17:45
sdaguejkt: right, because we're thinking about this mostly in the periodic job sense17:45
pabelangerfungi, fair enough. Either way, anything I can do to help with troubleshooting in -infra land, feel free to ask17:45
pabelangerI don't mind chipping in around here17:45
fungipabelanger: i'll make sure to! ;)17:45
fungiand thanks!!!17:46
pleia2thanks pabelanger \o/17:46
anteayapabelanger: mostly just look at what questions folks ask17:46
*** ddieterl_ has quit IRC17:46
anteayapabelanger: and try to find your way around the infra repos to help point them to docs17:46
*** che-arne has quit IRC17:47
Clintanteaya: you are forgiven17:47
anteayapabelanger: it doesn't require root to help, it helps you learn yourself, and it they do need root help it gives them something to do until a root becomes avaialble17:47
anteayaClint: thanks, you are a generous soul17:47
mordredShrews, jeblair, fungi, pleia2: I intend to cut a new shade release. I expect it to have zero impact on nodepool, as there are no changes in the release that touch the parts of shade that nodepool uses, and also we have testing17:47
mordredbut I thought I'd give a heads up just in case17:47
anteayapabelanger: as a starter, any time someone posts a log link in the channel, look at it and follow the conversation17:47
*** rkukura has joined #openstack-infra17:47
anteayapabelanger: if you can learn the log structure and teach others that goes a long way17:48
*** koolhead17 has joined #openstack-infra17:48
*** koolhead17 has quit IRC17:48
*** koolhead17 has joined #openstack-infra17:48
openstackgerritMerged openstack-infra/project-config: Handle incorrect project names more carefully.  https://review.openstack.org/18947617:50
jeblairmordred: ack, thanks17:51
*** dannywilson has quit IRC17:51
*** dannywilson has joined #openstack-infra17:52
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for oslo.middleware namespace bug 1463478  https://review.openstack.org/18983317:52
openstackbug 1463478 in oslo.middleware "liberty side of grenade fails to start ceilometer-api due to bad oslo.middleware entry in api-paste.ini" [High,In progress] https://launchpad.net/bugs/1463478 - Assigned to Doug Hellmann (doug-hellmann)17:52
openstackgerritDoug Hellmann proposed openstack/requirements: Raise minimums for oslo libs without namespace packages  https://review.openstack.org/18977517:52
pabelangeranteaya, right, I wasn't aware the problem pleia2 was having required root access. Mostly, if it was public I'd chip in to help. Your description is good about contributing BTW17:52
dhellmannsdague: ^^ includes a block for middleware17:52
*** Sukhdev has quit IRC17:52
anteayapabelanger: okay thanks, glad you see it as supportive and not putting you off17:53
sdaguedhellmann: +217:53
dhellmannsdague: ty17:53
anteayapabelanger: pleia2 needs to check the logs from a run of manage-projects (needs server access) and then potential refire manage-projects to run again, again root access to manually fire the script on the gerrit server17:53
wznoinskfungi: how is it looking now?17:54
openstackgerritPaul Belanger proposed openstack-infra/infra-specs: Host Stackalytics Service  https://review.openstack.org/18771517:55
*** vahidh has quit IRC17:55
anteayapabelanger: manage-projects is a script in the jeeby repo: http://git.openstack.org/cgit/openstack-infra/jeepyb/tree/jeepyb/cmd/manage_projects.py17:55
anteayapabelanger: it fires when changes are made to projects, as in creating a new one17:55
*** vahidh has joined #openstack-infra17:55
anteayapabelanger: it consumes a library that gerrit wrote to interface with gerrit and extends it17:56
*** dims has joined #openstack-infra17:56
anteayapabelanger: sometimes we have races when a project is created on gerrit but hasn't yet mirrored to git and github so it falls over17:56
*** achanda has joined #openstack-infra17:57
*** achanda has quit IRC17:57
anteayapabelanger: so if a patch to create a project merged to project-config but the project isn't on git.o.o it is a manage-projects issue17:57
openstackgerritJeremy Stanley proposed openstack-infra/project-config: Add periodic propose-requirements-constraints jobs  https://review.openstack.org/18951117:57
*** bhuvan has joined #openstack-infra17:58
*** achanda has joined #openstack-infra17:58
fungiwznoinsk: i don't see a stream-events task in the process queue, though i do see an established connection17:58
openstackgerritMatt Riedemann proposed openstack/requirements: Block oslo.middleware 2.0.0  https://review.openstack.org/18983917:58
*** ivar-lazzaro has quit IRC18:00
fungiwznoinsk: i've also added a comment to a change and the resulting event hasn't caused any stream-events task to remain in the queue either18:00
*** tvanderwiel has quit IRC18:00
*** ivar-lazzaro has joined #openstack-infra18:00
pabelangeranteaya, Yup, have it running locally in a few places. So experienced on how it work. For the most part, one of the reason I offered to help.18:00
dhellmannmriedem: I think your patch duplicates a similar change in https://review.openstack.org/18977518:00
fungiwznoinsk: though there might be some buffers at play, where we need to get over a certain volume of data in the outbound stream before it begins to back up. no idea18:00
pabelangeractually looking forward to the ansible stuff that mordred is writing for the manage-projects / rename stuff18:01
anteayapabelanger: ah cool, have you encountered any races in project creation, with manage projects?18:01
pabelangerwould be hot to get the race conditions fixed and automated18:01
*** tvanderwiel has joined #openstack-infra18:01
anteayapabelanger: agreed18:01
*** ivar-lazzaro has quit IRC18:01
sdaguefungi: so https://review.openstack.org/189775 is the new fix for the gate18:01
wznoinskfungi: i've added a new gerrit review and I saw my script catching it, I see events in my script now when I'm adding comments to changes as well so I'm listening to the stream18:01
*** ivar-lazzaro has joined #openstack-infra18:01
mriedemdhellmann: ack, dropped it18:01
*** tiswanso_ has quit IRC18:01
fungiwznoinsk: or it could have to do with there being more hung stream-events sessions than workers before the blocked ones start to become apparent in the process queue18:01
*** puranamr has quit IRC18:02
fungiwznoinsk: yeah, i have a session attached to stream-events as well18:02
*** gyee_ has joined #openstack-infra18:02
*** maishsk_afk has joined #openstack-infra18:02
fungipleia2: i'm looking into the new project problem now. what was the change number that should have added it?18:02
anteayaI can stream events from review-dev if that will help18:02
*** tiswanso_ has joined #openstack-infra18:02
*** thedodd has quit IRC18:03
*** puranamr has joined #openstack-infra18:03
fungianteaya: can't hurt!18:03
pleia2fungi: https://review.openstack.org/#/c/186839/ I'm browsing /var/log/manage_projects.log.1.gz and can find the intel-nfv-ci-tests bit but not what went wrong exactly18:03
openstackgerritMerged openstack-infra/elastic-recheck: Add query for oslo.middleware namespace bug 1463478  https://review.openstack.org/18983318:04
openstackbug 1463478 in oslo.middleware "liberty side of grenade fails to start ceilometer-api due to bad oslo.middleware entry in api-paste.ini" [High,In progress] https://launchpad.net/bugs/1463478 - Assigned to Doug Hellmann (doug-hellmann)18:04
*** hichtakk has joined #openstack-infra18:04
wznoinskfungi: I have 3 stopped jobs with stream-events listeners in each of them, any issues your side?18:06
anteayafungi: I've checked my ssh public key which was there and added it again and can't ssh in, not sure why18:06
fungiwznoinsk: none yet, no18:06
anteayafungi: not worth it to trouble shoot me though I don't think18:06
fungianteaya: wrong username?18:06
*** root has joined #openstack-infra18:06
*** ihrachyshka has quit IRC18:06
*** root is now known as Guest7405118:06
*** Guest74051 has left #openstack-infra18:06
anteayaokay streaming18:07
*** timcline has joined #openstack-infra18:07
*** sdake has joined #openstack-infra18:07
wznoinskfungi: added a few 'naughty' connections, can you see them?18:07
*** ihrachyshka_ has quit IRC18:07
*** bradjones is now known as bradjones|away18:07
fungiwznoinsk: i have 7 open connections for you now but still no running stream-events tasks in the process queue18:08
jeblairpabelanger: what is your goal with the stackalytics topic for todays meeting?  are you proposing it for final approval, or do you want to discuss something about it in the meeting?18:08
*** dguerri` is now known as dguerri18:08
*** amitgandhinz has joined #openstack-infra18:09
wznoinskfungi: so i'll do some gerrit queries now18:09
*** geoffarnold has joined #openstack-infra18:09
asselinjeblair, there are 2 downstream puppet reviews that are ready to merge IMO. Should I request via agent, or just ask in channel?18:09
asselinsorry 3*18:10
jeblairasselin: what do you mean via agent?18:10
*** baoli has quit IRC18:10
*** JamesK1 has quit IRC18:10
*** dguerri is now known as dguerri`18:11
pabelangerjeblair, okay, I was actually just wanting to get an update18:11
asselinjeblair, sorry...agenda18:11
pabelangerbut, we did get one today18:11
*** JamesK has joined #openstack-infra18:11
jeblairasselin: in channel please18:11
pabelangerjeblair, so, I can remove it from the list.  Until there is more movement from mirantis on it18:11
jeblairpabelanger: sounds reasonable18:12
wznoinskfungi: ran 3 gerrit queries over ssh18:12
*** soren has joined #openstack-infra18:12
fungiwznoinsk: no change18:12
pabelangerjeblair, removed18:13
mordredShrews, SpamapS, dguerri`: ok. released 0.7.0 of shade - we can start using those in the places18:13
nibalizerasselin: links?18:13
* asselin looks them up18:13
wznoinskI wasn't doing anything diff. last time I messed on prod, I'll leave the jobs running same as last time so we may see something later18:13
asselinshoudl be ready to merge: https://review.openstack.org/17557118:14
asselinshould be ready to merge: https://review.openstack.org/#/c/177373/18:14
asselinshould be ready to merge: https://review.openstack.org/#/c/180624/18:14
fungipleia2: it looks like it applied that change on the server at 20:26:24 yesterday according to /var/log/syslog.118:15
fungipleia2: unfortunately the logging for puppet execs leaves something to be desired18:15
pleia2fungi: yeah, and the manage projects logs show timestamps from 20:3318:15
pleia2but the manage project logs aren't so helpful either18:16
fungipleia2: and that exec completes successfully at 20:37:0018:16
fungipleia2: you can try running `sudo manage-projects stackforge/intel-nfv-ci-tests` to see what happens18:16
fungipleia2: oh, i see the issue18:17
pleia2hm, that didn't go as expected18:18
fungicompare the filename of their acl with the others in the same directory18:18
wznoinskfungi: I have to go now, I'll get back to you tomorrow, ping me if you find something18:18
fungiwznoinsk: thanks for having a look at it with us!18:18
pabelangerSeems pbr is not happy: http://logs.openstack.org/15/187715/5/check/gate-infra-specs-docs/a8cae26/console.html18:18
*** harlowja_ has quit IRC18:18
*** chris6131 has left #openstack-infra18:18
*** ZZelle_ has joined #openstack-infra18:18
anteayapabelanger: that is a sphnix error: sphinx.errors.SphinxWarning: WARNING: no "man_pages" config value found; no manual pages will be written18:19
*** puranamr has quit IRC18:19
pabelangeranteaya, right but pbr is invoking sphinx IIRC.18:19
pabelangerso either one is problematic18:19
anteayais it?18:19
pabelangerchecking code18:19
wznoinskfungi: on the intel-nfv-ci-tests groups creation issue then, is there something we need to fix change in18:20
anteayaI thought pbr was to limit itself rather than continue to try to do everything18:20
fungipabelanger: anteaya: we just started seeing it in openstack/ossa repos too. it's delaying an advisory release18:20
lifelessmordred: so - https://review.openstack.org/#/c/189491/2 please18:20
fungipabelanger: anteaya: i suspect a new library release may be incompatible with our sphinx configs... was oslosphinx in today's release batch?18:20
anteayayes, 3.0.018:21
lifelessmordred: and https://review.openstack.org/#/c/189491/218:21
pabelangerfungi, Yup18:21
anteayaI checked the wall of doug18:21
*** isviridov is now known as isviridov_away18:21
mordredanteaya: we've always done man pages18:21
anteayaI thought the error was a bit odd18:21
lifelesspabelanger: thats the new man page support18:21
fungiwznoinsk: yeah, submit a change updating the acl file from your original to include the same filename suffix as the other acl files in the same directory18:21
*** achanda has quit IRC18:21
pabelangerlifeless, ya, just reading code updates now18:21
lifelesspabelanger: erm, no we didn't land that. as mordred says its the existing stuff18:21
*** tonytan4ever has joined #openstack-infra18:22
openstackgerritsebastian marcet proposed openstack-infra/openstackid: Migration of resource server  https://review.openstack.org/17885418:22
* lifeless goes to crosscheck too18:22
openstackgerritJoe Gordon proposed openstack/requirements: Drop SQLAlchemy upper bound.  https://review.openstack.org/18984718:22
pabelangerso pinning oslosphinx<3.0.0 fixes18:22
pabelangerlet me see what is causing the problem18:22
mordredlifeless: I don't see anything new we landed that would be an issue18:22
pleia2fungi: ah, I'll submit a patch to .config it18:22
davidlenwelljeblair fungi https://review.openstack.org/#/c/18820718:23
anteayamordred lifeless oslosphnix 3.0.0 was released18:23
lifelessthe only related thing I see is 1dfe9ef Remove sphinx_config.init_values() manual call18:23
wznoinskfungi, pleia2 we'll do, thanks18:23
lifelessand yes oslosphinx3 released18:23
lifelessabout 3 hours ago18:24
lifelessanteaya: pbr isn't limiting itself per se; we're making it be compatible with all the relevant upstreams18:24
openstackgerritElizabeth K. Joseph proposed openstack-infra/project-config: Add .config to intel-nfv-ci-tests ACL  https://review.openstack.org/18984818:24
pleia2wznoinsk: ^^ this change should fix the issue we're seeing with intel-nfv-ci-tests gerrit groups not being created18:25
lifelessanteaya: so that we're driving ecosystem alignment not divergence, but as the only permitted build dependency for all of openstack, it has to do what openstack needs.18:25
anteayalifeless: oh okay, I have no concept of what that means18:25
*** harlowja has joined #openstack-infra18:25
anteayabut apparently that includes interactions with sphnix18:26
lifelessbecause we build docs18:26
anteayaokay good to know, thanks18:26
anteayapabelanger: I was wrong18:26
lifelesslikewise if openstack python packages start building C extensions, pbr will be involved18:26
*** mwagner_afk has joined #openstack-infra18:26
wznoinskpleia2: thanks for that18:26
mordredyup. although so will grumbling a lot18:26
lifelessbut olso sphinx is the big andidate I think18:27
*** hdd has quit IRC18:27
mordredlifeless: I just read through the diff since 2.5.018:27
fungiyep, we're seeing it across multiple projects, and more recently than the pbr release18:27
mordredthe only thing that looks even partially related is that oslosphinx's pbr dep changed18:27
wznoinski like the turnaround time thanks fungi anteaya pleia218:27
fungilike, just since the oslo release announcements18:27
dhellmannfungi, mordred : have an error log?18:27
mordreddhellmann: http://logs.openstack.org/15/187715/5/check/gate-infra-specs-docs/a8cae26/console.html18:27
*** ddieterly has joined #openstack-infra18:28
*** hdd has joined #openstack-infra18:28
mordredso - I wonder if the old pin was downgrading pbr or something18:28
* mordred takes random stabs at thin air18:28
anteayawznoinsk: thanks for paying attention and helping us, we appreciate it18:28
wznoinskheading off, have a good rest of the day everyone18:29
anteayawznoinsk: thanks you too18:29
lifelessmordred: Sphinx==1.1.318:29
pabelangerlifeless, ya, pretty sure https://github.com/openstack-dev/pbr/commit/1dfe9ef348c777bef67b2c2b1d35e50ebc720333 is the issue.  reverting the patch fixed my error18:30
lifelessI'm expecting to see 1.2 or 1.318:30
lifelessbut hey pabelanger ^ yay for testing18:30
*** annegentle has quit IRC18:30
lifelesspabelanger: ok, push up a revert, we'll +A and release asap.18:30
jgriffithfungi: are you already working on somethign for those Tag patches for Grenade and pylint tests?18:31
pabelangerlifeless, sure, give me a second. Just testing patch locally18:31
jgriffithfungi: ie https://review.openstack.org/#/c/128920/18:32
fungijgriffith: sdague has a change proposed to solve the oslo.messaging issue for grenade18:33
jgriffithfungi: thanks18:33
dhellmannjgriffith, fungi : we're reverting the oslo.middleware change and I'll cut a new release as soon as that lands18:33
sdaguefungi: actually, dhellmann has a requirements change18:33
dhellmannalso https://review.openstack.org/18982918:33
fungijgriffith: the pylint failure is odd and looks like a bug18:34
sdagueit's still getting test results18:34
*** ddieterly has quit IRC18:34
dhellmannfungi: requirements update to block middleware is https://review.openstack.org/18977518:34
fungisdague: ahh, cool thanks18:34
*** ddieterly has joined #openstack-infra18:34
fungijgriffith: i have suspicions about the correctness of the tools/lintstack.sh in cinder18:35
jgriffithfungi: maybe, but the CONFLICT messages made me suspect18:35
sdaguefungi: so.... https://review.openstack.org/189775 has passed where it previously had died18:35
*** mrmartin has joined #openstack-infra18:35
openstackgerritPaul Belanger proposed openstack-dev/pbr: Revert "Remove sphinx_config.init_values() manual call"  https://review.openstack.org/18985118:35
sdaguewhich means that I'm pretty confident in it being the correct fix18:35
sdagueit would be nice to promote that18:36
pabelangerlifeless, ^18:36
jgriffithfungi: we have multiple merge conflicts trying to apply that18:36
*** soren has quit IRC18:36
lifelessits odd that that passed our integration tests18:36
pabelangerapplied patch locally and reconfimed18:36
lifelessit implies that all of the things we tested have man pages setup ?18:36
pabelangerlifeless, ya, I was going to look and see how it passed too18:36
fungijgriffith: the "fatal: unable to auto-detect email address (got 'jenkins@bare-trusty-1433772880.(none)')" looks like a similar problem we worked around in git-review. something about git not being able to successfully guess a proper e-mail address if the hostname is incomplete (which is the case in hpcloud, where this job ran)18:36
lifelesspabelanger: please do; I'll leave that with you18:36
*** koolhead17 has quit IRC18:37
pabelangershouldn't be too hard to add a test for it18:37
*** puranamr has joined #openstack-infra18:37
*** ddieterly has quit IRC18:37
fungijgriffith: probably needs to do something like we did in https://review.openstack.org/18816518:38
jgriffithfungi: I was more interested in these: http://paste.openstack.org/show/278478/18:38
*** crc32 has joined #openstack-infra18:38
lifelesspabelanger: yeah, I'm interested in the integration test coverage here, as well as a functional test18:39
sdaguefungi: ok, walking away from computer for a bit, promoting https://review.openstack.org/189775 would get a lot of folks moving again.18:39
fungijgriffith: where?18:39
*** ildikov_ has joined #openstack-infra18:39
pabelangerlifeless, Ya, going to take a little to see how test coverage works for pbr.  But agree, more coverage is good18:39
fungijgriffith: i didn't see that in the pylint log18:39
jgriffithfungi: I pulled the currently runnign one: https://jenkins03.openstack.org/job/gate-cinder-pylint/203/console18:40
*** dprince has quit IRC18:40
*** sasi has left #openstack-infra18:40
openstackgerritMichal Rostecki proposed openstack-infra/project-config: Determining Rally-Devstack integration on pre_test_hook  https://review.openstack.org/18985318:41
*** dannywilson has quit IRC18:41
jgriffithfungi: so interesting... https://review.openstack.org/#/c/179287/218:42
fungijgriffith: taking a look in a sec once i have the requirements gate fix moved to the front of the line18:42
lifelesspabelanger: 'coverage' is likely the wrong word :)18:42
jgriffithfungi: notice that the only thing updated since the initial submittal on Apr 30'th is the req file18:42
lifelesspabelanger: we're hugely influence by external data rather than branch coverage inside the pbr code18:42
jgriffithfungi: sure... sorry, just more talking out loud :)18:42
lifelesspabelanger: because e.g. sphinx internals in this case18:43
*** dannywilson has joined #openstack-infra18:43
*** crc32 has quit IRC18:44
*** crc32 has joined #openstack-infra18:45
*** markvoelker has joined #openstack-infra18:45
*** liam_ has joined #openstack-infra18:47
*** liam_ is now known as Guest4176518:48
*** openstackgerrit has quit IRC18:48
*** tim_o has joined #openstack-infra18:48
*** dannywilson has quit IRC18:48
*** openstackgerrit has joined #openstack-infra18:48
*** puranamr has quit IRC18:48
jeblairfungi: i'm going to remove some merger git repos to speed them up; they haven't been cleaned up in a while (and we haven't automated that yet)18:49
fungijeblair: sounds good18:49
jeblairthis might cause a few spurious failures18:49
fungioh, ick18:49
*** markvoelker has quit IRC18:49
lifelessgreghaynes: https://github.com/pypa/pip/issues/288018:49
fungihopefully not too many18:50
greghaynesresource bug whack-a-mole18:51
*** dtantsur|afk is now known as dtantsur18:51
*** dannywilson has joined #openstack-infra18:52
pabelangerlifeless, okay, ya, there is like zero build_sphinx coverage in tests.18:53
pabelangerGoing to add a simple test right now18:53
mordredpabelanger: well, that would explain why it didn't fail our tests18:53
fungijgriffith: so the main thing to check with those tag merge changes is that the parent shas are the origin's branch tip (at the time it was pushed) and the tagged commit, and that a diff between the merge commit and the branch tip at the origin are empty18:53
pabelangerlifeless, I would expect potentially multiple new testpackage(s) being created to toggle different flags.18:54
pabelangerhowever, just going to use cookie-cutter for the moment18:54
*** ujuc has quit IRC18:54
fungilast check job running on 189775,2 just passed and it got enqueued, so i'm promoting it to the front of the gate now18:55
jgriffithfungi: that's kinda weird then18:55
jeblairfungi: if i stop the mergers and wait a few minutes, hopefully jobs will have fetched refs by then, and i can identify the change from the logs for the merger that was interrupted and recheck it manually18:55
lifelesspabelanger: well, for now18:55
fungijeblair: yeah, that's not so bad then18:55
lifelesspabelanger: add calls to build_sphinx to the integration tests18:55
openstackgerritRussell Bryant proposed openstack-infra/project-config: Add check-grenade-dsvm-partial-ncpu-neutron  https://review.openstack.org/18942618:55
openstackgerritRussell Bryant proposed openstack-infra/project-config: Add check-grenade-dsvm-partial-ncpu-neutron-dvr  https://review.openstack.org/18972718:55
lifelesspabelanger: that should give us good coverage right up18:56
pabelangerlifeless, right, that is what I am doing. into test_core18:56
pabelangerneed to add some doc bits too18:56
lifelesspabelanger: no18:56
lifelesspabelanger: I mean https://git.openstack.org/cgit/openstack-dev/pbr/tree/tools/integration.sh?id=bb838190eebbaf11887f89d8ef265f0924f19ae418:56
pabelangerlifeless, okay, checking18:56
lifelesspabelanger: and https://git.openstack.org/cgit/openstack-dev/pbr/tree/pbr/tests/test_integration.py?id=bb838190eebbaf11887f89d8ef265f0924f19ae418:56
pabelangerfirst time diving into the heart of pbr18:56
lifelesspabelanger: which together create a venv for every package in projects.txt18:57
lifelesspabelanger: and then build and install it a few different ways18:57
lifelesspabelanger: but as you say, we're not testing sphinx there yet.18:57
lifelesspabelanger: or are we: thats why I'm puzzled about this not triggering failures.18:57
fungijgriffith: lintstack.sh could stand to have some debugging details, but i suspect it's doing something bad/shortsighted. looking through the source for it now18:57
lifelesspabelanger: I would have expected building sdists or wheels to have included the docs18:58
lifelessperhaps I'm mistaken18:58
*** dizquierdo has joined #openstack-infra18:58
jgriffithfungi: yes, it looks like it.  I'm trying to figure it out18:59
*** wenlock has joined #openstack-infra18:59
jgriffithfungi: but still coming up to speed on what we're doing here anyway :)18:59
openstackgerritDoug Hellmann proposed openstack-infra/project-config: Change tag ACLs for library projects to library-release team  https://review.openstack.org/18985618:59
fungijgriffith: yeah, looks like this is from the naive git merge it attempts18:59
jgriffithfungi: which now I kinda get... because it does the little back/forward branch thing to compare deltas18:59
jeblairwow, creating the ref in git is taking the zuul mergers a huge amount of time; we should look into that18:59
jeblair(that should be really fast)19:00
jeblairmeeting time!19:00
* Clint nods.19:00
fungijgriffith: also some of that script is just plan wrong. especially "he HEAD is not a merge commit. This won't happen on gerrit." patently false19:00
jeblair(when keystone is finished)19:00
*** ianw has quit IRC19:00
openstackgerritMerged openstack-infra/project-config: Add upper-constraints generation glue  https://review.openstack.org/18949119:01
*** ianw has joined #openstack-infra19:01
*** soren has joined #openstack-infra19:01
fungiyeah, i suspect the massive gate resets are causing a lot of additional high-priority demand on the mergers19:02
*** tiswanso_ has quit IRC19:02
jgriffithfungi: sigh19:02
fungimy promote request has been sitting waiting in the queue for 10+ minutes already19:02
fungithere it finally went!19:02
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Add release_many.sh  https://review.openstack.org/18985819:02
fungiokay, fingers crossed 189775,2 doesn't hit any new problems19:02
*** wenlock_ has joined #openstack-infra19:03
*** ivar-lazzaro has quit IRC19:03
*** ivar-lazzaro has joined #openstack-infra19:04
*** tiswanso has joined #openstack-infra19:04
*** teran has joined #openstack-infra19:04
openstackgerritRichard Pijnenburg proposed openstack-infra/jenkins-job-builder: Adding support for Hipchat plugin 0.1.9 as publisher  https://review.openstack.org/17403619:05
*** Guest41765 has quit IRC19:06
*** smarcet has joined #openstack-infra19:06
electricalzxiiro: thanks for the review on the hipchat publisher. updated it19:06
openstackgerrityolanda.robla proposed openstack-infra/system-config: Don't hardcode pip.conf values  https://review.openstack.org/16625219:06
*** mwagner_afk has quit IRC19:06
*** otter768 has joined #openstack-infra19:06
smarcetfungi: i am started to get a weird error regarding doc builds here http://logs.openstack.org/54/178854/10/check/gate-openstackid-docs/bf826a7/console.html19:06
*** annegentle has joined #openstack-infra19:06
smarcetfungi: hello :)19:07
*** puranamr has joined #openstack-infra19:07
fungismarcet: yeah, that seems to be related to today's pbr 1.1.0 release. the offending patch is being reverted and an emergency release of pbr should be along shortly19:07
AJaegerleseb_: It looks ok but I'd like some more eyes on it. fungi, sdague, could you review https://review.openstack.org/#/c/188768/, please? It's a devstack plugin change...19:07
smarcetfungi: oh ok cool thx for the info :)19:08
*** spzala has quit IRC19:11
*** otter768 has quit IRC19:11
zxiiroelectrical: my pleasure19:13
*** annegentle has quit IRC19:14
fungizaro: meeting!19:15
*** e0ne has quit IRC19:15
*** rlandy has quit IRC19:16
openstackgerrityolanda.robla proposed openstack-infra/system-config: Start using puppet-puppet module to configure puppet  https://review.openstack.org/17111319:16
pabelangerlifeless, mordred so, the issue seems limited to sphinx <1.2 for man_pages19:17
*** mrunge has quit IRC19:17
mordredpabelanger: spectacular19:17
pabelangerusing 1.2.3 doesn't appear to be affected19:17
*** vahidh has quit IRC19:19
*** rlandy has joined #openstack-infra19:20
ekarlsofungi: what code search thing ?19:21
openstackgerritPaul Belanger proposed openstack-dev/pbr: WIP: sphinx test coverage  https://review.openstack.org/18988419:21
*** dprince has joined #openstack-infra19:21
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Add release_many.sh  https://review.openstack.org/18985819:22
openstackgerritJohn Griffith proposed openstack-infra/project-config: Make gate-cinder-pylint non-voting  https://review.openstack.org/18988519:22
fungiekarlso: it's described in https://review.openstack.org/18857419:22
*** achanda has joined #openstack-infra19:24
pabelangerlifeless, ^ thats what I was thinking for a set of tests too19:24
pabelangerstill working on how integration.sh runs19:24
*** marcusvrn has joined #openstack-infra19:24
*** thedodd has joined #openstack-infra19:27
*** marcusvrn1 has quit IRC19:28
flaper87fungi: I checked nodepool's logs and it seems that the image build ended but IIUC, it seems that it was built without the uuid-dev package. Any hint on how I can know if the package was installed ?19:29
*** achanda has quit IRC19:29
openstackgerritMerged openstack-infra/project-config: Add .config to intel-nfv-ci-tests ACL  https://review.openstack.org/18984819:30
fungiflaper87: i'll need to look after meetings wrap up19:30
*** annegentle has joined #openstack-infra19:30
*** dtantsur is now known as dtantsur|afk19:30
*** sdake has quit IRC19:31
flaper87fungi: ok, no rush. let me know when you know something :)19:31
*** freerunner has quit IRC19:32
openstackgerrityolanda.robla proposed openstack-infra/nodepool: Update nodepool to use nodepool.conf file  https://review.openstack.org/18976219:33
*** dboik has quit IRC19:34
pleia2zaro: re: earlier in the meeting, want to have a look at https://review.openstack.org/#/c/178887 and see if your questions were answered? we're moving to voting phase on that spec19:34
*** dboik has joined #openstack-infra19:34
*** dabos has joined #openstack-infra19:35
fungiflaper87: looks like we successfully updated the bare-trusty image in all rackspace regions but only one of our five hpcloud neutron nets19:36
zaropleia2: +1 and +119:36
*** puranamr has quit IRC19:36
fungiflaper87: i'll try to get the remaining four restarted here in a bit and keep on them until they succeed19:36
*** puranamr has joined #openstack-infra19:36
*** achanda has joined #openstack-infra19:36
flaper87fungi: ah got it, thanks a lot for your help.19:37
*** dizquierdo has quit IRC19:37
*** gyee_ has quit IRC19:38
fungiflaper87: i've got them running now so if we're particularly lucky, might be in place within half an hour19:39
electricalwho's Darragh btw? for the life of me i'm unable to link the irc name vs review name :-)19:40
nibalizerelectrical: thats pelix19:40
electricalnibalizer: ah!19:41
electricalthanks :-)19:41
*** mrmartin has quit IRC19:41
*** jtomasek has joined #openstack-infra19:41
*** maishsk_afk has quit IRC19:42
flaper87fungi: roger that, should I check that same nodepool link to get logs? Or can I shamelessly ping you ?19:42
*** maishsk_afk has joined #openstack-infra19:43
*** tonytan4ever has quit IRC19:44
fungiflaper87: the manual image updates don't write to the logs, but i'll give you a heads up as soon as they all finish19:44
*** bookwar has left #openstack-infra19:44
*** tonytan4ever has joined #openstack-infra19:44
flaper87fungi: roger19:45
fungiyou're welcome19:45
openstackgerritEmilien Macchi proposed openstack-infra/project-config: puppet: add 3.8 release tests support  https://review.openstack.org/18989019:45
EmilienMnibalizer, Hunner, crinkle: you might want to see that ^19:45
*** puranamr has quit IRC19:46
nibalizerEmilienM: i think 3.8.1 is latet19:47
EmilienMnibalizer: it's just we don't pin so far for other releases19:47
EmilienMI wanted to be consistent19:48
*** maishsk_afk has quit IRC19:48
EmilienMnibalizer: also, packaging does not provide 3.8.1 but 3.8 (in Fedora at least)19:48
*** tonytan4ever has quit IRC19:48
crinkle~>3.8.0 will be latest 3.8.x19:48
*** tonytan4ever has joined #openstack-infra19:48
crinkleoh but that's just for puppet syntax check?19:49
*** marun has quit IRC19:49
nibalizertwo things in that change19:49
nibalizerI'm wondering about the export PUPPET_VERSION line19:49
EmilienMcrinkle: the one on top yes, on bottom is for unit19:49
EmilienMthis is for syntax-future19:49
nibalizerthe 3.8 in the job-group is known to DTRT19:49
*** smarcet has quit IRC19:50
*** sshnaidm has quit IRC19:50
EmilienMdo the right thing19:50
*** devvesa has quit IRC19:50
EmilienMI had to go on google19:50
electricallol :-)19:50
crinklenibalizer: for the export puppet gem version it will use latest 3.8.x19:50
EmilienMif one day french is official language, I'll make sure to speak like this too19:50
dhellmanncan anyone tell me if the puppet-apply-centos6 failure on https://review.openstack.org/#/c/189856/ is in a known issue? It looks like a failure because SELinux is disabled.19:50
zaroelectrical: darragh==pelix19:51
electricalzaro: thanks :-)19:51
EmilienMnibalizer: do you think we can drop some old releases?19:51
electricalI'm still testing against 3.2.x19:51
nibalizermaybe just keep whats in trusty?19:52
nibalizer3.4 iirc ?19:52
crinklei think that's right19:52
crinkleEmilienM: but probably in a different patch?19:52
EmilienManyway, we can wait a bit.19:52
EmilienMcrc32: oh for sure19:52
EmilienMcrinkle: ^19:52
nibalizerI gues one thing to watch out for is if you drop testing on 3.2 and 3.3 someone will eventually use contain() and then you can never go back19:52
crinklelet's just keep them all for now and then someday when puppet 5 is out we can drop all the 3.x :)19:54
anteayadhellmann: I'm sorry I have no idea19:54
EmilienMcrinkle: +119:54
*** maishsk_afk has joined #openstack-infra19:54
anteayanibalizer: do we expect puppet-apply to work on centos? it is a voting job19:54
nibalizerfungi: ya did you see that? puppet 5 is comming :D19:54
*** hashar has joined #openstack-infra19:55
EmilienMnibalizer: it's not Friday.19:55
openstackgerritMerged openstack/requirements: Raise minimums for oslo libs without namespace packages  https://review.openstack.org/18977519:55
funginibalizer: le sigh19:55
nibalizeranteaya: so I think we stand apart from most of openstack on this19:55
anteayanibalizer: do explain19:55
nibalizerhow I think it works, is dev writes feature for nova, then test, then those tests run on centos and ubuntu19:55
nibalizerwhat we do is we write some code and some tests, then flag it with 'test this on ubuntu' or 'test this on centos' or both19:56
nibalizerwe actually drill more and say precise and trusty19:56
openstackgerritOpenStack Proposal Bot proposed openstack-dev/hacking: Updated from global requirements  https://review.openstack.org/18682919:56
fungidhellmann: it was a timeout19:56
fungidhellmann: "Build timed out (after 30 minutes). Marking the build as failed."19:56
nibalizeranteaya: ideally we'd test all code on all operating systems, but getting the code to where it will work on those oses and the testing in place doesn't provide direct advantage19:56
nibalizerso the thinking is the community can add those things when they need them19:57
fungidhellmann: puppet remote operations seem to sometimes hang on centos 6 for no discernable reason. i can't reliably reproduce it to debug19:57
dhellmannfungi: http://logs.openstack.org/56/189856/1/check/gate-infra-puppet-apply-centos6/eccfa2a/console.html#_2015-06-09_19_20_27_17419:57
nibalizermake sense?19:57
dhellmannfungi: ok, I'll recheck if you think it's transient19:57
fungidhellmann: the selinux but is a benign error which puppet is reporting as an actual error19:57
fungier, selinux message19:58
*** sputnik13 has quit IRC19:58
dhellmannfungi: ok, this one looks interesting, too, but is probably not related: http://logs.openstack.org/56/189856/1/check/gate-infra-puppet-apply-centos6/eccfa2a/console.html#_2015-06-09_19_20_29_57619:58
*** rossella_s has quit IRC19:58
anteayanibalizer: yes, all except the voting part19:58
anteayanibalizer: if a test is expect to fail or pass based upon the work someone from that distro puts into it, in my mind it shouldn't be a voting job19:59
fungidhellmann: yep, that's just vcsrepo being noisy. happens every time a vcsrepo class is instantiated in the manifest19:59
*** rossella_s has joined #openstack-infra19:59
anteayanibalizer: as the author has no ability to address the failure19:59
*** baoli has joined #openstack-infra19:59
*** sputnik13 has joined #openstack-infra19:59
mordredzaro: I may not have actually done any work on that20:00
nibalizerfor puppet-apply and beaker-rspec, in both cases centos tests pass and can vote20:00
jeblairmordred, zaro: but it's also not blocking our 2.10 upgrade20:00
nibalizerthe testing coverage is a lower % than the ubuntu testing coverage20:01
nibalizerbut that doesn't prevent voting20:01
*** ajmiller has joined #openstack-infra20:01
jeblairthat coverage can expand as people put effort into it, but we're only starting with the minimal effort20:01
nibalizerso if centos-friendly community members want to enhance the covereage % they can do20:01
*** jamielennox|away is now known as jamielennox20:01
*** MarkAtwood has quit IRC20:03
fungizaro: mordred is only allowed to submit 5 specs per meeting, so he used up his quota for the week anyway20:03
anteayanibalizer: oh I see what you are saying20:04
*** puranamr has joined #openstack-infra20:04
anteayanibalizer: was just trying to help doug, which fungi has done/is doing20:04
*** ayoung has quit IRC20:04
zaromordred: ok, just didn't know where you were with that.  can help there if you need.20:05
*** puranamr has quit IRC20:05
*** fallenpegasus has quit IRC20:07
*** kmartin has quit IRC20:08
*** dimtruck is now known as zz_dimtruck20:11
*** baoli has quit IRC20:11
*** Rockyg has joined #openstack-infra20:11
notmynamefungi: did the jerasure patch land? if so, was it soon enough to catch the 1414 rebuild time?20:12
*** baoli has joined #openstack-infra20:12
*** achanda has quit IRC20:13
jktcan the Zuul's QueueItem have ever jobs from more than one pipeline at a time?20:17
*** kmartin has joined #openstack-infra20:17
jktlooking at model.py, it seems that it really cannot20:18
*** ayoung has joined #openstack-infra20:20
*** timcline has quit IRC20:20
*** enikanorov_ has joined #openstack-infra20:20
*** enikanorov__ has quit IRC20:20
*** hdd has quit IRC20:21
funginotmyname: looks like no, but i've gone ahead and exercised my "i'm a core reviewer and there's another +2 on my change" prerogative. i'll rebuild all the bare-trusty inages again here in a bit to incorporate it as soon as it merges20:21
*** freerunner has joined #openstack-infra20:21
*** simonmcc has quit IRC20:21
notmynamefungi: thanks20:21
*** ajmiller has quit IRC20:21
*** bradjones|away has quit IRC20:21
*** samueldmq has quit IRC20:21
notmynamefungi: can you ping me when the images are ready so I can queue some patches?20:22
*** boris-42 has quit IRC20:22
funginotmyname: i absolutely will20:22
*** tchaypo has quit IRC20:22
*** dizquierdo has joined #openstack-infra20:23
*** tchaypo has joined #openstack-infra20:23
*** bradjones has joined #openstack-infra20:23
*** bradjones has quit IRC20:23
*** bradjones has joined #openstack-infra20:23
*** simonmcc has joined #openstack-infra20:24
*** samuelBartel has joined #openstack-infra20:24
*** boris-42 has joined #openstack-infra20:24
*** freerunner has quit IRC20:25
openstackgerritIan Wienand proposed openstack-infra/devstack-gate: [WIP] run centos7 with neutron  https://review.openstack.org/17968920:26
*** timcline has joined #openstack-infra20:26
*** freerunner has joined #openstack-infra20:26
*** dizquierdo has quit IRC20:28
fungiflaper87: all the bare-trusty images have been updated now, and the new nodes i've checked have uuid-dev installed, so try again20:29
ianwcrinkle: are you familiar with puppet release schedules?  wondering when a change merged into stable to make f22 work will likely be available20:30
*** maishsk_afk has quit IRC20:30
flaper87fungi: awesome, thanks a lot!20:30
*** abregman has quit IRC20:30
ianwhttps://github.com/puppetlabs/puppet/pull/3993 in particular20:31
*** freerunner has quit IRC20:31
*** fallenpegasus has joined #openstack-infra20:31
jesusaurusnibalizer: i was looking at your change to pull o_p::server out of o_p::elasticsearch_node and its pretty messy having all that iptables logic in manifests/site.pp, whats your opinion of https://github.com/arusso/puppet-iptables to clean up the iptables?20:32
*** dannywilson has quit IRC20:32
*** dabos has quit IRC20:32
*** freerunner has joined #openstack-infra20:32
*** maishsk_afk has joined #openstack-infra20:32
*** dannywilson has joined #openstack-infra20:33
nibalizerwell idunno about that module, but YES YES YES PLZ YES to having application classes manage their own rules20:33
nibalizerI haven't dug into it too much20:33
*** patrickeast_ has joined #openstack-infra20:33
nibalizerbut I expect we can use our own iptables module in this way20:33
*** patrickeast has quit IRC20:33
*** patrickeast_ is now known as patrickeast20:33
*** markvoelker has joined #openstack-infra20:34
jesusaurusyeah, whatever we do, it would be great if we some puppet defines that let various modules add rules instead of a single class managing all the rules as a single file resource20:35
jesusaurusthat module I linked looks like theyve done most of the heavy lifting to get there and has recent activity20:36
openstackgerritJames E. Blair proposed openstack-infra/project-config: Update project-team-guide groups  https://review.openstack.org/18992620:37
*** miqui has quit IRC20:39
*** markvoelker has quit IRC20:39
nibalizerjesusaurus: so I think we can just do that20:39
*** TravT has joined #openstack-infra20:40
*** abitha has joined #openstack-infra20:40
mordrednibalizer: I believe we do not use puppet-iptables for specific reasons20:40
mordrednibalizer, jesusaurus: so - after the TC meeting, you should connect with jeblair20:40
*** zz_dimtruck is now known as dimtruck20:42
jesusaurusnibalizer: oh, huh, that would work?20:42
nibalizerjesusaurus: i mean in theory20:43
nibalizeri have a vm now gonna do some testing++20:43
fungisdague: mrodden: ianw: did you try out https://github.com/koalaman/shellcheck when embarking on bashate? looks like it's been around for a few years and is available in debian/ubuntu stable releases20:44
flaper87fungi: would you hate me if I come up with another patch to have swig?20:44
fungijust curious20:45
* flaper87 hides20:45
fungiflaper87: go for it. please propose an update to the bindep-fallback.txt file in project-config too though20:45
sdaguefungi: it does very different things20:45
flaper87fungi: awesome, thanks!20:45
sdagueit's a static analysis tool20:45
fungisdague: good point, not a style checker. so potentially complimentary20:46
*** dimtruck is now known as zz_dimtruck20:46
openstackgerritMerged openstack-infra/project-config: Make gate-cinder-pylint non-voting  https://review.openstack.org/18988520:46
*** maishsk_afk has quit IRC20:48
*** dizquierdo has joined #openstack-infra20:48
*** weshay has joined #openstack-infra20:49
crinkleianw: i have no information, you could ask kylo in #puppet-dev20:49
*** dizquierdo has quit IRC20:49
ianwfungi: i forget how bashate got embarked, but i wasn't there for that.  parsing looks "interesting"20:49
ianwcrinkle: thanks20:50
sdagueianw: it was a hack in the devstack tree, then people asked, a lot, to have it extracted into a dedicated tool20:50
sdaguewhich, is actually how we got hacking as well (it was a script in the nova tree for a while)20:50
nibalizerjesusaurus: lets play a game called 'i type a collection of symbols and we see if it is valid puppet'20:51
openstackgerritMerged openstack-infra/system-config: Add libjerasure dev packages on workers  https://review.openstack.org/18947220:51
*** patrickeast has quit IRC20:51
*** stevemar has joined #openstack-infra20:51
*** fallenpegasus has quit IRC20:52
greghaynesnibalizer: kleene closure coding is what we call that20:52
openstackgerritMerged openstack-dev/pbr: Revert "Remove sphinx_config.init_values() manual call"  https://review.openstack.org/18985120:53
*** samuelBartel has quit IRC20:53
ianwsdague: yeah, i mean just the initial hack part & what was considered.  i've long thought we needed more haskell :)20:54
*** dannywilson has quit IRC20:54
*** patrickeast has joined #openstack-infra20:55
*** dannywilson has joined #openstack-infra20:56
openstackgerritFlavio Percoco proposed openstack-infra/system-config: Add swig packages  https://review.openstack.org/18993620:56
sdagueianw: ikr?20:56
fungiianw: "requires 3gb to compile" is quite a testament to code quality for a command-line shell linter20:57
openstackgerritColleen Murphy proposed openstack-infra/project-config: Turn debug logging on for beaker jobs  https://review.openstack.org/18993720:57
*** puranamr has joined #openstack-infra20:57
fungiianw: oh, sorry, it was 2gb20:57
fungiso i exaggerate (slightly)20:57
mordredClint: nice20:58
EmilienMpuppet team want to enable debug to beaker jobs so we can debug Puppet catalogs easily: https://review.openstack.org/#/c/18993720:58
*** julim has quit IRC20:58
*** zz_dimtruck is now known as dimtruck21:00
EmilienMfungi: is there any particular reason to pin pbr>=0.8.2,<1.0 jenkins-job-builder ? can't we use 1.X ?21:00
EmilienMin https://github.com/openstack-infra/jenkins-job-builder/blob/master/requirements.txt I mean21:00
*** annegentle has quit IRC21:00
mordredjeblair, jesusaurus, nibalizer: so - iptables rules21:00
mordredand things in site.pp21:00
fungiEmilienM: i bet we could just fine and nobody has thought to propose it yet. give it a shot21:00
*** dprince has quit IRC21:00
mordredI am not following the details21:01
EmilienMfungi: ack21:01
*** baoli has quit IRC21:01
mordredbut I'd like to say that I'm wary of too much puppet in site.pp ... which is not to say you're going down the wrong path - just that I'm wary that we don't unwind too much and wind up with important information in site.pp21:01
*** aysyanne has quit IRC21:01
openstackgerritFlavio Percoco proposed openstack-infra/project-config: Add swig packages  https://review.openstack.org/18993921:02
mordred(I know it might be silly to say "wary of too much puppet in the main puppet file" ... but I think you know what I mean)21:02
*** xyang1 has quit IRC21:03
nibalizermordred: no i totes do21:03
nibalizerits sortof unfashionable for us to have such a verbose site.pp as it is21:03
nibalizerand I want those iptables things out21:03
mordredand I love fashion21:03
nibalizerwell when you're hip, you define fashion21:03
*** dguerri` is now known as dguerri21:04
mordredI'd honestly love for "list of remote servers" to be a parameter to the ES stuff21:04
openstackgerritEmilien Macchi proposed openstack-infra/jenkins-job-builder: requirements: pin pbr>=0.8.2,<2.0  https://review.openstack.org/18994021:04
flaper87fungi: done, added you to both. Thanks21:04
mordredand have the ES thing take that list and turn it into parameters to iptables or something21:04
nibalizerso yea my plan is for o_p::server to define the 'core set of iptables rules' and then the elasticearch role looks up in hiera the list of servers and pokes the right holes in the firewall21:04
jesusaurusmordred: yeah21:04
nibalizerright so you're gonna keep the state of things in openstack right? so we'll need a hiera->openstack backend (not hard)21:05
*** amitgandhinz has quit IRC21:05
mordredso long as the ES role has the logic of what to do with the data, which si the list of remote servers21:05
mordrednibalizer: aroo?21:05
ams__Any pointers on JJB and the Build Promotion plugin? I manually configured 1 promotion, I was hoping it would show up for all my jobs (which have the same name, from JJB) but no luck21:05
nibalizerand i'll need to refactor puppet-iptables to support external resources21:05
*** tiswanso has quit IRC21:05
mordrednibalizer: so - yes - but I don't think you need to refactor21:05
nibalizeri do21:05
mordrednibalizer: when ansible runs it will get the entire data set into the ansible inventory21:05
nibalizeri mean puppet-iptables module literall has this line hardcoded: -A openstack-INPUT -m udp -p udp --dport 161 -s -j ACCEPT21:06
mordredand can easily write out input params when it goes to run puppet21:06
mordredno need to have nodes be able to call back out to the openstack endpoint21:06
nibalizeruh okay21:06
mordredin fact, I thin kit would be highly undesirable for each node to know how to talk to openstack21:06
*** baoli has joined #openstack-infra21:06
mordredthe puppetmaste will have the creds for that21:06
mordredand will have already slurped all the data before the run21:06
nibalizeroh okay, so mabye ansible puts that data into yaml under /etc/puppet/hierdata ?21:06
mordredsomething like that is exactly what I was thinking21:07
mordred*details to be worked out later*21:07
nibalizer*waves hands*21:07
* mordred throws a bucket of snails on nibalizer's head and giggles21:07
nibalizerbut yea I'm gonna redo puppet-iptables to be concaty21:07
*** puranamr has quit IRC21:07
*** otter768 has joined #openstack-infra21:07
* mordred supports puppet-iptables being good-er21:08
nibalizerthen the ::elasticsearch or o_p::elasticsearch modules can just ask for iptables resources21:08
nibalizerunless we want to use a public iptables module like jesusaurus was suggesting21:08
clarkbplease no (I am not here but have to jump in). Exterbal resources are bad because it requires >1 run to converge and concat is so chatty and yyou cant reliably make changes (see haproxy)21:08
nibalizerclarkb: external resources are not exported resources21:08
greghaynesbahahaha, I was wondering how long until that happened21:08
clarkbwe dont need unreliable firewall updates21:08
pleia2clarkb :D21:09
nibalizerconcat 'chatty' ? no?21:09
clarkbnibalizer yes every single pieve seems to trigger an update21:09
clarkbtheb even if there is no diff at the end it makes a refresh21:10
ianwis anyone around who can look at the console of devstack-centos7-rax-dfw-2952112 ?21:10
clarkbwhich restarts your services wheb you dont want to21:10
clarkband it completely breaks haproxy21:10
mordredZOMG A clarb21:10
openstackgerritEmilien Macchi proposed stackforge/python-jenkins: requirements: pin pbr>=0.8.2,<2.0  https://review.openstack.org/18994221:10
anteayaclarkb: hello21:10
mordredZOMG A clarkb21:10
*** annegentle has joined #openstack-infra21:10
clarkbalso my 'b' key is very near my 'n' key21:10
nibalizerwell i am highly confident that I can make it not do that, so I'm happy to accept it as a requirement21:10
anteayaclarkb: I was wondering if no sleep was affecting your typing21:11
*** jerryz has quit IRC21:11
anteayaclarkb: I'm confident you have no sleep21:11
clarkbbasically firewall updates should be both atomic and deterministic and I think concat breaks both pieces?21:11
clarkbanteaya very little21:11
jeblairclarkb: hey, nice to see you!  congrats!21:11
*** otter768 has quit IRC21:12
nibalizerclarkb: sorry to pull you out of your cave21:12
clarkbeveryone else is asleep right niw so I can follow irc for a bit21:12
jeblairclarkb: this will last for... 5-10 minutes?21:12
mordredclarkb: we rewrote a bunch of things in ruby while you were gone, hope that's ok21:12
nibalizerclarkb: so I haven't done the work yet, but I would consider my implementation broken if it wasnt atomic and deterministic21:12
anteayaclarkb: you should be sleeping21:12
clarkbjeblair yes twins should be demanding food shortly21:12
anteayaclarkb: child sleeps you sleep21:12
*** ivar-lazzaro has quit IRC21:12
nibalizerso i think I'm in agreement with you on expected end state and I don't agree with you that concat is a bad approach21:13
clarkbanteaya its too hot to sleep :/21:13
crinklenibalizer: clarkb no one would use concat if it behaved that way21:13
jeblairanteaya: child talks on irc, you talk on irc?21:13
clarkbcrinkle then why is haproxy so broken?21:13
anteayajeblair: maybe?21:13
*** ivar-lazzaro has joined #openstack-infra21:13
*** jtomasek has quit IRC21:13
clarkbcrinkle you cannot make sane changes to haproxy config due to concat21:13
anteayahub_cap had his 2 year old trolling trove channel at one point21:13
jeblairclarkb: the weather here just went back to bay area standard summer (after a hot yesterday and morning today)21:13
EmilienMmordred, fungi, jeblair: puppet group is also ready to move our repos to OpenStack namespace, see https://review.openstack.org/#/c/176326/21:13
clarkbso instead you manually make changes and reload then wheb you can take a downtime you run concat +pupprt21:14
EmilienMcrinkle: ^21:14
*** ivar-lazzaro has quit IRC21:14
*** baoli has quit IRC21:14
nibalizerclarkb: are you talking about 'reload' ?21:14
*** ivar-lazzaro has joined #openstack-infra21:14
jeblairEmilienM: ah cool, can you add that to our meeting wikipage?21:14
clarkbnibalizer its related conxat refreshes when there is no diff21:14
clarkbnibalizer so restart hapoens21:14
EmilienMjeblair: sure thing21:14
nibalizeruhhhh idunno I can look at that, but it sounds like a bug in how we're using these things and not a bug baked into concat21:15
clarkbbut you dont kniw when it will do that because its updates that trigger refresh dobt appear deterministic21:15
clarkbat least they are not based on diff21:15
clarkbbecause diff is nil but refresh occurs21:15
sdagueso... anyone know why jenkins is periodically reporting "can't merge" on this - https://review.openstack.org/#/c/188148/ ?21:15
nibalizerclarkb: so I hear what you're saying21:15
EmilienMjeblair: like, do you want me to add the repos ?21:16
pleia2EmilienM: like the rest here https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Upcoming_Project_Renames21:16
pleia2old name > new name change-url21:16
EmilienMpleia2: I was not sure I could edit it, thanks21:16
pleia2ah, yes :)21:16
EmilienMpleia2: but infra needs to merge https://review.openstack.org/#/c/176326/ before all21:17
EmilienMoh, maybe they do that in one shot21:17
*** marcusvrn1 has joined #openstack-infra21:17
*** dimtruck is now known as zz_dimtruck21:17
*** zz_dimtruck is now known as dimtruck21:17
jeblairanyone can edit; that's the best and worst thing about wikis :)21:17
pleia2EmilienM: yeah, merging that change is part of the process we'll do on friday21:17
clarkbnibalizer basically I cant manually update + reload to avoid downtimw because cpncat refreshes when diff is nil. that tells me that its methods of refreshing are either not deterministic or completely unintuitive21:17
EmilienMpleia2: ack21:18
nibalizerclarkb: so i believe you that concat + openstack + haproxy is broken21:18
fungiianw: sorry, paying attention to the cross-project meeting and just saw your request. the console for that is blank (black) and not responding to carriage return keypresses. i suspect it's hung or resource-starved and the "display" has been blanked for inactivity or something21:18
clarkbnibalizer this is bad when you want firewall updates to happen in a clear manner21:18
nibalizerbut it doesn't have to be that way21:18
*** puranamr has joined #openstack-infra21:18
nibalizerclarkb: so I am happy to accept that any changes to iptables management must be 1) deterministic 2) atomic21:18
clarkbniablizer you could not subscrive to any concat changes21:19
nibalizerthose are easy requirements to meet21:19
clarkbbut otherwise hiw do you fix this without concat changes?21:19
*** peristeri has quit IRC21:19
ianwfungi: that's ok, figured most were away.  but yeah, i setup https://review.openstack.org/#/c/179689/ to send telemetry out, the last report it had 1,361,472 available ... so i don't think it's memory constrained21:19
nibalizerbecause you have an implementation bug somewhere clark21:19
nibalizerits not a problem in the library21:19
*** freerunner has quit IRC21:19
clarkbits a subscribe...21:19
*** alivigni has quit IRC21:19
*** Sukhdev has joined #openstack-infra21:19
clarkbthats nothing ti do with puppet-haproxy21:19
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632621:19
clarkbservice sibscribes cobfig which is managed by conxat21:20
crinkleclarkb: what you are seeing is either a bug in concat or a bug in the implementation, either way we can fix the bug, it's not a reason to avoid the 2nd most commonly used puppet library21:20
ianwfungi: i'm wondering if the network config has gone away21:20
*** baoli has joined #openstack-infra21:20
clarkbit is when it makes the specific thing you want to work not work...21:20
mordredyah. I'd suggest taht nibalizer makes a patch and we empirically test that it does not have the bad behavior that clarkb is worried about21:20
* anteaya believes clarkb is using his fat albert keyboard21:20
nibalizerclarkb: howabout I just build it then you can take it out and kick it a bunch and show me wrong?21:20
clarkbwhich is update config properly21:20
mordrednibalizer: yah. that21:20
pleia2EmilienM: it's also helpful if you keep the topic as project-rename (that's the series of patches we'll review to prep for the rename), you can change it through the web UI without a new upload if you want21:20
EmilienMpleia2: cool, doing it right now!21:21
pleia2EmilienM: thanks!21:21
*** annegentle has quit IRC21:21
*** rkukura_ has joined #openstack-infra21:21
ianwfungi: i had a syntax error in my telemetry anyway, i'm going to restart the whole thing21:21
*** ayoung has quit IRC21:21
*** rkukura has quit IRC21:22
*** rkukura_ is now known as rkukura21:22
*** annegentle has joined #openstack-infra21:22
openstackgerritIan Wienand proposed openstack-infra/devstack-gate: [WIP] run centos7 with neutron  https://review.openstack.org/17968921:22
*** pc_m has quit IRC21:22
*** tvanderwiel has quit IRC21:23
*** jerryz has joined #openstack-infra21:23
ianwfungi: ok, it's queued, i'll ping you in a bit and hopefully we can see if we can keep the console alive; or see if the console ever has anything on it...21:27
*** tim_o has quit IRC21:27
openstackgerritEmilien Macchi proposed openstack-infra/project-config: Puppet: move from Stackforge to OpenStack  https://review.openstack.org/17632621:28
fungiianw: unfortunately rackspace closes the virtual terminal session after a few minutes of no keyboard activity, so i'd have to sit in front of it and act like a dead-man switch21:28
j^2is there a readonly gerrit user?21:28
*** earlephilhower has joined #openstack-infra21:29
j^2it seems jenkins is a username21:29
*** puranamr has quit IRC21:29
anteayaj^2: jenkins is the account to report the job status via zuul21:29
anteayaj^2: why do you want a readonly gerrit user?21:30
j^2yeah, i’m trying to check out a patch and converge:21:30
j^2anteaya: ^^21:30
anteayayou can download any patch with git review -d <patchnumber>21:30
j^2oh nice21:31
clarkbnibalizer https://github.com/puppetlabs/puppetlabs-haproxy/blob/master/manifests/init.pp#L137-L138 fwiw21:31
j^2anteaya: thanks!21:31
asselingreghaynes, or other dib folks. Any idea how to fix this? http://paste.openstack.org/show/278912/21:31
ianwfungi: yeah, that sucks with the rax java thingy.  but i think i'd like to confirm *something* actually comes up to start with...21:31
fungiianw: sure21:31
*** JamesK has quit IRC21:32
fungithat's easy enough21:32
*** geoffarnold has quit IRC21:32
ianwasselin: old files in /opt/dib_cache/source-repositories/repositories_flock ?21:32
*** marcusvrn2 has joined #openstack-infra21:33
asselinianw, should I just delete that folder?21:33
*** marcusvrn1 has quit IRC21:33
asselinacutally...not there21:33
*** eharney has quit IRC21:34
*** freerunner has joined #openstack-infra21:35
*** puranamr has joined #openstack-infra21:35
ianwasselin: hmm, well that's the "real" error ... d-i-b can output a confusing amount after an error while it cleans up.  can you replicate it?  some reason the flock call fails ... multiple concurrent builds?21:35
ianwfungi: ok, devstack-centos7-rax-iad-2954595 is running21:36
asselinianw, my images stopped building 7-8 days ago, so should be reproducible. I will double-check there's nothing concurrent.21:36
*** marcusvrn2 has quit IRC21:37
*** ayoung has joined #openstack-infra21:37
fungiianw: yep, there's cloud-init spew on the console of that one right now21:38
fungiianw: and if i feed it a carriage return, i get a login prompt21:38
ianwfungi: ok, but does it put up a getty at any point?21:38
*** pikos has joined #openstack-infra21:38
fungiianw: yes (login prompt)21:38
ianwok, well that's a start... at least we know it was there21:39
fungicloud-init had simply scrolled the original one off the screen because of console logging21:39
ianwfungi: ok, so maybe disable consoleblank (in fact, that sounds like something we should do in the scripts ...)21:40
openstackgerritEmilien Macchi proposed openstack-infra/jenkins-job-builder: requirements: pin pbr>=0.8.2,<2.0  https://review.openstack.org/18994021:40
*** rakhmerov has quit IRC21:41
*** kgiusti has left #openstack-infra21:42
*** samueldmq has joined #openstack-infra21:42
ianwasselin: yeah, my guess is that there's an old dib or something hanging around that's got the flock() on the dir ... /proc/locks might help21:42
*** TravT has quit IRC21:43
fungiianw: asselin: one problem we've identified which can _look_ like that is if you accidentally sudo nodepool to create new diskimages rather than sudo -u nodepool nodepool21:43
greghaynesasselin: huh21:44
fungithen the root-owned files sit around in the cache and the nodepool user can't remove/replace them21:44
*** rakhmerov has joined #openstack-infra21:44
greghaynesasselin: so I dont think we guard against killed things that didnt remove their lock21:44
asselininteresting...well I see some dib processes running since jun 2....so probably need to kill those21:44
asselinis there a 'safe' way to do it?21:44
greghaynesah, or that21:44
greghaynesjust kill kill kill21:45
fungiasselin: also if the issue persists once there's no more processes running, try chowning all the contents of the dib cache to nodepool:nodepool (or whatever user/group your nodepoold runs under)21:45
*** mwagner_afk has joined #openstack-infra21:45
asselinfungi, ok thanks will do that too21:45
*** dustins has quit IRC21:47
fungiianw: that console is still responding for the moment, though i did discover that the rackspace console timeout is ~5 minutes before it disconnects my session21:47
ianwfungi: ok, thanks, it's still chugging along https://jenkins05.openstack.org/job/check-tempest-dsvm-centos7/40/console21:48
ianw*if* it fails, which is about 75% i think, it will be at a random spot in tempest (at least, i can't find any smoking gun as to what stops it, yet)21:49
*** mriedem has quit IRC21:49
*** puranamr has quit IRC21:50
*** puranamr has joined #openstack-infra21:52
openstackgerritLouis Taylor proposed openstack-infra/project-config: Add non-voting python34 job for glance_store  https://review.openstack.org/18995421:53
fungiianw: so now the console seems to be frozen for me21:54
fungiit hasn't blanked, it's still got the repeated login prompts from me hitting enter over and over, but it no longer responds to keypresses now21:54
ianwfungi: well tempest just started...21:55
ianwi am getting my telemetry from the host ok21:55
fungiianw: i disconnected and reconnected the console session, came right back to where it was frozen. still doesn't respond to keyboard input21:55
openstackgerritDavide Guerri proposed openstack-infra/shade: Add get_server_external_ipv4() to meta  https://review.openstack.org/18903021:56
ianwfungi: can you log in?  i can not think of anything that would even remotely fiddle with console type settings21:57
*** dboik has quit IRC21:58
*** pikos has quit IRC21:59
fungiianw: yeah, ssh still seems to work21:59
openstackgerritDoug Hellmann proposed openstack/requirements: Block oslo.messaging 1.12.0  https://review.openstack.org/18780822:00
*** freerunner has quit IRC22:00
openstackgerritDoug Hellmann proposed openstack/requirements: Block oslo.messaging 1.12.0  https://review.openstack.org/18780822:00
j^2so anteaya i was wrong. It turns out we’re attempting to do a query against gerrit22:01
j^2is there an account i can do a query against it to get the data?22:01
openstackgerritEmilien Macchi proposed stackforge/python-jenkins: requirements: pin pbr>=0.8.2,<2.0  https://review.openstack.org/18994222:02
openstackgerritJustin Shepherd proposed openstack-infra/project-config: Adding doc gate job for openstack/defcore  https://review.openstack.org/18995622:02
fungij^2 what data do you need, and from where do you need to obtain it?22:02
*** marun has joined #openstack-infra22:03
j^2fungi: it seems i need something like: https://gist.github.com/33cca91f182c0c15ded922:03
j^2our scripts take data from that and put it in our testing-co-stack thinngy22:04
fungij^2 which part of that do you need? we already expose some of that detail via envvars22:04
fungij^2: what testing co-stack? an integration testsuite?22:05
j^2yep. for all the chef cookbooks22:05
j^2we’ve basicly recreated devstack with the chef cookbooks22:05
asselincan I get +2/+A to this 1 line devstack-gate change (already has +2) https://review.openstack.org/#/c/188604/22:05
fungij^2 we already have a lot of integration testing capability built into our ci... for example devstack-based jobs don't query gerrit, they just get the info they need for what changes and branches they're integrating from environment variables22:06
*** freerunner has joined #openstack-infra22:06
*** hashar has quit IRC22:07
j^2can you show me some examples. It would be great for us not to have to reinvent the wheel22:07
fungij^2: and to answer your initial question, no our nodepool workers don't have accounts to log into gerrit to query it22:07
*** alexpilotti has quit IRC22:07
j^2awwww grrrrr that’s frustrating...22:08
*** annegentle has quit IRC22:08
*** baoli has quit IRC22:08
j^2any suggestings on how i can get around this?22:08
anteayaI think you're frustrated because you are trying to start in the middle22:09
anteayaat least this is how it looks to me22:09
anteayahave you spent any time looking at how our system works?22:09
*** jgrimm_out is now known as jgrimm22:09
anteayarather than trying to get around it?22:09
fungij^2: a simple example is http://git.openstack.org/cgit/openstack-infra/system-config/tree/tools/apply-test.sh22:09
j^2anteaya: I’ve tried the best i can, i’d love to know where i’m wrong and what I’ve missed22:10
fungij^2: that performs an integration test involving our ~60 infra puppet modules22:10
j^2oh nice22:10
anteayawell it is tough for me to help you since I don't know what you have done, what your current understanding is or what your goal is22:10
*** changbl_ has quit IRC22:11
j^2anteaya: agreed, that’s the problem, i don’t know what i don’t know, and i’m attempting to learn the best i can.22:11
anteayaj^2: what have you done so far to understand our system?22:11
fungij^2: the interesting parts are the clonemap.yaml and use of the zuul-cloner utility already installed on the workers22:11
anteayaI ask because I wonder how discoverable our docs are22:11
anteayaor as a new project what your method of understanding infra is22:11
jeblairyeah, lines 22-29 and 52-55 of that file22:12
fungij^2: a more complex example is https://git.openstack.org/cgit/openstack-infra/devstack-gate/ which does this for devstack, basically22:12
j^2anteaya: yeah i’m trying to process all this information one sec22:12
*** puranamr has quit IRC22:12
jeblairbut without zuul-cloner, so don't mimic devstack-gate :)22:12
EmilienMjeblair: I'm having hard time on python-jenkins & hacking: https://review.openstack.org/#/c/189942/22:13
*** bknudson has quit IRC22:13
fungiianw: my ssh session is now hung, and i see the last 24 lines of a kernel panic (i think) on the console22:13
j^2anteaya: I’ve only really learned stuff from ci.openstack.org and conversations with people here. greghaynes has been unbelievablly helpful too22:13
jeblairEmilienM: i can't help with hacking, sorry.  we don't use it for any infra projects22:14
*** evm_ has joined #openstack-infra22:14
jeblair(python-jenkins is not an infra project)22:14
j^2fungi: thanks, i’ll see what i can pull out of this for us22:14
EmilienMjeblair: ok but it's in test-requirements. Should I just drop it?22:14
anteayaj^2: okay have you found the infra manual in your travels?22:14
jeblairj^2: do you have a jenkins job already that you are working on or are you starting a new one?22:15
j^2jeblair: already have one22:15
j^2anteaya: i don’t think so, i don’t recall it at all22:15
jeblairEmilienM: we use hacking to get its dependencies, and then ignore H* rules in tox.ini22:15
*** tonytan4ever has quit IRC22:15
jeblairj^2: what's the job?22:15
fungiianw: and now the worker seems to be gone, so i guess nodepool deleted it after the job ended22:15
*** davideagnello has quit IRC22:16
*** freerunner has quit IRC22:16
j^2jeblair: the full job?22:16
j^2because basicly the job just calls rake22:16
j^2and we’re doing everything underneth it22:16
anteayaj^2: okay so somehow as a new openstack project having a link to the infra manual somewhere discoverable for you would be helpful22:16
j^2anteaya: nice!22:16
j^2i’ll try and track it down in a sec22:16
jeblairj^2: i'm asking for the name of the job so i can look at its configuration :)22:16
anteayaj^2: http://docs.openstack.org/infra/manual/22:17
*** evm has quit IRC22:17
j^2jeblair: gate-cookbook-openstack-common-chef-rake22:17
j^2anteaya: thanks!22:17
j^2i have some reading to do22:17
*** wuhg has joined #openstack-infra22:17
jeblairj^2: i can't find that job :(22:17
*** jsavak has joined #openstack-infra22:17
anteayaj^2: hopefully that helps with an overview22:17
j^2jeblair: https://review.openstack.org/#/c/188924 is the review that i’m working off22:18
j^2anteaya: any more information i can get, ie this, will signifgantly help me22:18
* docaedo is enjoying this "getting help" conversation (as someone who's about to ask for some infra help coincidentally)22:18
*** puranamr has joined #openstack-infra22:18
jeblairj^2: oh interesting; that job only runs on the cookbook-openstack-common repo22:19
*** davideagnello has joined #openstack-infra22:19
j^2yeah it’s our inital job before we spread it out to our other cookbooks22:19
fungiwow. no wonder i can't update images in ord because of over-quota error now. 43 aliens there according to nodepool22:19
jeblairj^2: makes sense22:20
jeblairj^2: for our puppet integration test (and similarly, for openstack integration tests with devstack-gate), there is a repo where we maintain a list of all the included projects22:20
jeblairj^2: eg http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules.env#n6122:21
jeblairj^2: and http://git.openstack.org/cgit/openstack-infra/devstack-gate/tree/devstack-vm-gate-wrap.sh#n3522:21
j^2jeblair, anteaya, fungi: basicly we have some rake tasks that can build everything for us. We wanted to check out the patch someone puts in, makes the change then run the rake task, then run Rally to confirm everything works as expected. that’s our ultimate goal at least.22:21
j^2jeblair: looking22:21
*** freerunner has joined #openstack-infra22:22
jeblairj^2: is there a repo that has, or will have, or can have, such a list of all of your chef repos that you want to test together?  if not, that's okay, we can keep it in jjb, it just means that changes have to go through project-config review.22:22
j^2that’s a great question22:22
fungij^2: in fact, the proposed change should already be checked out locally (if you use the gerrit-git-prep builder macro), so not sure why you need to try to find out what it is and check it out again22:22
jeblair(zuulv3 idea: provide this data to jobs based on the zuul config)22:22
j^2jeblair: i dont think we have any plan to have a “list” we have a chef-repo: https://github.com/stackforge/openstack-chef-repo that basicly does it all for us22:23
j^2fungi: interesting22:24
*** jsavak has quit IRC22:24
fungij^2: that's why i was originally trying to find out why you wanted your job to query gerrit22:24
*** jsavak has joined #openstack-infra22:24
j^2fungi: hmm, yeah oooh! yeah it’s inside the vm that we create that we want to check it out22:25
j^2and we get the data from the query to pass it to that22:25
openstackgerritEmilien Macchi proposed stackforge/python-jenkins: requirements: pin pbr>=0.8.2,<2.0  https://review.openstack.org/18994222:25
j^2it’s very “yo dawg” virtualization22:25
fungij^2: you create a vm inside a vm? do you have any way to pass things into that vm?22:25
j^2fungi: yep22:25
j^2or we don’t pass it per-se, we checkout the recheckout of the berksfile and build the vm with the changes22:26
j^2it would be soooo much easier with a whiteboard22:26
fungij^2: so there are definitely envvars present in the calling environment (your hypervisor layer i guess) which include the change number, a clone url to get it from, et cetera22:26
j^2oh nice!22:26
jeblairj^2: so here's what zuul provides, which i think you should take advantage of: it can set up a number of git repos that reflect the proposed future state of the world based on the changes currently in queue to be merged and user-specified dependencies (ie, Depends-On: headers).  if your job can avoid doing any git operations on its own, and simply use the git repos provided by zuul, it will do all that work for you22:26
j^2yeah that’s more or less what i’m looking for22:26
j^2jeblair: i think i understand this. i really need to read what anteaya sentme a bit ago to fill in gaps.22:27
jeblairj^2: moreover, it will perform the checkouts in the most optimized way, using local caches, etc, to ensure that your jobs succeed and avoid DoSing our git servers22:27
fungii guess in your case the trick, if you want to avoid git operations entirely, is having a way to copy local repos from the hypervisor environment into the guests22:27
jeblairfungi: yes that22:27
jeblair(copy, or else, bind-mount, etc)22:28
j^2yeah and that’s what the test_patch.rb does iirc22:28
j^2at least that’s what it does locally22:28
j^2test_patch = https://github.com/stackforge/openstack-chef-repo/blob/master/tools/test_patch.rb#L44-L6022:29
*** dizquierdo has joined #openstack-infra22:29
greghaynesj^2: glad to hear youre making progress :)22:29
j^2greghaynes: :D22:29
j^2re-creating devstack with chef, gotta love it22:29
greghaynesheh, im all for testing more ways of deploying22:30
j^2ditto :D22:30
jeblairj^2: right, we just need to remove the gerrit and git operations from that :)22:30
jeblairj^2: zuul-cloner will do it all for you22:30
j^2jeblair: ahhhhh22:31
*** bhuvan has quit IRC22:31
j^2ok, let me read up on somethings, and i’ll take another stab at it, then come back?22:31
anteayamorning jhesketh22:31
jeblairj^2: sure thing.  please pester me about it; this is complicated and there are not many examples of it, and it's important to get right.  :)22:31
j^2jeblair, anteaya, greghaynes, fungi: thanks for the advice/direction i’ve got a lot to process :D22:31
j^2jeblair: oh i will don’t worry22:32
anteayaj^2: it helps to have the bigger picture22:32
anteayafungi: my ssh connection to review-dev was closed by remote host22:32
jeblairi'm about to add a subsection to the infra manual about it too; ETA 10 minutes :)22:32
anteayafungi: not sure how long ago22:32
*** marun has quit IRC22:32
fungianteaya: mine too, a couple hours ago22:33
anteayais that significant?22:33
anteayashould I reconnect?22:33
fungianteaya: no idea if that's significant or not. i've been too swamped with other things to look further at it for now22:34
fungidoesn't look like the service was restarted though. running since 5 days ago22:34
anteayaI will reconnect and see what happens22:34
asselinfungi, could you review this 1 line?  https://review.openstack.org/18860422:34
fungiasselin: lgtm22:35
asselinfungi, YEAH! :)22:35
*** stevemar has quit IRC22:35
fungilooks like zuul is finally catching back up with the change backlog now that most of the day's firedrills are over22:36
*** evm_ has quit IRC22:42
*** HeOS has joined #openstack-infra22:42
alexsyipHi, we’ve been trying to setup a jenkins job for a tempest test and we ran into a failure while devstack is trying to clone noVNC.22:42
alexsyipYou can see the error at the bottom of this file: http://logs.openstack.org/15/186615/1/experimental/gate-congress-dsvm-api/d88f120/logs/devstacklog.txt.gz22:42
*** evm_ has joined #openstack-infra22:43
openstackgerritThomas Goirand proposed openstack/requirements: Tries removing the upper bound of SQLA  https://review.openstack.org/18996822:43
alexsyipEarlier in the log devstack is able to clone keystone, and cinder, etc, but noVNC had a problem.22:43
fungisdague: pymysql warning shot fired over the bow of the -dev ml just now22:43
*** ayoung has quit IRC22:44
alexsyipWe tried to figure out why, and it looks like ERROR_ON_CLONE is true for noVNC but false for the other repos.  Could this be the reason for the failure?22:44
openstackgerritJames E. Blair proposed openstack-infra/infra-manual: Add information about integration tests  https://review.openstack.org/18997022:45
jeblairokay 13 mins22:45
jeblairfungi, anteaya, j^2: ^22:45
j^2ha nice!22:45
anteayajeblair: thanks22:46
jeblairj^2: i'm particularly interested in knowing if that answers any of your questions :)22:46
j^2jeblair: absolutely, let me catch up make sure i get the venactuar correct, and then come back22:46
jeblairj^2: np, thanks22:47
j^2i think that _might_ be the answer, but i need to make sure i don’t go back into the weeds…I’ve been known to do that22:47
fungijeblair: raw version lgtm. i'm lazy so waiting for the docs-draft to come back so i can check postprocessed layout22:48
anteayaalexsyip: I'm looking at the most recent patch for the job: https://review.openstack.org/#/c/180726/3/jenkins/jobs/congress.yaml22:49
anteayaalexsyip: I don't see ERROR_ON_CLONE set22:49
alexsyipWe didn’ do anything with ERROR_ON_CLONE.22:50
alexsyipI could be wrong, but I looked at the log and saw [[ True = \T\r\u\e ]] and figured that was the test for ERROR_ON_CLONE.22:50
*** timcline has quit IRC22:51
alexsyipThat test was right before die 440 'Cloning not allowed in this configuration'22:51
fungianteaya: it's on by default to prevent cloning new repos on our job workers, because that has a tendency to break. i'm guessing something happened22:51
fungiwith the worker setup22:51
anteayafungi: yes, was trying to work backwards22:51
fungianteaya: i'm not sure off the top of my head how devstack or devstack-gate supplies novnc (presumably it does?). it's not a project we host, so has to be pulled in from somewhere22:53
fungibut i need to run away in moments to go get dinner22:53
anteayaif memory serves it has something to do with nova?22:53
jeblairfungi: i don't think it does22:53
anteayaI will see if I can run around in circles looking busy while you dine22:53
fungijeblair: in that case, interesting that it suddenly tried. maybe gate-congress-dsvm-api is a broken job definition and tries to obtain it?22:54
alexsyipThe call trace after the error does say it was trying to install nova.22:54
jeblairyes, i think it specifies that it wants it22:54
*** bhuvan has joined #openstack-infra22:54
jeblairbut we can not use it in the gate for $REASONS22:54
anteayasomehow it is reaching out to a github repo for it22:55
*** sarob has quit IRC22:55
jeblairbasically, no gate job should inclued it22:55
jeblairbut this one does22:55
fungi"$REASONS" including, not packaged anywhere, only available via github22:55
jeblairfungi: something like that i believe22:55
ianwfungi: did you manage to grab a screenshot or anything of that panic?22:55
jeblairanteaya, alexsyip: so remove the novnc enablement lines from that defn and should work22:56
fungiianw: no, because the session ended as i was looking at it (nodepool deleting the worker)22:56
anteayaalexsyip: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/congress.yaml#n2222:56
openstackgerritDavid Lenwell proposed openstack-infra/infra-specs: Spec to host refstack.org in infra  https://review.openstack.org/18820722:56
anteayajeblair: okay thanks22:56
fungiianw: and also it was only the last ~20 lines of a much, much, much longer system dump because this isn't a serial terminal it's an emulated vga monitor22:56
alexsyipok, I can remove it.22:56
fungiianw: basically worthless for debugging kernel panics22:57
fungiianw: compiling in a remote kernel debugger might be possible? would almost certainly need a custom worker though22:57
*** chlong-zzz has quit IRC22:58
fungiianw: more likely easier to recreate it interactively with whatever hacked up kernel we need to dump into a debugger or create a local corefile before rebooting (can linux do that? or am i spoiled from years of bsd bliss?)22:58
anteayaalexsyip: I have to go for a walk now, I will look for your patch when I return22:58
alexsyipok thanks!22:58
ianwfungi: i think figuring out a serial console might be the best first step; chances are once we get a handle what's going on there's a bugzilla22:59
mordredfungi, jeblair: I thought there was some discussion a while back about openstack forking or taking over novnc22:59
jeblairmordred: i don't see an entry for it in project-config22:59
docaedoI could use a litle help again :)  I read how to add a new server:23:00
docaedoThen I think fungi created a new repo, and I added stuff to it (puppet-apps_site)23:00
ianwnetconsole might be an option, never really used it on linux23:00
docaedoI tested that by cloning system-config, running install_puppet.sh and23:00
docaedoinstall_module.sh, then copied puppet-apps_site to23:00
docaedo/etc/puppet/modules/apps_site. then created apps_site/manifests/file.pp with:23:00
docaedoclass {'apps_site':23:00
docaedothen I ran:23:01
docaedopuppet apply --verbose --modulepath=modules:/etc/puppet/modules manifests/file.pp23:01
docaedoThat ran, and turned the VM I ran it on into the app-catalog server (which I23:01
docaedothink is exactly what it should do, right?)23:01
jeblairdocaedo: yep23:01
docaedoSo now I am trying to make https://review.openstack.org/#/c/187863/ right, and23:01
docaedofeel lost. My first question - are the contents of the puppet-apps_site repo at23:01
docaedosome point magically pullet in to system-config?  Or should my change to system23:01
docaedoconfig clone that repo at some point?23:01
openstackgerritAlexander Yip proposed openstack-infra/project-config: Remove novnc service from congress-dsvm-api jenkins job  https://review.openstack.org/18997223:02
*** thedodd has quit IRC23:02
openstackgerritwill soula proposed openstack-infra/jenkins-job-builder: Add support for the Naginator Plugin  https://review.openstack.org/17936623:02
*** ZZelle_ has quit IRC23:02
alexsyipanteaya: I uploaded a patch to gerrit.23:02
jeblairdocaedo: gimme one sec, i think i'm about to show you something really cool23:02
mordreddocaedo: you want to add an ntry here: http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules.env#n6123:02
mordredjeblair has you23:02
docaedosweet, thanks to all :) standing by...23:02
jeblairmordred: yeah, that's part of it, carry on23:02
jeblairmordred: oh, yeah, that's in that change already23:03
*** puranamr has quit IRC23:03
* mordred goes to look at the change23:03
openstackgerritJames E. Blair proposed openstack-infra/system-config: Add Apps Site to System Config  https://review.openstack.org/18786323:03
jeblairdocaedo: i made a one-line change to the commit message ^23:03
funginotmyname: okay, i finally managed to get the last of the bare-trusty images rebuilt after a few tries and some issues with the quota in that provider, but now it's taking forever to snapshot. until it does (i hope it does!) any rechecks will have a ~90% chance of running on a node where the jerasure lib/headers are present. i need to break away to get dinner before my wife kills me though23:04
*** hdd has joined #openstack-infra23:04
jeblairdocaedo: i told the system that change "Depends-On" the change to puppet-apps_site23:04
mordreddocaedo: it's mostly great23:04
jeblairdocaedo: which means that now, when zuul tests https://review.openstack.org/187863 it will include the change to apps_site in its testing, even though it hasn't landed yet23:04
jeblairdocaedo: (but it will make sure that it does land before the change to system-config lands, since that's the right order for those two)23:04
docaedoah ha!23:04
mordreddocaedo: the only thing you need is to do the comments here:  https://review.openstack.org/#/c/187863/6/modules/openstack_project/manifests/apps_site.pp23:04
*** puranamr has joined #openstack-infra23:05
mordreddocaedo: you've already got ::server in site.pp - so you can just delete it from that file23:05
funginotmyname: i take that back, looks like it just went ready, so you should be 100% set now unless there's still a handful nodes built from earlier images and haven't been used up from the pool yet23:05
docaedomordred: ok thanks, I think I follow23:05
mordred... I kinda thing you may not even need the openstack_project::apps_site since there is no content in it other than ::apps_site23:05
mordredjeblair: ^^ thoughts?23:05
* fungi disappears for a while23:05
ianwfungi: can you get me a rax centos7 host?  i can try with the crash-dump reporting turned on in the kernel.  i haven't actually replicated by hand, yet, but it's worth a try23:05
jeblairdocaedo: as you address the comments on the change as mordred suggests above, make sure you first download the latest patchset of the system-config change since i technically just changed it by modifying the commit message.  use "git review -d 187863" to do that.23:06
docaedojeblair: got it23:06
jeblairmordred, docaedo: yes, i think you can drop the openstack_project::apps_site file completely23:07
*** dimtruck is now known as zz_dimtruck23:07
jeblairdocaedo: we consider the openstack_project files to be legacy stuff that needs to be refactored; you are ahead of the curve on where we want our puppet to be, congratulations :)23:07
*** bhuvan has quit IRC23:08
*** otter768 has joined #openstack-infra23:08
docaedodocaedo: haha, funny23:09
*** fallenpegasus has joined #openstack-infra23:09
docaedoer. jeblair: funny (already been awake too long today!)23:09
mordreddocaedo: yah man. you're talking to yourself!23:09
*** notnownikki has quit IRC23:09
*** samueldmq has quit IRC23:11
*** otter768 has quit IRC23:13
*** Sukhdev has quit IRC23:14
*** Sukhdev has joined #openstack-infra23:16
*** ashleighfarnham has quit IRC23:18
*** ams__ has quit IRC23:20
*** markvoelker has joined #openstack-infra23:21
*** yamamoto has joined #openstack-infra23:21
*** markvoelker has quit IRC23:26
notmynamefungi: thanks. I just got out of a meeting, so I'll kick off a job23:28
*** mikedillion has joined #openstack-infra23:31
*** mikedillion has quit IRC23:32
*** dkranz has quit IRC23:33
*** stevemar has joined #openstack-infra23:33
*** mikedillion has joined #openstack-infra23:36
*** yamamoto has quit IRC23:37
docaedofungi: ok regarding the comments here: https://review.openstack.org/#/c/187863/6/modules/openstack_project/manifests/apps_site.pp23:37
docaedothat class is already at the bottom of site.pp file23:37
asselinjhesketh, you around?23:38
docaedofungi: and if I understand you right, you're saying I can entirely drop the file modules/openstack_project/manifests/apps_site.pp23:38
*** dizquierdo has quit IRC23:40
*** oomichi has quit IRC23:40
*** dguerri is now known as dguerri`23:40
*** ociuhandu has joined #openstack-infra23:41
asselinfungi, would now be a good time to merge 180624, 175571, 177373?23:41
*** Zhongjun has joined #openstack-infra23:42
jeblairthe mergers are idle now, so i have stopped all of them and am removing all of their git repos (i only cleaned up 2 of them earlier)23:42
*** harlowja has quit IRC23:42
jeblairthat should speed things up a lot23:43
*** harlowja has joined #openstack-infra23:43
jeblairand they are running again now23:43
*** dangers is now known as dangers_away23:44
asselinjeblair, what do you think? would now be a good time to merge 180624, 175571, 177373?23:45
asselinwell I guess nees another core...23:45
*** Sukhdev has quit IRC23:45
jeblairasselin: i think those are more beginning of day changes rather than end of day :)23:45
asselinwho's beggining of day? :)23:45
*** Sukhdev has joined #openstack-infra23:45
asselinok I'll ping tomorrow about it, unless jhesketh gets to it first :)23:46
jeblairasselin: SergeyLukjanov or fungi; we're entering the period of weak infra-root coverage23:46
docaedojeblair: ah, meant to ask you that question - looks to me like I can drop apps_site.pp, and the comment about adding that to site.pp file seems to already by addressed (which is why I am feeling confused)23:47
*** crc32 has quit IRC23:48
*** mikedillion has quit IRC23:48
jheskethasselin: I'm not infra root so I can't unroll things easily unfortunately23:48
*** freerunner has quit IRC23:48
*** sarob has joined #openstack-infra23:48
asselinjhesketh, ok, well then please review/comment and we can plan for tomorrow23:49
mordreddocaedo: yup23:49
*** bhuvan has joined #openstack-infra23:49
*** freerunner has joined #openstack-infra23:50
*** yamamoto has joined #openstack-infra23:50
jheskethasselin: will do23:50
mordreddocaedo: except you'll need to change manifests/site.pp to reference ::apps_site23:50
*** ivar-laz_ has joined #openstack-infra23:51
docaedomordred: thanks, so drop "openstack_project" from that last class right?23:51
docaedolike http://paste.openstack.org/show/279267/23:53
*** jsavak has quit IRC23:53
jeblairdocaedo: i think you need commit => 'master',23:53
jeblairdocaedo: but otherwise yes23:53
docaedojeblair: ok awesome, thanks23:53
jeblair(unless that's a default?)23:54
jeblairdocaedo: it is actually23:54
jeblairdocaedo: so that's not even necessary, your paste is fine as-is23:54
docaedojeblair: it is the default (that's probably why Clint asked about that line at one point)23:54
*** jsavak has joined #openstack-infra23:54
jeblairyep, listen to him not me :)23:54
*** ivar-lazzaro has quit IRC23:55
anteayaalexsyip: so you did, thank you23:55
*** dannywilson has quit IRC23:56
openstackgerritChristopher Aedo proposed openstack-infra/system-config: Add Apps Site to System Config  https://review.openstack.org/18786323:56
alexsyipanteaya: thanks for the review!23:56
anteayathank you23:57
*** rossella_s has quit IRC23:58
*** rossella_s has joined #openstack-infra23:58
*** bradjones has quit IRC23:58
notmynamefungi: looks like it worked https://jenkins07.openstack.org/job/gate-swift-python27/1269/console23:58
notmynamefungi: thanks again23:59

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