| fungi | asselin__: obviously that's going to need a nodepoold restart since it modifies the rpclistener | 00:00 |
|---|---|---|
| fungi | er, not nodepoold rather zuul scheduler daemon | 00:00 |
| asselin__ | ok | 00:01 |
| fungi | the brain-finger interconnect here is bug-ridden | 00:01 |
| *** dimtruck is now known as zz_dimtruck | 00:01 | |
| ctlaugh | asselin__, I'm working with your fork (on 12.04) and am getting some puppet module version warnings/errors about not being able to install something because a newer version is already installed. Are these a problem? I don't have specific module names on-hand atm. | 00:01 |
| *** emagana has quit IRC | 00:02 | |
| asselin__ | ctlaugh, I think those are ok. They started happending since the pupet module split | 00:02 |
| *** zz_dimtruck is now known as dimtruck | 00:02 | |
| *** abhirc has joined #openstack-infra | 00:02 | |
| asselin__ | ctlaugh, at least I haven't noticed any issues | 00:02 |
| *** markvoelker has quit IRC | 00:03 | |
| ctlaugh | asselin__, I'll disregard them for now, then. | 00:03 |
| fungi | ctlaugh: asselin__: yeah, we see it everywhere... even in our apply jobs http://logs.openstack.org/99/155099/2/check/gate-infra-puppet-apply-precise/f10743e/console.html#_2015-02-11_23_46_52_213 | 00:03 |
| *** dims__ has joined #openstack-infra | 00:04 | |
| *** ghostpl has quit IRC | 00:04 | |
| *** hdd has quit IRC | 00:05 | |
| ctlaugh | fungi, asselin__, yes - those are some of the same ones I'm seeing. | 00:06 |
| fungi | it's benign as far as we've been able to ascertain | 00:07 |
| asselin__ | ctlaugh, fungi, yup, b/c the script I use is the same one https://github.com/rasselin/os-ext-testing/blob/master/puppet/install_master.sh#L39 | 00:07 |
| *** baoli has quit IRC | 00:08 | |
| *** emagana has joined #openstack-infra | 00:08 | |
| *** bswartz has joined #openstack-infra | 00:09 | |
| openstackgerrit | Jay Faulkner proposed openstack-infra/project-config: Make devstack check ironic agent_ssh tempest job https://review.openstack.org/155117 | 00:10 |
| *** dangers is now known as dangers_away | 00:10 | |
| *** emagana has quit IRC | 00:12 | |
| *** dimtruck is now known as zz_dimtruck | 00:12 | |
| timrc | fungi, mordred Sorry I was helping my 3 year old daughter making "inbavatations" (invitations) -- to what, I don't know. | 00:14 |
| timrc | Reading up. | 00:14 |
| timrc | fungi, Oh yeah. Hi. | 00:14 |
| *** MarkAtwood has quit IRC | 00:14 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email base plugin. https://review.openstack.org/151948 | 00:15 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email SMTP Sender https://review.openstack.org/149792 | 00:15 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: [WIP] Email Send Cron Worker https://review.openstack.org/151416 | 00:15 |
| timrc | fungi, I um wasn't able to get my test to pass. I felt like my thinking was correct, but what's actually happening in the test puts that in question. | 00:15 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email Working Directory Utility Method https://review.openstack.org/151947 | 00:15 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email Outbox https://review.openstack.org/151413 | 00:15 |
| *** mrda has joined #openstack-infra | 00:15 | |
| *** mriedem has joined #openstack-infra | 00:15 | |
| *** achanda has quit IRC | 00:16 | |
| mordred | jlk: heyho | 00:16 |
| jesusaurus | timrc: i was also having trouble wrapping my head around the test and what i was expecting to happen | 00:16 |
| *** afazekas has quit IRC | 00:16 | |
| timrc | jesusaurus, I believe that I had 3 changes added to the gate pipeline with an A -> B -> C dependency and that by dequeing A all three would be dequeued | 00:16 |
| timrc | but that's not what's happening. | 00:17 |
| *** dalgaaf has quit IRC | 00:17 | |
| *** otter768 has joined #openstack-infra | 00:18 | |
| fungi | sdague: your testing spec has reached 100 inline comments! | 00:18 |
| *** tim_o has quit IRC | 00:24 | |
| *** tsg has quit IRC | 00:24 | |
| *** emagana has joined #openstack-infra | 00:26 | |
| clarkb | timc need to dequeue C | 00:26 |
| *** unicell has joined #openstack-infra | 00:27 | |
| *** r-daneel has joined #openstack-infra | 00:28 | |
| *** abhirc has quit IRC | 00:29 | |
| *** Longgeek has quit IRC | 00:30 | |
| clarkb | timrc C and B can still merge if C is the closest to HEAD | 00:30 |
| *** ddieterly has joined #openstack-infra | 00:30 | |
| *** claudiub has joined #openstack-infra | 00:31 | |
| *** tsg has joined #openstack-infra | 00:33 | |
| *** wenlock has quit IRC | 00:34 | |
| *** vigneshvar has quit IRC | 00:34 | |
| *** joesavak has quit IRC | 00:35 | |
| asselin__ | I'm having issues cloning from zuul on 14.04. I see this error in apache: [Wed Feb 11 16:29:59.345015 2015] [authz_core:error] [pid 9482:tid 139853983373056] [client 10.52.13.1:38324] AH01630: client denied by server configuration: /usr/lib/git-core/git-http-backend | 00:37 |
| asselin__ | but that configuration file seems to be binary...? | 00:38 |
| clarkb | thats a command not config | 00:38 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/system-config: Stop using the puppet-pip module https://review.openstack.org/153305 | 00:38 |
| clarkb | almost looks like something is misconfigured? | 00:38 |
| *** avozza is now known as zz_avozza | 00:39 | |
| asselin__ | clarkb, thanks for the hint | 00:39 |
| *** hdd has joined #openstack-infra | 00:39 | |
| *** amitgandhinz has joined #openstack-infra | 00:40 | |
| fungi | does our zuul puppet module do apache 2.4-isms for its vhost config yet? | 00:41 |
| clarkb | fungi likely no | 00:41 |
| fungi | might just not be getting served right, for example lacking a docroot which 2.4 needs even if you're not serving content out of it | 00:42 |
| jogo | so bug target openstack-gate not an openstack project | 00:42 |
| fungi | or the changes in how it does its grant allow/deny acls | 00:42 |
| jogo | mordred: can we get that fixed ? | 00:42 |
| fungi | jogo: did i not add it to openstack? i'll do that | 00:42 |
| nibalizer | fungi: did you get answers to your puppet-httpd questioins? | 00:42 |
| jogo | fungi: thanks | 00:42 |
| clarkb | jogo fungi ? | 00:42 |
| *** VijayTripathi has joined #openstack-infra | 00:42 | |
| jogo | clarkb: the LP groups | 00:42 |
| clarkb | oh | 00:43 |
| jogo | clarkb: openstack-gate is not associated with 'openstack' | 00:43 |
| fungi | nibalizer: not entirely, i just randomly peppered a vhost template with them (see 155099 for details) | 00:43 |
| nibalizer | kk | 00:43 |
| fungi | jogo: it's set as part of openstack now | 00:44 |
| clarkb | nibalizer did you see my -1 on your its all puppet3 change? | 00:44 |
| nibalizer | also, we currently use puppetlabs-apache, we plan to use openstack-httpd | 00:44 |
| nibalizer | clarkb: just getting to a computer now, will review | 00:44 |
| *** amitgandhinz has quit IRC | 00:44 | |
| jogo | fungi: thanks, retesting now | 00:45 |
| asselin__ | fungi, clarkb I added some, but I guess there's a bit more needed. https://review.openstack.org/#/c/141518/ | 00:45 |
| asselin__ | clarkb, got past that error and on to the next | 00:45 |
| nibalizer | we wont be switched over to httpd until the type changes to httpd::vhsot | 00:45 |
| fungi | asselin__: oh, right, i thought my question sounded familiar. deja vu | 00:45 |
| fungi | nibalizer: it's all about the v-shot | 00:46 |
| *** VijayTripathi has quit IRC | 00:46 | |
| *** otter768 has quit IRC | 00:46 | |
| clarkb | jogo you dug out from stable/ yet? | 00:47 |
| fungi | shovt? trying and failing to come up with good anagrams for vhost | 00:47 |
| jogo | clarkb: actually yes! | 00:47 |
| clarkb | jogo curious how multinode tempest full is going | 00:47 |
| jogo | clarkb: https://review.openstack.org/#/c/147451/ is working too | 00:47 |
| clarkb | also yay | 00:47 |
| *** patrickeast has quit IRC | 00:47 | |
| jogo | fungi: got another: bug target grenade not an openstack project | 00:47 |
| fungi | jogo: you need dtroyer or other members of grenade-core for that one | 00:48 |
| jogo | fungi: oh I am a grenade core | 00:48 |
| jogo | maybe not on LP | 00:48 |
| fungi | jogo: which seems to just be bcwaldon | 00:48 |
| nibalizer | fungi: in this case, and in most cases, srvname, vhost_name and name will all the be the same | 00:48 |
| fungi | jogo: https://launchpad.net/~grenade-core/+members | 00:48 |
| jogo | ugg | 00:48 |
| jogo | oh dtroyer is there too | 00:49 |
| fungi | nibalizer: yeah, that's what i figured, just didn't know if i should use one throughout or if they're differently adapted such that i should use different ones for different things | 00:49 |
| fungi | jogo: yeah, i meant in additiojn | 00:49 |
| jogo | fungi: heh, I'll split this patch up into two part and deal with this tomorrow | 00:50 |
| jogo | fungi: thanks | 00:50 |
| fungi | jogo: any time | 00:50 |
| *** abhirc has joined #openstack-infra | 00:51 | |
| nibalizer | fungi: i also agree with you on that patch that sni is the road to madness | 00:51 |
| *** markvoelker has joined #openstack-infra | 00:52 | |
| clarkb | asselin__ I will fix that image update nodepool change as soon as I get back home | 00:53 |
| clarkb | I see the problem | 00:53 |
| fungi | nibalizer: yep. we want to avoid wildcard certs for obvious security isolation reasons, and subjectaltname for multiple distinct subdomains is a pain to get issued | 00:53 |
| clarkb | thank you for giving it a good testing | 00:53 |
| asselin__ | clarkb, cool thanks | 00:53 |
| fungi | though i have to _asssume_ most modern browsers deal well with name-based virtual hosts matching subjectaltnames, and i know modern apache releases are fine serving things that way | 00:54 |
| fungi | it's just the ca end of things that makes it painful | 00:54 |
| fungi | in the "old" days i would just bind additional ip aliases to the interface. meh, clouds | 00:55 |
| openstackgerrit | Khai Do proposed openstack-infra/system-config: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155135 | 00:56 |
| zaro | clarkb: ^ | 00:56 |
| clarkb | zaro reviewing | 00:57 |
| *** markvoelker has quit IRC | 00:57 | |
| clarkb | zaro does puppet-gerrit support it yet? | 00:57 |
| zaro | ohh, looks like i need to make a depends-on connection. | 00:58 |
| *** mase_home has joined #openstack-infra | 00:58 | |
| zaro | i have it setup but didn't push yet. | 00:58 |
| clarkb | yup | 00:58 |
| *** dims__ has quit IRC | 00:58 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Make test_queries runnable as gating test https://review.openstack.org/150186 | 00:58 |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Add launchpad test to test_queries https://review.openstack.org/155138 | 00:58 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Use python3 compatible httpclient https://review.openstack.org/155139 | 00:59 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Use python3 compatible unicode types https://review.openstack.org/155140 | 00:59 |
| *** markvoelker has joined #openstack-infra | 00:59 | |
| openstackgerrit | Khai Do proposed openstack-infra/puppet-gerrit: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155142 | 01:01 |
| openstackgerrit | Khai Do proposed openstack-infra/system-config: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155135 | 01:02 |
| *** ChuckC has quit IRC | 01:02 | |
| *** ChuckC has joined #openstack-infra | 01:03 | |
| *** bhuvan has quit IRC | 01:03 | |
| zaro | clarkb: ok, try again. | 01:04 |
| *** camunoz has quit IRC | 01:04 | |
| *** dmorita has quit IRC | 01:04 | |
| clarkb | zaro the .pp file needs to accept the parameter too | 01:04 |
| clarkb | in puppet-gerrit | 01:04 |
| *** markvoelker has quit IRC | 01:04 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 01:06 | |
| *** lnxnut has joined #openstack-infra | 01:06 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Remove Puppet 2/Puppet 3 logic https://review.openstack.org/154788 | 01:07 |
| asselin__ | clarkb, what does this do exactly? https://github.com/openstack-infra/puppet-zuul/blob/master/templates/zuul.vhost.erb#L27 This is my current error message: AH00130: File does not exist: /usr/lib/git-core/git-http-backend/python-3parclient/info/refs | 01:07 |
| *** tsg has quit IRC | 01:07 | |
| fungi | asselin__: that's the smart git backend | 01:08 |
| *** dmorita has joined #openstack-infra | 01:09 | |
| asselin__ | this is the command: git clone --progress -o origin http://10.52.13.1/p//python-3parclient | 01:09 |
| *** ashleighfarnham has quit IRC | 01:10 | |
| *** asettle-mtg is now known as asettle | 01:10 | |
| *** zhiwei has joined #openstack-infra | 01:11 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 01:11 | |
| asselin__ | fungi, so that's expected? there's no file there | 01:11 |
| openstackgerrit | Khai Do proposed openstack-infra/puppet-gerrit: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155142 | 01:11 |
| *** signed8bit is now known as signed8bit_ZZZzz | 01:11 | |
| *** zhiwei has left #openstack-infra | 01:12 | |
| fungi | asselin__: yeah, that's a program which implements a git-over-http protocol rather than having apache serve the git files itself | 01:12 |
| *** davideagnello has quit IRC | 01:13 | |
| fungi | asselin__: and in the vhost you configure environment variables apache hands off to that backend so that it knows where to find your git repositories on the filesystem itself | 01:13 |
| fungi | asselin__: but to apache it looks more or less like those files are a "subdirectory" of that executable | 01:13 |
| *** bdpayne has quit IRC | 01:14 | |
| asselin__ | ok I see...so somehow those files didn't make it to the subdirectly, so [core:info] complained it's missing | 01:14 |
| fungi | asselin__: in the vhost, make sure that GIT_PROJECT_ROOT is set to the directory which contains python-3parclient | 01:16 |
| *** achanda has joined #openstack-infra | 01:16 | |
| asselin__ | SetEnv GIT_PROJECT_ROOT /var/lib/zuul/git/ | 01:16 |
| asselin__ | it's there: /var/lib/zuul/git/python-3parclient | 01:17 |
| fungi | then yeah, there's likely some new apache 2.4 related permissions problem there | 01:18 |
| fungi | anyway, i'm going mostly offline for the night, but good luck... hopefully you can run down the cause | 01:18 |
| *** achanda has quit IRC | 01:18 | |
| asselin__ | fungi, thanks. yeah I fixed a few permissions. seems there's a bit more | 01:19 |
| *** achanda has joined #openstack-infra | 01:19 | |
| jogo | clarkb: we are still seeing http://logs.openstack.org/04/136504/15/experimental/check-tempest-dsvm-neutron-aiopcpu/6263aca/ :/ | 01:19 |
| jogo | http://logs.openstack.org/04/136504/15/experimental/check-tempest-dsvm-neutron-aiopcpu/6263aca/console.html#_2015-02-11_00_22_57_878 | 01:19 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Use python3 compatible urllib https://review.openstack.org/155150 | 01:19 |
| jogo | we need to get that sorted out before we can think about gating | 01:19 |
| clarkb | jogo: I have yet to even debug it | 01:19 |
| clarkb | and yes we do | 01:19 |
| jogo | clarkb: do you know if we have a filed bug? | 01:20 |
| clarkb | I don't would be in storyboard under nodepool if so | 01:20 |
| clarkb | like I said I really haven't looked into it | 01:20 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Use python3 compatible unicode types https://review.openstack.org/155140 | 01:22 |
| jogo | clarkb: https://storyboard.openstack.org/#!/story/2000159 filed | 01:22 |
| clarkb | jogo: thanks | 01:22 |
| clarkb | jogo: I think it may be nodepool inadverdently cleaning up subnodes while the primary node stays around or it could be cloud issues | 01:23 |
| jeblair | we can eliminate or confirm the nodepool hypothesis quickly. i will do so. | 01:23 |
| jogo | clarkb: cool, maybe we can dig into this later this week | 01:23 |
| jogo | jeblair: thanks! | 01:23 |
| *** sputnik13 has quit IRC | 01:23 | |
| jogo | I think there is a unrelated neutron tempest failure we need to resolve as well. AFAIK those are the only two issues | 01:24 |
| *** amotoki has joined #openstack-infra | 01:27 | |
| jeblair | jogo, clarkb: nodepool deleted correctly. | 01:27 |
| jeblair | debug.log.2015-02-10_23:2015-02-11 00:24:03,070 DEBUG nodepool.NodePool: Deleting node id: 791563 which has been in used state for 0.114464105301 hours | 01:28 |
| jeblair | debug.log.2015-02-10_23:2015-02-11 00:24:09,390 DEBUG nodepool.NodePool: Deleting server 74406613-d73c-45fc-805d-cb939553b431 for subnode id: 152 of node id: 791563 | 01:28 |
| *** amotoki has quit IRC | 01:28 | |
| jeblair | debug.log.2015-02-10_23:2015-02-11 00:24:09,818 DEBUG nodepool.NodePool: Deleting server 005e1852-4228-4c75-b106-e30be448198c for node id: 791563 | 01:28 |
| jogo | could it be the second node isn't actually ready? | 01:29 |
| *** tqtran is now known as tqtran_afk | 01:30 | |
| clarkb | jeblair: how did it end up running a job then? | 01:30 |
| clarkb | oh or do you mean it deleted at a normal time and not when the job ran? | 01:32 |
| clarkb | asselin__: I almost have the fix ready for you | 01:32 |
| clarkb | ugh git why are you fighting me | 01:33 |
| asselin__ | clarkb, ok thanks. I think I got zuul working on trusty, this time....manually...still need to figure out a few more details | 01:33 |
| *** sdake has quit IRC | 01:33 | |
| jeblair | clarkb: yes, it deleted when the job was complete | 01:34 |
| *** dmsimard_away is now known as dmsimard | 01:35 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Remove unneeded Label.*_providers https://review.openstack.org/152757 | 01:35 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Better image checking in update_image command https://review.openstack.org/151749 | 01:35 |
| clarkb | asselin__: ^ that should properly check for the image being available by the provider, was relying on the image in use flag too much | 01:35 |
| clarkb | asselin__: curious to see what the apache changes are | 01:36 |
| asselin__ | <Directory /usr/lib/git-core> | 01:36 |
| asselin__ | Options +ExecCgi -MultiViews +SymLinksIfOwnerMatch | 01:36 |
| asselin__ | Require all granted | 01:36 |
| asselin__ | </Directory> | 01:36 |
| *** sarob has quit IRC | 01:36 | |
| asselin__ | sudo a2enmod cgi | 01:37 |
| JayF | Is there any way to get a devstack image rebuilt ahead of time? I'm looking at http://nodepool.openstack.org/ and it looks like Ironic's gate will be broken for ~14 more hours if there's not an image pushed | 01:37 |
| asselin__ | well...I'm working backwards now...but seems that's the minimum | 01:37 |
| *** zz_dimtruck is now known as dimtruck | 01:37 | |
| jeblair | jogo, clarkb: http://paste.openstack.org/show/171691/ | 01:38 |
| clarkb | JayF: iirc its broken due to devstack not any image | 01:38 |
| asselin__ | clarkb, do you know where to do "sudo a2enmod cgi" in the config? | 01:38 |
| jeblair | jogo: nodepool was able to ssh into it (which it does before declaring nodes and subnodes 'ready') | 01:38 |
| JayF | clarkb: I thought devstack was baked into the image? | 01:38 |
| JayF | clarkb: maybe I misunderstand? | 01:38 |
| *** Daviey has quit IRC | 01:38 | |
| clarkb | asselin__: ya puppet has an a2mod provider if you grep for that you shoud find examples | 01:38 |
| clarkb | JayF: nope not at all | 01:38 |
| *** sarob has joined #openstack-infra | 01:38 | |
| clarkb | JayF: devstack is self testing | 01:38 |
| JayF | oh, well time to recheck some things then. Thanks! | 01:38 |
| jogo | hmm we don't test 10.209.65.144 | 01:39 |
| *** asettle is now known as asettle-lurnch | 01:39 | |
| *** asettle-lurnch is now known as asettle-lernch | 01:39 | |
| clarkb | jogo: no becaue that may be in a different datacenter | 01:40 |
| *** markvoelker has joined #openstack-infra | 01:40 | |
| clarkb | so I guess its possibly a network thing in rax | 01:40 |
| jeblair | clarkb: they shouldn't be? | 01:40 |
| jeblair | clarkb: subnodes are supposed to come from the same provider | 01:40 |
| clarkb | jeblair: sorry different region than noepool | 01:41 |
| clarkb | jeblair: so nodepool can't ssh to that IP | 01:41 |
| jeblair | clarkb: ah yes, sorry :) | 01:41 |
| jeblair | clarkb: hpcloud nodes in particular are rarely in rax dcs. | 01:42 |
| clarkb | yup :) | 01:42 |
| *** sarob has quit IRC | 01:43 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 01:43 | |
| jeblair | i'd be really interested to interactively debug this, because on the surface there seems to be little reason that if ssh answers on the public ip, it would not on the private. | 01:44 |
| jogo | can we test using public IP to login to main node and from there use private to subnode? | 01:44 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/puppet-zuul: Now can clone from zuul using apache 2.4 https://review.openstack.org/155154 | 01:44 |
| jogo | although not sure if that is even the issue | 01:44 |
| jogo | anyway have to head out | 01:44 |
| asselin__ | clarkb, ^^ still need to test it though. I willl continue tomorrow & look at your nodepool change as well. I need to go home now. | 01:44 |
| clarkb | asselin__: ok thanks again. I will look at 155154 | 01:44 |
| *** dmsimard is now known as dmsimard_away | 01:45 | |
| jeblair | jogo, clarkb: also worth noting is that it was 5 minutes after the nodes were online before devstack even tried to ssh into the other one. so system startup timing seems highly unlikely. | 01:45 |
| *** dmsimard_away is now known as dmsimard | 01:46 | |
| *** ChuckC_ has joined #openstack-infra | 01:46 | |
| *** r-daneel has quit IRC | 01:46 | |
| clarkb | jeblair: ya maybe we should add some debug code to d-g where if that ssh connection fails then try public ip then if that fials try ping | 01:46 |
| clarkb | then sit in a holding pattern for human interaction | 01:47 |
| *** camunoz has joined #openstack-infra | 01:47 | |
| clarkb | asselin__: no rush, get home :) but I notice most of that chnage is ocmmented out which seems wrong. I will leave a note in the change | 01:47 |
| *** dmorita has quit IRC | 01:48 | |
| *** ChuckC has quit IRC | 01:49 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config: Zuul status: only count live changes https://review.openstack.org/155155 | 01:50 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Status: only count live changes https://review.openstack.org/155157 | 01:52 |
| *** ghostpl has joined #openstack-infra | 01:53 | |
| *** TheJulia has quit IRC | 01:54 | |
| *** dannywilson has quit IRC | 01:58 | |
| *** dannywilson has joined #openstack-infra | 01:58 | |
| jeblair | jhesketh: how important is it to include parameters in the status.json? | 02:01 |
| *** ghostpl has quit IRC | 02:01 | |
| jhesketh | jeblair: which parameters sorry? | 02:01 |
| jeblair | jhesketh: i was thinking we should try to get to the point where they are considered more secure (right now jenkins exposes them, so that's not the case) | 02:01 |
| jeblair | jhesketh: but if we clean that up, then we can pass more interesting things to the jobs | 02:01 |
| jeblair | jhesketh: oh sorry, i meant job parameters | 02:02 |
| jeblair | jhesketh: like queue['jobs'][0]['parameters'] | 02:02 |
| jeblair | jhesketh: i think you added it in the zuul client list jobs change | 02:02 |
| *** r-daneel has joined #openstack-infra | 02:03 | |
| jhesketh | Hmm, not sure. I don't /think/ they are used anywhere.. Somebody doing an odd deployment maybe but I doubt it | 02:03 |
| * jhesketh is in a meeting now sorry | 02:03 | |
| jeblair | jhesketh: np. i'll push a change and you can review at your leisure :) | 02:03 |
| *** camunoz has quit IRC | 02:04 | |
| jhesketh | Shiny | 02:04 |
| *** mtanino has quit IRC | 02:04 | |
| *** shashankhegde has quit IRC | 02:04 | |
| *** camunoz has joined #openstack-infra | 02:04 | |
| jhesketh | Oh my change added it? Interesting. Probably because of wanting to expose it via rpc and the json was a mistake (would have to check) | 02:05 |
| *** achanda has quit IRC | 02:06 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Remove job parameters from status.json https://review.openstack.org/155160 | 02:06 |
| *** melwitt has quit IRC | 02:06 | |
| *** claudiub has quit IRC | 02:06 | |
| clarkb | doesn't look like mordred got around to writing the interesting ansible change :) or at least didn't push it | 02:06 |
| *** melwitt has joined #openstack-infra | 02:06 | |
| jeblair | clarkb: he wrote at least part of it, did you see that? | 02:06 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Switch from mysql-python to PyMySQL https://review.openstack.org/152339 | 02:06 |
| mordred | clarkb: I wrote the front end - and the POC of the backend | 02:07 |
| mordred | I need to tie up a few other things though because I can clean up the POC and push it up | 02:08 |
| mordred | (I opened too many cans all at once today - need to close some of them) | 02:08 |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Reorder py27 test to happen after py3 tests https://review.openstack.org/155161 | 02:08 |
| mordred | clarkb: but I have the needed backend parts working on my laptop | 02:08 |
| mordred | just not integrated into this specific flow yet | 02:08 |
| clarkb | mordred: see my comment on the system-config change. I don't know that it is worth a -1 but I was explicitly trying to avoid the way you did it | 02:09 |
| mordred | looking | 02:09 |
| clarkb | jeblair: ya I saw the system-config changes | 02:09 |
| jeblair | clarkb: i think that's a good idea; it would be good to have project-config at the same ref globally. i'm okay if we baby-step there though :) | 02:10 |
| mordred | clarkb: nod. I want to achieve the same thing | 02:10 |
| *** nelsnelson has joined #openstack-infra | 02:10 | |
| openstackgerrit | greghaynes proposed openstack-infra/storyboard: Use python3 compatible httpclient https://review.openstack.org/155139 | 02:10 |
| clarkb | mordred: I think it belongs in run_all.sh then | 02:10 |
| jeblair | clarkb, mordred: same thing for system-config if we go masterless. | 02:10 |
| *** ashleighfarnham has joined #openstack-infra | 02:10 | |
| mordred | clarkb: I think run_All.sh should not exist | 02:10 |
| jeblair | mordred: because ansible should be able to do express that itself? | 02:11 |
| *** signed8bit_ZZZzz is now known as signed8bit | 02:11 | |
| mordred | yah | 02:11 |
| clarkb | mordred: we can collapse the playbooks and invoke ansible-playbook directly | 02:11 |
| mordred | yah | 02:11 |
| clarkb | mordred: BUT that seems like a completely different problem to solve :) | 02:11 |
| mordred | part of the problem is that we're doing this kinda half-way and kinda-backwards which makes it painful | 02:11 |
| clarkb | mordred: I dunno that I agree but ok | 02:12 |
| clarkb | mordred: I think the puppet role is the painful bit | 02:12 |
| mordred | so I'd like to spend a minute (probably on the plane in the morning) | 02:12 |
| *** achanda has joined #openstack-infra | 02:12 | |
| mordred | yeah - Iv'e got a fix for that too | 02:12 |
| clarkb | which aiui is very ansibley | 02:12 |
| mordred | so I think I'll hunker down in the morning and get you a set of patches of what's in my brainhole | 02:12 |
| clarkb | ok | 02:12 |
| jeblair | mordred: ++ | 02:12 |
| *** camunoz has quit IRC | 02:13 | |
| clarkb | mordred: also fwiw I would avoid calling it a sha simply because it could be any ref but I don't care too much about the color of that shed | 02:13 |
| mordred | but part of fixing the puppet-ansible thing is actually to (and I'm sorry for how ridiculous this is about to sound) make it an ansible-puppet repo ... | 02:14 |
| *** shashankhegde has joined #openstack-infra | 02:14 | |
| mordred | but that that repo is a standalone ansible role like you can register with galaxy similar to what we just did with our puppet module splitout | 02:14 |
| clarkb | I don't understand why that is part of the fixing :) | 02:15 |
| mordred | because then, similar to install_modules.sh - you can actually keep a list of ansible roles to be installed from git repos and whatnot in a yaml file | 02:15 |
| *** sdake has joined #openstack-infra | 02:15 | |
| *** Longgeek has joined #openstack-infra | 02:15 | |
| clarkb | mordred: I think you are conflating fix this race with make ansible more ansibley | 02:15 |
| mordred | there are 8 things I'm conflating | 02:15 |
| *** salv-orlando has quit IRC | 02:15 | |
| mordred | but I think I can put patches up for all of them in the morning because none are hard patches | 02:15 |
| mordred | I've just been putting them all off | 02:15 |
| mordred | but I think they'll help the "working on this feels janky" problem | 02:15 |
| * nibalizer chucles | 02:16 | |
| * mordred throws an aardvark at nibalizer | 02:16 | |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config: Zuul status: only display time on live changes https://review.openstack.org/155162 | 02:18 |
| openstackgerrit | Joshua Harlow proposed openstack/requirements: Bump six up to 1.9.0 https://review.openstack.org/155163 | 02:18 |
| nibalizer | mordred: clarkb so, could we just put project-config on AFS and only have one? | 02:19 |
| clarkb | nibalizer: no, because then anyone using this model would need a filesystem like afs which isn't really friendly | 02:20 |
| * nibalizer prepares to duck additional animals | 02:20 | |
| mordred | clarkb: but anyone using this model could actually totally just run from the copy of the repo that we're running from. not a clone of it - they could actually reference /afs/openstack.org/project-config in their code ... | 02:21 |
| mordred | it would be distributed synchonicity! | 02:21 |
| * mordred hands nibalizer a nicely plated pheasant | 02:21 | |
| clarkb | mordred: not really | 02:21 |
| clarkb | mordred: project_config is where you put your site specific things | 02:21 |
| mordred | clarkb: I belive nibalizer and I Are both trolling | 02:21 |
| clarkb | its the one bit thta you can't count on that for :) | 02:21 |
| nibalizer | I'm more serious than trolling | 02:21 |
| jeblair | nibalizer: i think mordred was trolling you | 02:22 |
| nibalizer | clarkb: your argument is interesting to me, im not exactly sure how I feel about it | 02:22 |
| *** signed8b_ has joined #openstack-infra | 02:22 | |
| greghaynes | I think the real answer is to put project-config in mongodb and have users replicate off of you | 02:22 |
| nibalizer | greghaynes: you're thinking of couch i think | 02:22 |
| * mordred throws a very shiny worst cat at greghaynes | 02:22 | |
| clarkb | greghaynes: isn't couch what npm uses? we should use that | 02:22 |
| nibalizer | clarkb: do you have additional arguments against using AFS? | 02:23 |
| clarkb | nibalizer: no other than that I think it would work well | 02:23 |
| mordred | greghaynes: http://33.media.tumblr.com/8d9778a4b1accd7445bb27fb1071d21c/tumblr_njmlnwPgqT1tlnaoto1_r3_400.gif | 02:23 |
| greghaynes | haha | 02:23 |
| *** dannywilson has quit IRC | 02:23 | |
| *** signed8bit has quit IRC | 02:24 | |
| mordred | clarkb: ultimately, once we get our crap together there - if we can make infra easy for people to reconsume (or even _doable_) then I have full faith that we could make AFS push-button for people | 02:25 |
| lifeless | sdague: mtreinish: just putting this out there, but why do we install tempest at all - if we had it already running somewhere we could just deploy the cloud and ask it to hammer it | 02:25 |
| *** signed8b_ is now known as signed8bit | 02:25 | |
| *** camunoz has joined #openstack-infra | 02:25 | |
| nibalizer | tempest as a service? | 02:25 |
| mordred | lifeless: you just described refstack I think? | 02:25 |
| lifeless | mordred: ish, yes | 02:25 |
| lifeless | mordred: except tied into CI, elasticsearch etcetc | 02:26 |
| lifeless | also I know nothing about how refstack manages its workers | 02:26 |
| lifeless | are they containers? vms? processes? etc | 02:26 |
| mordred | *shrug* | 02:27 |
| nibalizer | clarkb: so for the ambitious people consuming infra, they could set up AFS | 02:27 |
| *** mase_home has quit IRC | 02:27 | |
| nibalizer | othrewise they could set up NFS/Samba (which is probably already set up) | 02:27 |
| lifeless | SFS! | 02:27 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Status: only display time on live changes https://review.openstack.org/155165 | 02:28 |
| *** armax has quit IRC | 02:29 | |
| *** mase_home has joined #openstack-infra | 02:29 | |
| nibalizer | point is 'make this dir the same on all hosts' could be the requirement, not AFS, which deals AFS back into the game | 02:29 |
| *** woodster_ has quit IRC | 02:30 | |
| clarkb | nibalizer: sure and we already have tools to do that | 02:30 |
| clarkb | called ansible and puppet | 02:30 |
| nibalizer | and, AFS is there, its only a matter of time before something ends up hard requiring it, so we can push that conversation down the road, but we can't ignore it forever | 02:30 |
| clarkb | I don't think adding another requirement is necessary or advised | 02:30 |
| clarkb | nibalizer: wat | 02:30 |
| nibalizer | clarkb: well your point is basically 'we dont want to have a hard rrequirement that you must set up AFS in order to run infra' right? | 02:31 |
| nibalizer | and im saying 'it seems like that won't be true forever' | 02:32 |
| *** dims__ has joined #openstack-infra | 02:32 | |
| clarkb | nibalizer: no what I am saying is for this specific thing which knows how to configure stuff don't depend on afs to configure things | 02:33 |
| *** Daviey has joined #openstack-infra | 02:33 | |
| nibalizer | ok | 02:35 |
| *** sdake has quit IRC | 02:38 | |
| *** sarob has joined #openstack-infra | 02:39 | |
| *** hdd has quit IRC | 02:42 | |
| *** sarob has quit IRC | 02:44 | |
| *** ashleighfarnham has quit IRC | 02:45 | |
| *** masayukig has quit IRC | 02:46 | |
| *** masayukig has joined #openstack-infra | 02:46 | |
| mtreinish | lifeless: we've thought about that before for gating it would be interesting because of the overlap in development | 02:46 |
| mtreinish | nibalizer: that was something we've actually discussed before: https://etherpad.openstack.org/p/juno-summit-tempest-gui-client-server | 02:46 |
| timrc | clarkb, Hm. Okay. | 02:47 |
| lifeless | mtreinish: ok cool. If you want to rap about it someday let me know :) | 02:47 |
| *** otter768 has joined #openstack-infra | 02:47 | |
| mtreinish | nibalizer: and someone proposed a new design summit session in paris which was very similar to that | 02:48 |
| *** Longgeek has quit IRC | 02:48 | |
| mtreinish | lifeless: at least the last time I looked refstack deploys a container then installs and runs tempest from inside that container against a deployment somewhere else | 02:49 |
| mtreinish | or at least that was one way they did it | 02:49 |
| *** shakamunyi has joined #openstack-infra | 02:49 | |
| mtreinish | granted it's been almost a year since I looked at anything refstack | 02:50 |
| *** thingee has quit IRC | 02:50 | |
| *** rodrigod` has joined #openstack-infra | 02:51 | |
| *** loquacity has joined #openstack-infra | 02:51 | |
| *** shashankhegde has quit IRC | 02:52 | |
| *** gyee has quit IRC | 02:52 | |
| *** rwsu has quit IRC | 02:52 | |
| *** vponomaryov has quit IRC | 02:52 | |
| *** loquacities has quit IRC | 02:52 | |
| *** rodrigods has quit IRC | 02:52 | |
| *** otter768 has quit IRC | 02:52 | |
| *** spzala has quit IRC | 02:53 | |
| *** emagana has quit IRC | 02:53 | |
| *** emagana has joined #openstack-infra | 02:54 | |
| *** Longgeek has joined #openstack-infra | 02:55 | |
| *** ghostpl has joined #openstack-infra | 02:57 | |
| *** gyee has joined #openstack-infra | 02:57 | |
| *** TheJulia has joined #openstack-infra | 02:58 | |
| *** rwsu has joined #openstack-infra | 02:58 | |
| *** emagana has quit IRC | 02:58 | |
| zaro | i'm seeing this error "Error: Could not install module 'puppetlabs-stdlib' (v4.3.2)" | 02:59 |
| mordred | davidlenwell: ^^ we're talking about your stuff all up in here ^^ | 02:59 |
| zaro | anybody know about this? | 02:59 |
| zaro | log is here: http://logs.openstack.org/35/155135/2/check/gate-infra-puppet-apply-precise/1fc56d3/console.html | 02:59 |
| *** david-lyle is now known as david-lyle_afk | 02:59 | |
| *** woodster_ has joined #openstack-infra | 02:59 | |
| zaro | ohh maybe that's not the error, but this one: "Could not find init script or upstart conf file for 'exim4'" | 03:00 |
| clarkb | zaro: thats not it either | 03:01 |
| clarkb | zaro: http://logs.openstack.org/35/155135/2/check/gate-infra-puppet-apply-precise/1fc56d3/console.html#_2015-02-12_01_09_23_035 thats it, the thing I was trying to point out from the bus | 03:01 |
| *** ghostpl has quit IRC | 03:02 | |
| ctlaugh | asselin__, where do I point nodepool to my openstack install? Is it in etc/nodepool/nodepool.yaml, providers: auth-url? | 03:02 |
| zaro | clarkb: thanks. i'll look into that. what's up with all the errors in the log though? are those errors expected? | 03:03 |
| clarkb | zaro: they are "soft" errors that result from running in noop mode | 03:03 |
| clarkb | zaro: basically things that depend on side effects of puppet actually running but those don't occur so things error | 03:03 |
| clarkb | ctlaugh: yes | 03:03 |
| ctlaugh | clarkb, thank you | 03:04 |
| *** yamahata has quit IRC | 03:06 | |
| *** murkk has quit IRC | 03:08 | |
| *** MarkAtwood has joined #openstack-infra | 03:13 | |
| *** MarkAtwood has quit IRC | 03:14 | |
| *** vponomaryov has joined #openstack-infra | 03:15 | |
| *** salv-orlando has joined #openstack-infra | 03:16 | |
| *** shashankhegde has joined #openstack-infra | 03:17 | |
| *** otter768 has joined #openstack-infra | 03:19 | |
| *** salv-orlando has quit IRC | 03:21 | |
| clarkb | mordred: ok I ranted a greghaynes enough to better understand ansible a bit. Also I grepped through their source code | 03:21 |
| mordred | clarkb: ohmy | 03:21 |
| clarkb | mordred: please tell me one of the things you want to do is remove the puppet module and just have a puppet role that does the right things? | 03:21 |
| mordred | ansible role | 03:21 |
| mordred | yes | 03:22 |
| clarkb | er ya | 03:22 |
| clarkb | so remove the special python sauce and rely on ansible primitives | 03:22 |
| mordred | NOOOOOOO | 03:22 |
| clarkb | no? | 03:22 |
| mordred | because puppet is broken | 03:22 |
| clarkb | mordred: I think we can fix that | 03:22 |
| clarkb | using primitives | 03:22 |
| mordred | and handling the error codes in anbisle itself is just bananas | 03:22 |
| *** nelsnelson has quit IRC | 03:22 | |
| mordred | we can | 03:22 |
| mordred | it'll just be uglier than the python code | 03:22 |
| *** melwitt has quit IRC | 03:22 | |
| clarkb | mordred: I don't think so | 03:22 |
| * mordred is willing to be proven wrong | 03:22 | |
| clarkb | you just register the puppet commmand as a var then check the return code value | 03:23 |
| greghaynes | why not register the return code and then operate on it separately? | 03:23 |
| clarkb | greghaynes: y aexactly | 03:23 |
| mordred | yeah - but theyn you're doing all of the logic about the return code in yaml | 03:23 |
| mordred | and we've already written that in python | 03:23 |
| clarkb | mordred: except with it in python we end up needing to do a bunch of other things | 03:23 |
| mordred | like? | 03:23 |
| greghaynes | you could still operate on the registerd value in a module | 03:23 |
| clarkb | mordred: like understand environment ourselves | 03:23 |
| *** doude has quit IRC | 03:23 | |
| mordred | we don't need ot do anthing with the environment | 03:24 |
| clarkb | mordred: you do for puppet facts | 03:24 |
| mordred | nope. | 03:24 |
| clarkb | but back up | 03:24 |
| mordred | I've already got that sorted | 03:24 |
| mordred | wait - let me push up the next patches and respond to them | 03:24 |
| clarkb | just have a second task that does command test {{return.rc}} != 0 && {return.rc != 2 | 03:24 |
| clarkb | er lost some {}s but you get the gist | 03:25 |
| clarkb | mordred: how are you setting puppet facts without env vars? | 03:25 |
| *** nelsnelson has joined #openstack-infra | 03:25 | |
| mordred | sure - but then your'e doing jinja + yaml for complex logic - and one of the nice things about ansible is that it's super easy to drop in new python code to handle actual logic and let the ansible yaml be about clear and clean lists of tasks to perform | 03:25 |
| greghaynes | or even echo "0 2 ..." | grep {{ return.rc }} | 03:25 |
| *** doude has joined #openstack-infra | 03:26 | |
| clarkb | mordred: right but I think the cost is lower to do that rc thing once | 03:26 |
| clarkb | mordred: than it is to carry a 200 line python file | 03:26 |
| mordred | ok. I don't want to die on a hill on that point | 03:26 |
| mordred | but I don't want to get rid of it because passing env vars in from a wrapper shell script is hard | 03:26 |
| mordred | if there is a DIFFERENT reason to ditch it, ok | 03:26 |
| mordred | but that particular flow is almost never the thing we want to do | 03:26 |
| mordred | that said - I agree, we can totaly re-do the puppet role to be a pure-yaml puppet role | 03:27 |
| clarkb | mordred: for me the biggest issue is the ansible programming api is really not well documented | 03:27 |
| mordred | and it might be easier to grok | 03:27 |
| mordred | OH | 03:27 |
| clarkb | mordred: so we have this script | 03:27 |
| mordred | it's the best part about ansible | 03:27 |
| clarkb | mordred: and I have very little idea of what it is doing or how to extend it or how to fix it | 03:27 |
| mordred | it's SO GOOD | 03:27 |
| clarkb | mordred: I really disagree | 03:27 |
| clarkb | at least from a reading docs on a bus perspective I wanted to rage quit | 03:28 |
| mordred | nononoono | 03:28 |
| mordred | screw the docs | 03:28 |
| clarkb | there is a ton of hand waving around the things that are actually important | 03:28 |
| mordred | they need to be fixed | 03:28 |
| *** tqtran_afk has quit IRC | 03:28 | |
| greghaynes | heh, I do really like it, IMO you need a reason for the added complexity not a reason against having it | 03:28 |
| *** EmilienM is now known as EmilienM|afk | 03:28 | |
| mordred | that's why I'm saying - Iv'e been deep in that world and the API is actually great | 03:28 |
| mordred | because it's dead simple | 03:28 |
| clarkb | mordred: ok | 03:28 |
| mordred | the biggest problem is grokking that it's dead simple- and there are about 5 thigns that need better docs | 03:29 |
| mordred | and I should be ashamed because I've learned them all and not fixed the docs | 03:29 |
| clarkb | mordred: well I identified at least 3 things. What ansibles metavars things are, where they apply, and how they apply | 03:29 |
| *** sarob has joined #openstack-infra | 03:29 | |
| clarkb | mordred: eg environment: which when digging into environment: I found out there is massive list of these things | 03:29 |
| clarkb | serial, any_errors_fatal, no_log, max_fail_percentage etc | 03:30 |
| mordred | yes. those need much better docs | 03:31 |
| clarkb | but also AnsibleModule | 03:31 |
| clarkb | I couldn't find docs on its api, haven't gone spelunking to figure it out yet now that I am not on a bus | 03:32 |
| mordred | well - you should look at my openstack modules patches rather than going full spelunking | 03:32 |
| clarkb | in particular I was tring to figure out if run_command is effected by environment: | 03:33 |
| clarkb | through inference I think it is | 03:33 |
| mordred | it is not | 03:33 |
| clarkb | well thats bonghits | 03:33 |
| mordred | because it's running somewhere else | 03:33 |
| mordred | no. it's COMPLETELY the sane thing | 03:33 |
| clarkb | mordred: no environment: sets the env | 03:33 |
| mordred | oh | 03:33 |
| mordred | I missed the : | 03:33 |
| jhesketh | jeblair: back home now, will look at all the zuul changes shortly | 03:34 |
| clarkb | mordred: what if, and tell me if this is worse | 03:34 |
| clarkb | mordred: we just extend shell/command to accept a set of return codes and reject others | 03:34 |
| clarkb | mordred: that is bonghits but processes run that way apparently | 03:34 |
| jhesketh | mordred, jeblair: any chance you guys have had an opportunity to give thought to the bower mirror stuff? I've updated the spec here: https://review.openstack.org/#/c/154297/ | 03:35 |
| clarkb | mordred: so rather than a puppet module its just command: rc_ok=[0,2] | 03:35 |
| mordred | clarkb: yeah - you can totally do that - I just don't understand which problem you're solving | 03:36 |
| clarkb | mordred: I don't think we need a puppet module | 03:36 |
| *** salv-orlando has joined #openstack-infra | 03:36 | |
| clarkb | mordred: I think we should solve this purely with existing primitives. The puppet module is just checking return code sand setting default values | 03:36 |
| clarkb | setting default values is easy in ansible as is leaving us with just the error checking | 03:36 |
| mordred | (just to be clear, I originally wrote it with pure primitives) | 03:37 |
| *** dims__ has quit IRC | 03:37 | |
| mordred | and then rewrote it in python because it felt cleaner | 03:37 |
| mordred | so yes, you can do that and *shrug* as to whether it's better - if you like it better the other way, cool! | 03:37 |
| clarkb | mordred: well 99% of my confusion the last 2 days would go away if I didn't have to tackle the api this script was written against | 03:38 |
| clarkb | mordred: so maybe the other thing is to just go learn the damn api via spelunkin | 03:38 |
| *** melwitt has joined #openstack-infra | 03:39 | |
| *** salv-orlando has quit IRC | 03:40 | |
| *** unicell has quit IRC | 03:40 | |
| clarkb | mordred: also earlier you implied you weren't setting a puppet fact via the environment? how are you passing into puppet? | 03:40 |
| mordred | puppet can also read facts from yaml/json files | 03:41 |
| *** salv-orlando has joined #openstack-infra | 03:41 | |
| greghaynes | heh, this is exactly the evolution the tripleo heat+puppet integration had | 03:41 |
| clarkb | mordred: so you want to write to a file instead of setting the env? | 03:41 |
| clarkb | mordred: that seems wrong to me but ok | 03:41 |
| mordred | clarkb: yes - because I ultimately want to potentially set 50 settings | 03:41 |
| mordred | or somethign | 03:41 |
| clarkb | mordred: right but they can leak | 03:42 |
| mordred | aroo? | 03:42 |
| clarkb | whereas chances of the env doing that are smaller | 03:42 |
| clarkb | mordred: you have to be very careful when running puppet to currate those files | 03:42 |
| mordred | so - if you do env, you will be unable to run puppet without doing it from the master host via ansible | 03:43 |
| clarkb | wheras if I neglect to set an env var it should error on undef | 03:43 |
| mordred | if you write out root protected files | 03:43 |
| mordred | then for debug cycle you can TOTALLY just shell onto the box and run puppet apply | 03:43 |
| clarkb | mordred: you have no garuntee those files are correct though | 03:43 |
| clarkb | mordred: I get what you are saying but I think bailing out when you haven't explicitly said what to do is better | 03:43 |
| mordred | they will be as correct as the last time puppet ran on the bix | 03:43 |
| mordred | clarkb: I think we need to have beer to dive in to this :) | 03:44 |
| clarkb | probably I just opened one :) | 03:44 |
| clarkb | mordred: I think this would be better because when I run by hand usually its not with the last puppet run in mind, its some new config or testin new puppet etc | 03:45 |
| mordred | clarkb: nod | 03:45 |
| mordred | well - my food just arrived- so I will contineu later... | 03:45 |
| clarkb | mordred: kk | 03:46 |
| clarkb | also http://docs.ansible.com/developing_modules.html#common-module-boilerplate is the section in need of an update | 03:46 |
| clarkb | "There are also other useful functions in the module class, such as module.sha1(path). See lib/ansible/module_common.py in the source checkout for implementation details." its python autodoc please | 03:47 |
| clarkb | mordred: let me check resolution order on facts though. it will probably work ok if env vars override files | 03:48 |
| clarkb | then a manual run is same as last time, but can be easily modified to test something new | 03:48 |
| *** asettle-lernch is now known as asettle | 03:48 | |
| *** melwitt has quit IRC | 03:51 | |
| *** teran has joined #openstack-infra | 03:52 | |
| *** yamahata has joined #openstack-infra | 03:53 | |
| *** mriedem has quit IRC | 03:55 | |
| clarkb | mordred: at least between external dir facts and env facts env seems to win | 03:56 |
| clarkb | so this may work | 03:56 |
| *** salv-orlando has quit IRC | 03:57 | |
| *** shashankhegde has quit IRC | 03:57 | |
| *** Mr0rris0 has quit IRC | 03:58 | |
| *** TheJulia has quit IRC | 03:58 | |
| *** achanda has quit IRC | 03:59 | |
| clarkb | mordred: this seems to work with the default external facts location dir too | 04:00 |
| *** nelsnelson has quit IRC | 04:00 | |
| *** gyee has quit IRC | 04:00 | |
| clarkb | mordred: which is /etc/facter/facts.d for root | 04:00 |
| *** salv-orlando has joined #openstack-infra | 04:01 | |
| *** achanda has joined #openstack-infra | 04:02 | |
| *** armax has joined #openstack-infra | 04:04 | |
| *** salv-orlando has quit IRC | 04:05 | |
| *** harlowja is now known as harlowja_away | 04:08 | |
| *** TheJulia has joined #openstack-infra | 04:09 | |
| *** achanda has quit IRC | 04:10 | |
| *** Longgeek has quit IRC | 04:10 | |
| *** otter768 has quit IRC | 04:13 | |
| *** BharatK has joined #openstack-infra | 04:17 | |
| *** bhuvan has joined #openstack-infra | 04:17 | |
| *** doude has quit IRC | 04:18 | |
| *** doude has joined #openstack-infra | 04:20 | |
| *** Longgeek has joined #openstack-infra | 04:25 | |
| openstackgerrit | Joe Gordon proposed openstack/requirements: Bump openstackclient version min 1.0.2 https://review.openstack.org/155182 | 04:25 |
| *** Longgeek has quit IRC | 04:26 | |
| *** tiswanso has quit IRC | 04:27 | |
| *** sdake has joined #openstack-infra | 04:27 | |
| *** doude has quit IRC | 04:30 | |
| *** doude has joined #openstack-infra | 04:30 | |
| *** bhuvan has quit IRC | 04:34 | |
| *** ayoung has quit IRC | 04:34 | |
| *** barra204 has joined #openstack-infra | 04:35 | |
| *** salv-orlando has joined #openstack-infra | 04:37 | |
| *** ivar-lazzaro has quit IRC | 04:38 | |
| *** shakamunyi has quit IRC | 04:39 | |
| *** sileht has quit IRC | 04:40 | |
| *** salv-orlando has quit IRC | 04:41 | |
| *** ashleighfarnham has joined #openstack-infra | 04:41 | |
| *** ghostpl has joined #openstack-infra | 04:46 | |
| *** dmsimard is now known as dmsimard_away | 04:47 | |
| *** harlowja_at_home has joined #openstack-infra | 04:49 | |
| *** Alexandra_ has joined #openstack-infra | 04:49 | |
| *** doude has quit IRC | 04:50 | |
| *** ghostpl has quit IRC | 04:51 | |
| *** doude has joined #openstack-infra | 04:52 | |
| *** BharatK has quit IRC | 04:52 | |
| *** barra204 has quit IRC | 04:53 | |
| *** r-daneel has quit IRC | 04:54 | |
| *** hdd has joined #openstack-infra | 04:54 | |
| *** abhirc has quit IRC | 04:59 | |
| *** salv-orlando has joined #openstack-infra | 04:59 | |
| *** salv-orlando has quit IRC | 05:02 | |
| *** salv-orlando has joined #openstack-infra | 05:05 | |
| *** barra204 has joined #openstack-infra | 05:07 | |
| *** signed8b_ has joined #openstack-infra | 05:08 | |
| *** signed8bit has quit IRC | 05:08 | |
| *** BharatK has joined #openstack-infra | 05:08 | |
| *** sarob has quit IRC | 05:09 | |
| *** garyh has quit IRC | 05:09 | |
| *** woodster_ has quit IRC | 05:10 | |
| *** bhunter71 has quit IRC | 05:11 | |
| *** salv-orlando has quit IRC | 05:12 | |
| *** koolhead17 has joined #openstack-infra | 05:13 | |
| *** ashleighfarnham has quit IRC | 05:14 | |
| *** salv-orlando has joined #openstack-infra | 05:15 | |
| *** signed8b_ has quit IRC | 05:18 | |
| *** harlowja_at_home has quit IRC | 05:18 | |
| *** mwagner_lap has quit IRC | 05:19 | |
| *** coolsvap has joined #openstack-infra | 05:19 | |
| *** coolsvap has quit IRC | 05:21 | |
| *** sputnik13 has joined #openstack-infra | 05:21 | |
| *** dimtruck is now known as zz_dimtruck | 05:22 | |
| *** coolsvap has joined #openstack-infra | 05:23 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 05:24 | |
| *** amotoki has joined #openstack-infra | 05:25 | |
| *** sarob has joined #openstack-infra | 05:28 | |
| *** salv-orlando has quit IRC | 05:30 | |
| *** sarob has quit IRC | 05:32 | |
| *** Alexandra_ has quit IRC | 05:36 | |
| openstackgerrit | Khai Do proposed openstack-infra/puppet-gerrit: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155142 | 05:39 |
| *** salv-orlando has joined #openstack-infra | 05:40 | |
| *** unicell has joined #openstack-infra | 05:40 | |
| *** unicell has quit IRC | 05:41 | |
| *** unicell has joined #openstack-infra | 05:41 | |
| openstackgerrit | Khai Do proposed openstack-infra/system-config: Add the Gerrit replication plugin paramater 'defaultForceUpdate' https://review.openstack.org/155135 | 05:41 |
| *** yfried_ has quit IRC | 05:44 | |
| *** carl_baldwin has joined #openstack-infra | 05:46 | |
| *** salv-orlando has quit IRC | 05:52 | |
| *** achanda has joined #openstack-infra | 05:52 | |
| *** BharatK has quit IRC | 05:52 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-logstash: Modernize kibana vhost template https://review.openstack.org/153819 | 05:54 |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-logstash: Locking down apache module version https://review.openstack.org/153818 | 05:54 |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-logstash: Enhance and refactor head-of-file docs https://review.openstack.org/153817 | 05:54 |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-logstash: Flesh out the readme https://review.openstack.org/153820 | 05:54 |
| *** achanda has quit IRC | 05:56 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/publications: Consuming Open Source Infrastructure https://review.openstack.org/155193 | 05:58 |
| *** sileht has joined #openstack-infra | 06:04 | |
| *** carl_baldwin has quit IRC | 06:04 | |
| *** Longgeek has joined #openstack-infra | 06:07 | |
| *** garyh has joined #openstack-infra | 06:07 | |
| *** achanda has joined #openstack-infra | 06:08 | |
| *** salv-orlando has joined #openstack-infra | 06:08 | |
| *** ashleighfarnham has joined #openstack-infra | 06:11 | |
| *** otter768 has joined #openstack-infra | 06:13 | |
| *** BharatK has joined #openstack-infra | 06:14 | |
| *** crc32 has joined #openstack-infra | 06:15 | |
| *** salv-orlando has quit IRC | 06:15 | |
| *** salv-orlando has joined #openstack-infra | 06:16 | |
| *** BharatK has quit IRC | 06:18 | |
| *** otter768 has quit IRC | 06:18 | |
| *** salv-orlando has quit IRC | 06:18 | |
| *** hdd has quit IRC | 06:19 | |
| *** salv-orlando has joined #openstack-infra | 06:21 | |
| *** salv-orlando has quit IRC | 06:25 | |
| *** Ryan_Lane has joined #openstack-infra | 06:28 | |
| *** BharatK has joined #openstack-infra | 06:30 | |
| *** boris-42 has joined #openstack-infra | 06:33 | |
| *** zz_avozza is now known as avozza | 06:34 | |
| *** ghostpl has joined #openstack-infra | 06:35 | |
| *** barra204 has quit IRC | 06:36 | |
| *** dims__ has joined #openstack-infra | 06:38 | |
| *** ghostpl has quit IRC | 06:39 | |
| *** mwagner_lap has joined #openstack-infra | 06:42 | |
| stevemar | looks like https://jenkins02.openstack.org/ went down? | 06:42 |
| *** jerryz has joined #openstack-infra | 06:42 | |
| stevemar | and it's causing the patch at the top of the gate to hang? | 06:42 |
| *** dims__ has quit IRC | 06:42 | |
| stevemar | since it's been 6 hrs... | 06:42 |
| *** ashleighfarnham has quit IRC | 06:44 | |
| *** salv-orlando has joined #openstack-infra | 06:49 | |
| *** yfried_ has joined #openstack-infra | 06:50 | |
| *** vigneshvar has joined #openstack-infra | 06:52 | |
| *** yfried__ has joined #openstack-infra | 06:53 | |
| *** mrunge has joined #openstack-infra | 06:53 | |
| *** yfried_ has quit IRC | 06:54 | |
| *** yamamoto_ has joined #openstack-infra | 06:56 | |
| *** salv-orlando has quit IRC | 06:56 | |
| *** oomichi_ has joined #openstack-infra | 06:58 | |
| openstackgerrit | Merged openstack-infra/storyboard: Reject false creator_id from requests https://review.openstack.org/151650 | 06:58 |
| *** pcrews has quit IRC | 07:04 | |
| *** afazekas has joined #openstack-infra | 07:06 | |
| *** shashankhegde has joined #openstack-infra | 07:10 | |
| *** scheuran has joined #openstack-infra | 07:12 | |
| *** sressot has quit IRC | 07:12 | |
| openstackgerrit | Masayuki Igawa proposed openstack-infra/subunit2sql: Add db api methods for getting recent run uuids https://review.openstack.org/153040 | 07:13 |
| openstackgerrit | Khai Do proposed openstack-infra/project-config: Update javamelody build to ver 2.9 https://review.openstack.org/155208 | 07:15 |
| *** shashankhegde has quit IRC | 07:17 | |
| openstackgerrit | Khai Do proposed openstack-infra/project-config: Update javamelody build to ver 2.9 https://review.openstack.org/155208 | 07:18 |
| *** sressot has joined #openstack-infra | 07:19 | |
| *** kapil___ has joined #openstack-infra | 07:25 | |
| *** yfried__ is now known as yfried|afk | 07:26 | |
| *** yfried|afk is now known as yfried__ | 07:29 | |
| *** sarob has joined #openstack-infra | 07:30 | |
| *** achanda has quit IRC | 07:31 | |
| *** avozza is now known as zz_avozza | 07:33 | |
| *** sarob has quit IRC | 07:34 | |
| *** jcoufal has joined #openstack-infra | 07:35 | |
| *** achanda has joined #openstack-infra | 07:37 | |
| *** HeOS has quit IRC | 07:37 | |
| *** ildikov has quit IRC | 07:38 | |
| *** camunoz has quit IRC | 07:39 | |
| *** e0ne has joined #openstack-infra | 07:40 | |
| *** 20WAAL1Y7 has joined #openstack-infra | 07:41 | |
| *** flaper87 has quit IRC | 07:41 | |
| *** flaper87 has joined #openstack-infra | 07:41 | |
| *** Ryan_Lane has quit IRC | 07:47 | |
| *** Ryan_Lane has joined #openstack-infra | 07:47 | |
| *** sdake has quit IRC | 07:48 | |
| *** sdake has joined #openstack-infra | 07:48 | |
| *** coolsvap has quit IRC | 07:48 | |
| *** coolsvap has joined #openstack-infra | 07:48 | |
| *** BharatK has quit IRC | 07:49 | |
| *** BharatK has joined #openstack-infra | 07:49 | |
| *** boris-42 has quit IRC | 07:49 | |
| *** boris-42 has joined #openstack-infra | 07:49 | |
| *** yfried__ has quit IRC | 07:49 | |
| *** yfried__ has joined #openstack-infra | 07:49 | |
| *** afazekas has quit IRC | 07:49 | |
| *** afazekas has joined #openstack-infra | 07:49 | |
| *** scheuran has quit IRC | 07:49 | |
| *** scheuran has joined #openstack-infra | 07:49 | |
| *** kapil___ has quit IRC | 07:49 | |
| *** kapil___ has joined #openstack-infra | 07:49 | |
| *** jcoufal has quit IRC | 07:49 | |
| *** jcoufal has joined #openstack-infra | 07:49 | |
| *** Ryan_Lane has quit IRC | 07:51 | |
| *** Longgeek has quit IRC | 07:53 | |
| *** sandywalsh_ has joined #openstack-infra | 07:53 | |
| *** Longgeek has joined #openstack-infra | 07:53 | |
| *** Longgeek has quit IRC | 07:53 | |
| *** doude has quit IRC | 07:54 | |
| *** sandywalsh has quit IRC | 07:55 | |
| *** doude has joined #openstack-infra | 07:55 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Use local git path in DEVSTACK_LOCAL_CONFIG https://review.openstack.org/155217 | 07:55 |
| *** Mr0rris0 has joined #openstack-infra | 08:01 | |
| *** Mr0rris0 has quit IRC | 08:01 | |
| openstackgerrit | Anusha Ramineni proposed openstack-infra/project-config: Allow proliantutils to be published on PyPI https://review.openstack.org/155218 | 08:02 |
| *** dtantsur|afk is now known as dtantsur | 08:02 | |
| openstackgerrit | Merged openstack-infra/project-config: Add Fuel LMA collector Plugin to Stackforge https://review.openstack.org/153570 | 08:03 |
| *** stevemar has quit IRC | 08:04 | |
| *** sandywalsh_ has quit IRC | 08:04 | |
| *** mrmartin has joined #openstack-infra | 08:05 | |
| openstackgerrit | Merged openstack-infra/storyboard: Add test_after_publishes_payload to test_notification_hook https://review.openstack.org/155017 | 08:05 |
| *** sandywalsh has joined #openstack-infra | 08:05 | |
| *** salv-orlando has joined #openstack-infra | 08:11 | |
| *** arxcruz has joined #openstack-infra | 08:14 | |
| *** otter768 has joined #openstack-infra | 08:14 | |
| *** e0ne has quit IRC | 08:17 | |
| *** otter768 has quit IRC | 08:19 | |
| *** ghostpl has joined #openstack-infra | 08:20 | |
| *** openstackgerrit has quit IRC | 08:21 | |
| *** openstackgerrit has joined #openstack-infra | 08:21 | |
| *** hashar has joined #openstack-infra | 08:22 | |
| *** luqas has joined #openstack-infra | 08:24 | |
| *** ghostpl has quit IRC | 08:25 | |
| *** liam_ has joined #openstack-infra | 08:27 | |
| *** ashleighfarnham has joined #openstack-infra | 08:30 | |
| *** kapil___ has quit IRC | 08:34 | |
| *** achanda has quit IRC | 08:35 | |
| *** ildikov has joined #openstack-infra | 08:39 | |
| *** jgallard__ has joined #openstack-infra | 08:39 | |
| *** mpaolino has joined #openstack-infra | 08:40 | |
| *** zz_avozza is now known as avozza | 08:41 | |
| *** chlong has quit IRC | 08:42 | |
| *** hashar has quit IRC | 08:45 | |
| *** yamamoto_ has quit IRC | 08:46 | |
| *** jlibosva has joined #openstack-infra | 08:48 | |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added branches to storyboard https://review.openstack.org/150447 | 08:50 |
| *** andreykurilin_ has joined #openstack-infra | 08:50 | |
| *** Ala has joined #openstack-infra | 08:52 | |
| *** Longgeek has joined #openstack-infra | 08:54 | |
| *** jp_at_hp has joined #openstack-infra | 08:54 | |
| *** jistr|off is now known as jistr | 08:54 | |
| *** MaxV has joined #openstack-infra | 08:55 | |
| *** HeOS has joined #openstack-infra | 08:57 | |
| *** andreykurilin__ has joined #openstack-infra | 09:01 | |
| *** andreykurilin_ has quit IRC | 09:01 | |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added branches to storyboard https://review.openstack.org/150447 | 09:01 |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Datetime instances from SQLAlchemy now all contain timezone. https://review.openstack.org/155083 | 09:01 |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added milestones https://review.openstack.org/152500 | 09:02 |
| *** ashleighfarnham has quit IRC | 09:02 | |
| *** hamzy has quit IRC | 09:02 | |
| *** markus_z has joined #openstack-infra | 09:03 | |
| *** ghostpl has joined #openstack-infra | 09:03 | |
| *** ramineni has joined #openstack-infra | 09:05 | |
| *** jpich has joined #openstack-infra | 09:06 | |
| *** Longgeek has quit IRC | 09:07 | |
| *** jistr has quit IRC | 09:08 | |
| *** jistr has joined #openstack-infra | 09:11 | |
| *** hashar has joined #openstack-infra | 09:12 | |
| *** esmute has quit IRC | 09:14 | |
| *** timrc has quit IRC | 09:15 | |
| *** esmute has joined #openstack-infra | 09:16 | |
| *** timrc has joined #openstack-infra | 09:16 | |
| *** BharatK has quit IRC | 09:16 | |
| *** enikanorov_ has quit IRC | 09:19 | |
| *** enikanorov_ has joined #openstack-infra | 09:20 | |
| *** yamamoto has joined #openstack-infra | 09:22 | |
| *** vipul has quit IRC | 09:22 | |
| *** juice has quit IRC | 09:22 | |
| *** esmute has quit IRC | 09:23 | |
| *** amotoki has quit IRC | 09:23 | |
| *** esmute has joined #openstack-infra | 09:24 | |
| *** juice has joined #openstack-infra | 09:25 | |
| *** vipul has joined #openstack-infra | 09:25 | |
| *** yamamoto has quit IRC | 09:27 | |
| *** amotoki has joined #openstack-infra | 09:28 | |
| *** BharatK has joined #openstack-infra | 09:30 | |
| *** scheuran_ has joined #openstack-infra | 09:31 | |
| *** e0ne has joined #openstack-infra | 09:31 | |
| *** scheuran has quit IRC | 09:32 | |
| *** markus_z has quit IRC | 09:32 | |
| *** yamamoto has joined #openstack-infra | 09:33 | |
| *** armax has quit IRC | 09:33 | |
| *** markus_z has joined #openstack-infra | 09:34 | |
| *** xyang1 has quit IRC | 09:34 | |
| *** doug-fish has quit IRC | 09:38 | |
| *** Guest41759 has joined #openstack-infra | 09:43 | |
| *** hashar has quit IRC | 09:43 | |
| *** ghostpl has quit IRC | 09:44 | |
| *** nikil89_ has joined #openstack-infra | 09:44 | |
| *** nikil89_ is now known as naggappan | 09:44 | |
| *** ghostpl has joined #openstack-infra | 09:44 | |
| *** hashar has joined #openstack-infra | 09:45 | |
| *** yamamoto has quit IRC | 09:46 | |
| *** enikanorov_ has quit IRC | 09:51 | |
| *** enikanorov_ has joined #openstack-infra | 09:51 | |
| *** teran has quit IRC | 09:56 | |
| *** enikanorov__ has joined #openstack-infra | 09:56 | |
| *** hichihara has quit IRC | 09:56 | |
| *** enikanorov_ has quit IRC | 09:57 | |
| *** skolekonov has joined #openstack-infra | 09:57 | |
| *** ssam2 has joined #openstack-infra | 09:58 | |
| *** yamamoto has joined #openstack-infra | 10:00 | |
| *** ashleighfarnham has joined #openstack-infra | 10:01 | |
| openstackgerrit | Rossella Sblendido proposed openstack-infra/elastic-recheck: Add e-r query for test_rebuild_availability_range https://review.openstack.org/155234 | 10:01 |
| *** avozza is now known as zz_avozza | 10:02 | |
| *** Longgeek has joined #openstack-infra | 10:04 | |
| *** ashleighfarnham has quit IRC | 10:05 | |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added branches to storyboard https://review.openstack.org/150447 | 10:06 |
| *** dtantsur is now known as dtantsur|bbl | 10:09 | |
| openstackgerrit | Masayuki Igawa proposed openstack-infra/subunit2sql: Add db api methods for getting recent run uuids https://review.openstack.org/153040 | 10:15 |
| *** otter768 has joined #openstack-infra | 10:15 | |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added milestones https://review.openstack.org/152500 | 10:16 |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard: Added new paginate query https://review.openstack.org/153231 | 10:19 |
| *** raghav_mulugu has joined #openstack-infra | 10:19 | |
| *** otter768 has quit IRC | 10:20 | |
| *** zz_avozza is now known as avozza | 10:21 | |
| *** e0ne is now known as e0ne_ | 10:21 | |
| *** mpaolino has quit IRC | 10:21 | |
| *** yamamoto has quit IRC | 10:21 | |
| *** teran has joined #openstack-infra | 10:23 | |
| *** chmouel_ has joined #openstack-infra | 10:24 | |
| *** chmouel_ has quit IRC | 10:25 | |
| *** teran has quit IRC | 10:27 | |
| *** e0ne_ is now known as e0ne | 10:30 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Fix error notifications https://review.openstack.org/155242 | 10:33 |
| *** koolhead17 has quit IRC | 10:37 | |
| *** Longgeek has quit IRC | 10:38 | |
| *** ldnunes has joined #openstack-infra | 10:38 | |
| *** mpaolino has joined #openstack-infra | 10:39 | |
| *** leopoldj has joined #openstack-infra | 10:40 | |
| *** mpaolino has quit IRC | 10:41 | |
| *** rameshg87 has joined #openstack-infra | 10:43 | |
| *** cdent has joined #openstack-infra | 10:44 | |
| *** Longgeek has joined #openstack-infra | 10:44 | |
| *** yfried__ is now known as yfried|afk | 10:44 | |
| *** yfried|afk is now known as yfried__ | 10:45 | |
| *** mpaolino has joined #openstack-infra | 10:46 | |
| *** aysyd has joined #openstack-infra | 10:46 | |
| *** mpaolino has quit IRC | 10:47 | |
| *** ghostpl has quit IRC | 10:48 | |
| *** mpaolino has joined #openstack-infra | 10:48 | |
| *** ghostpl has joined #openstack-infra | 10:49 | |
| openstackgerrit | Marc Koderer proposed openstack-infra/project-config: Add telco working group usecase repo https://review.openstack.org/155248 | 10:50 |
| *** ihrachyshka has joined #openstack-infra | 10:51 | |
| *** pelix has joined #openstack-infra | 10:51 | |
| *** salv-orlando has quit IRC | 10:53 | |
| *** rfolco has joined #openstack-infra | 10:55 | |
| *** yfried__ is now known as yfried|afk | 10:55 | |
| *** yamahata has quit IRC | 10:59 | |
| *** dtantsur|bbl is now known as dtantsur | 10:59 | |
| *** teran has joined #openstack-infra | 11:01 | |
| *** koolhead17 has joined #openstack-infra | 11:02 | |
| *** ramineni has left #openstack-infra | 11:03 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/jeepyb: Enforce manage_projects to be run with specified user https://review.openstack.org/152085 | 11:04 |
| *** adalbas has joined #openstack-infra | 11:06 | |
| *** Liu has joined #openstack-infra | 11:06 | |
| Liu | Hi | 11:08 |
| Liu | Our CI build failed when build devstack | 11:09 |
| *** rlandy has joined #openstack-infra | 11:13 | |
| *** hamzy has joined #openstack-infra | 11:14 | |
| *** yfried|afk is now known as yfried__ | 11:15 | |
| *** dims__ has joined #openstack-infra | 11:15 | |
| *** kmartin has quit IRC | 11:16 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/jeepyb: Enforce manage_projects to be run with specified user https://review.openstack.org/152085 | 11:17 |
| *** rameshg87 has quit IRC | 11:17 | |
| *** derekh has joined #openstack-infra | 11:18 | |
| *** mpaolino has quit IRC | 11:23 | |
| HenryG | Everything that depends on Neutron will fail until this merges: https://review.openstack.org/155152 | 11:24 |
| HenryG | Is anyone around to unstick the gate? | 11:25 |
| *** niska has quit IRC | 11:27 | |
| *** salv-orlando has joined #openstack-infra | 11:27 | |
| *** sdake has quit IRC | 11:28 | |
| *** BharatK has quit IRC | 11:30 | |
| *** markvoelker has quit IRC | 11:34 | |
| *** rlandy has quit IRC | 11:35 | |
| *** BharatK has joined #openstack-infra | 11:35 | |
| *** niska has joined #openstack-infra | 11:36 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/nodepool: Move image uploads out of ProviderManager https://review.openstack.org/137401 | 11:37 |
| *** claudiub has joined #openstack-infra | 11:38 | |
| openstackgerrit | Guido Günther proposed stackforge/python-jenkins: Add BadHTTPException https://review.openstack.org/155262 | 11:38 |
| openstackgerrit | Guido Günther proposed stackforge/python-jenkins: Allow to wait for jenkins to enter normal operation https://review.openstack.org/155263 | 11:38 |
| *** jgallard__ has quit IRC | 11:46 | |
| SergeyLukjanov | fungi, thx for info | 11:47 |
| openstackgerrit | yolanda.robla proposed openstack-infra/infra-specs: Add nodepool image testing spec. https://review.openstack.org/139598 | 11:48 |
| SergeyLukjanov | fungi, it's just to ensure that if we'll need it than I know how to do it | 11:48 |
| *** e0ne is now known as e0ne_ | 11:48 | |
| *** rodrigod` is now known as rodrigods | 11:49 | |
| openstackgerrit | Guido Günther proposed openstack-infra/jenkins-job-builder: Allow to wait for jenkins to start up https://review.openstack.org/152892 | 11:50 |
| openstackgerrit | Guido Günther proposed openstack-infra/jenkins-job-builder: Use open() instead of file() https://review.openstack.org/153359 | 11:50 |
| openstackgerrit | Guido Günther proposed openstack-infra/jenkins-job-builder: Lazy load plugin list https://review.openstack.org/153396 | 11:50 |
| *** rlandy has joined #openstack-infra | 11:51 | |
| *** asselin__ has quit IRC | 11:52 | |
| *** e0ne_ has quit IRC | 11:58 | |
| *** amotoki has quit IRC | 11:59 | |
| *** BharatK has quit IRC | 12:02 | |
| *** pc_m has joined #openstack-infra | 12:06 | |
| *** hashar has quit IRC | 12:11 | |
| *** otter768 has joined #openstack-infra | 12:16 | |
| *** ociuhandu has joined #openstack-infra | 12:16 | |
| *** BharatK has joined #openstack-infra | 12:17 | |
| *** otter768 has quit IRC | 12:21 | |
| *** _nadya_ has joined #openstack-infra | 12:22 | |
| *** otter768 has joined #openstack-infra | 12:23 | |
| *** bswartz has quit IRC | 12:23 | |
| *** salv-orlando has quit IRC | 12:29 | |
| openstackgerrit | Guido Günther proposed stackforge/python-jenkins: Add BadHTTPException https://review.openstack.org/155262 | 12:30 |
| openstackgerrit | Guido Günther proposed stackforge/python-jenkins: Allow to wait for jenkins to enter normal operation https://review.openstack.org/155263 | 12:30 |
| *** markvoelker has joined #openstack-infra | 12:35 | |
| *** dims_ has joined #openstack-infra | 12:37 | |
| *** dims__ has quit IRC | 12:38 | |
| *** markvoelker has quit IRC | 12:40 | |
| *** yfried__ is now known as yfried|afk | 12:41 | |
| *** yfried|afk is now known as yfried__ | 12:44 | |
| *** doude has quit IRC | 12:52 | |
| *** doude has joined #openstack-infra | 12:53 | |
| *** weshay has joined #openstack-infra | 12:55 | |
| *** dprince has joined #openstack-infra | 12:56 | |
| *** _nadya_ has quit IRC | 12:57 | |
| *** dims_ has quit IRC | 12:57 | |
| sdague | SergeyLukjanov / fungi could you guys land this voting change while my updown.sh job passes - https://review.openstack.org/#/c/154994/ | 12:59 |
| sdague | so we can not regress :) | 12:59 |
| *** markvoelker has joined #openstack-infra | 12:59 | |
| *** woodster_ has joined #openstack-infra | 13:00 | |
| *** _amrith_ is now known as amrith | 13:01 | |
| *** bswartz has joined #openstack-infra | 13:02 | |
| *** baoli has joined #openstack-infra | 13:03 | |
| *** salv-orlando has joined #openstack-infra | 13:03 | |
| *** yfried has joined #openstack-infra | 13:04 | |
| *** yfried__ has quit IRC | 13:04 | |
| *** dkranz has joined #openstack-infra | 13:05 | |
| *** markvoelker has quit IRC | 13:05 | |
| *** kgiusti has joined #openstack-infra | 13:05 | |
| *** dims__ has joined #openstack-infra | 13:05 | |
| *** e0ne has joined #openstack-infra | 13:05 | |
| *** eharney has joined #openstack-infra | 13:05 | |
| *** slagle has quit IRC | 13:08 | |
| *** slagle has joined #openstack-infra | 13:09 | |
| *** alexpilotti has joined #openstack-infra | 13:09 | |
| *** EmilienM|afk is now known as EmilienM | 13:11 | |
| *** mwagner_lap has quit IRC | 13:11 | |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Fix in project groups validation https://review.openstack.org/152133 | 13:12 |
| *** hashar has joined #openstack-infra | 13:12 | |
| *** dims__ has quit IRC | 13:13 | |
| *** yfried_ has joined #openstack-infra | 13:14 | |
| *** yfried has quit IRC | 13:14 | |
| *** koolhead17 has quit IRC | 13:17 | |
| *** jcoufal_ has joined #openstack-infra | 13:17 | |
| *** dims__ has joined #openstack-infra | 13:19 | |
| *** jcoufal has quit IRC | 13:20 | |
| HenryG | Sorry for the annoying dumb question, but what is wrong with the check gate? | 13:23 |
| *** oomichi_ has quit IRC | 13:23 | |
| sdague | wrong? | 13:23 |
| HenryG | Several jobs have been stuck for 11 or 12 hours. | 13:24 |
| sdague | looks like jenkins02 fell over | 13:25 |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 13:26 |
| *** ZZelle has quit IRC | 13:26 | |
| *** ZZelle has joined #openstack-infra | 13:26 | |
| *** bhunter71 has joined #openstack-infra | 13:27 | |
| pblaho | pls, where are results of coverage jobs in post hook published? | 13:28 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Add service_catalog property https://review.openstack.org/155304 | 13:30 |
| mordred | SergeyLukjanov, fungi: ^^ I'd love a quite apprv on that ... | 13:31 |
| anteaya | Liu: we need more information that that to help you | 13:32 |
| anteaya | Liu: do you have a paste of the stacktrace of the failure? | 13:33 |
| openstackgerrit | yolanda.robla proposed openstack-infra/infra-specs: Spec for nodepool image update retries https://review.openstack.org/155305 | 13:33 |
| anteaya | Liu: http://paste.openstack.org/ | 13:33 |
| *** scheuran_ has quit IRC | 13:33 | |
| *** e0ne is now known as e0ne_ | 13:35 | |
| *** ssam2 has quit IRC | 13:36 | |
| *** jgjl has joined #openstack-infra | 13:38 | |
| *** eharney has quit IRC | 13:39 | |
| openstackgerrit | Marc Koderer proposed openstack-infra/project-config: Add telco working group usecase repo https://review.openstack.org/155248 | 13:39 |
| *** mbacchi has joined #openstack-infra | 13:42 | |
| *** jgjl has quit IRC | 13:47 | |
| *** koolhead17 has joined #openstack-infra | 13:47 | |
| *** radez_g0n3 is now known as radez | 13:48 | |
| *** amotoki has joined #openstack-infra | 13:48 | |
| *** e0ne_ is now known as e0ne | 13:48 | |
| *** dhritishikhar_ has joined #openstack-infra | 13:48 | |
| *** jgjl has joined #openstack-infra | 13:49 | |
| *** Mr0rris0 has joined #openstack-infra | 13:49 | |
| *** Mr0rris0 has quit IRC | 13:49 | |
| *** Mr0rris0 has joined #openstack-infra | 13:50 | |
| *** Mr0rris0 has quit IRC | 13:50 | |
| *** Mr0rris0 has joined #openstack-infra | 13:50 | |
| *** Mr0rris0 has quit IRC | 13:51 | |
| *** Mr0rris0 has joined #openstack-infra | 13:51 | |
| *** Mr0rris0 has quit IRC | 13:51 | |
| *** Mr0rris0 has joined #openstack-infra | 13:52 | |
| *** Mr0rris0 has quit IRC | 13:52 | |
| *** zz_dimtruck is now known as dimtruck | 13:52 | |
| *** Mr0rris0 has joined #openstack-infra | 13:52 | |
| *** eharney has joined #openstack-infra | 13:53 | |
| *** Mr0rris0 has quit IRC | 13:53 | |
| *** Mr0rris0 has joined #openstack-infra | 13:53 | |
| *** Mr0rris0 has quit IRC | 13:54 | |
| *** 20WAAL1Y7 has left #openstack-infra | 13:54 | |
| *** Mr0rris0 has joined #openstack-infra | 13:54 | |
| *** jgallard__ has joined #openstack-infra | 13:54 | |
| *** Mr0rris0 has quit IRC | 13:54 | |
| *** ssam2 has joined #openstack-infra | 13:54 | |
| *** Mr0rris0 has joined #openstack-infra | 13:55 | |
| sdague | mordred: hey, can I get you to +A this - https://review.openstack.org/#/c/154994/ ? | 13:55 |
| *** Mr0rris0 has quit IRC | 13:55 | |
| sdague | it makes a devstack job voting so we are less likely to mess up unstack.sh and clean.sh | 13:55 |
| *** Mr0rris0 has joined #openstack-infra | 13:55 | |
| *** Mr0rris0 has quit IRC | 13:56 | |
| *** doug-fish has joined #openstack-infra | 13:56 | |
| *** Mr0rris0 has joined #openstack-infra | 13:56 | |
| *** Mr0rris0 has quit IRC | 13:56 | |
| *** Mr0rris0 has joined #openstack-infra | 13:57 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Add service_catalog property https://review.openstack.org/155304 | 13:57 |
| mordred | sdague: done | 13:58 |
| mordred | fungi, SergeyLukjanov ^^ (oops, there was a pep8 issue before) | 13:58 |
| *** mwagner_lap has joined #openstack-infra | 13:58 | |
| *** amotoki_ has joined #openstack-infra | 14:00 | |
| *** ssam2 has quit IRC | 14:00 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:01 | |
| *** amotoki__ has joined #openstack-infra | 14:02 | |
| *** jcoufal_ has quit IRC | 14:02 | |
| *** jcoufal has joined #openstack-infra | 14:02 | |
| *** amotoki has quit IRC | 14:02 | |
| *** amotok___ has joined #openstack-infra | 14:02 | |
| *** dimtruck is now known as zz_dimtruck | 14:03 | |
| openstackgerrit | MORITA Kazutaka proposed openstack-infra/project-config: Add test job for sheepdog devstack plugin https://review.openstack.org/154605 | 14:03 |
| *** ddieterly has quit IRC | 14:03 | |
| *** vigneshvar has quit IRC | 14:05 | |
| *** amotoki_ has quit IRC | 14:05 | |
| *** thorst has joined #openstack-infra | 14:06 | |
| *** miqui has joined #openstack-infra | 14:06 | |
| *** amotoki__ has quit IRC | 14:06 | |
| *** _nadya_ has joined #openstack-infra | 14:07 | |
| *** dmsimard_away is now known as dmsimard | 14:07 | |
| *** denis_makogon has joined #openstack-infra | 14:08 | |
| *** salv-orlando has quit IRC | 14:08 | |
| *** salv-orlando has joined #openstack-infra | 14:08 | |
| *** vhoward has joined #openstack-infra | 14:09 | |
| openstackgerrit | Sean Dague proposed openstack-infra/project-config: remove check-neutron-dsvm-functional from devstack gates https://review.openstack.org/155323 | 14:10 |
| *** dhritishikhar_ has quit IRC | 14:11 | |
| openstackgerrit | Merged openstack-infra/project-config: make devstack updown job voting on master https://review.openstack.org/154994 | 14:11 |
| *** avozza is now known as zz_avozza | 14:11 | |
| *** salv-orlando has quit IRC | 14:13 | |
| *** dalgaaf has joined #openstack-infra | 14:13 | |
| *** miqui_ has joined #openstack-infra | 14:13 | |
| *** mjturek1 has joined #openstack-infra | 14:13 | |
| openstackgerrit | Guido Günther proposed stackforge/python-jenkins: Allow to wait for jenkins to enter normal operation https://review.openstack.org/155263 | 14:14 |
| *** radez is now known as radez_g0n3 | 14:15 | |
| *** dmsimard is now known as dmsimard_away | 14:15 | |
| *** ssam2 has joined #openstack-infra | 14:17 | |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 14:18 |
| *** sdake has joined #openstack-infra | 14:19 | |
| *** dkliban_afk is now known as dkliban | 14:21 | |
| *** joesavak has joined #openstack-infra | 14:24 | |
| *** david-lyle_afk is now known as david-lyle | 14:25 | |
| *** naggappan has quit IRC | 14:26 | |
| fungi | mordred: still fails pep8 | 14:27 |
| fungi | checking out jenkins02 now | 14:27 |
| anteaya | so I see there are some stackforge repos that are using openstack project names | 14:28 |
| anteaya | like neutron and cinder | 14:28 |
| anteaya | I had been working to keep the use of those names out of stackforge | 14:28 |
| BobBall | How large are the images that the offical nodepool creates for devstack tests? I'm creating images that are ~5GB which seems huge and it's taking ~20 minutes to download the image to each host when spinning up a node | 14:28 |
| *** alexpilotti has quit IRC | 14:28 | |
| anteaya | unfortunately some got in | 14:28 |
| *** jgjl has quit IRC | 14:28 | |
| anteaya | so how do I go about createing a proposal to rename the stackforge repos to change the name to networking and block-storage | 14:29 |
| fungi | aha, looks like jenkins02 is also why 155152 isn't coming back from check | 14:29 |
| *** alexpilotti has joined #openstack-infra | 14:29 | |
| anteaya | as those had been the names that, at least with neutron/networking, that folks had agreed to | 14:29 |
| *** ghostpl has quit IRC | 14:29 | |
| *** ssam2 has quit IRC | 14:30 | |
| fungi | since i can't get into the jenkins02 webui, i'm stopping it from the operating system side | 14:31 |
| *** boris-42 has quit IRC | 14:32 | |
| *** gordc has joined #openstack-infra | 14:33 | |
| *** jcoufal has quit IRC | 14:33 | |
| *** jcoufal_ has joined #openstack-infra | 14:33 | |
| mordred | fungi: DOH. a debugging statement had crept in | 14:33 |
| *** ghostpl has joined #openstack-infra | 14:33 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Add service_catalog property https://review.openstack.org/155304 | 14:34 |
| *** r-daneel has joined #openstack-infra | 14:36 | |
| ctlaugh | Any tips for building scp-plugin 1.9? | 14:36 |
| fungi | okay, jenkins02 is back up and running with a clean slave list, all nodes assigned to it have been deleted from nodepool, and zuul appears to have requeued any jobs previously running there | 14:36 |
| fungi | HenryG: sdague: ^ | 14:36 |
| BobBall | ctlaugh: https://github.com/rasselin/os-ext-testing (See under "Usage" section | 14:37 |
| BobBall | ctlaugh: Do it on your host and it "just worked" for me. | 14:37 |
| BobBall | (not the puppet master - I meant local box) | 14:37 |
| *** cdent has quit IRC | 14:38 | |
| *** cdent has joined #openstack-infra | 14:39 | |
| *** ddieterly has joined #openstack-infra | 14:39 | |
| *** mriedem has joined #openstack-infra | 14:39 | |
| krtaylor | ctlaugh, keep in mind there are also swift options, vs scp, but maybe for a future enhancement to your setup | 14:40 |
| anteaya | ttx: would I be able to make a motion to the tc about using openstack project names in stackforge | 14:41 |
| anteaya | ttx: I'm strongly against it | 14:41 |
| *** boris-42 has joined #openstack-infra | 14:42 | |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 14:43 |
| *** ssam2 has joined #openstack-infra | 14:43 | |
| *** garyh has quit IRC | 14:43 | |
| *** ctlaugh_ has joined #openstack-infra | 14:44 | |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 14:44 |
| ctlaugh_ | I was building it on the puppet master that I was setting up. Maven kept complaining about a missing tools.jar 1.5 dependency. | 14:45 |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 14:45 |
| ctlaugh_ | I'll try on a different system. | 14:46 |
| BobBall | puppet master is less good because you need to upload it through the web ui - but it just worked(tm) for me. | 14:46 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Consume auth_token if it's passed in https://review.openstack.org/155332 | 14:46 |
| ttx | anteaya: what do you mean ? | 14:49 |
| ttx | anteaya: openstack projects are under openstack*/* | 14:49 |
| *** mtanino has joined #openstack-infra | 14:50 | |
| *** mase_home has quit IRC | 14:51 | |
| anteaya | right | 14:52 |
| fungi | anteaya: if you're complaining about what i think you're complaining about, we've had projects like stackforge/puppet-neutron and stackforge/neutron-salt-formula long before we added stackforge/fuel-plugin-neutron-fwaas or stackforge/neutron-powervm so it's not entirely clear where you want the line to be. should config management/orchestration tools be required to change the names of the repos to | 14:52 |
| fungi | not mention the codenames of the projects they configure? | 14:52 |
| anteaya | fungi: the configutation tools make sense, since taht is what they configure | 14:52 |
| anteaya | but the neutron plugins being moved out of tree | 14:52 |
| fungi | seems like it would be complex to draft a tc resolution/policy around it then | 14:52 |
| anteaya | had agreed to use networking | 14:52 |
| anteaya | as I discussed with kyle | 14:53 |
| anteaya | and he saw my point for not using neutron-driver in stackforge | 14:53 |
| anteaya | so I would like the one project that has that renamed | 14:53 |
| anteaya | all other plugins are using networking-plugin | 14:53 |
| *** tiswanso has joined #openstack-infra | 14:53 | |
| *** ghostpl has quit IRC | 14:54 | |
| openstackgerrit | Merged openstack-infra/shade: Make is_object_stale() a public method https://review.openstack.org/153898 | 14:54 |
| fungi | anteaya: that sounds more like something you should discuss with the neutron ptl and the members of https://review.openstack.org/#/admin/groups/592,members | 14:54 |
| anteaya | seems the repo in review saw some sense and looked at stackforge and changed the name | 14:54 |
| openstackgerrit | Merged openstack-infra/shade: Remove unnecessary container creation https://review.openstack.org/155063 | 14:54 |
| anteaya | fungi: I can start there | 14:54 |
| anteaya | and kyle was already in agreement | 14:55 |
| anteaya | I haven't found the patch for it yet, so I don't know if he weighed in the the review for it | 14:55 |
| *** sdake__ has joined #openstack-infra | 14:55 | |
| *** ghostpl has joined #openstack-infra | 14:55 | |
| *** sdake has quit IRC | 14:56 | |
| anteaya | but obviously the need to assess stackforge names isn't a group concern | 14:56 |
| anteaya | I had thought it was | 14:56 |
| *** ayoung has joined #openstack-infra | 14:56 | |
| fungi | i would personally rather not be the name police, and let projects have whatever names they want (within reason) so long as it's not going to cause trademark issues | 14:56 |
| *** ctlaugh_ has quit IRC | 14:57 | |
| anteaya | right | 14:57 |
| anteaya | me too | 14:57 |
| *** carl_baldwin has joined #openstack-infra | 14:57 | |
| anteaya | however I do believe that openstack project names used in stackforge projects need closer scrunity than a rubber stamp | 14:57 |
| *** amitgandhinz has joined #openstack-infra | 14:58 | |
| anteaya | if they want a rubber stamp, don't use an openstack project name in the stackforge project name | 14:58 |
| anteaya | I wish we didn't have to have a need for name police | 14:58 |
| fungi | anteaya: the change adding that project was https://review.openstack.org/150989 and mestery doesn't seem to have commented in it | 14:59 |
| anteaya | fungi: thanks for finding that | 14:59 |
| fungi | looks like it was in review for ~5 days | 14:59 |
| anteaya | so yeah, I'm going with kyle didn't know | 14:59 |
| * mordred does not really think it's important to police this - although I agree a cursory check to make sure it's not completely bonghits may be in order if it's refrencing a name | 14:59 | |
| *** _nadya_ has quit IRC | 14:59 | |
| anteaya | and he isn't online atm so I can't ask him | 14:59 |
| anteaya | well the thing is that a name is only as valuable as how it is used | 15:00 |
| fungi | i can get behind "if you're adding a project which includes the name of official project "foo" as a substring, then get a +1 from the ptl of "foo" before approving as a sanity check | 15:00 |
| mordred | nova-docker is a great example - does it need to be approved by mikal? nah. it's an out of tree drive for nova. however - if it was called nova-docker and was actually a rewrite of nova based on docker - then I think we might want to say something | 15:00 |
| anteaya | so if it is used well, the name or any word has value | 15:00 |
| anteaya | fungi: great thanks | 15:00 |
| anteaya | fungi: that will be sufficient | 15:00 |
| anteaya | would it be so hard as to indicate that mikal at least knew? | 15:01 |
| fungi | but adding arbitrary reviewing rules around these sorts of changes does increase complexity for reviewers and so we should do something to offset that (like a concise checklist maybe) | 15:01 |
| anteaya | that would indicate to me that some line of communication was established if only for cursory purposes | 15:01 |
| anteaya | fungi: yes, but I don't think sanity checking for names using openstack project names is arbitrary | 15:02 |
| anteaya | I'm looking for things to indicate co-operation | 15:02 |
| anteaya | not co-opting | 15:03 |
| sigmavirus24 | So I know library releases have been causing a big headache for us. Glance is planning one for next week and I'm thinking of pre-emptively capping it in stable/juno's g-r so that it doesn't cause a huge headache for anyone until we can ensure it doesn't break stable/juno | 15:03 |
| sigmavirus24 | Would the noise of capping and then potentially changing the cap be too annoying and not worth it, or would we rather keep the gate in working order through that release? | 15:03 |
| fungi | anteaya: sure, just one of us should try to encode these sanity checks in some sort of a checklist when there's time, maybe in the readme for the project-config repo like we do for requirements repo reviewer guidelines | 15:04 |
| anteaya | but I don't know how to state "show me you are co-operating with the project in question, not co-opting" other than name checking | 15:04 |
| anteaya | fungi: I have no argument with that | 15:04 |
| * anteaya creates an etherpad | 15:04 | |
| fungi | sigmavirus24: what library? | 15:04 |
| sigmavirus24 | glance_store | 15:04 |
| fungi | sigmavirus24: what projects use it? are any of them uncapped in juno? | 15:05 |
| sigmavirus24 | glance should be the sole consumer and I'm 99% sure it isn't capped yet. I'll double check though | 15:05 |
| sigmavirus24 | Oh it's already capped. Nevermind | 15:06 |
| sigmavirus24 | Sorry for the noise | 15:06 |
| fungi | sigmavirus24: but yes, coordinate with jogo in case he already has a patch to do that, otherwise i'd say it's a good idea to get the cap into global-requirements.txt soon so that the cap trickles down to glance stable/juno before you tag a new glance_store | 15:06 |
| *** prad has joined #openstack-infra | 15:06 | |
| *** dmsimard_away is now known as dmsimard | 15:06 | |
| *** signed8bit has joined #openstack-infra | 15:06 | |
| fungi | sigmavirus24: oh, all set then. good work! ;) | 15:06 |
| sigmavirus24 | Hah | 15:06 |
| sigmavirus24 | I feel like it wasn't capped the last time I looked at it, but I can't remember when that was exactly | 15:06 |
| sigmavirus24 | So clearly I need to consume more coffee | 15:07 |
| anteaya | https://etherpad.openstack.org/p/new-repo-reviewing-sanity-checks | 15:07 |
| fungi | sigmavirus24: yeah, jogo has been on a frenzied stable capping rampage recently, so that's entirely plausible | 15:08 |
| anteaya | ttx this etherpad will probably come into play as project-config reviewers try to navigate the new waters: https://etherpad.openstack.org/p/new-repo-reviewing-sanity-checks | 15:08 |
| anteaya | ttx would be nice to see you are at least aware the etherpad exists | 15:08 |
| BobBall | Setting up a 3rd party CI: Nodepool creates the nodes in RAX (with hostname including openstack.org), seems that puppet downloads the keys for jenkins from puppetmaster.openstack.org... Is the only way to set up a CI with having a domain and a puppetmaster on that domain to send out the keys? | 15:08 |
| *** hashar has quit IRC | 15:09 | |
| *** hashar has joined #openstack-infra | 15:09 | |
| * sigmavirus24 just wants to make all of your lives easier =P | 15:11 | |
| ttx | anteaya: looking | 15:11 |
| *** marun has joined #openstack-infra | 15:12 | |
| clarkb | BobBall I think jhesketh patched the domain issue (or someone did pretty sure you can set that now) and ssh keys are configurable in puppet and the puppet master should not be used for image builds. Does a puppet apply which is all local | 15:13 |
| ttx | anteaya: in those cases (3rd-party drivers in stackforge) I think they would stay in stackforge because they probably fail to follow open design / open development -- but I see now your concern is about reusing openstack project names in stackforge git repository names | 15:13 |
| *** salv-orlando has joined #openstack-infra | 15:14 | |
| clarkb | BobBall and our image build sizes were about 3GB last I checked | 15:15 |
| clarkb | qcow2 are smaller | 15:15 |
| *** stevemar has joined #openstack-infra | 15:15 | |
| BobBall | Hmmm - I see. Then I'll have to look into that because it definitely says it's got a bunch of keys from somewhere else... | 15:15 |
| BobBall | and, for example, you've got an account on my nodes ;) | 15:16 |
| *** pblaho_ has joined #openstack-infra | 15:16 | |
| fungi | BobBall: yeah, there's a boolean class variable you can pass in when calling puppet agent on the manifest instructing it to not create those accounts | 15:16 |
| clarkb | BobBall http://ci.openstack.org/nodepool/configuration.html#providers template hostname there for not using opemstack.org | 15:17 |
| *** pblaho has quit IRC | 15:17 | |
| *** salv-orlando has quit IRC | 15:17 | |
| BobBall | ok thanks both. *digs more* | 15:17 |
| fungi | BobBall: we tried to make it as easy as possible for people outside our infrastructure to not give us access to their servers | 15:18 |
| BobBall | *grin* | 15:18 |
| *** bkero has joined #openstack-infra | 15:19 | |
| anteaya | ttx yes, no concern about the drivers moving from stackforge to openstack, my concern is proper use of openstack project names | 15:20 |
| anteaya | fungi: thank you, yes service name was what I was grasping for | 15:21 |
| *** rkukura has quit IRC | 15:22 | |
| *** koolhead17 has quit IRC | 15:23 | |
| *** pblaho_ is now known as pblaho | 15:23 | |
| *** e0ne is now known as e0ne_ | 15:23 | |
| *** ghostpl has quit IRC | 15:23 | |
| fungi | anteaya: i stole the term from the governance projects.yaml file | 15:23 |
| fungi | (also yay for now having _three_ projects.yaml files which do entirely different things!) | 15:24 |
| *** jgrimm is now known as zz_jgrimm | 15:24 | |
| *** zz_dimtruck is now known as dimtruck | 15:24 | |
| *** timcline has joined #openstack-infra | 15:24 | |
| *** raghav_mulugu has quit IRC | 15:25 | |
| *** garyh has joined #openstack-infra | 15:26 | |
| anteaya | making grep fun | 15:26 |
| anteaya | fungi: thanks, was mid email on something else so thanks for beating me to that | 15:27 |
| openstackgerrit | Matthias Runge proposed openstack/requirements: Raise cap for Django https://review.openstack.org/155353 | 15:28 |
| *** lnxnut_ has joined #openstack-infra | 15:28 | |
| *** derekh has quit IRC | 15:29 | |
| *** asselin has joined #openstack-infra | 15:30 | |
| *** jpich has quit IRC | 15:30 | |
| *** mfink_ has joined #openstack-infra | 15:31 | |
| *** ghostpl has joined #openstack-infra | 15:32 | |
| *** lnxnut has quit IRC | 15:32 | |
| BobBall | OK - tracked it down... Seems I must use "service nodepool start" (NODEPOOL_SSH_KEY is defined in /etc/default/nodepool which I think does the magic) - but that doesn't work with https://github.com/rasselin/os-ext-testing telling me to start nodepoold directly. | 15:32 |
| *** sputnik13 has quit IRC | 15:33 | |
| mkoderer | anteaya: thx for the review on https://review.openstack.org/#/c/155248/2 | 15:33 |
| openstackgerrit | Brian Hunter proposed openstack-infra/project-config: gate devstack on gate-trove-functional-dsvm-mysql https://review.openstack.org/155356 | 15:33 |
| asselin | BobBall, hi | 15:33 |
| *** rkukura has joined #openstack-infra | 15:33 | |
| BobBall | howdy | 15:33 |
| *** hashar has quit IRC | 15:34 | |
| mkoderer | anteaya: I need to talk to the ppl about moving them into an different IRC channel | 15:34 |
| asselin | BobBall, I lost my scroll back, so I need to go look at evesdrop | 15:34 |
| *** baoli has quit IRC | 15:34 | |
| *** hashar has joined #openstack-infra | 15:34 | |
| *** e0ne_ is now known as e0ne | 15:35 | |
| anteaya | mkoderer: sure, I think that having repo name/channel name agreement really helps peole | 15:35 |
| *** baoli_ has joined #openstack-infra | 15:35 | |
| *** hdd has joined #openstack-infra | 15:36 | |
| *** achanda has joined #openstack-infra | 15:36 | |
| anteaya | otherwise they eiether disappear or ask in here | 15:36 |
| *** wenlock has joined #openstack-infra | 15:36 | |
| anteaya | and personally I just search on repo name | 15:36 |
| *** garyh has quit IRC | 15:37 | |
| *** abhirc has joined #openstack-infra | 15:37 | |
| *** lnxnut_ has quit IRC | 15:37 | |
| openstackgerrit | Antoine Musso proposed openstack-infra/zuul: Update merge status after merge:merge is submitted https://review.openstack.org/128921 | 15:38 |
| *** jerryz has quit IRC | 15:39 | |
| *** unicell1 has joined #openstack-infra | 15:40 | |
| *** unicell has quit IRC | 15:41 | |
| dtantsur | hi infra! does anybody know if 'jq' utility is installed on slaves where (unit) tests a run? | 15:43 |
| *** ssam2 has quit IRC | 15:43 | |
| *** funzo has quit IRC | 15:43 | |
| *** ssam2 has joined #openstack-infra | 15:43 | |
| *** funzo has joined #openstack-infra | 15:43 | |
| *** achanda has quit IRC | 15:44 | |
| *** crc32 has quit IRC | 15:45 | |
| BobBall | asselin: Is this something you've seen before then? | 15:46 |
| asselin | BobBall, actually I didn't realize I could start nodepool as a service...it was on my todo list..I guess it got fixed a while back :) | 15:47 |
| BobBall | It doesn't work for me though :) | 15:47 |
| *** ParsectiX has joined #openstack-infra | 15:47 | |
| asselin | BobBall, But otherwise I don't follow the exact question | 15:47 |
| asselin | BobBall, oh, I manually start it in screen: nodepoold -d | 15:48 |
| asselin | as nodepool user: sudo su - nodepool | 15:48 |
| *** amotok___ has quit IRC | 15:48 | |
| asselin | BobBall, but looking now to see about using it as a service since that's much better | 15:48 |
| BobBall | Please tell me if it works :) | 15:50 |
| *** hdd has quit IRC | 15:52 | |
| *** lxsli has quit IRC | 15:53 | |
| *** yfried_ is now known as yfried|afk | 15:54 | |
| *** hashar has quit IRC | 15:54 | |
| *** devvesa has joined #openstack-infra | 15:54 | |
| *** salv-orlando has joined #openstack-infra | 15:54 | |
| *** BharatK has quit IRC | 15:55 | |
| fungi | dtantsur: jq is not installed, no | 15:55 |
| *** jlibosva has quit IRC | 15:55 | |
| dtantsur | that's pity.. thanks anyway | 15:56 |
| *** ssam2 has quit IRC | 15:56 | |
| fungi | BobBall: you should be able to export that envvar in the terminal session before interactively invoking nodepoold, as a workaround | 15:56 |
| openstackgerrit | Merged openstack-infra/subunit2sql: Add db api methods for getting recent run uuids https://review.openstack.org/153040 | 15:56 |
| *** AJaeger has joined #openstack-infra | 15:57 | |
| BobBall | I'm doing that now. | 15:57 |
| fungi | dtantsur: it could probably be installed if projects are going to find it useful, though with most of our unit tests being written in python which already has a built-in json parser, nobody's likely needed it until now | 15:57 |
| AJaeger | hi, https://review.openstack.org/153237 is a requirements job that fails, logs are at http://logs.openstack.org/37/153237/3/check/check-requirements-integration-dsvm/8128bae/console.html | 15:58 |
| AJaeger | Seems that installneutron is setup wrongly - is anybody looking at that already? | 15:58 |
| *** jlibosva has joined #openstack-infra | 15:58 | |
| dtantsur | fungi, I see... the problem is I have a test for stackforge/ironic-discoverd that also involves running it's ramdisk in a fake environment - that's where I need jq... | 15:58 |
| *** jedimike has joined #openstack-infra | 15:59 | |
| *** lxsli has joined #openstack-infra | 15:59 | |
| *** lxsli has quit IRC | 15:59 | |
| anteaya | AJaeger: I haven't heard of anyone mentioning it prior to you | 16:00 |
| fungi | dtantsur: how do you build the fake environment? | 16:00 |
| *** jgjl has joined #openstack-infra | 16:00 | |
| *** skolekonov has quit IRC | 16:01 | |
| fungi | AJaeger: interesting--that's trying to use a very old pbr | 16:01 |
| dtantsur | fungi, build a temporary dir, create bash scripts that emulate hardware utilities like lscpu, fdisk etc, run ramdisk script with PATH=this temp dir | 16:01 |
| AJaeger | fungi: we merged yesterday to requirements a patch by sdague called " don't install pbr from upstream source" | 16:01 |
| fungi | dtantsur: okay, so you want to copy jq into that chroot i guess? | 16:01 |
| *** doude has quit IRC | 16:01 | |
| *** yfried|afk is now known as yfried_ | 16:01 | |
| *** dangers_away is now known as dangers | 16:02 | |
| fungi | AJaeger: i suspect pbr is being installed on trusty from ubuntu distro packages as a dependency of something else | 16:02 |
| clarkb | AJaeger looks like neutron.hooks.setup_hook is busted | 16:02 |
| fungi | AJaeger: and so pip isn't upgradint it | 16:02 |
| dtantsur | fungi, well... yeah I probably can just download the binary and put it there. that's an idea. | 16:02 |
| clarkb | fails to import | 16:02 |
| fungi | dtantsur: well, having it available locally is one option, but if the job already does things as root it can just apt-get install jq while it's doing other things | 16:03 |
| sdague | fungi: ok, so perhaps we backed off too much there and still install a pbr install -U at the beginning | 16:03 |
| *** doude has joined #openstack-infra | 16:03 | |
| *** marcusvrn1 has joined #openstack-infra | 16:03 | |
| dtantsur | fungi, I don't think our unit tests (and it's implemented as a unit-test-alike Python job) have root access, do they? | 16:03 |
| *** marcusvrn has quit IRC | 16:04 | |
| fungi | dtantsur: i | 16:04 |
| fungi | er | 16:04 |
| fungi | dtantsur: i'm probably misunderstanding how you restrict execution to a specific environment in that case | 16:04 |
| fungi | dtantsur: not using chroot or anything? | 16:04 |
| fungi | dtantsur: you mention running a ramdisk, which seems like it would need elevated privileges | 16:04 |
| dtantsur | fungi, PATH=/dir bash -eux ramdisk.script :) | 16:04 |
| dtantsur | fungi, that's why I use fake tools like fdisk | 16:05 |
| fungi | you use a faked fdisk? neat | 16:05 |
| dtantsur | just bash scripts outputting pre-defined "correct" output | 16:05 |
| *** zz_jgrimm is now known as jgrimm | 16:05 | |
| fungi | so you're simply restricting execution evironment by limiting path, that seems like it would still go well outside your minimal environment for any dynamically-linked libraries | 16:06 |
| AJaeger | sdague: will you look into that requirements problem, please? I know too little to help and I should be in my bed soon again (got the flu ;( | 16:06 |
| *** claudiub has quit IRC | 16:06 | |
| dtantsur | fungi, yep, especially since jq does not have any dependencies. thanks! | 16:06 |
| fungi | dtantsur: do you also set restrictive envvars for the dynamic linker? or you just audit all the executables to make sure they're statically compiled? (not even using libc?) | 16:07 |
| *** BharatK has joined #openstack-infra | 16:07 | |
| anteaya | AJaeger: :( | 16:07 |
| asselin | BobBall, sorry, starting nodepool as a service doesn't work for me either.... | 16:07 |
| *** lnxnut has joined #openstack-infra | 16:07 | |
| *** salv-orlando has quit IRC | 16:08 | |
| *** garyh has joined #openstack-infra | 16:08 | |
| dtantsur | fungi, no, I don't touch LD_LIBRARY_PATH and make fake utilities are in bash (that's not really a lot of them) | 16:08 |
| asselin | BobBall, but you should be able to start it manually in screen | 16:08 |
| asselin | BobBall, and it should have the Nodepool ssh key | 16:08 |
| asselin | BobBall, just make sure it's in your env as nodepool user before starting nodepoold | 16:09 |
| fungi | dtantsur: so like, for example, tools you're calling are using the libc.so.6 from /usr/lib rather than the one from your "ramdisk"? | 16:09 |
| dtantsur | fungi, I don't download the whole ramdisk, only the specific bash script that is its core, i.e. https://raw.githubusercontent.com/openstack/diskimage-builder/master/elements/ironic-discoverd-ramdisk/init.d/80-ironic-discoverd-ramdisk | 16:11 |
| fungi | dtantsur: anyway, your design choices aside, you can simply add a builder to apt-get install that package before the revoke-sudo builder in your job (which may require a separate job definition if you're currently sharing an existing job template with other projects) | 16:11 |
| fungi | dtantsur: or if it's a utility likely to be used by other jobs too we could just preinstall it on the system | 16:11 |
| *** lnxnut has quit IRC | 16:12 | |
| dtantsur | hmm... I will think about it | 16:12 |
| anteaya | I never knew we had these metrics | 16:13 |
| anteaya | http://activity.openstack.org/dash/browser/irc.html | 16:13 |
| anteaya | I be chatty | 16:14 |
| *** kmartin has joined #openstack-infra | 16:14 | |
| *** salv-orlando has joined #openstack-infra | 16:14 | |
| jlk | hah, jroll is on a roll | 16:14 |
| jlk | neat bit of metrics. Nice spike of activity last year. | 16:15 |
| nibalizer | cool! | 16:16 |
| anteaya | funny decrease in january traffic this year | 16:17 |
| anteaya | compared to last | 16:17 |
| openstackgerrit | Mehdi Abaakouk proposed openstack-infra/project-config: oslo.messaging: enabling fonctionnal tests https://review.openstack.org/155389 | 16:17 |
| *** AJaeger has quit IRC | 16:18 | |
| openstackgerrit | Mehdi Abaakouk proposed openstack-infra/project-config: gnocchi: enabling fonctionnal tests https://review.openstack.org/155391 | 16:19 |
| *** jlibosva has quit IRC | 16:20 | |
| mtreinish | anteaya: they're always watching... | 16:20 |
| *** pcrews has joined #openstack-infra | 16:22 | |
| anteaya | mtreinish: and measuring apparently | 16:22 |
| *** ihrachyshka has quit IRC | 16:22 | |
| *** Parsecti_ has joined #openstack-infra | 16:22 | |
| anteaya | why can I never find shoes that fit, that is my question | 16:22 |
| anteaya | well rarely, rarely find shoes that fit | 16:23 |
| *** arxcruz has quit IRC | 16:24 | |
| *** ParsectiX has quit IRC | 16:24 | |
| *** SlickNik has quit IRC | 16:24 | |
| *** yamahata has joined #openstack-infra | 16:26 | |
| *** sdake__ is now known as sdake | 16:26 | |
| *** claudiub has joined #openstack-infra | 16:27 | |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: MidoNet's neutron plugin to StackForge https://review.openstack.org/155299 | 16:27 |
| *** dimtruck is now known as zz_dimtruck | 16:28 | |
| mtreinish | anteaya: I'm guessing if you ask for shoes that fit frequently on random irc channels one day they'll just show up at your front door | 16:28 |
| *** hashar has joined #openstack-infra | 16:29 | |
| *** ildikov has quit IRC | 16:29 | |
| *** mbacchi has quit IRC | 16:29 | |
| *** morgabra has quit IRC | 16:29 | |
| *** mbacchi has joined #openstack-infra | 16:30 | |
| *** unicell1 has quit IRC | 16:30 | |
| *** SlickNik has joined #openstack-infra | 16:30 | |
| *** andreaf has quit IRC | 16:30 | |
| anteaya | 5 size men's shoes | 16:30 |
| anteaya | they fit perfect | 16:30 |
| anteaya | noone makes them | 16:30 |
| *** andreaf has joined #openstack-infra | 16:31 | |
| anteaya | here is an interesting metric: http://activity.openstack.org/dash/browser/ | 16:31 |
| anteaya | 17,000 people use our irc channels | 16:31 |
| anteaya | 4,000 post to the mailing lists | 16:31 |
| anteaya | so 25% of folks post to the mailing list | 16:32 |
| *** carl_baldwin has quit IRC | 16:32 | |
| anteaya | of folks that use irc | 16:32 |
| *** zz_dimtruck is now known as dimtruck | 16:32 | |
| *** otter768 has quit IRC | 16:33 | |
| anteaya | since you can't measure how many people recieve/read a mailing list post | 16:33 |
| *** morgabra has joined #openstack-infra | 16:33 | |
| *** zz_avozza is now known as avozza | 16:33 | |
| anteaya | I'm going to have to start skimming the irc vs mailing list thread | 16:34 |
| anteaya | if I make a point that the thread refutes please link me | 16:34 |
| jeblair | i don't read that as a "vs". more of a "+". | 16:34 |
| mtreinish | anteaya: there are subscriber lists available in mailman | 16:34 |
| *** imcsk8 has quit IRC | 16:34 | |
| *** imcsk8 has joined #openstack-infra | 16:34 | |
| mtreinish | but I think you need special perms to be able to view it | 16:35 |
| anteaya | jeblair: great, glad there are different ways of looking at it | 16:35 |
| *** avozza is now known as zz_avozza | 16:35 | |
| *** dannywilson has joined #openstack-infra | 16:35 | |
| anteaya | mtreinish: funny I would think if the subscriber lists were part of the calculation of mailing list participant the numbers would be closer to irc particpation | 16:35 |
| *** jgallard__ has quit IRC | 16:35 | |
| anteaya | jeblair: to be clear, I feel it is a '+' but I'm feeling some folks feel it isn't | 16:36 |
| hashar | good to see you guys using the VizGrimoire activity system :] | 16:36 |
| jeblair | anteaya: i'm skeptical that 17,000 isn't the result of counting errors | 16:36 |
| anteaya | jeblair: I'm open to your skeptisim | 16:36 |
| mtreinish | anteaya: well it's there, but I don't think anyone except infra has access | 16:36 |
| anteaya | jeblair: I just have nothing to check to refute that number | 16:36 |
| mtreinish | anteaya: I think what's more interesting in the irc graphs the clear dips for summits | 16:36 |
| jeblair | anteaya: perhaps all those "Guest1234" nicks from the web interface, etc... | 16:36 |
| anteaya | jeblair: ah okay, yeah that would drop it down a couple thousand | 16:37 |
| anteaya | mtreinish: yes | 16:37 |
| jeblair | anteaya: and 'jeblair_' and 'jeblair' counting twice. and zz_. and _afk...etc | 16:37 |
| anteaya | jeblair: yeah, okay so drop by another third maybe | 16:37 |
| anteaya | so still hitting around 5000 in my head | 16:38 |
| *** ghostpl has quit IRC | 16:38 | |
| *** dannywilson has quit IRC | 16:38 | |
| *** jgallard__ has joined #openstack-infra | 16:38 | |
| anteaya | which is closer to the mailing list count, if that can be considered reasonable | 16:38 |
| jeblair | i mean, i'm making assumptions that it's not tracking nick changes. i do not know that. | 16:38 |
| anteaya | jeblair: fair | 16:38 |
| *** dannywilson has joined #openstack-infra | 16:38 | |
| *** patrickeast has joined #openstack-infra | 16:39 | |
| pleia2 | good morning | 16:40 |
| mtreinish | http://activity.openstack.org/dash/browser/demographics.html this is not what I was expecting from the title | 16:40 |
| *** hashar has quit IRC | 16:40 | |
| *** sarob has joined #openstack-infra | 16:40 | |
| jeblair | mtreinish: "of or relating to the study of changes that occur in large groups of people over a period of time" | 16:42 |
| anteaya | morning pleia2 | 16:42 |
| markus_z | god I love dashboards. I have no idea what it means but its beautiful :) | 16:42 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/puppet-zuul: Now can clone from zuul using apache 2.4 https://review.openstack.org/155154 | 16:43 |
| jeblair | mtreinish: seems to fit the bill :) | 16:43 |
| mtreinish | jeblair: heh, thanks :) | 16:43 |
| *** jgjl has quit IRC | 16:44 | |
| anteaya | mtreinish: where you thinking more like geographical demographics or something? | 16:44 |
| anteaya | were | 16:44 |
| *** dustins has joined #openstack-infra | 16:44 | |
| mtreinish | anteaya: yeah, when I saw demographics I just wasn't thinking retention rates | 16:44 |
| anteaya | in the spell checking metric for openstack communication, I'd be at the bottom | 16:44 |
| *** sarob has quit IRC | 16:45 | |
| anteaya | mtreinish: I understand your expectation, I also understand jeblair's posted definition | 16:45 |
| *** erlon has joined #openstack-infra | 16:45 | |
| *** yamahata has quit IRC | 16:45 | |
| *** lxsli has joined #openstack-infra | 16:46 | |
| *** afazekas has quit IRC | 16:46 | |
| *** yamahata has joined #openstack-infra | 16:46 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/puppet-zuul: Now can clone from zuul using apache 2.4 https://review.openstack.org/155154 | 16:48 |
| *** ghostpl has joined #openstack-infra | 16:48 | |
| *** ChuckC_ has quit IRC | 16:51 | |
| *** ChuckC_ has joined #openstack-infra | 16:51 | |
| *** ChuckC_ is now known as ChuckC | 16:51 | |
| *** leopoldj has quit IRC | 16:52 | |
| *** ihrachyshka has joined #openstack-infra | 16:53 | |
| *** e0ne has quit IRC | 16:54 | |
| *** MaxV has quit IRC | 16:54 | |
| *** sputnik13 has joined #openstack-infra | 16:54 | |
| *** gyee has joined #openstack-infra | 16:54 | |
| *** lnxnut has joined #openstack-infra | 16:54 | |
| *** e0ne has joined #openstack-infra | 16:55 | |
| *** dims__ has quit IRC | 16:56 | |
| *** ashleighfarnham has joined #openstack-infra | 16:56 | |
| *** koolhead17 has joined #openstack-infra | 16:57 | |
| *** bdpayne has joined #openstack-infra | 16:57 | |
| *** dims__ has joined #openstack-infra | 16:57 | |
| * tchaypo reads about cross-repo dependencies in zuul | 16:58 | |
| * tchaypo gets happy | 16:58 | |
| * anteaya likes it when tchaypo gets happy | 16:58 | |
| tchaypo | not as much as i like it | 16:59 |
| anteaya | you are quiet likely correct | 16:59 |
| jeblair | oh i was hoping you were going to fight it out and it would become an epic battle | 16:59 |
| tchaypo | but a quiet one | 17:00 |
| *** hdd has joined #openstack-infra | 17:00 | |
| jeblair | "no, i'm happier!" "no, I'M happier!!" | 17:00 |
| jeblair | :) | 17:00 |
| tchaypo | I can’t imagine an epic battle involving anteaya though | 17:00 |
| *** baoli_ has quit IRC | 17:00 | |
| *** ddieterly has quit IRC | 17:00 | |
| tchaypo | unless her weapon of choice is a quiet gentle disagreement | 17:00 |
| *** ddieterly has joined #openstack-infra | 17:01 | |
| anteaya | funny thing been a while since I was able to employ that tactic | 17:01 |
| anteaya | though that is my tactic of choice | 17:01 |
| jeblair | anteaya: it's a good one | 17:01 |
| tchaypo | you’ve never employed anything more on me... | 17:01 |
| anteaya | jeblair: I've never seen it fail | 17:01 |
| anteaya | tchaypo: you bring out the quiet gentle side of me | 17:02 |
| anteaya | both in agreement and disagreement | 17:02 |
| tchaypo | I’m going to need to dig into the CRD stuff to see if I can use it to replace prep_source_repos’ logic | 17:02 |
| *** dmorita has joined #openstack-infra | 17:03 | |
| * anteaya hands tchaypo a flashlight and lunch | 17:03 | |
| tchaypo | I’m guessing there’s something that’s constructing the DAG of the dependent changes based on the Depends-On: headers | 17:03 |
| BobBall | asselin: So I started "nodepoold -d" with NODEPOOL_SSH_KEY defined but it seems not to get passed through... How can I test/where can I see where it _should_ be passed through (e.g. log files etc?) | 17:03 |
| tchaypo | if I can replace that with something that creates a DAG based on the the yaml file of changes I’ve got, it sounds like I should be able to plug right in.. | 17:03 |
| jeblair | tchaypo: i'm not sure what you're using prep-source-repos for, but if you're working with zuul and you use zuul-cloner, it will do the right thing. | 17:04 |
| asselin | BobBall, It should be seen when nodepool creates the image | 17:04 |
| BobBall | oh d'oh how stupid of me | 17:04 |
| BobBall | I didn't regenerate the image | 17:04 |
| * BobBall hides under a rock | 17:04 | |
| BobBall | I should have thought of that one. | 17:04 |
| BobBall | I was just launching nodes and hitting my head... | 17:05 |
| *** e0ne has quit IRC | 17:05 | |
| *** e0ne has joined #openstack-infra | 17:06 | |
| asselin | BobBall, hmm...seems to no longer be here: https://github.com/openstack-infra/project-config/blob/master/nodepool/scripts/prepare_node.sh#L67 | 17:06 |
| *** abhirc has quit IRC | 17:07 | |
| *** jcoufal_ has quit IRC | 17:07 | |
| *** tkelsey has joined #openstack-infra | 17:08 | |
| asselin | BobBall, seems to have gotten removed recently: https://github.com/openstack-infra/project-config/commit/90e3c200bd51940fe58f9640621d5bfa97e28d48#diff-aeed75bb52c22c9082d39c2b15f342a7L71 | 17:08 |
| tchaypo | jeblair: one simple use case is when I want to run gerrit with 6 different un-landed patches - I list them in the yaml, and p_s_r checks out master then applies the patches for me | 17:08 |
| jeblair | tchaypo: run gerrit? | 17:08 |
| tchaypo | a more complex use case is trying to build a CI cluster that depends on ~12 unlanded patches in tripleo-image-elements and diskimage-builder and tripleo-incubator and heat - it checks them all out and applies the list of patches to each, then exports DIB variables to make DIB use the patched repos | 17:09 |
| tchaypo | s/gerrit/gertty/g | 17:09 |
| BobBall | asselin: Where's it removed? | 17:09 |
| asselin | BobBall, nevermind...need to scroll to the right: https://github.com/openstack-infra/project-config/blob/master/nodepool/scripts/prepare_node.sh#L73 | 17:09 |
| asselin | BobBall, it's still there | 17:09 |
| BobBall | Ah good - I saw it there so was getting confused :) | 17:10 |
| jeblair | tchaypo: i'm still not following -- are you trying to test these patches with zuul (maybe in the tripleo pipelines)? or are you trying to run them locally? | 17:10 |
| *** pblaho_ has joined #openstack-infra | 17:10 | |
| *** jlibosva has joined #openstack-infra | 17:10 | |
| *** Parsecti_ is now known as ParsectiX | 17:11 | |
| asselin | BobBall, manual nodepool updates in currently broken. Fix is here or you can go back to an older revision. https://review.openstack.org/#/c/151749/ | 17:11 |
| BobBall | tsok - I'm just deleting the image and let nodepool bake a new one. | 17:11 |
| asselin | BobBall, that works toos | 17:12 |
| *** coolsvap is now known as coolsvap_ | 17:12 | |
| tchaypo | jeblair: locally | 17:13 |
| *** pblaho has quit IRC | 17:13 | |
| *** thorst has quit IRC | 17:13 | |
| jeblair | tchaypo: gotcha. you could _probably_ extract the code from zuul, but there's quite a bit and its spread throughout. it's not in a convenient "make DAG from gerrit query" function. | 17:14 |
| tchaypo | but what p_s_r does is quite similar to what zuul does. I’d really like to make p_s_r a wrapper around zuul_merger/zuul_cloner | 17:14 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/puppet-zuul: Now can clone from zuul using apache 2.4 https://review.openstack.org/155154 | 17:14 |
| jeblair | tchaypo: but most of what you would need is in lib/gerrit. that's the part that crawls gerrit to contstruct the dependency relationships. | 17:15 |
| *** carl_baldwin has joined #openstack-infra | 17:15 | |
| jeblair | tchaypo: and trigger/gerrit | 17:15 |
| openstackgerrit | Sean Dague proposed openstack/requirements: also install a new pbr into the venvs https://review.openstack.org/155410 | 17:15 |
| tchaypo | jeblair: <3 | 17:16 |
| *** Ala has quit IRC | 17:17 | |
| *** hdd has quit IRC | 17:18 | |
| *** unicell has joined #openstack-infra | 17:20 | |
| *** hdd has joined #openstack-infra | 17:21 | |
| *** tqtran_afk has joined #openstack-infra | 17:23 | |
| *** tqtran_afk is now known as tqtran | 17:23 | |
| *** jgallard__ has quit IRC | 17:23 | |
| *** jgallard__ has joined #openstack-infra | 17:24 | |
| *** EmilienM is now known as EmilienM|afk | 17:25 | |
| anteaya | reed: any idea what the difference is between a code developer and a code submitter? http://activity.openstack.org/dash/browser/ | 17:26 |
| anteaya | reed if you aren't the correct person to ask, do re-direct | 17:26 |
| *** dtantsur is now known as dtantsur|afk | 17:26 | |
| *** jistr has quit IRC | 17:27 | |
| *** salv-orlando has quit IRC | 17:29 | |
| *** markus_z has quit IRC | 17:29 | |
| *** armax has joined #openstack-infra | 17:29 | |
| *** salv-orlando has joined #openstack-infra | 17:29 | |
| openstackgerrit | Merged openstack-infra/project-config: gnocchi: enabling fonctionnal tests https://review.openstack.org/155391 | 17:29 |
| *** devvesa has quit IRC | 17:30 | |
| *** liam_ has joined #openstack-infra | 17:30 | |
| *** liam_ is now known as Guest13534 | 17:30 | |
| *** Guest13534 has quit IRC | 17:30 | |
| *** Guest41759 has quit IRC | 17:31 | |
| jroll | jlk: lol | 17:32 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: prepare_node.sh line lengths are too long https://review.openstack.org/155417 | 17:33 |
| openstackgerrit | Merged openstack-infra/shade: Add service_catalog property https://review.openstack.org/155304 | 17:33 |
| *** jgallard__ has quit IRC | 17:33 | |
| openstackgerrit | Merged openstack-infra/nodepool: Fail gracefully if disk-image-create is missing https://review.openstack.org/153937 | 17:35 |
| openstackgerrit | Merged openstack-infra/project-config: Allow proliantutils to be published on PyPI https://review.openstack.org/155218 | 17:35 |
| openstackgerrit | Merged openstack-infra/project-config: remove check-neutron-dsvm-functional from devstack gates https://review.openstack.org/155323 | 17:36 |
| anteaya | sdague: does anyone from neutron know about https://review.openstack.org/#/c/155323/ ? | 17:37 |
| sputnik13 | gerrit/git-review question... if I downloaded a review with `git review -d <review number>`, how can I get new patch sets for the review? | 17:38 |
| *** pblaho__ has joined #openstack-infra | 17:39 | |
| sputnik13 | short of deleting the branch and re-executing git review -d | 17:39 |
| anteaya | I realize it is a change to testing for devstack, but it would be nice to see someone was at least aware in the review | 17:39 |
| sdague | anteaya: don't know, but none of the other jobs that are using devstack to just install packages co-gate with devstack, so I don't think it's appropriate for this one to | 17:39 |
| anteaya | oh okay | 17:39 |
| sdague | though, honestly, I didn't expect it to merge that fast, because it's usually many days for those things to merge | 17:40 |
| anteaya | sdague: yeah, the speed was kind of what caused me to ask as well | 17:40 |
| fungi | HenryG: looks like 155152,1 just merged so neutron should be set now. sorry about the delays | 17:40 |
| clarkb | sputnik13 just rerun git review -d | 17:40 |
| anteaya | I'm all for fast turn over but not at the expense of excluding impacted people/projects | 17:41 |
| sputnik13 | clarkb: ah cool, thanks | 17:41 |
| HenryG | fungi: thanks! Don't be sorry | 17:41 |
| sdague | yeh, once marun pops up will make sure to heads up him on it | 17:41 |
| anteaya | sdague: thanks | 17:42 |
| *** ChuckC has quit IRC | 17:42 | |
| openstackgerrit | Merged openstack-infra/zuul: Remove lockfile from requirements https://review.openstack.org/148252 | 17:42 |
| *** emagana has joined #openstack-infra | 17:42 | |
| *** ChuckC has joined #openstack-infra | 17:42 | |
| openstackgerrit | Merged openstack-infra/nodepool: Add doc for 'pool' parameter for nodepool provider config https://review.openstack.org/151286 | 17:42 |
| *** pblaho_ has quit IRC | 17:43 | |
| openstackgerrit | Merged openstack-infra/nodepool: Remove lockfile from requirements https://review.openstack.org/148565 | 17:43 |
| marun | sdague: uh | 17:43 |
| marun | I wish you hadn't done that | 17:43 |
| marun | sdague: our functional job depends heavily on implementation details of devstack | 17:43 |
| marun | sdague: unless you'd like to see us remove that dependency entirely - and duplicate effort in the process - I think we need to continue gating | 17:44 |
| *** jlibosva has quit IRC | 17:45 | |
| sdague | marun: it's not sustainable for devstack to run ever single functional job for every project that uses devstack. And not be able to merge devstack code because projects have racy tests | 17:46 |
| marun | sdague: ok | 17:46 |
| marun | sdague: then I'll remove all dependency | 17:46 |
| marun | including installation of system packages | 17:46 |
| sdague | so let's figure out what behavior in devstack is required | 17:46 |
| jeblair | marun, sdague: (maybe this already exists), but we could define what the devstack API is that neutron is using, devstack could test that to make sure we don't break neutron or anything else, yeah? | 17:46 |
| sdague | jeblair: right, exactly | 17:47 |
| sdague | like any other functional testing | 17:47 |
| *** bknudson has quit IRC | 17:47 | |
| *** ildikov has joined #openstack-infra | 17:47 | |
| *** BobBall is now known as BobBall_AWOL | 17:48 | |
| *** SumitNaiksatam has joined #openstack-infra | 17:52 | |
| *** marun has quit IRC | 17:52 | |
| *** andreykurilin__ has quit IRC | 17:52 | |
| *** baoli has joined #openstack-infra | 17:53 | |
| *** melwitt has joined #openstack-infra | 17:53 | |
| *** dimtruck is now known as zz_dimtruck | 17:54 | |
| pc_m | infra folks... I've got a series of general questions related to Tempest tests, and some specific questions with several (conflicting) tests. Need advice. Anyone have some time to answer? | 17:55 |
| anteaya | pc_m: post away | 17:55 |
| anteaya | though tempest is the perview of -qa | 17:55 |
| pc_m | OK. Here goes... easy stuff first... | 17:55 |
| anteaya | purview | 17:55 |
| *** coolsvap_ is now known as coolsvap | 17:55 | |
| pc_m | What is the difference between a Tempest test and a functional test? | 17:55 |
| * anteaya continues to work on her poor spelling metric | 17:55 | |
| anteaya | pc_m: a functional test exercises the service | 17:56 |
| anteaya | pc_m: tempest exercises all the services | 17:56 |
| * anteaya trusts others will correct where she is wrong | 17:56 | |
| fungi | pc_m: also tempest is a specific tool/suite not necessarily a "type" of test | 17:56 |
| jeblair | clarkb, fungi, mordred, flaper87: i kind of want to weigh in on the "lets keep our community open" thread and say that, absolutely, in no uncertain terms, should discussion of barbecue happen in private channels. i would hate to miss out on knowing about a good bbq joint. but i feel like that thread is really good and i don't want to derail it. so i'll just say so here. :) | 17:56 |
| *** baoli has quit IRC | 17:57 | |
| anteaya | jeblair: I support your ability to access knowledge of good bbq joints | 17:57 |
| fungi | jeblair: i am 100% on board with insisting that _all_ discussions of barbecue happen in #openstack-infra specifically | 17:57 |
| flaper87 | jeblair: hahahaha, I kinda implied that in my last answer | 17:57 |
| flaper87 | :D | 17:57 |
| anteaya | jeblair: I will not be jointing you there | 17:57 |
| anteaya | joining either | 17:57 |
| fungi | jeblair: so mentioning it in the ml thread would actually be in violation of that requirement | 17:57 |
| pc_m | fungi: ok, but the tempest tests are in their own repo currently, and functional are in tree. | 17:58 |
| clarkb | jeblair: ha | 17:58 |
| flaper87 | jeblair: fungi bbq discussions are the reason I'm always around in this channel | 17:58 |
| fungi | flaper87: definitely among the better reasons to /join us | 17:58 |
| clarkb | I wonder if I can get a smoked brisket through canadian customs | 17:58 |
| anteaya | pc_m: yes, functional tests are for exercising the service of which they are in | 17:58 |
| flaper87 | I'm happy with an ascii pork knee if I can't fly where it is happening | 17:58 |
| anteaya | clarkb: no, no meat allowed to cross the boarder | 17:58 |
| pc_m | fungi: anteaya: So let me get to more specific points... | 17:58 |
| clarkb | anteaya: even cooked? | 17:58 |
| anteaya | clarkb: correct | 17:59 |
| fungi | clarkb: i wonder if we can find smoked elk brisket in bc | 17:59 |
| clarkb | anteaya: wow you all have mad cow we don't :P | 17:59 |
| fungi | clarkb: or maybe moose? | 17:59 |
| pc_m | For VPN, there is a desire for a scenario test that exercises an end-to-end IPSec connection. | 17:59 |
| anteaya | clarkb: asian folks like beef candy and someone is being charged for not telling customs it was beef, since they consider it candy | 17:59 |
| anteaya | clarkb: I know, it makes zero sense to me | 17:59 |
| clarkb | anteaya: though I think that was a few years back unsure if still an issue with canadian beef stock | 17:59 |
| pc_m | I guess this would really be a functional test, then? | 17:59 |
| *** zz_dimtruck is now known as dimtruck | 18:00 | |
| fungi | pc_m: you should talk to mtreinish and sdague but i believe their suggestion is that it can be implemented outside of tempest while still using some of that framework | 18:00 |
| anteaya | fungi clarkb moose and elk are different than beef, they can only be handled by special folks | 18:00 |
| pc_m | mtreinish: sdague: ^^ | 18:00 |
| anteaya | which is why you don't see the meat in restaurants | 18:00 |
| anteaya | clarkb: the canaidan beef industry is a mess | 18:00 |
| fungi | anteaya: aww, do bison range that far north? | 18:01 |
| fungi | i've had smoked bison brisket before, and it's definitely tasty too | 18:01 |
| openstackgerrit | Jaime Flynn proposed openstack-infra/jenkins-job-builder: adds support for not condition in conditional_step https://review.openstack.org/155423 | 18:01 |
| * jlk ponders skipping the beef tonight then. | 18:01 | |
| anteaya | clarkb: due to our foreign worker laws (which amount to slavery) and foreign ownship of beef manufacuting plants (brazil) and food security people not doing their job (federal government) | 18:01 |
| jlk | <--- on his way to canada right now | 18:01 |
| *** jackie_acosta has joined #openstack-infra | 18:02 | |
| *** teran has quit IRC | 18:02 | |
| pc_m | fungi: I guess I'm trying to understand if this test should be in the tempest repo or under the functional test section. | 18:02 |
| jeblair | jlk: you're allowed to bring it in if you've already eaten it! :) | 18:02 |
| clarkb | anteaya: so meat is allowed in fresh frozen and chilled, but must be packaged with identifying marks indicating what it is and proof of country origin may be required | 18:02 |
| clarkb | anteaya: those are rules specific to USA -> Canada travel | 18:02 |
| anteaya | fungi: not sure, but I think folks own bison in canada | 18:02 |
| anteaya | not sure about wild bison | 18:02 |
| clarkb | anteaya: cooked food items appear to be ok as long as they are not meat | 18:02 |
| fungi | pc_m: in general, discussion of patches to tempest or reusing parts of tempest usually goes on in #openstack-qa since that's where their developer community primarily congregate | 18:02 |
| pc_m | fungi: It needs to create routers, nets, subnets, and then create VPN connection and check ping across. | 18:03 |
| jlk | food items? Not meat? I can't compute. | 18:03 |
| *** _buzz has joined #openstack-infra | 18:03 | |
| anteaya | clarkb: then you know more than I, all I can advise is claim it all | 18:03 |
| anteaya | clarkb: if you claim it worst they can do is confiscate | 18:03 |
| *** tiswanso has quit IRC | 18:03 | |
| anteaya | clarkb: if you don't claim it, then they get very upset | 18:03 |
| mtreinish | pc_m: so I can say the adv services w/ regard to tempest that the tests don't belong in tempest | 18:03 |
| clarkb | anteaya: well I think a smoked brisket would not be allowed | 18:03 |
| clarkb | anteaya: so will skip it | 18:03 |
| fungi | pc_m: can it do that entirely via api calls to neutron et cetera? otherwise you're talking about something broader than tempest (probably a custom devstack plugin?) | 18:03 |
| jlk | pc_m: you might also be interested in Rally, which wraps around tempest and does many things like that | 18:04 |
| *** tiswanso_ has joined #openstack-infra | 18:04 | |
| anteaya | clarkb: yeah, food across the border is a pain, the only universal currency is chocolate, so I stick to that | 18:04 |
| mtreinish | pc_m: the medium term plan is to migrate any existing adv. services tempests tests into the neutron-*aaS repos | 18:04 |
| *** claudiub has quit IRC | 18:04 | |
| pc_m | mtreinish: Should I move this to openstack-qa? | 18:04 |
| jlk | although multi-instance communications with rally are a bit future looking as well. | 18:04 |
| *** jp_at_hp has quit IRC | 18:04 | |
| pc_m | or proceed here? | 18:04 |
| mtreinish | pc_m: we can, but I'm going to be afk for a bit to eat | 18:05 |
| anteaya | clarkb: and canadian customs can tolerate a lot, except dishonesty | 18:05 |
| mtreinish | pc_m: I would check with dougwig first, because he was spearheading the test migrations | 18:05 |
| anteaya | clarkb: so tell them about everything | 18:05 |
| pc_m | mtreinish: Well this is more involved that just the migration. Should I elaborate on the quagmire I have? | 18:06 |
| anteaya | clarkb: I bore them until the wave me on when I return home | 18:06 |
| *** ivar-lazzaro has joined #openstack-infra | 18:06 | |
| pleia2 | nibalizer: can you have a look at my comment on 153305? | 18:06 |
| clarkb | anteaya: ya typical customs dealings are "be honest and don't have anything on your person that is illegal in either source or destination for all transit points" | 18:06 |
| pc_m | mtreinish: Or wait till you're available and discuss on QA IRC | 18:06 |
| anteaya | clarkb: you are fine to travel then | 18:06 |
| mtreinish | pc_m: sure, I'll ping you on -qa when I'm back | 18:07 |
| pc_m | mtreinish: OK. Thanks! | 18:07 |
| anteaya | oh and general note, guns are not permissable in canada, not for carrying around | 18:07 |
| anteaya | for anyone for whom that is news | 18:07 |
| nibalizer | pleia2: sure thing | 18:07 |
| *** ivar-lazzaro has quit IRC | 18:07 | |
| clarkb | krotscheck: re 154672 that explanation clarify things? /me wants to install pleia2 rootkit already | 18:07 |
| *** ivar-lazzaro has joined #openstack-infra | 18:08 | |
| clarkb | btw jeblair fungi SergeyLukjanov mordred jhesketh 154672 is a fun one :) | 18:08 |
| krotscheck | clarkb: You can always override me. I’m not core. | 18:08 |
| *** ghostpl has quit IRC | 18:08 | |
| anteaya | krotscheck: I think he would prefer to include you, not override | 18:08 |
| pc_m | anteaya: fungi: jlk: I'll wait till mtreinish is available and post on Q/A. Let me know, if you want me to ping you as well. | 18:08 |
| pleia2 | I didn't mean to cause contention with this | 18:08 |
| anteaya | pleia2: well I think mtreinish is your best contact | 18:09 |
| anteaya | pc_m: that was for you | 18:09 |
| clarkb | krotscheck: yup was hoping that nibalizer's comment clarified the change and why it is being made | 18:09 |
| anteaya | pc_m: and I would support you following his advice and including dougwig in your musings | 18:09 |
| openstackgerrit | Merged openstack-infra/system-config: Remove Puppet 2/Puppet 3 logic https://review.openstack.org/154788 | 18:09 |
| pc_m | anteaya: Sure, though the issue is much broader than the migration effort. | 18:10 |
| anteaya | pc_m: fair enough | 18:10 |
| SergeyLukjanov | clarkb, yeah, fun one :) | 18:10 |
| openstackgerrit | Andrew Lazarev proposed openstack/requirements: Bump sahara client version https://review.openstack.org/155428 | 18:10 |
| *** salv-orlando has quit IRC | 18:10 | |
| openstackgerrit | Merged openstack-infra/system-config: HPCloud has deprecated another Precise image https://review.openstack.org/155062 | 18:10 |
| anteaya | pc_m: no problem making me aware, happy to know about stuff | 18:10 |
| openstackgerrit | Merged openstack-infra/system-config: Update email_stats.py for the big tent https://review.openstack.org/154719 | 18:10 |
| anteaya | pc_m: and thanks for thinking of me :) | 18:10 |
| *** salv-orlando has joined #openstack-infra | 18:10 | |
| krotscheck | clarkb: So why am I admin? I have no status more special than NikitaKonovalov? | 18:11 |
| pc_m | anteaya: sure np. Will ping you, when we discuss in openstack-qa later | 18:11 |
| anteaya | pc_m: thanks | 18:11 |
| jeblair | pleia2, krotscheck: adding plea2 is the right thing -- root is root, there is no further subdivision and to exclude her would be unfair. considering that she will have server-level root acces and responsibility for the operation of the service, it makes sense to give her that level of access. | 18:11 |
| krotscheck | jeblair: So why am I being special cased? | 18:12 |
| *** salv-orl_ has joined #openstack-infra | 18:12 | |
| *** tkelsey has quit IRC | 18:12 | |
| *** salv-orl_ has quit IRC | 18:12 | |
| *** salv-orl_ has joined #openstack-infra | 18:13 | |
| krotscheck | jeblair, clarkb: If the policy is clear, and it is that “All infra-core are admins on storyboard”, great. But then I shouldn’t be core, and neither should ttx (though he’s DTech, so that can be a different rule. | 18:13 |
| jeblair | krotscheck: how are you being special cased? | 18:13 |
| *** bknudson has joined #openstack-infra | 18:13 | |
| krotscheck | jeblair: I’m admin on storyboard. NikitaKonovalov is not. Neither is yolanda. | 18:13 |
| krotscheck | What rule applies to me? | 18:13 |
| krotscheck | Actually, let me just fix this problem, just a sec. | 18:13 |
| *** mwagner_lap has quit IRC | 18:16 | |
| clarkb | fungi: your comment on 151749 has me thinking about ways to test nodepool commads. I guess they just operate against a config and a database so should be relatively easy to put together some tests for that using the existing test fakes | 18:16 |
| nibalizer | pleia2: okay commented | 18:16 |
| nibalizer | you were super right | 18:16 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Removed krotscheck@gmail.com from SB Admin https://review.openstack.org/155429 | 18:16 |
| nibalizer | also thats concerning that our testing sucks | 18:17 |
| clarkb | fungi: would need to fake out sys.argv but thats not too terrible | 18:17 |
| *** whoops has joined #openstack-infra | 18:17 | |
| *** afazekas has joined #openstack-infra | 18:17 | |
| krotscheck | jeblair, clarkb ^^. There, that removes the special treatment that i’ve gotten. | 18:17 |
| nibalizer | krotscheck: i think ttx is also sb admin | 18:17 |
| pleia2 | nibalizer: thanks, want to divide and conquer with these patches? | 18:17 |
| krotscheck | nibalizer: I know, but he’s DTech for the foundation. I think that constitutes a good reason to be an admin. | 18:17 |
| fungi | krotscheck: not necessarily | 18:18 |
| jeblair | krotscheck: oh, i did not realise that. i agree that the rule should be clear and consistent. | 18:18 |
| *** vigneshvar has joined #openstack-infra | 18:18 | |
| krotscheck | fungi: Well, we can argue that independently. | 18:18 |
| fungi | krotscheck: there's nothing special about the foundation either where our infrastructure is concerned | 18:18 |
| jeblair | krotscheck: (and there is no special rule for ttx either) | 18:18 |
| jeblair | krotscheck: what do superusers do in storyboard? | 18:18 |
| krotscheck | jeblair: Should I pull him off then? | 18:18 |
| krotscheck | jeblair: Right now? Nothing. | 18:18 |
| *** baoli has joined #openstack-infra | 18:18 | |
| krotscheck | jeblair: Everyting that’s an admin function is overridden by yaml script loads. | 18:19 |
| krotscheck | jeblair: So you _could_ add a project. | 18:19 |
| fungi | krotscheck: what do you _expect_ sb admins to eventually do via the webui/api? | 18:19 |
| fungi | krotscheck: modelling how we decide who to put there may be easier if we know what the permission is supposed to accomplish for them | 18:19 |
| *** thingee has joined #openstack-infra | 18:19 | |
| krotscheck | fungi: Honestly? Set admins, set projects, set project groups, define system-level tags, maybe even manage storyboard’s configuration. | 18:20 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: prepare_node.sh line lengths are too long https://review.openstack.org/155417 | 18:20 |
| fungi | i mean if sb admin doesn't grant any actual useful capability currently, then there's no _immediate_ need for anyone to be in it | 18:20 |
| *** salv-orlando has quit IRC | 18:20 | |
| krotscheck | fungi: Maybe ACL's? | 18:20 |
| *** ZZelle_ has joined #openstack-infra | 18:20 | |
| fungi | krotscheck: so, those are tasks the infra team and release manager currently handle in other venues (gerrit, launchpad, et cetera) | 18:21 |
| jeblair | krotscheck: is it conceivable that an admin may be required to do something through the ui or api to fix an error with the system? | 18:21 |
| krotscheck | fungi: In line with jeblair’s statement in that old review, I’m a fan of keeping things restricted until there’s a present need to elevate. | 18:21 |
| krotscheck | jeblair: Only if we move the configuration into the database? | 18:21 |
| *** harlowja_away is now known as harlowja | 18:21 | |
| krotscheck | jeblair: Eventually, yes. | 18:21 |
| krotscheck | jeblair: I think. | 18:22 |
| krotscheck | jeblair: Honestly, I think I can’t answer that question until a feature ships, right? | 18:22 |
| *** baoli has quit IRC | 18:22 | |
| *** EmilienM|afk is now known as EmilienM | 18:22 | |
| nibalizer | pleia2: sure, mind if I don't finish mine until a bit later today/tonight? | 18:22 |
| krotscheck | jeblair: There is one thing though- the only place where you can search for user accounts is in the admin. | 18:22 |
| fungi | i just want to make sure we step back and consider that putting people in positions of power over some piece of infrastructure is not a badge, reward, et cetera. it's simply acknowledging that we expect them to help fix problems that arise which that (whatever) permission would allow them to do | 18:22 |
| krotscheck | jeblair: So, for example, if a user record becomes messed up and prevents someone from logging in, or if an admin wants to merge user accounts or something like that, that’s something you’d need to be an admin for. | 18:23 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/zuul: Zuul sometimes fails to update remotes https://review.openstack.org/149336 | 18:23 |
| jeblair | my thoughts are that, if it's useless, then it could be empty afaic. if we might need to "fix" something in the way that we sometimes have to log into servers to fix some stuck or broken process, then we should make it infra-root. and if there are other actions that may need to be performed (acls, global tags, etc), then we can create a clear list of criterea for who should additionally have that access. | 18:23 |
| pleia2 | nibalizer: no problem, do we just want to be adding the dependency to metadata.json, or..? | 18:23 |
| jeblair | fungi: indeed | 18:23 |
| krotscheck | jeblair, fungi: As far as storyboard-core goes, we should have admin access on the soon-to-be dev instance, because in that we’ll need to create testing user accounts to debug ACL errors and things like that. | 18:24 |
| jeblair | krotscheck: i am supportive of that | 18:24 |
| krotscheck | jeblair, fungi: Because we need to fix things. But on the prod instance? No need. | 18:24 |
| *** shashankhegde has joined #openstack-infra | 18:24 | |
| nibalizer | pleia2: there is a longer answer here | 18:24 |
| pleia2 | nibalizer: we can wait until you have time | 18:25 |
| fungi | as an example, all the infra sysadmins get operator/founder access to make changes to official irc channels, but because there aren't enough of us to take on day-to-day management of goings on within all the channels that permission is also granted to additional volunteers | 18:25 |
| *** achanda has joined #openstack-infra | 18:25 | |
| jeblair | krotscheck: so for prod, it sounds like we're somewhere between 'nobody' and 'infra-root' at the moment. | 18:26 |
| jeblair | krotscheck: yeah? | 18:26 |
| fungi | krotscheck: also giving the developers of the tool control (even ssh access if it helps) to a dev deployment of that while not granting them similar access to the production deployment has existing precedent (there are several examples) | 18:27 |
| krotscheck | jeblair: And, honestly, that’s your decision. I just disagreed with pleia2’s escalation because previous policy was unclear and contradictory. | 18:28 |
| *** HeOS has quit IRC | 18:28 | |
| krotscheck | jeblair: So, for now, I’d say-> to get the pleia2 rootkit installed, we can make the policy infra-core, and then argue about whether that’s necessary later. | 18:29 |
| openstackgerrit | Merged openstack-infra/puppet-elasticsearch: Make the number of shards and replicas tunable https://review.openstack.org/154711 | 18:29 |
| *** emagana has quit IRC | 18:29 | |
| asselin | anyone know if evesdrop support links to timestamps? | 18:29 |
| *** emagana has joined #openstack-infra | 18:29 | |
| anteaya | asselin: I haven't seen that it does, but oh my it would be wonderful | 18:30 |
| *** jedimike has quit IRC | 18:30 | |
| *** dimtruck is now known as zz_dimtruck | 18:30 | |
| *** zz_avozza is now known as avozza | 18:30 | |
| *** emagana has quit IRC | 18:30 | |
| jeblair | krotscheck: k. just trying to gather as much info as possible. based on what i know, i think infra-root is the best choice at the moment. i could fairly easily be convinced of 'nobody', but if there's a slight chance that something may need fixing, it's better to have someone able to do that without too much trouble. | 18:30 |
| *** emagana has joined #openstack-infra | 18:30 | |
| *** Guest41759 has joined #openstack-infra | 18:31 | |
| krotscheck | jeblair: Let me put it like this: I want storyboard to be so easy to manage that you can fix all problems from the admin UI. | 18:31 |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/project-config: Add lyz account to accessbot config https://review.openstack.org/155438 | 18:31 |
| pleia2 | ^^ thanks fungi, forgot that one | 18:31 |
| krotscheck | jeblair: Based on that, infra core makes sense. | 18:31 |
| jeblair | i'd also be happy for the feature to go away if not needed, or, if it becomes operationally necessary, to add others when we have a reason and can establish criteria. | 18:31 |
| *** annegent_ has joined #openstack-infra | 18:32 | |
| krotscheck | yep. | 18:32 |
| fungi | i would be happy to eventually see the global admins group in sb simply get the ability to bootstrap more appropriate rbac delegations | 18:32 |
| *** boris-42 has quit IRC | 18:32 | |
| jeblair | krotscheck: just a nit, but just so we're on the same page, we're actually talking about infra-root, which has server access, vs infra-core which has gerrit approval access | 18:32 |
| jeblair | fungi: ++ | 18:32 |
| krotscheck | Righto | 18:32 |
| *** Ryan_Lane has joined #openstack-infra | 18:32 | |
| fungi | but of course, that's future state | 18:32 |
| jeblair | krotscheck: (infra-root is a subset of infra-core) | 18:32 |
| fungi | likely quite a ways out i'd imagine | 18:32 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config: Remove ttx from SB admin https://review.openstack.org/155440 | 18:33 |
| jeblair | krotscheck: so you don't have to propose it ^ | 18:33 |
| *** e0ne is now known as e0ne_ | 18:33 | |
| *** otter768 has joined #openstack-infra | 18:33 | |
| *** yfried_ has quit IRC | 18:34 | |
| *** zz_dimtruck is now known as dimtruck | 18:35 | |
| *** e0ne_ is now known as e0ne | 18:35 | |
| krotscheck | jeblair: Thanks :) | 18:35 |
| *** jaypipes has quit IRC | 18:36 | |
| krotscheck | I can’t believe I just rageargued myself into _less_ access. | 18:36 |
| anteaya | krotscheck: you are quite the debater | 18:36 |
| jeblair | krotscheck: you're living the dream. | 18:36 |
| anteaya | asselin: jenkins is still unhappy: https://review.openstack.org/#/c/155417/ | 18:37 |
| yjiang5 | fungi: hi, do we have multiple node test in gate now? I'm trying to add some test for live migration. | 18:37 |
| *** otter768 has quit IRC | 18:38 | |
| fungi | yjiang5: i think clarkb and jogo have the experimental nova multi-node job in a mostly working state but believe they're still working through some minor hurdles | 18:38 |
| asselin | anteaya, apparently I don't know how to multiply | 18:38 |
| fungi | yjiang5: they could probably better report on status than i | 18:39 |
| anteaya | asselin: that is okay, it happens | 18:39 |
| clarkb | yjiang5: yes we have a job and jogo is working to make live migration tests work , let me get a link to the change stack | 18:39 |
| anteaya | asselin: looking forward to results on patchset 3! | 18:39 |
| * krotscheck would make a good manager, he’s good at arguing himself out of responsibility :) | 18:39 | |
| *** mriedem has quit IRC | 18:40 | |
| *** baoli has joined #openstack-infra | 18:40 | |
| anteaya | krotscheck: well done | 18:40 |
| fungi | yjiang5: worth noting that the current example assumes a 2-node devstack deployment, and we'd need to make changes elsewhere to accommodate 3 or more | 18:40 |
| clarkb | yjiang5: https://review.openstack.org/#/c/141530/ that is the end of the stack | 18:40 |
| yjiang5 | fungi: clarkb, thanks for your information. I'm trying to figure out if I can have guard for https://bugs.launchpad.net/nova/+bug/1082414 scenerio in gate , but possibly even with multiple-node support, possibly we can't achieve it. | 18:40 |
| openstack | Launchpad bug 1082414 in OpenStack Compute (nova) "Live migration between hosts with differents CPU models fails" [Medium,In progress] - Assigned to jiang, yunhong (yunhong-jiang) | 18:41 |
| clarkb | yjiang5: there are a lot of little tweak changes to make live migration work, need to set cpu model, update /etc/hosts, setup ssh keys, etc | 18:41 |
| *** tiswanso_ has quit IRC | 18:41 | |
| openstackgerrit | Merged openstack-infra/project-config: Added Python34 Jobs as non-voting to storyboard and python-storyboardclient https://review.openstack.org/155086 | 18:41 |
| krotscheck | Wooo | 18:41 |
| yjiang5 | clarkb: yes, exactly. That's what I did for my local test environment. | 18:41 |
| krotscheck | greghaynes: ^^ | 18:41 |
| fungi | clarkb: sounds like yjiang5 could just change the cpu model on one of the two subnodes to test that then? i wonder if that can be done without having to restack? | 18:42 |
| clarkb | fungi: its the cpu model that qemu exposes not for the underlying host | 18:42 |
| clarkb | though you don't want it to be different on different nodes, I am missing something | 18:43 |
| fungi | clarkb: i think that's what the bug yjiang5 linked is about | 18:43 |
| clarkb | oh I see the bug now | 18:43 |
| clarkb | so I think thats a libvirt issue | 18:43 |
| clarkb | at least it was libvirt exploding at us when we had this issue in rax with different cpu models | 18:43 |
| *** abhirc has joined #openstack-infra | 18:43 | |
| clarkb | the difficulty is that you never know which direction you are going to migrate because of the scheduler | 18:43 |
| yjiang5 | fungi: I'm not use the infra to achieve it, I use my own script to achieve it. | 18:44 |
| *** avozza is now known as zz_avozza | 18:44 | |
| clarkb | so how do you enforce that you always migrate to a node with a subset of functionality? | 18:44 |
| clarkb | I am not a nova dev but that almost appears to be not a bug to me | 18:44 |
| *** e0ne has quit IRC | 18:44 | |
| *** zz_avozza is now known as avozza | 18:44 | |
| JayF | We could use more eyes on https://review.openstack.org/#/c/155357/1 (devstack) to unbreak the Ironic gate | 18:44 |
| fungi | at least in other (commercial) competitors i know the control plane tracks cpu flags and can determine when the glags for one model are a subset of another | 18:44 |
| *** abhirc has quit IRC | 18:44 | |
| fungi | s/glags/flags/ | 18:45 |
| yjiang5 | clarkb: for that bug, the thing is it will fail even with the cpu_model. | 18:45 |
| clarkb | yjiang5: ya I am reading more :) so it fails if the VMs have the same cpu model but the hypervisor models are different | 18:45 |
| clarkb | yjiang5: we don't have much control over what the hypervisor nodes report, sometimes they are different in rackspace but not always and they are always the same in hpcloud | 18:46 |
| yjiang5 | clarkb: the cpu_mode/cpu_model is to make sure we can migrate either way, if both host has same cpu_mode/cpu_model set. But it failed now. | 18:46 |
| fungi | but yeah, that seems like something nova would have to query from libvirt and refuse the migrate call rather than finding out partway in and giving up while leaving things broken | 18:46 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: prepare_node.sh line lengths are too long https://review.openstack.org/155417 | 18:46 |
| clarkb | yjiang5: so I don't think we have control of the bit you need to toggle | 18:46 |
| asselin | anteaya, ok I see, you need multiples of 4 of spaces, so cursor should be at 13, not 12. ^^ | 18:47 |
| yjiang5 | clarkb: so possibly we can't do it in gate yet, possibly I can try to squeeze it into our 3rd party testing? We have hardware based testing environment. Your idea? | 18:47 |
| *** abhirc has joined #openstack-infra | 18:47 | |
| clarkb | yjiang5: we can test live migration in the gate once all of these fixes get through, but I don't think we can test the scenario you are interested in because we have no control of the cpu model for the hypervisor/compute nodes | 18:48 |
| anteaya | asselin: ah ha, third time's the charm | 18:49 |
| * anteaya waits for jenkins | 18:49 | |
| yjiang5 | clarkb: yes, got it. | 18:49 |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard-webclient: Fix error notifications https://review.openstack.org/155242 | 18:49 |
| *** ihrachyshka has quit IRC | 18:50 | |
| *** sdake__ has joined #openstack-infra | 18:50 | |
| *** mdenny has joined #openstack-infra | 18:50 | |
| fungi | ahh, i see yjiang5's proposed https://review.openstack.org/53746 fix is likely valid, but agree trying to work a regression test for that into tempest is likely not going to be possible even with our multi-node configuration | 18:50 |
| *** Guest41759 has quit IRC | 18:51 | |
| *** Longgeek has quit IRC | 18:51 | |
| fungi | may simply need to stick to lower-level testing for something like that | 18:51 |
| fungi | (e.g. not end-to-end) | 18:51 |
| *** Longgeek has joined #openstack-infra | 18:52 | |
| yjiang5 | fungi: yes, this is in fact more a hardware-based test scenerio. | 18:52 |
| *** ParsectiX has quit IRC | 18:53 | |
| *** abhirc has quit IRC | 18:53 | |
| *** sdake has quit IRC | 18:53 | |
| *** ParsectiX has joined #openstack-infra | 18:53 | |
| *** ihrachyshka has joined #openstack-infra | 18:54 | |
| fungi | well, we've said in the past that something like the tripleo cloud would be one way to have upstream gating on hardware-specific jobs, but that presumes someone maintaining a suitably large deployment giving our infrastructure the access needed to build workers and assign jobs there | 18:55 |
| *** Longgeek has quit IRC | 18:56 | |
| *** coolsvap is now known as coolsvap_ | 18:57 | |
| *** doude has quit IRC | 18:58 | |
| *** tiswanso has joined #openstack-infra | 18:58 | |
| *** avozza is now known as zz_avozza | 18:59 | |
| yjiang5 | fungi: I don't think one company can provide enough large deployment, but if we can have a way to combine the machines donated by different company transparently, that will be possible. | 18:59 |
| zaro | morning | 18:59 |
| *** eharney has quit IRC | 19:01 | |
| fungi | yjiang5: well, some companies can but so far those have all been public cloud providers. though what you describe is sort of the case with the tripleo cloud (hewlett-packard and red hat are both providing hardware to that cloud as separate regions) | 19:01 |
| *** dustins has quit IRC | 19:01 | |
| yjiang5 | fungi: thanks for your discussion, I will have a deep look on the tripleo cloud. | 19:02 |
| jroll | fungi: y'all could use our bare metal stuff, it's homogenous at least for now | 19:03 |
| fungi | yjiang5: https://wiki.openstack.org/wiki/TripleO/TripleOCloud | 19:03 |
| jroll | "our bare metal stuff" being rackspace onmetal | 19:03 |
| *** dustins has joined #openstack-infra | 19:03 | |
| fungi | jroll: it's tempting to take it for a test drive some time, though i don't think we've had a compelling use case for it yet | 19:04 |
| jroll | fungi: if you want two machines from a cloud with the same CPU... | 19:04 |
| yjiang5 | jroll: I want two machines with different CPU ... | 19:05 |
| jroll | oh :| | 19:05 |
| jroll | ignore me then | 19:05 |
| *** doude has joined #openstack-infra | 19:06 | |
| devananda | fungi: elastic search backend on a high IO instance? though the size of the index might make it untenable | 19:06 |
| *** thedodd has joined #openstack-infra | 19:06 | |
| fungi | devananda: that's one possible use, agreed | 19:06 |
| *** pelix has quit IRC | 19:06 | |
| fungi | those nodes are using the largest vm flavor rackspace offers at the moment | 19:07 |
| devananda | fungi: they're using cinder volumes for the index store though, right? | 19:08 |
| *** davideagnello has joined #openstack-infra | 19:09 | |
| clarkb | ya baremetal for es would be nice | 19:10 |
| clarkb | but aiui we dno't have free baremetal :) | 19:10 |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/puppet-graphite: Replace pip class with conditional package install https://review.openstack.org/155449 | 19:10 |
| * nibalizer has baremetal, would trade for cloud | 19:11 | |
| *** achanda has quit IRC | 19:11 | |
| jroll | clarkb: what do we give you, specifically? is it a $ discount or x GB of RAM or? | 19:11 |
| *** alexpilotti has quit IRC | 19:11 | |
| *** rlandy has quit IRC | 19:13 | |
| *** bswartz has quit IRC | 19:13 | |
| clarkb | jroll: I actually don't know, but I remember when baremetal became a thing it was off limits but maybe that has changed | 19:13 |
| jroll | wat | 19:13 |
| jroll | that's messed up | 19:13 |
| jlk | early on I think it was, RAX didn't want infra filling it up | 19:13 |
| clarkb | jroll: mordred and jeblair likely know all the details of how its setup | 19:13 |
| JayF | I think it's more likely whoever told you that | 19:14 |
| JayF | didn't ask us :) | 19:14 |
| jroll | right | 19:14 |
| jroll | we have plenty | 19:14 |
| JayF | and just went "metal is expensive, you can not has" | 19:14 |
| *** rlandy has joined #openstack-infra | 19:14 | |
| openstackgerrit | Khai Do proposed openstack-infra/project-config: Make gerritbot report changes from Gerrit 2.9.4 branch https://review.openstack.org/155450 | 19:14 |
| fungi | i believe they just 100% comp anything we add in our tenant and then complain to us if we're using things they don't want (or adjust quotas/options so that we can't exceed what they are comfortable with) | 19:14 |
| jroll | aha | 19:14 |
| jroll | I vote just do it then | 19:14 |
| *** baoli has quit IRC | 19:15 | |
| fungi | i too would be fine with the "easier to ask forgiveness than permission" model there | 19:15 |
| fungi | but it would kind of suck to put a lot of work into configuring and migrating to a resource that they then tell us we can't hold onto | 19:16 |
| jroll | <- some percentage of the onmetal ops team :P | 19:16 |
| jroll | fungi: should be fine, at worst they'll keep your quota at default (10 instances of each flavor) | 19:17 |
| *** mwagner_lap has joined #openstack-infra | 19:18 | |
| *** eharney has joined #openstack-infra | 19:18 | |
| *** abhirc has joined #openstack-infra | 19:18 | |
| clarkb | this might be the incentive I need to upgrade our es cluster :) | 19:19 |
| jroll | fungi: got an "I approve" from someone, gogogo | 19:20 |
| *** larsks has joined #openstack-infra | 19:20 | |
| jeblair | clarkb: what do you think would be good for es? http://www.rackspace.com/cloud/servers/onmetal | 19:20 |
| jroll | jeblair: onmetal-io1 | 19:20 |
| jroll | 2x 1.6TB PCI-E flash | 19:20 |
| jroll | the others only have a 32GB SATADOM, and we don't have cinder support yet :( | 19:21 |
| jroll | unless you want to do onmetal-memory1 and keep everything in a tmpfs | 19:21 |
| jeblair | clarkb, jroll: it looks like our typical es member is using ~6tb on / | 19:22 |
| jroll | guh | 19:22 |
| jeblair | clarkb: is most of that data? (as opposed to logs, etc) | 19:22 |
| sdague | jeblair: it's data | 19:22 |
| jeblair | clarkb: (also, es2 is an outlier at 20g, may need to take a look at that) http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=123 | 19:22 |
| jeblair | es3 is typical: http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=124 | 19:23 |
| *** eharney has quit IRC | 19:23 | |
| jeblair | jroll: no cinder, right? | 19:24 |
| jroll | jeblair: correct | 19:24 |
| sdague | jeblair: but isn't a big part of our issue io perf | 19:24 |
| sdague | so the point was to get fast local disk | 19:24 |
| jeblair | sdague: not necessarily; i believe clarkb experimented with ssd and saw no substantial improvement | 19:25 |
| sdague | ok | 19:25 |
| jroll | these are super ssd | 19:25 |
| jroll | (fwiw) | 19:25 |
| *** teran has joined #openstack-infra | 19:25 | |
| * jeblair hopes clarkb comes back soon in case jeblair is making things up | 19:25 | |
| * jroll bbiab | 19:25 | |
| clarkb | sorry I am making nodepoolcmd tests :) | 19:25 |
| fungi | huh, puppetboard claims several of our jenkins masters haven't reported for days | 19:25 |
| fungi | nearly a week | 19:25 |
| clarkb | ya the ssd volume didn't seem to make a whole lot of difference | 19:25 |
| clarkb | but I think that may have been due to contention? | 19:25 |
| clarkb | whereas with baremetal we would be the only thing accessing the ssds so in theory will help | 19:26 |
| *** emagana has quit IRC | 19:26 | |
| *** emagana has joined #openstack-infra | 19:26 | |
| openstackgerrit | Joe Gordon proposed openstack/requirements: Import cap.py tool to cap explicit dependencies https://review.openstack.org/155454 | 19:27 |
| jeblair | clarkb, jroll, sdague: i guess all we can say is that the two different flavors of cinder did not help. there's still a lot of variables that onmetal would change | 19:27 |
| clarkb | yup | 19:27 |
| jeblair | but it seems like we would need to ~2x the cluster size from 6 to 12? | 19:27 |
| clarkb | jeblair: we might be able to go smaller since the metal nodes are bigger | 19:28 |
| *** dustins has quit IRC | 19:28 | |
| clarkb | jeblair: or are you looking at the 32GB flavor? | 19:28 |
| sdague | so I also had this conversation with rax folks at the nova midcycle | 19:28 |
| jeblair | erm wait | 19:28 |
| jeblair | i think we may have a units problem | 19:28 |
| jeblair | and i may be responsible | 19:29 |
| jeblair | sorry | 19:29 |
| jeblair | we're using 6 _gb_ on each node, right? | 19:29 |
| sdague | jeblair: for the root volume | 19:29 |
| *** dustins has joined #openstack-infra | 19:29 | |
| sdague | what's the cinder volume? | 19:29 |
| clarkb | jeblair: its a 1TB cinder volume per node | 19:29 |
| *** zul has quit IRC | 19:29 | |
| sdague | because that's what we'd need to move onto root | 19:29 |
| clarkb | and we use about 2/3s of that which is healthy (gives room for when you lose a node) | 19:30 |
| jeblair | clarkb: oh, that is not showing in cacti for some reason | 19:30 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/project-config: Switch on python34 job for oslo.versionedobjects https://review.openstack.org/155455 | 19:30 |
| jeblair | okay, so yeah, onmetal io has enough room to hold 2-3 current-sized nodes worth of data | 19:30 |
| *** ashleighfarnham has quit IRC | 19:31 | |
| *** emagana has quit IRC | 19:31 | |
| jeblair | (assuming we don't raid1, but we don't need to since ES is raiding for us) | 19:31 |
| jeblair | jroll: ^ (sorry i got crosseyed earlier) | 19:32 |
| clarkb | ya no raid1 necessary | 19:32 |
| jeblair | clarkb: so, for starters, do you think doing a 3-node cluster would make sense? | 19:33 |
| clarkb | jeblair: yes, I think 3 is a good start, that way if you lose a node you still end up with some redundancy post recovery | 19:33 |
| *** achanda has joined #openstack-infra | 19:33 | |
| mordred | Wow. I'm offline for a couple of hours ... | 19:33 |
| fungi | mordred: that was your second mistake | 19:34 |
| anteaya | mordred: life moves pretty fast... | 19:34 |
| fungi | (your first mistake was looking at scrollback once you weren't) | 19:34 |
| mordred | fungi: ++ | 19:34 |
| *** hdd has quit IRC | 19:35 | |
| jeblair | mordred: have some ansible show and tell? | 19:36 |
| mordred | No, I'm walking in to a meeting ... Will show/tell in a few | 19:36 |
| *** Sukhdev has joined #openstack-infra | 19:37 | |
| *** HeOS has joined #openstack-infra | 19:41 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Add nodepool command tests https://review.openstack.org/155458 | 19:41 |
| clarkb | fungi: ^ | 19:41 |
| fungi | clarkb: ooh! | 19:41 |
| clarkb | note I did not use testscenarios due to maruns argument against them | 19:41 |
| *** e0ne has joined #openstack-infra | 19:41 | |
| clarkb | seems like a simple enough thing to accomodate if others prefer explicit test cases | 19:42 |
| *** zul has joined #openstack-infra | 19:43 | |
| jroll | jeblair: heh, cool | 19:43 |
| *** emagana has joined #openstack-infra | 19:44 | |
| *** sarob has joined #openstack-infra | 19:46 | |
| openstackgerrit | Khai Do proposed openstack-infra/system-config: Upgrade Gerrit from 2.8 to ver 2.9 https://review.openstack.org/155463 | 19:46 |
| *** zul has quit IRC | 19:46 | |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/puppet-zanata: Initial commit of Zanata-specific sections https://review.openstack.org/147947 | 19:46 |
| *** dprince has quit IRC | 19:47 | |
| *** zul has joined #openstack-infra | 19:47 | |
| *** lnxnut has quit IRC | 19:47 | |
| clarkb | asselin: you might also be interested in 155458 | 19:47 |
| *** dims__ has quit IRC | 19:48 | |
| *** emagana has quit IRC | 19:49 | |
| *** emagana has joined #openstack-infra | 19:49 | |
| *** dims__ has joined #openstack-infra | 19:50 | |
| *** liam_ has joined #openstack-infra | 19:51 | |
| *** liam_ is now known as Guest73606 | 19:51 | |
| *** Longgeek has joined #openstack-infra | 19:53 | |
| *** emagana has quit IRC | 19:54 | |
| *** zz_avozza is now known as avozza | 19:54 | |
| clarkb | asselin: I am pretty sure it covers your case that failed in an earlier patchset of my changes to the image update command | 19:55 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/project-config: Switch on python34 job for oslo.versionedobjects https://review.openstack.org/155455 | 19:57 |
| *** achanda has quit IRC | 19:58 | |
| *** bogdando has quit IRC | 19:58 | |
| *** achanda has joined #openstack-infra | 19:58 | |
| *** jackie_acosta has quit IRC | 19:59 | |
| *** spzala has joined #openstack-infra | 20:00 | |
| *** jsavak has joined #openstack-infra | 20:00 | |
| *** diva_melani has joined #openstack-infra | 20:01 | |
| *** zul has quit IRC | 20:01 | |
| *** joesavak has quit IRC | 20:02 | |
| *** Longgeek has quit IRC | 20:02 | |
| diva_melani | hi i need some help with jenkins-job-builder .. when i run 'jenkins-jobs test jobs' where job is folder i defined all jobs and macros , im getting this """Unknown entry point or macro 'd' for component type: 'wrapper'."""" | 20:02 |
| *** rfolco has quit IRC | 20:02 | |
| diva_melani | this is the macro i defined | 20:03 |
| diva_melani | - wrapper: name: autodeploy_common_wrappers wrappers: - defaultwrappers - workspace-cleanup | 20:03 |
| *** yfried_ has joined #openstack-infra | 20:04 | |
| diva_melani | not sure where im getting this """Unknown entry point or macro 'd' for component type: 'wrapper'." | 20:04 |
| diva_melani | any help appreciated | 20:04 |
| *** yfried_ has quit IRC | 20:05 | |
| *** briancurtin has quit IRC | 20:07 | |
| *** zhiyan has quit IRC | 20:08 | |
| *** ctracey has quit IRC | 20:08 | |
| *** jraim has quit IRC | 20:09 | |
| *** simonmcc has quit IRC | 20:09 | |
| *** emagana has joined #openstack-infra | 20:09 | |
| *** thingee has quit IRC | 20:10 | |
| *** doude has quit IRC | 20:10 | |
| *** ameade has quit IRC | 20:10 | |
| *** zul has joined #openstack-infra | 20:10 | |
| *** hdd has joined #openstack-infra | 20:11 | |
| *** doude has joined #openstack-infra | 20:12 | |
| *** jraim has joined #openstack-infra | 20:15 | |
| *** ldnunes has quit IRC | 20:15 | |
| clarkb | diva_melani: it would probably help to paste your entire configs (sanitized of any sensitive data) on a paste service so we can see the context | 20:15 |
| *** ddieterly has quit IRC | 20:15 | |
| diva_melani | clarkb: ok let me try to link you | 20:16 |
| *** ctracey has joined #openstack-infra | 20:16 | |
| *** briancurtin has joined #openstack-infra | 20:17 | |
| *** ddieterly has joined #openstack-infra | 20:17 | |
| *** zhiyan has joined #openstack-infra | 20:18 | |
| *** ameade has joined #openstack-infra | 20:18 | |
| *** aysyd has quit IRC | 20:18 | |
| *** simonmcc has joined #openstack-infra | 20:18 | |
| diva_melani | clarkb: here is the link to my files : http://paste.openstack.org/show/172329/ | 20:19 |
| *** gyee has quit IRC | 20:20 | |
| diva_melani | the issue is probably is file 2..line 24 in that paste. | 20:20 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Add nodepool command tests https://review.openstack.org/155458 | 20:21 |
| *** VijayTripathi has joined #openstack-infra | 20:22 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 20:22 | |
| clarkb | diva_melani: and where is the job or job-template that is breaking? | 20:23 |
| clarkb | diva_melani: the error should at least tell you the file that it broke on | 20:23 |
| *** abhirc has quit IRC | 20:24 | |
| diva_melani | clarkb: thats the thing..the error traceback is not showing me line number of file... im testing the whole folder named 'jobs' and i get this error...when i test individual files, i dont see any error | 20:24 |
| clarkb | diva_melani: can you paste the traceback too then? | 20:25 |
| diva_melani | clarb: here is traceback http://paste.openstack.org/show/172335/ | 20:25 |
| diva_melani | clarkb: as you can see, nothing is shown where the issue is :/ all i have is 3 files .. in folder named 'jobs' | 20:26 |
| clarkb | so its saying it doesn't know how to expand 'd', there is no macro with that name and there is no entrypoint with that name with a type of wrapper | 20:27 |
| *** shashankhegde has quit IRC | 20:27 | |
| diva_melani | anyone can try copy pasting my first paste into three files.. add them to folder and run 'jenkins-jobs test folder' ..it throws the same error | 20:27 |
| clarkb | that plus the info ou have pasted indicates to me that it is in your jobs somewhere that a thing called 'd' is being used under wrappers | 20:27 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-zuul: Option to allow HTTPS for the proxy https://review.openstack.org/155471 | 20:28 |
| fungi | jhesketh: ^ (corresponding system-config patch on the way) | 20:28 |
| *** ashleighfarnham has joined #openstack-infra | 20:30 | |
| *** annegent_ has quit IRC | 20:32 | |
| *** ociuhandu has quit IRC | 20:33 | |
| *** otter768 has joined #openstack-infra | 20:34 | |
| *** spzala has quit IRC | 20:36 | |
| diva_melani | clarkb: thanks..i found the issue ..i was calling the wrapper in wrong way | 20:36 |
| pleia2 | clarkb: looks like https://review.openstack.org/#/c/154188 can be reverted now, thoughts? https://github.com/puppetlabs/puppetlabs_spec_helper/pull/90 merged yesterday | 20:36 |
| diva_melani | clarkb: thanks for pointing out..there was error in the job i defined..not the macro itself .. appreciate your help :) | 20:36 |
| clarkb | diva_melani: no problem | 20:37 |
| clarkb | pleia2: ya if gem install puppet_spec_helper works on a fresh node now we should revert my change | 20:37 |
| *** salv-orlando has joined #openstack-infra | 20:38 | |
| *** prad has quit IRC | 20:38 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/system-config: Also serve zuul status.json via HTTPS https://review.openstack.org/155474 | 20:38 |
| fungi | jhesketh: ^ second piece | 20:38 |
| *** diva_melani has quit IRC | 20:38 | |
| clarkb | pleia2: I can figure out how to test that, my ruby foo is not strong and it should be doable without building an entire image | 20:39 |
| * pleia2 nods | 20:39 | |
| pleia2 | need to run off to an appointment now | 20:39 |
| *** otter768 has quit IRC | 20:39 | |
| *** tkelsey has joined #openstack-infra | 20:39 | |
| *** bogdando has joined #openstack-infra | 20:39 | |
| anteaya | clarkb: if you want to manually test it, just hold a node and run 'gem install puppet_spec_helper' | 20:41 |
| *** salv-orl_ has quit IRC | 20:41 | |
| *** markvoelker has joined #openstack-infra | 20:41 | |
| clarkb | anteaya: well it should already be installed on those nodes. I will just spin up a container locally | 20:41 |
| anteaya | clarkb: ah then 'gem list' should show it installed | 20:41 |
| *** BharatK has quit IRC | 20:42 | |
| *** cdent has quit IRC | 20:42 | |
| fungi | um... | 20:43 |
| fungi | _SO_ | 20:43 |
| clarkb | fungi: aroo? | 20:43 |
| fungi | the reason several jenkins masters aren't checking into puppetdb any longer is that puppetmaster doesn't recognize their host keys | 20:43 |
| *** tkelsey has quit IRC | 20:43 | |
| fungi | oh, wait | 20:43 |
| fungi | that may be pebcak | 20:43 |
| asselin | clarkb, ok looking now | 20:44 |
| fungi | yeah, that was my bad, digging deeper for actual cause | 20:44 |
| clarkb | pleia2: Unable to resolve dependencies: rspec requires rspec-core (~> 2.99.0), rspec-expectations (~> 2.99.0), rspec-mocks (~> 2.99.0) is still what I get | 20:45 |
| clarkb | I am going to guess we ned Hunner or crinkle to make a release | 20:45 |
| *** yamahata has quit IRC | 20:46 | |
| *** mriedem has joined #openstack-infra | 20:46 | |
| *** sdake__ is now known as sdake | 20:46 | |
| *** dmorita has quit IRC | 20:46 | |
| *** dmorita has joined #openstack-infra | 20:47 | |
| *** abhirc has joined #openstack-infra | 20:47 | |
| *** koolhead17 has quit IRC | 20:48 | |
| fungi | clarkb: nibalizer: so _this_ is why some of our jenkins masters haven't been reporting for ~week: http://paste.openstack.org/show/172341/ | 20:48 |
| fungi | Error: Could not prefetch package provider 'pip': #<Puppet::ExecutionFailure:0x7f3badfcbc88> | 20:48 |
| clarkb | fungi: I think that means it wants to copy the pip provider from the master but can't | 20:49 |
| clarkb | possibly a permissions issue on fails on the master? | 20:49 |
| anteaya | it thinks pip is a vcsrepo argument? | 20:49 |
| fungi | strangely, about half of our jenkins masters do not have this problem | 20:49 |
| anteaya | or location? | 20:49 |
| clarkb | anteaya: I think vcsrepo may depend on it? | 20:49 |
| fungi | anteaya: nope, that's just noise from another part of the manifest getting processed | 20:49 |
| crinkle | clarkb: yep the fix was merged, need hunner to release | 20:50 |
| anteaya | fungi: oh | 20:50 |
| *** pc_m has quit IRC | 20:50 | |
| nibalizer | anteaya: the vcsrepo thing is innocuous | 20:52 |
| nibalizer | fungi: pip works on that host right? | 20:52 |
| *** Guest73606 has quit IRC | 20:53 | |
| *** bdpayne has quit IRC | 20:53 | |
| fungi | nibalizer: yep, but looks like something has downgraded it | 20:53 |
| zaro | clarkb, fungi, jeblair, pleia2: https://etherpad.openstack.org/p/gerrit-2.9-upgrade | 20:54 |
| *** zul has quit IRC | 20:55 | |
| anteaya | zaro: I have never seen that colour of green before | 20:55 |
| nibalizer | fungi: i would also explore pathing, etc | 20:55 |
| zaro | i think everything has been prepped for the upgrade. | 20:55 |
| anteaya | the etherpad colour lottery, some folks just win | 20:55 |
| anteaya | oh keep it, don't change it | 20:56 |
| fungi | nibalizer: yeah, it looks like we have a fairly broad variety of pip versions on our jenkins masters and the ones where this is happening have pip <1.5.1 while the ones where it is working have pip>=1.5.1 | 20:56 |
| mtreinish | anteaya: it would be better with white font :) | 20:57 |
| anteaya | mtreinish: ha ha ha | 20:57 |
| anteaya | troll | 20:57 |
| *** thingee has joined #openstack-infra | 20:57 | |
| nibalizer | clark may be right though | 20:58 |
| nibalizer | about it being some kind of comm with puppetmaster problem | 20:58 |
| *** annegent_ has joined #openstack-infra | 20:58 | |
| nibalizer | since if the provider was pulled down properly and pip wasn't there we'd get an error like 'provider "pip" is not functional on this hist' | 20:59 |
| dstufft | dhellmann: why does stevedore depend on setuptools without declaring that it depends on setuptools? | 20:59 |
| fungi | dstufft: i thought having explicit dependencies on setuptools was bad form? | 21:00 |
| dstufft | fungi: having an explicit dependency on setuptools because you use it in your setup.py is bad form | 21:01 |
| dstufft | having an explicit dependency on setuptools because you use it in your runtime is the right thing to do | 21:01 |
| fungi | dstufft: ahh, fair point. if you use it runtime then yes i guess it's warranted | 21:01 |
| *** joesavak has joined #openstack-infra | 21:01 | |
| dstufft | as it stands right now pip can successfully install stevedore but not all the required dependencies | 21:01 |
| fungi | nibalizer: upgrading to current pip on all the jenkins masters seems to have solved the puppet package provider issue--thanks for the nudge in the right direction | 21:02 |
| dstufft | since pip doesn't mandate setuptools anymore an can install from wheels without it | 21:02 |
| dstufft | (Is stevedore limited by global requirements? I can submit a CR but idk if I need to get something into global requirements first) | 21:02 |
| *** e0ne has quit IRC | 21:03 | |
| *** dmellado has quit IRC | 21:03 | |
| nibalizer | fungi: awesome | 21:03 |
| dstufft | whoops just noticed the time -> goes to get daughter from bus | 21:03 |
| *** jsavak has quit IRC | 21:04 | |
| *** dmellado has joined #openstack-infra | 21:04 | |
| fungi | dstufft: yeah, looks like it's in sync with and checked against the global requirements list, which does not currently list setuptools | 21:04 |
| dhellmann | dstufft: I think I had fungi's understanding that we shouldn't declare a dependency on setuptools. I suspect setuptools would need to be listed in g-r to add it to stevedore's requirements list. | 21:05 |
| *** baoli has joined #openstack-infra | 21:06 | |
| fungi | dstufft: though stevedore _does_ declare a dependency on pbr which _does_ import from setuptools runtime | 21:06 |
| *** mriedem has quit IRC | 21:07 | |
| fungi | so 1) stevedore might be able to reach what it needs through pbr, but 2) maybe pbr should now declare a dependency on setuptools? | 21:07 |
| *** mriedem has joined #openstack-infra | 21:07 | |
| fungi | mordred: ^ ? | 21:07 |
| dhellmann | fungi: stevedore uses pkg_resources directly, so it's probably appropriate to have an explicit requirement there if we do need it | 21:07 |
| *** claudiub has joined #openstack-infra | 21:07 | |
| *** zul has joined #openstack-infra | 21:07 | |
| fungi | dhellmann: yeah, just worried about install-time implications under older pip where upgrading setuptools while installing caused *boom* | 21:08 |
| *** ghostpl has joined #openstack-infra | 21:08 | |
| jogo | so it turns out we aren't gating projects on having 'python setup.py install' work | 21:09 |
| dhellmann | fungi: yeah, me, too. I think that's how we got to where we are now. dstufft, are you seeing an error somewhere? | 21:09 |
| clarkb | fungi so last time that happened they fixed setuptools | 21:09 |
| greghaynes | jogo: yep, this was a source of much tripleo breakage | 21:09 |
| fungi | jogo: right, i've been thinking for a while that it would be nice to test install, sdist, bdist_wheel, et cetera targets explicitly | 21:09 |
| *** kgiusti has left #openstack-infra | 21:09 | |
| dhellmann | jogo: yeah, we have the install step turned off in tox | 21:09 |
| dhellmann | fungi: ++ | 21:09 |
| *** boris-42 has joined #openstack-infra | 21:09 | |
| jogo | fungi: ++ it would be an easy test to set up | 21:09 |
| clarkb | fungi and including setuptools is technically the right thing to do | 21:09 |
| *** zul has quit IRC | 21:10 | |
| *** ddieterly has quit IRC | 21:10 | |
| jogo | tox -evenv -- python setup.py install | 21:10 |
| fungi | clarkb: yeah, agreed. mainly just wanting to make sure it won't significantly increase our risk of getting wedged again | 21:10 |
| *** ddieterly has joined #openstack-infra | 21:10 | |
| *** zul has joined #openstack-infra | 21:10 | |
| *** baoli has quit IRC | 21:11 | |
| fungi | jogo: yep, the impetus for me is the number of projects which tag a release for the first time and then come to us wanting to know why it didn't end up on pypi, only to find out that their packaging was never corectly done | 21:11 |
| clarkb | jogo or just remove use develop | 21:11 |
| *** baoli has joined #openstack-infra | 21:11 | |
| jogo | clarkb: well using develop is faster right? | 21:12 |
| clarkb | yes but thats why this isnt tested | 21:12 |
| jogo | hmm so the failing case doesn't use develop | 21:12 |
| fungi | jogo: also may want to do it as part of a packaging check job which also lints the trove classifiers so that we'll have some confidence the pypi api won't barf when we try to upload the result | 21:12 |
| jogo | (oslo.policy) | 21:12 |
| *** shashankhegde has joined #openstack-infra | 21:13 | |
| jogo | fungi: good idea, so a packaging test | 21:13 |
| fungi | (their trove not our trove, obviously) | 21:13 |
| jogo | I'll take a whack at adding a new test | 21:13 |
| fungi | jogo: yeah, likely should be added to the release-to-pypi template in the zuul layout too so that everyone who thinks they're developing something they're going to be uploading there actually can | 21:14 |
| clarkb | jogo so I would just diaable use dwvelop on pep8 checks | 21:14 |
| *** ghostpl has quit IRC | 21:14 | |
| clarkb | no one runs them anywaus | 21:14 |
| jogo | fungi: the 'venv' env is required right? | 21:14 |
| jogo | clarkb: trying that, not working | 21:14 |
| fungi | jogo: it gets used in the tarball and wheel building scripts, so yes | 21:14 |
| clarkb | jogo can you show me? | 21:14 |
| fungi | huh, and now on to the reason pypi.region-b.geo-1.o.o isnn't reporting into puppetdb for a good chunk of today... | 21:15 |
| fungi | ssh: connect to host pypi.region-b.geo-1.openstack.org port 22: Connection timed out | 21:15 |
| fungi | to the virtual console, robin! | 21:15 |
| jogo | clarkb: http://paste.openstack.org/show/172357 | 21:16 |
| clarkb | now I have that song stuck in my head | 21:16 |
| jogo | is the failure I am seeing in oslo.policy | 21:16 |
| fungi | clarkb: does the nova console via novaclient work with hpcloud? | 21:16 |
| dhellmann | jogo: we're looking at the problem in openstack-oslo now | 21:17 |
| clarkb | fungi: not for interactive console | 21:17 |
| clarkb | fungi: you just get the log iirc | 21:17 |
| jogo | clarkb: but can't reproduce that failure in with tox (doesn't use develop etc.) | 21:17 |
| jogo | dhellmann: thanks, looking into adding a test | 21:17 |
| jogo | for this | 21:17 |
| dhellmann | jogo: good idea | 21:17 |
| fungi | grr... the openstackci2 credentials for hpcloud region b aren't right on the puppetmaster | 21:18 |
| fungi | i'll see if i can update them correctly | 21:18 |
| clarkb | jogo: but it runs sdist then installs the sdist, likely there is a bug somewhere else if tox doesn't catch it | 21:18 |
| clarkb | jogo: 'oslo_policy/tests/var/policy.d/fake_link' not a regular file -- skipping is what sdist does | 21:19 |
| jogo | running egg_info | 21:19 |
| jogo | is the last thing that happens | 21:19 |
| clarkb | now why would that be any different than install | 21:19 |
| jogo | clarkb: pbr? | 21:19 |
| clarkb | possibly | 21:19 |
| clarkb | oh I remember | 21:19 |
| clarkb | dstufft: still around? looks like setuptools ignores the non normal file rather than exploding | 21:20 |
| clarkb | dstufft: when making an sdist but not when you do install | 21:20 |
| clarkb | neutron ran into similar with a symlink | 21:20 |
| clarkb | and I think it has to do with setuptools | 21:20 |
| sputnik13 | Hi, I have a change that's failing tempest tests and the logs aren't providing enough information... I was told it might be possible to get temporary access to the VM that most recently ran the tempest test | 21:21 |
| jogo | clarkb: so you think we can get this fixed upstream in setuptools | 21:21 |
| sputnik13 | can anyone help me with this? | 21:21 |
| clarkb | jogo: ya I think setuptools should use consistent behavior across the commands | 21:21 |
| anteaya | sputnik13: what tests are failing, what is the failure? | 21:21 |
| clarkb | jogo: but we also probably need to confirm that pbr isn't doing anything silly | 21:21 |
| jogo | clarkb fungi: still think its worth adding a packaging check job anyway? | 21:21 |
| anteaya | sputnik13: can you link to the logs of the running job? | 21:21 |
| sputnik13 | anteaya: https://review.openstack.org/#/c/152357/ | 21:21 |
| jogo | (can check trove classify stuff too | 21:21 |
| sputnik13 | http://logs.openstack.org/57/152357/5/check/gate-tempest-dsvm-neutron-src-oslo.utils/97d8991/console.html | 21:22 |
| clarkb | jogo: maybe I really don't think an explicit setup.py install test is necessary if we stop using develop everywhere | 21:22 |
| fungi | jogo: yeah, i'll eventually get around to it if you don't though, so no rush there | 21:22 |
| clarkb | or at least use develop more sparingly | 21:22 |
| *** claudiub has quit IRC | 21:22 | |
| jogo | fungi: sound good, I will leave it up to you | 21:22 |
| anteaya | sputnik13: the jobs aren't running right now | 21:22 |
| fungi | jogo: my interest is more in making sure packaging is at least nominally tested before we try to upload | 21:22 |
| jogo | clarkb: yeah I agree this sounds like a mostly fixable thing | 21:22 |
| jogo | upstream | 21:22 |
| sputnik13 | anteaya: sorry, it ran before, I can submit a recheck and have it spin up again | 21:22 |
| anteaya | sputnik13: you need a running failing job log for us to hold a node, but we also need a compelling reason to hold it | 21:23 |
| jogo | fungi: sounds reasonable | 21:23 |
| sputnik13 | anteaya: ok, actually I think I might have something else I can look at first | 21:23 |
| anteaya | sputnik13: what have you tried to isolate the issue | 21:23 |
| anteaya | sputnik13: exhaust all local possiblities first please | 21:23 |
| anteaya | holding a node is a last resort | 21:24 |
| sputnik13 | anteaya: I looked through the log I linked to you, I didn't know that server logs were available under /logs | 21:24 |
| sputnik13 | one of my co-workers is showing me now, thank you | 21:24 |
| fungi | sputnik13: often these sorts of situations are a good reason to identify what other data you'd want to look at and make sure that gets logged/collected in the future. with cross-repo-dependencies working now you may even be able to rerun your change's jobs on top of an updated devstack-gate or devstack or tempest or whatever | 21:24 |
| clarkb | sputnik13: http://logs.openstack.org/57/152357/5/check/gate-tempest-dsvm-neutron-src-oslo.utils/97d8991/logs/screen-c-api.txt.gz?level=INFO#_2015-02-11_07_43_59_768 is that not sufficient? | 21:24 |
| fungi | even before it gets approved | 21:24 |
| clarkb | oh /me catches up on scrollback | 21:24 |
| *** dmellado has quit IRC | 21:25 | |
| sputnik13 | clarkb: that looks sufficient, thank you | 21:25 |
| sputnik13 | I'm still relatively new, thank you for your help | 21:26 |
| clarkb | no problem | 21:26 |
| *** dmellado has joined #openstack-infra | 21:26 | |
| clarkb | it can be hard to dig through those logs so wasn't sure if you had found that or not | 21:26 |
| anteaya | sputnik13: just for context we tend to be willing to hold a node for a long standing bug that is blocking the gate that can't be reproduced anywhere else | 21:27 |
| dstufft | dhellmann: fungi pbr is a weird case where it's generally only useful if you already have setuptools installed right? since it's a setuptools extension? | 21:27 |
| sputnik13 | clarkb: I didn't realize it was even there, but now I know about how logs are being captured so I'll start looking through these first before I ask for a VM to be held | 21:27 |
| dstufft | clarkb: I'm back now | 21:27 |
| sputnik13 | :) | 21:27 |
| anteaya | sputnik13: then when someone spots it we hold the node for debugging | 21:27 |
| anteaya | sputnik13: for context | 21:27 |
| sputnik13 | anteaya: ic ic, thank you | 21:27 |
| anteaya | sputnik13: np | 21:27 |
| clarkb | dstufft: so symlinks break python setup.py install but not python setup.py sdist | 21:28 |
| clarkb | dstufft: yes this is with pbr so may not be setuptools fault | 21:28 |
| fungi | dstufft: special case perhaps, but since pbr uses setuptools at runtime (imports it in pbr.core) it makes sense that anything else depending on pbr can pretty much be guaranteed that setuptools has to be there too, right? | 21:28 |
| dhellmann | dstufft: true, it is a little different -- I wasn't clear on the preferred best practices | 21:28 |
| *** sdake has quit IRC | 21:28 | |
| clarkb | dstufft: just curious if you had any additional insight into the symlink issue since you had commented on it last time when neutron broke installs with a symlink | 21:29 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul: Layout: adds include layout-dir https://review.openstack.org/152290 | 21:29 |
| *** timcline has quit IRC | 21:29 | |
| fungi | clarkb: ooh, was this another symlink in the source tree? fun | 21:29 |
| clarkb | fungi: I think so | 21:29 |
| dstufft | clarkb: not offhand, i don't think distutils supports symlinks via install either, it might be an accident that it works with sdist at all | 21:30 |
| clarkb | yup confirmed | 21:30 |
| *** timcline has joined #openstack-infra | 21:30 | |
| clarkb | dstufft: 'oslo_policy/tests/var/policy.d/fake_link' not a regular file -- skipping tht is what sdist says | 21:30 |
| clarkb | dstufft: seems to skip it rather than error but install errors | 21:30 |
| *** emagana has quit IRC | 21:30 | |
| fungi | clarkb: jeblair: mordred: the credentials for hpcloud openstackci2 in our encrypted list match what's in our environment file, but doesn't seem to work with novaclient... any idea what's going on there? | 21:31 |
| dstufft | fungi: dhellmann So I generally think that projects should declare their own runtime dependencies regardless of what else exists in the dependency graph, (sort of), basically is your use of pbr and your use of setuptools linked together, or are you using them seperately? | 21:31 |
| *** emagana has joined #openstack-infra | 21:31 | |
| clarkb | fungi: no | 21:31 |
| clarkb | fungi: I don't think they have been changed since heartbleed | 21:31 |
| jhesketh | Morning | 21:32 |
| fungi | dstufft: i generally agree (if for no other reason that changes in transitive deps can spontaneously kill you by ceasing to depend on something you were using but not declaring as an explicit dependency) | 21:32 |
| jeblair | fungi: is that pypi server up? | 21:32 |
| anteaya | morning jhesketh | 21:32 |
| dhellmann | dstufft: sure, that makes sense -- but it wasn't clear to me that it was safe to do that with setuptools in stevedore because of issues with setuptools updating itself, etc. | 21:32 |
| jeblair | fungi: i'm wondering if the account is toast | 21:32 |
| dhellmann | dstufft: it ought to be safe to list a bare setuptools dependency, without a version | 21:32 |
| fungi | jeblair: i can't tell. i can't ssh or http o it and can't authenticate to the hpcloud api | 21:33 |
| dstufft | clarkb: it's probably a distutils behavior tbh, I don't have much insight into why it's that way though, that precedes my involvement by about 15 years I believe | 21:33 |
| clarkb | dstufft: fun, thanks for the input | 21:33 |
| *** eharney has joined #openstack-infra | 21:33 | |
| fungi | jeblair: it seems entirely likely that someone could have deactivated or deleted our tenant out from under us | 21:33 |
| fungi | tteggel: gilliard: ^ | 21:33 |
| clarkb | jogo: so it is entirely possible we will need explicit install checks | 21:33 |
| dstufft | dhellmann: it should be safe, we're doing it in pyca/cryptography and as far as I know we've had no problems | 21:33 |
| clarkb | fungi: are jobs failing? /me looks at zuul status | 21:34 |
| clarkb | dstufft: you have had problems | 21:34 |
| clarkb | dstufft: setuptools broke setuptools installing itself for a while and cryptography could not be installed | 21:34 |
| *** emagana has quit IRC | 21:34 | |
| clarkb | I complained over in the cryptography channel but was basically told cryptography is correct and later setuptools was fixed | 21:34 |
| dstufft | clarkb: wellt hat's not really specific to setuptools itself, any dependency you have can break it's ability to be installed and need fixed | 21:35 |
| *** emagana has joined #openstack-infra | 21:35 | |
| jeblair | clarkb, fungi: horizon says | 21:35 |
| jeblair | Invalid user name or password. | 21:35 |
| clarkb | dstufft: no this was special because it was setuptools installing itself | 21:35 |
| clarkb | dstufft: it has to do with setuptools chicken and egging | 21:35 |
| fungi | jeblair: sounds like someone said "what's this? i guess nobody's using it... BALETED" | 21:35 |
| dstufft | clarkb: sure that particular case, but i mean the general problem of "a depenency released a version that can't be installed for $reasons" | 21:35 |
| clarkb | sure | 21:36 |
| fungi | jeblair: and yeah, timeouts for ping, http, ssh | 21:36 |
| fungi | seems conclusive | 21:36 |
| dstufft | in general the case should work baring setuptools releasing a version that's buggy about being installed | 21:36 |
| jeblair | clarkb, fungi: i would like to know why this hasn't taken out half our system | 21:37 |
| *** baoli has quit IRC | 21:37 | |
| clarkb | jeblair: yes that is a good question | 21:37 |
| clarkb | I am trying to find an hpcloud job now | 21:37 |
| *** Sukhdev has quit IRC | 21:37 | |
| dstufft | dhellmann: I just noticed it because I was looing at stevedore for others reasons and when I saw it didn't have a dep on setuptools I went looking for how you were loading entry points without it :) | 21:37 |
| *** baoli has joined #openstack-infra | 21:37 | |
| fungi | jeblair: agreed, looking in logs for potential signs that it's falling back on pypi.python.org or something | 21:38 |
| dstufft | dhellmann: you can have an environment without setuptools installed into it, and you can do ``pip install thing-that-depends-on-setuptools`` and get a bunch of things installed without ever having setuptools installed | 21:38 |
| *** sdake has joined #openstack-infra | 21:38 | |
| dstufft | and at some point get-pip.py and ensurepip will stop installing setuptools by default (this'll be when we'll automatically load is as a build dependency for sdists, but we won't fully install it for a build dependency, it'll be installed into a temporary location) | 21:39 |
| dhellmann | dstufft: right, and in that case would pip break or get confused? at one point we thought it would, and that's why we avoided being explicit. I honestly don't know if it's safe, but if it is then we should add the dependency. | 21:39 |
| *** jaypipes has joined #openstack-infra | 21:39 | |
| clarkb | http://logs.openstack.org/51/154851/1/gate/gate-requirements-python27/afe29e8/tox/py27-1.log | 21:39 |
| dstufft | dhellmann: if everything is available via wheels it'll work just fine | 21:39 |
| clarkb | fungi: jeblair ^ that log indicates it is using the mirror | 21:39 |
| dstufft | until you import stevedore | 21:39 |
| dhellmann | dstufft: that's a big if, though | 21:39 |
| *** davideagnello has quit IRC | 21:39 | |
| dstufft | dhellmann: and in the future when we switch setuptools to an automatically and temporarily installed stevedore it'll be for any case sdist or wheels | 21:40 |
| dstufft | er | 21:40 |
| dstufft | automatically and temporarily installed buiuld dependency* | 21:40 |
| dhellmann | dstufft: I'm not arguing that you're wrong in principle, just trying to make clear why we chose not to do it already and seeking to understand if it's safe to do it now | 21:40 |
| *** jaypipes has quit IRC | 21:40 | |
| dstufft | dhellmann: it should be roughly as safe to list setuptools as a dependency as it is to list anything else as a dependency | 21:41 |
| clarkb | fungi: jeblair but I can't get it either | 21:41 |
| *** dmellado has quit IRC | 21:41 | |
| jeblair | me neither | 21:41 |
| clarkb | fungi: jeblair: maybe this is filtering traffic from the internet? | 21:41 |
| dhellmann | dstufft: ok | 21:41 |
| clarkb | including to the api server? | 21:41 |
| fungi | clarkb: yeah i'm trying from an hpcloud instance now | 21:41 |
| clarkb | and nodepool only works due to an existing connection? | 21:41 |
| dhellmann | dstufft: will you file a bug? or a patch? | 21:41 |
| *** davideagnello has joined #openstack-infra | 21:42 | |
| jeblair | clarkb, fungi: it worked for me from a random hpcloud-b5 instance | 21:42 |
| fungi | clarkb: nodepool is connecting to another tenant | 21:42 |
| clarkb | fungi: oh right | 21:42 |
| dstufft | dhellmann: yea, is it via gerrit or github or? | 21:42 |
| *** dmellado has joined #openstack-infra | 21:42 | |
| *** sputnik13 has quit IRC | 21:42 | |
| jeblair | clarkb, fungi: dns results match on my workstation and the hpcloud-b5 host | 21:42 |
| dstufft | PyPI lists github.com/dreamhost/stevedore, btu that hasn't gotten any commits since 2013 so I'm not sure that's accurate :) | 21:43 |
| dhellmann | dstufft: stevedore is an oslo project, so bugs are in lp (python-stevedore) and patches are in gerrit | 21:43 |
| fungi | jeblair: clarkb: could be something up with routing. traceroutes to it die for me in Level3.net | 21:43 |
| *** zul has quit IRC | 21:44 | |
| jeblair | fungi: can you ssh to any hpcloud instance? | 21:44 |
| clarkb | fungi same thing here | 21:44 |
| jeblair | also same here | 21:44 |
| fungi | i was able to ssh to a nodepool node 15.126.244.43 | 21:45 |
| *** davideagnello has quit IRC | 21:45 | |
| clarkb | jeblair: I seem to get through to 15.126.247.33 | 21:45 |
| *** e0ne has joined #openstack-infra | 21:45 | |
| *** davideagnello has joined #openstack-infra | 21:45 | |
| fungi | and traceroutes for it die at the same place, but then pick up again at the last hop | 21:45 |
| openstackgerrit | Donald Stufft proposed openstack/requirements: Add setuptools to global requirements https://review.openstack.org/155498 | 21:45 |
| *** sputnik13 has joined #openstack-infra | 21:45 | |
| *** davideagnello has quit IRC | 21:46 | |
| clarkb | fungi: ya | 21:46 |
| jeblair | i'm going to try agent-forwarding through a personal hpcloud dev host | 21:46 |
| *** dkliban is now known as dkliban_afk | 21:46 | |
| clarkb | jeblair: gl | 21:46 |
| *** davideagnello has joined #openstack-infra | 21:46 | |
| fungi | i can also ssh to 15.126.223.209 which is in the same /24 as that server | 21:47 |
| *** Longgeek has joined #openstack-infra | 21:47 | |
| jeblair | clarkb, fungi: that works | 21:48 |
| fungi | makes me think they've deleted the tenant but not the server and due to quirks of their routing it's still reachable from within region-b but not through their border gateways | 21:48 |
| dstufft | dhellmann: https://review.openstack.org/#/c/155498/ there's the patch to add setuptools to glboal requirements | 21:48 |
| clarkb | so must be on the non hp side of hpclouds internet presence | 21:48 |
| jeblair | just going to make sure there's no crazy fw rule or something | 21:48 |
| fungi | though it's also entirely possible that the credentials we have on file for that tenant are simply wrong | 21:48 |
| clarkb | fungi: and you compared agaust the ci-launch files too? | 21:49 |
| dhellmann | dstufft: we should probably have a comment above the new line explaining briefly why we don't specify a version like we do with the other libs | 21:49 |
| fungi | yep | 21:49 |
| jeblair | fw looks reasonable | 21:49 |
| *** jaypipes has joined #openstack-infra | 21:49 | |
| fungi | also, useful datapoint, the last time this system was reachable from the puppetmaster was 15:10 utc | 21:49 |
| dhellmann | jogo: when dstufft's patch lands we'll want to exclude it from caps in your requirements script ^^ | 21:49 |
| fungi | so whatever happened was shortly after that | 21:49 |
| dstufft | dhellmann: do we need that? lots of other things in here don't have versions | 21:49 |
| jeblair | fungi: thanks. i'll hop over to the hpcloud noc and start asking around | 21:50 |
| dhellmann | dstufft: "write it down" -- more documentation is better, especially in special cases | 21:50 |
| *** daneyon_ has joined #openstack-infra | 21:50 | |
| *** daneyon_ has left #openstack-infra | 21:50 | |
| fungi | jeblair: sword and shield recommended | 21:50 |
| clarkb | and maybe a holy hand grenade or two | 21:50 |
| *** zul has joined #openstack-infra | 21:51 | |
| ekarlso | supersheep as well ? | 21:51 |
| fungi | amusingly, this hasn't tanked 2/3 of our ci due to sheer luck | 21:51 |
| dstufft | dhellmann: I'm not sure what you want me to write though, the answer is basically "because the libraries that use this can use any you want", it's perfectly reasonable to put a runtime version constraint on what version of setuptools you support, pip will upgrade/dowgrande it as needed, you just can't control what version is used to _install_ your thing very well | 21:52 |
| dhellmann | dstufft: I'm not convinced that putting a version in there and having pip change the version of setuptools installed on a system is safe. We've just seen it not work, and I haven't tried it myself so I'm wary. I'll see if I can come up with some wording and comment on the review. | 21:53 |
| jogo | dhellmann: o_O yeah | 21:53 |
| jogo | although that can break a stable branch but yeah | 21:53 |
| dhellmann | jogo: you have the overrides thing, so it should be easy, we just have to remember to do it. Is that script in master? | 21:53 |
| *** jsavak has joined #openstack-infra | 21:54 | |
| *** ihrachyshka has quit IRC | 21:55 | |
| clarkb | note I approved the two sb access changes, figuring the long discussion earlier today covered the why and how and if anyone disagrees with the approval they know where to bring the discussion up again | 21:55 |
| clarkb | also the affected users is tiny so don't think its worthy of a ML thread | 21:55 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 21:55 | |
| clarkb | if anyone disagrees feel free to let me know | 21:55 |
| *** bogdando has quit IRC | 21:55 | |
| dstufft | dhellmann: fwiw i totally don't mind putting words there I just don't know what to put :) Though maybe a realistic reason is because the openstack cI is hardcoded to force it to the latest version anyways | 21:55 |
| *** joesavak has quit IRC | 21:55 | |
| dhellmann | dstufft: see my comment -- and that thing about the ci system is a reasonable thing to add to it, for context | 21:56 |
| jogo | dhellmann: there is a patch up to forward port it to master right now | 21:57 |
| jogo | broke on oslo though ;) | 21:57 |
| dhellmann | dstufft: I'm heading over to the local meetup, but I'll be back online in a bit | 21:57 |
| dhellmann | jogo: ok, we have a patch fast-tracked to fix that | 21:57 |
| *** joesavak has joined #openstack-infra | 21:57 | |
| dhellmann | jogo: https://review.openstack.org/#/c/155485/ | 21:57 |
| *** ddieterly has quit IRC | 21:58 | |
| clarkb | fungi: pleia2 mordred jeblair jhesketh SergeyLukjanov 144167 has had its deps merge | 21:58 |
| openstackgerrit | Donald Stufft proposed openstack/requirements: Add setuptools to global requirements https://review.openstack.org/155498 | 21:59 |
| *** dmellado has quit IRC | 22:00 | |
| jeblair | clarkb: aprvd. i do not think it will interefere with the current issue | 22:00 |
| *** denis_makogon has quit IRC | 22:00 | |
| *** jsavak has quit IRC | 22:00 | |
| *** emagana has quit IRC | 22:01 | |
| *** dmellado has joined #openstack-infra | 22:01 | |
| *** rlandy has quit IRC | 22:02 | |
| rhe00_ | my ci posts a SUCCESS comment on the change but it does not add a +1. https://review.openstack.org/#/c/155493/ Any ideas what I am doing wrong? | 22:02 |
| jeblair | rhe00_: using zuul? | 22:03 |
| rhe00_ | yes | 22:05 |
| rhe00_ | do I need to be added to the sandbox project to be allowed to vote? | 22:05 |
| jogo | clarkb: any progress on https://storyboard.openstack.org/#!/story/2000159 ? | 22:05 |
| clarkb | jogo: ya I think we should trap that error and go into a deep sleep, then try to catch such a job and interact with it as humans | 22:06 |
| openstackgerrit | Merged openstack-infra/system-config: Removed krotscheck@gmail.com from SB Admin https://review.openstack.org/155429 | 22:06 |
| jogo | clarkb: good idea, how do we do that | 22:07 |
| *** dmellado has quit IRC | 22:07 | |
| jogo | in other news neutron job passed so looks like tempest-full may be working everywhere | 22:07 |
| jogo | running a recheck | 22:07 |
| openstackgerrit | Merged openstack-infra/system-config: Remove ttx from SB admin https://review.openstack.org/155440 | 22:07 |
| jogo | but looks like we should be able to move to tempest-full now | 22:07 |
| jogo | (since jobs are still experimental) | 22:08 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-zuul: Option to allow HTTPS for the proxy https://review.openstack.org/155471 | 22:08 |
| clarkb | jogo: NICE | 22:08 |
| jeblair | clarkb, fungi, mordred: the account is suspended. i am working on next steps. i asked but do not know when intra-hpcloud access might stop working. | 22:08 |
| anteaya | rhe00_: you need to be commenting on sandbox-ci which is set up for ci testing | 22:08 |
| *** Ryan_Lane has quit IRC | 22:08 | |
| *** dmellado has joined #openstack-infra | 22:08 | |
| clarkb | jeblair: fun | 22:08 |
| anteaya | rhe00_: http://git.openstack.org/cgit/openstack-dev/ci-sandbox/ | 22:08 |
| anteaya | pardon ci-sandbox | 22:09 |
| fungi | jeblair: wow, suspended. did its parents get a call from the school principal too? | 22:09 |
| anteaya | and no, you don't need any permissions to test your ci verification voting on that repo | 22:09 |
| jeblair | fungi: Suspended:3 even. | 22:09 |
| fungi | yikes | 22:10 |
| fungi | the 3 means 3 time more bad | 22:10 |
| jogo | clarkb: can you handle the trap that error side? | 22:10 |
| jesusaurus | jeblair: is that triple-secret-suspended? | 22:10 |
| clarkb | jogo: ya let me finish commenting on the thing I told bnemec I would comment on then I will look into trappign that so we can debug | 22:10 |
| clarkb | jesusaurus++ | 22:10 |
| jogo | clarkb: thanks! | 22:11 |
| *** emagana has joined #openstack-infra | 22:12 | |
| fungi | jesusaurus: indeed, the escalation after double secret probation | 22:12 |
| rhe00_ | anteaya: https://review.openstack.org/#/c/155493/ so my CI posting a SUCCESS status to the change is not the same as commenting? | 22:12 |
| *** annegent_ has quit IRC | 22:13 | |
| *** zul has quit IRC | 22:13 | |
| anteaya | rhe00_: which is your ci? | 22:13 |
| *** denis_makogon has joined #openstack-infra | 22:13 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Make aiopcpu jobs run tempest-full https://review.openstack.org/140445 | 22:14 |
| rhe00_ | anteaya: X-IO ISE ISCSI CI | 22:14 |
| openstackgerrit | Merged openstack-infra/system-config: Remove per-user pip settings https://review.openstack.org/144167 | 22:14 |
| anteaya | so posting success is commenting | 22:14 |
| anteaya | rhe00_: I'm letting you know we created a specific sandbox repo just for cis to test their systems | 22:15 |
| *** hdd has quit IRC | 22:15 | |
| anteaya | ci-sandbox | 22:15 |
| *** annegent_ has joined #openstack-infra | 22:15 | |
| rhe00_ | anteaya: I am using openstack-dev/sandbox from jay pipes instructions. I did not realize it had changed. I will try that one instead. | 22:16 |
| anteaya | thanks | 22:16 |
| anteaya | rhe00_: jaypipes instructions should have a note at the top redirecting to our documentation, does it not? | 22:16 |
| *** bdpayne has joined #openstack-infra | 22:16 | |
| *** tiswanso has quit IRC | 22:17 | |
| *** dalgaaf has quit IRC | 22:17 | |
| *** sputnik1_ has joined #openstack-infra | 22:18 | |
| *** harlowja_ has joined #openstack-infra | 22:18 | |
| *** e0ne has quit IRC | 22:18 | |
| *** salv-orlando has quit IRC | 22:19 | |
| *** salv-orlando has joined #openstack-infra | 22:19 | |
| rhe00_ | anteaya: I don't see any redirect instructions. | 22:19 |
| *** Longgeek has quit IRC | 22:19 | |
| anteaya | rhe00_: at the top did it not say the post is out of date and not maintained? | 22:20 |
| *** Apsu has quit IRC | 22:20 | |
| *** ddieterly has joined #openstack-infra | 22:20 | |
| *** zul has joined #openstack-infra | 22:21 | |
| *** Apsu has joined #openstack-infra | 22:21 | |
| anteaya | rhe00_: I had thought he added a note guiding folks to http://ci.openstack.org/third_party.html | 22:21 |
| *** sputnik13 has quit IRC | 22:21 | |
| *** harlowja has quit IRC | 22:21 | |
| rhe00_ | yes, it does say it is not maintained. I have been using it as a supplement to asselin's instructions | 22:21 |
| asselin | rhe00_, anteaya I just updated my sampe to point to the new ci-sandbox | 22:22 |
| *** dmellado has quit IRC | 22:22 | |
| anteaya | asselin: thank you | 22:22 |
| *** dmellado has joined #openstack-infra | 22:24 | |
| *** thedodd has quit IRC | 22:24 | |
| jeblair | clarkb, fungi, mordred: it has been reactivated. apparently the gratis period expired on that account. the owner will have to request extension periodically. there is no way to extend it indefinitely. | 22:24 |
| rhe00_ | asselin: thank you | 22:25 |
| jeblair | clarkb, fungi, mordred: i do not yet know how to do that process. i also do not know when the jenkins account will expire. | 22:25 |
| fungi | jeblair: any idea who the owner is, and what the gratis period duration is? | 22:25 |
| *** yamahata has joined #openstack-infra | 22:26 | |
| fungi | jeblair: also did they say when the vm would be reachable again? it's still not responding outside their network | 22:26 |
| clarkb | jeblair: pretty sure I am the person that signed up, but used gerrit's email addr? | 22:26 |
| *** vigneshvar has quit IRC | 22:27 | |
| *** e0ne has joined #openstack-infra | 22:27 | |
| *** mrmartin has quit IRC | 22:28 | |
| *** timcline has quit IRC | 22:28 | |
| jeblair | fungi: still working on the gratis thing. i'll ask about network access. i did verify api acess works now though. | 22:28 |
| anteaya | my cat is finally seeking attention again, took about a week | 22:30 |
| jeblair | clarkb: i think the email address makes all the difference. :/ hpcs email==indefinite gratis; though perhaps only as long as the person works for hpcs? i dunno... | 22:32 |
| fungi | can we rotate an hpcs address through the infra cores we keep rotating in and out of there? ;) | 22:33 |
| jeblair | fungi: ha! save on paperwork if we just assume identities | 22:33 |
| fungi | just always claim our name is "openstack jones" | 22:34 |
| clarkb | jeblair: well no one explained that to me ... :/ | 22:34 |
| anteaya | steve | 22:34 |
| *** ParsectiX has quit IRC | 22:34 | |
| anteaya | steve openstack | 22:34 |
| *** ParsectiX has joined #openstack-infra | 22:35 | |
| fungi | okay, popping out to dinner, back in a little while | 22:35 |
| *** sputnik1_ is now known as sputnik13 | 22:35 | |
| *** otter768 has joined #openstack-infra | 22:35 | |
| jeblair | clarkb: i'm changing the address to mine, as i am assured that will not impact my current account, and i feel like the other arrangement would be better... | 22:35 |
| *** davideagnello has quit IRC | 22:36 | |
| *** thedodd has joined #openstack-infra | 22:36 | |
| clarkb | ya I think that should be fine | 22:36 |
| jeblair | clarkb: we have the choice of forgetting to update something when monty or i leave, vs forgetting to submit a sharepoint form every X time period (still unknown) | 22:36 |
| clarkb | +1 :) | 22:37 |
| *** miqui is now known as miqui_away | 22:37 | |
| *** thedodd has quit IRC | 22:37 | |
| *** bogdando has joined #openstack-infra | 22:37 | |
| *** davideagnello has joined #openstack-infra | 22:37 | |
| *** mbacchi has quit IRC | 22:37 | |
| *** camunoz has joined #openstack-infra | 22:38 | |
| jeblair | pleia2: next time it's your turn ;) | 22:39 |
| *** otter768 has quit IRC | 22:40 | |
| *** carl_baldwin has quit IRC | 22:40 | |
| *** BobBall_AWOL is now known as BobBall | 22:41 | |
| *** dank_ has joined #openstack-infra | 22:42 | |
| BobBall | So... With libvirt+xen I want to bring up an Ubuntu node, which works fine, then run a script to install Xen but then we need to reboot the node. Suggested extension? Most obvious way to me is to use a 'ready' script and add a new 'reboot-after-ready' flag | 22:42 |
| BobBall | of course, checking SSH works still after reboot. Thoughts? | 22:42 |
| clarkb | BobBall: I think we were hoping that we could just use dib or similar and upload an image that straight boots xen | 22:43 |
| BobBall | unfortunately can't use DIB as we must boot without Xen in the first place | 22:43 |
| dank_ | Good Afternoon - i'm having some trouble with Gerrit, it was recommended I come here and talk to ClarkB or Fungi | 22:43 |
| *** e0ne has quit IRC | 22:43 | |
| BobBall | becausewe need to get the IP address which isn't accessible if Xen boots first | 22:43 |
| clarkb | dank_: hi there, best to just ask your question and we will do our best to help | 22:43 |
| openstackgerrit | Jaime Flynn proposed openstack-infra/jenkins-job-builder: updated conditional_publisher to support plugin changes https://review.openstack.org/155514 | 22:43 |
| clarkb | BobBall: you can config drive in rax | 22:44 |
| clarkb | BobBall: and get the ip from that | 22:44 |
| BobBall | Last I checked the IP wasn't in RAX config drive | 22:44 |
| BobBall | has that changed? | 22:44 |
| mordred | jeblair: back online ... are you telling me that our HP Cloud accoutn got suspended? | 22:44 |
| dank_ | OK - stupid really, i'm trying to insert my contact details into gerrit so I cansubmit something for review, my contact details aren't filled in, when I filled it in it tells me my details couldn't be saved | 22:44 |
| *** vigneshvar has joined #openstack-infra | 22:44 | |
| clarkb | BobBall: yup it was pretty recent. mordred has all the details | 22:45 |
| mordred | BobBall: it is | 22:45 |
| pleia2 | jeblair: sure :) (back from lunch+appointment now) | 22:45 |
| clarkb | fungi: have that ask answer handy? | 22:45 |
| mordred | BobBall: and I have a script you can use to consume it and produce network files even | 22:45 |
| clarkb | dank_: hold on we have this well documented at this point I just have to dig up a link | 22:45 |
| BobBall | wow - that would be useful mordred | 22:45 |
| mordred | BobBall: one sec - let me get you a link | 22:45 |
| mordred | BobBall: I was thinking about pulling it out and making it a thing you can install from pypi - if you're going to use it I might do that | 22:46 |
| BobBall | clarkb: Planning not to use DIB anyway - can have a prepare_node script that does the work | 22:46 |
| dank_ | Server Error: Cannot store contact information | 22:46 |
| clarkb | dank_: https://ask.openstack.org/en/question/56720/cannot-store-contact-information-when-updating-info-in-openstack-gerrit/ does that help? | 22:46 |
| clarkb | dank_: if not le us know and we can continue to debug | 22:46 |
| * dank_ looks | 22:46 | |
| BobBall | Do you happen to know which device the config drive shows up as? | 22:46 |
| BobBall | Just wondering if it's present in HVM mode with no PV drivers (which is how nested xen can boot) | 22:47 |
| *** mfink_ has quit IRC | 22:47 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate: Add trap handler for aiopcpu ssh errors https://review.openstack.org/155516 | 22:47 |
| clarkb | jogo: ^ I think that should do it | 22:47 |
| *** thrawn01 has joined #openstack-infra | 22:48 | |
| mordred | BobBall: https://review.openstack.org/#/c/154132/ | 22:48 |
| mordred | BobBall: there is a shell script anda python script in there | 22:48 |
| jogo | awesome | 22:48 |
| dstufft | dhellmann: Hmm, stevedore doesn't handle multiple extensions named the same thing at all | 22:48 |
| mordred | they don't have anything specific to dib in them | 22:48 |
| dstufft | or well I mean | 22:48 |
| BobBall | Just checked - looks like configdrive is device 3 so should work | 22:49 |
| mordred | BobBall: https://review.openstack.org/#/c/154132/4/nodepool/elements/read-vendor-json/bin/static-network-config.sh has the info about mounting the device if it's helpful | 22:49 |
| BobBall | mordred: wonderful! thanks. | 22:49 |
| mordred | BobBall: https://review.openstack.org/#/c/154132/4/nodepool/elements/read-vendor-json/bin/read-vendor-json.py reads the data and splats out network config | 22:49 |
| *** annegent_ has quit IRC | 22:49 | |
| mordred | BobBall: if you think it would be helpful I can work on splitting one or both of those into something independently installable | 22:50 |
| thrawn01 | anyone know what incantation is needed to allow me to update my contact information on gerrit? "Cannot store contact information" ( https://code.google.com/p/gerrit/issues/detail?id=2985 ) | 22:50 |
| BobBall | Love it - that looks perfect mordred | 22:51 |
| anteaya | that is two inquires in less than 10 minutes is the foundation db down? | 22:51 |
| BobBall | RAX supporting ips in config drive solves a lot of our problems | 22:51 |
| anteaya | fungi: ^^ | 22:51 |
| *** ddieterly has quit IRC | 22:52 | |
| clarkb | thrawn01: https://ask.openstack.org/en/question/56720/cannot-store-contact-information-when-updating-info-in-openstack-gerrit/ can you see if that fixes it, if not let us know | 22:52 |
| *** ddieterly has joined #openstack-infra | 22:52 | |
| anteaya | or is there an onboarding event happening somewhere? | 22:52 |
| *** amitgandhinz has quit IRC | 22:52 | |
| *** BobBall is now known as BobBall_AWOL | 22:54 | |
| anteaya | I can search the user db | 22:54 |
| rhe00_ | http://ci.openstack.org/third_party.html does not mention zuul. Has it been replaced by the gerrit trigger model? | 22:54 |
| thrawn01 | clarkb: thanks! that was it! | 22:54 |
| asselin | rhe00_, there's a pending change for that | 22:54 |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/system-config: Use ansible logging during puppet run_all.sh https://review.openstack.org/154283 | 22:55 |
| anteaya | perhaps just circumstance then | 22:55 |
| asselin | rhe00_, https://review.openstack.org/#/c/149032/ | 22:55 |
| *** zul has quit IRC | 22:55 | |
| asselin | anteaya, ^^ would interest you | 22:55 |
| *** zul has joined #openstack-infra | 22:55 | |
| anteaya | asselin: I'm reading | 22:56 |
| anteaya | rhe00_: you can also see the changes by reading the draft from the ci-docs job | 22:56 |
| anteaya | rhe00_: which results in http://docs-draft.openstack.org/32/149032/6/check/gate-ci-docs/0558253//doc/build/html/third_party.html#the-zuul-gerrit-trigger-way | 22:57 |
| rhe00_ | asselin: so zuul is the still the recommended way to go? | 22:57 |
| anteaya | rhe00_: it depends on your use case | 22:57 |
| anteaya | what repo are you planning on testing? | 22:57 |
| rhe00_ | cinder | 22:57 |
| asselin | rhe00_, i'm in cinder and I use zuul. | 22:58 |
| rhe00_ | asselin: ok | 22:58 |
| anteaya | then follow asselin's advice, since noone knows more about cinder testing with third party ci | 22:58 |
| *** dimtruck is now known as zz_dimtruck | 22:59 | |
| asselin | rhe00_, because you're going to want to use nodepool, and zuul and nodepool play nice together | 23:00 |
| rhe00_ | asselin: I modified my layout.yaml to point to openstack-dev/ci-sandbox and restarted jenkins, zuul, zuul-merger. it still does not pick up the update to the change in ci-sandbox. Any ideas what I have missed? | 23:00 |
| *** thrawn01 has left #openstack-infra | 23:01 | |
| asselin | rhe00_, take a look at /var/log/zuul/debug.log | 23:01 |
| *** zul has quit IRC | 23:01 | |
| *** jsavak has joined #openstack-infra | 23:01 | |
| jeblair | mordred: yes, hpcloud suspended the account with the pypi server | 23:02 |
| jeblair | mordred: only by accident did that not take out half the ci system | 23:02 |
| jeblair | mordred: (access to the server within hpcloud still worked, even while access from the Internet was disabled) | 23:03 |
| *** thingee has quit IRC | 23:03 | |
| mordred | jeblair: wow. there are so man things broken about that | 23:03 |
| jeblair | i am currently trying to convince the noc to look into that, but i'm being asked to talk to the support team since it's an "individual instance problem". | 23:04 |
| *** joesavak has quit IRC | 23:04 | |
| rhe00_ | look at that. it picked up 'recheck x-io', but not 'recheck' only. must be something wrong in my layout.yaml | 23:04 |
| *** dmellado has quit IRC | 23:05 | |
| *** zul has joined #openstack-infra | 23:05 | |
| pleia2 | is there anyone in particular we should wait for WRT this one? https://review.openstack.org/#/c/154474/1 (#openstack-stable logging) | 23:06 |
| *** jgrimm is now known as zz_jgrimm | 23:06 | |
| *** dmellado has joined #openstack-infra | 23:06 | |
| clarkb | pleia2: it will restart the meetbot so don't approve when meetings are going on or when they are soon to start | 23:07 |
| jogo | are we using standard pypi.python.org again ? http://logs.openstack.org/29/152329/22/gate/gate-tempest-dsvm-neutron-full/ab939f2/console.html#_2015-02-12_20_13_18_883 | 23:07 |
| *** ParsectiX has quit IRC | 23:07 | |
| *** dmellado has quit IRC | 23:07 | |
| clarkb | pleia2: otherwise no I think you can approve it | 23:07 |
| *** ParsectiX has joined #openstack-infra | 23:08 | |
| jogo | clarkb: ^ | 23:08 |
| clarkb | jogo: I think that means mordreds global config of pip doesn't work | 23:08 |
| *** dmellado has joined #openstack-infra | 23:08 | |
| pleia2 | seems all clear right now | 23:08 |
| * jogo throws a wet cat at mordred | 23:09 | |
| * anteaya blow dries the cat aimed at mordred | 23:09 | |
| pleia2 | pro tip: cats don't like hair dryers | 23:10 |
| clarkb | jogo: I will hop on a node and look in a sec | 23:10 |
| anteaya | not always, that is true | 23:10 |
| pleia2 | :) | 23:10 |
| anteaya | hehe | 23:11 |
| * anteaya offers snow lizards | 23:11 | |
| pleia2 | haha | 23:11 |
| anteaya | very rare | 23:11 |
| anteaya | :) | 23:11 |
| clarkb | yup confirmed /etc/pip.conf isn't being updated properly | 23:12 |
| * jogo wants a snow lizard | 23:12 | |
| clarkb | let me see what the ready script for nodepool does | 23:12 |
| zaro | clarkb: why don't i see a 'Continuous Integration Tools' Group in review.o.o ? | 23:13 |
| *** avozza is now known as zz_avozza | 23:13 | |
| *** thingee has joined #openstack-infra | 23:13 | |
| clarkb | zaro: its not visible | 23:14 |
| zaro | ok. thanks. | 23:14 |
| jogo | clarkb: tracking bug https://bugs.launchpad.net/openstack-gate/+bug/1421452 | 23:14 |
| openstack | Launchpad bug 1421452 in OpenStack-Gate "We reverted back to https://pypi.python.org" [Undecided,New] | 23:14 |
| * anteaya gives jogo a snow lizard | 23:14 | |
| clarkb | well that'll do it, doesn't look like the configure_mirror script was ever updated | 23:14 |
| jogo | \o/ | 23:14 |
| clarkb | 144165 | 23:15 |
| clarkb | fungi: ^ can your remove your -2? | 23:15 |
| *** diva_milani has joined #openstack-infra | 23:15 | |
| diva_milani | does jenkins-job-builder reconginze a parameters for github plugins? | 23:16 |
| clarkb | fungi: and approve it if you are still happy with it. that should get us working again | 23:16 |
| anteaya | pleia2: feel like updating your +1 to a +2? https://review.openstack.org/#/c/149032/ | 23:16 |
| anteaya | pleia2: then we just need one more | 23:16 |
| diva_milani | like i have a github plugin in jenkins, i want to pull a github private repo, but i am not sure how do i write a yaml format job for it? git scm does not have anything for github :// | 23:17 |
| *** chlong has joined #openstack-infra | 23:17 | |
| clarkb | diva_milani: if JJB doesn't support that plugin it will need to be updated to do so | 23:17 |
| pleia2 | anteaya: done, but I think input from the rest of infra is good WRT adding this stuff to the docs | 23:17 |
| clarkb | diva_milani: also the git scm plugin should work right? | 23:17 |
| clarkb | diva_milani: github isn't that special I hope | 23:17 |
| *** dustins has quit IRC | 23:17 | |
| pleia2 | I know there's some apprehension about the maintenance burden | 23:17 |
| anteaya | pleia2: sure | 23:17 |
| anteaya | pleia2: which is warrented | 23:18 |
| * pleia2 nods | 23:18 | |
| pleia2 | through my reviews of this one I tried to incorporate as much existing docs as possible, so I'm happy with it anyway :) | 23:18 |
| *** vigneshvar has quit IRC | 23:18 | |
| anteaya | I don't think there are any wrong turns in it | 23:18 |
| diva_milani | clarkb: lol ...reading this source, i cant seem to find github reference ...it talks all about git/svn/mercurial .. | 23:19 |
| diva_milani | http://ci.openstack.org/jenkins-job-builder/scm.html | 23:19 |
| anteaya | and I think folks reading it will have pain points we won't discover until they consume it | 23:19 |
| * pleia2 nods | 23:19 | |
| anteaya | I don't think it introduces anything new in what is happening, just documenting it | 23:19 |
| openstackgerrit | Merged openstack-infra/system-config: Collect #openstack-stable logs https://review.openstack.org/154474 | 23:19 |
| anteaya | and yes more eyes are better | 23:19 |
| clarkb | diva_milani: github uses git | 23:19 |
| clarkb | diva_milani: you should be able to use the git scm plugin to talk to github | 23:20 |
| diva_milani | yes i can but, not the private repos..for that i need to set credentials but not sure | 23:20 |
| clarkb | diva_milani: you should be able to provide those credentials to the git scm plugin, via an ssh key for example | 23:20 |
| dstufft | clarkb: oh, was that fake link thing you asked about earlier breaking the gate? | 23:20 |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Add query for bug 1421452 https://review.openstack.org/155527 | 23:21 |
| openstack | bug 1421452 in OpenStack-Gate "We reverted back to https://pypi.python.org" [Undecided,New] https://launchpad.net/bugs/1421452 | 23:21 |
| clarkb | dstufft: I am not sure, jogo pointed me at it | 23:21 |
| dstufft | https://review.openstack.org/#/c/155498/ failed for what looks like that reason | 23:22 |
| jogo | dstufft: that is exactly it | 23:22 |
| jogo | clarkb: query ^ | 23:22 |
| clarkb | also interesting that its failing a bunch | 23:23 |
| jogo | I have a clarkb any idea of when that change started? | 23:23 |
| clarkb | jogo: recently, I think last night when the d-g change merged | 23:24 |
| clarkb | but I could be wrong, thee are so many moving pieces here and the depends on were incomplete | 23:24 |
| *** hichihara has joined #openstack-infra | 23:24 | |
| jeblair | clarkb, fungi: i am attempting to contact the hpcloud support team. | 23:25 |
| *** VijayTripathi has quit IRC | 23:25 | |
| *** doug-fish has left #openstack-infra | 23:25 | |
| *** VijayTripathi has joined #openstack-infra | 23:26 | |
| *** andreaf_ has joined #openstack-infra | 23:27 | |
| *** zul has quit IRC | 23:27 | |
| *** gyee has joined #openstack-infra | 23:29 | |
| clarkb | jeblair: mordred fungi 144165 would be good to get in real soon now | 23:29 |
| clarkb | I can remove fungi's -2 if someone else wants to review and +2 | 23:29 |
| clarkb | (its a procedural -2) | 23:29 |
| jeblair | clarkb: why sed instead of cat? | 23:29 |
| jeblair | do we have something else in there now? | 23:29 |
| clarkb | jeblair: yes, we have the list of trusted nodes in there as well | 23:30 |
| jeblair | ah ok | 23:30 |
| clarkb | so all this does is update the default index | 23:30 |
| clarkb | s/default/only/ | 23:30 |
| jeblair | clarkb: i +2 and support your lifting the -2 | 23:31 |
| clarkb | thank you | 23:31 |
| jeblair | clarkb: which i would do myself but i did not add that feature to gertty | 23:31 |
| jeblair | and it sounds like you're already logged into gerrit | 23:31 |
| clarkb | yup | 23:31 |
| clarkb | all done | 23:32 |
| *** blackchip has joined #openstack-infra | 23:32 | |
| *** gyee has quit IRC | 23:33 | |
| openstackgerrit | Elizabeth K. Joseph proposed openstack-infra/reviewstats: Add lyz to core reviews in infra.json https://review.openstack.org/155532 | 23:33 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email base plugin. https://review.openstack.org/151948 | 23:34 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email SMTP Sender https://review.openstack.org/149792 | 23:34 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email Send Cron Worker https://review.openstack.org/151416 | 23:34 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email Working Directory Utility Method https://review.openstack.org/151947 | 23:34 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Email Outbox https://review.openstack.org/151413 | 23:34 |
| *** Mr0rris0 has quit IRC | 23:35 | |
| *** andreaf_ has quit IRC | 23:35 | |
| *** EmilienM is now known as EmilienM|afk | 23:35 | |
| *** gyee has joined #openstack-infra | 23:35 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for bug 1421452 https://review.openstack.org/155527 | 23:35 |
| *** gyee has quit IRC | 23:35 | |
| openstack | bug 1421452 in OpenStack-Gate "We reverted back to https://pypi.python.org" [Undecided,New] https://launchpad.net/bugs/1421452 | 23:36 |
| *** gyee has joined #openstack-infra | 23:36 | |
| *** asettle is now known as asettle-bfast | 23:38 | |
| *** gyee has quit IRC | 23:38 | |
| *** gyee has joined #openstack-infra | 23:38 | |
| openstackgerrit | Merged openstack-infra/project-config: Edit global pip.conf instead of per-user pip.conf https://review.openstack.org/144165 | 23:38 |
| *** yamahata has quit IRC | 23:39 | |
| *** openstack has joined #openstack-infra | 23:41 | |
| *** sdake has quit IRC | 23:41 | |
| pleia2 | \o/ | 23:41 |
| *** baoli has quit IRC | 23:42 | |
| *** bknudson has quit IRC | 23:42 | |
| *** baoli has joined #openstack-infra | 23:42 | |
| *** andreaf_ has joined #openstack-infra | 23:44 | |
| *** ashleighfarnham has quit IRC | 23:44 | |
| *** ashleighfarnham has joined #openstack-infra | 23:45 | |
| *** amitgandhinz has joined #openstack-infra | 23:47 | |
| *** andreaf_ has quit IRC | 23:47 | |
| *** oomichi has joined #openstack-infra | 23:50 | |
| *** amitgandhinz has quit IRC | 23:53 | |
| fungi | clarkb: jeblair: looking | 23:53 |
| fungi | cool, already taken care of | 23:53 |
| *** carl_baldwin has joined #openstack-infra | 23:54 | |
| *** blackchip has quit IRC | 23:54 | |
| *** salv-orlando has quit IRC | 23:54 | |
| clarkb | fungi: wasn't sure if we had lost you for the day so went ahead and removed the -2 | 23:56 |
| *** krtaylor has quit IRC | 23:57 | |
| *** ayoung has quit IRC | 23:57 | |
| *** diva_milani has quit IRC | 23:57 | |
| *** annegent_ has joined #openstack-infra | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
