| *** otter768 has joined #openstack-chef | 00:01 | |
| *** otter768 has quit IRC | 00:19 | |
| *** mattray has joined #openstack-chef | 00:20 | |
| *** ChanServ sets mode: +o mattray | 00:20 | |
| *** mattray has quit IRC | 00:30 | |
| *** otter768 has joined #openstack-chef | 00:40 | |
| *** zhiwei has joined #openstack-chef | 00:53 | |
| *** _joel has quit IRC | 00:57 | |
| *** tmichael has quit IRC | 01:18 | |
| *** _joel has joined #openstack-chef | 01:23 | |
| *** emagana has quit IRC | 01:27 | |
| *** kaseyalusi has quit IRC | 01:46 | |
| *** dvlwrk has quit IRC | 01:49 | |
| *** harlowja is now known as harlowja_away | 02:02 | |
| *** alanmeadows has quit IRC | 02:26 | |
| *** alanmeadows has joined #openstack-chef | 02:28 | |
| *** devlop has joined #openstack-chef | 02:40 | |
| *** wenchma has joined #openstack-chef | 02:50 | |
| openstackgerrit | Ma Wen Cheng proposed stackforge/openstack-chef-repo: Bump rabbitmq version https://review.openstack.org/176792 | 02:59 |
|---|---|---|
| *** devlop has quit IRC | 03:00 | |
| openstackgerrit | Ma Wen Cheng proposed stackforge/cookbook-openstack-bare-metal: Move to rhel 7.1 https://review.openstack.org/169482 | 03:00 |
| *** devlop has joined #openstack-chef | 03:00 | |
| *** _joel has quit IRC | 03:07 | |
| *** tmichael has joined #openstack-chef | 03:19 | |
| openstackgerrit | Ma Wen Cheng proposed stackforge/cookbook-openstack-identity: Fix for rabbit mq support https://review.openstack.org/177004 | 03:25 |
| *** GonZo2K has quit IRC | 03:31 | |
| *** otter768 has quit IRC | 03:54 | |
| *** devlop has quit IRC | 04:00 | |
| *** emagana has joined #openstack-chef | 04:45 | |
| *** nkrinner has joined #openstack-chef | 04:52 | |
| *** otter768 has joined #openstack-chef | 05:55 | |
| *** otter768 has quit IRC | 06:00 | |
| *** stevemar has quit IRC | 06:32 | |
| *** chlong has quit IRC | 06:39 | |
| *** wojdev has joined #openstack-chef | 06:40 | |
| *** wenchma has quit IRC | 07:27 | |
| openstackgerrit | Merged stackforge/cookbook-openstack-bare-metal: Move to rhel 7.1 https://review.openstack.org/169482 | 07:35 |
| *** jerome80 has joined #openstack-chef | 07:52 | |
| *** otter768 has joined #openstack-chef | 07:56 | |
| openstackgerrit | Zhiwei Chen proposed stackforge/cookbook-openstack-telemetry: Revert "Use new "service" role instead of "admin"" https://review.openstack.org/176663 | 07:58 |
| *** otter768 has quit IRC | 08:01 | |
| openstackgerrit | Adrian Smith proposed stackforge/cookbook-openstack-compute: Include neutron_url_timeout in nova.conf https://review.openstack.org/176696 | 08:02 |
| openstackgerrit | Adrian Smith proposed stackforge/cookbook-openstack-compute: Include neutron_url_timeout in nova.conf https://review.openstack.org/176696 | 08:05 |
| openstackgerrit | Imtiaz Chowdhury proposed stackforge/cookbook-openstack-image: Make SSL settings configurable for Glance https://review.openstack.org/175144 | 08:07 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: refactore libraries/network.rb https://review.openstack.org/176676 | 08:18 |
| openstackgerrit | Merged stackforge/cookbook-openstack-identity: Fix for rabbit mq support https://review.openstack.org/177004 | 08:32 |
| *** jerome80 has quit IRC | 09:28 | |
| *** otter768 has joined #openstack-chef | 09:57 | |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: refactor libraries/network.rb https://review.openstack.org/176676 | 09:59 |
| *** otter768 has quit IRC | 10:03 | |
| *** wojdev has quit IRC | 10:06 | |
| *** wojdev has joined #openstack-chef | 10:06 | |
| *** wojdev has quit IRC | 10:35 | |
| *** zhiwei has quit IRC | 11:08 | |
| *** otter768 has joined #openstack-chef | 11:58 | |
| *** otter768 has quit IRC | 12:02 | |
| *** wojdev has joined #openstack-chef | 12:20 | |
| *** andymccr has quit IRC | 12:24 | |
| openstackgerrit | Yuki Nishiwaki proposed stackforge/cookbook-openstack-client: Standardize cookbook name to openstack_client https://review.openstack.org/177209 | 12:25 |
| *** zhiwei has joined #openstack-chef | 12:30 | |
| openstackgerrit | Yuki Nishiwaki proposed stackforge/cookbook-openstack-client: Standardize cookbook name to openstack-client https://review.openstack.org/177215 | 12:31 |
| *** wojdev has quit IRC | 12:38 | |
| *** dvlwrk has joined #openstack-chef | 12:47 | |
| *** dvlwrk has quit IRC | 12:53 | |
| *** nkrinner has quit IRC | 13:03 | |
| *** neelashah has joined #openstack-chef | 13:05 | |
| *** libsysguy has joined #openstack-chef | 13:07 | |
| *** wenchma has joined #openstack-chef | 13:10 | |
| *** libsysguy has quit IRC | 13:23 | |
| *** libsysguy has joined #openstack-chef | 13:40 | |
| *** libsysguy has quit IRC | 13:42 | |
| *** andymccr has joined #openstack-chef | 13:42 | |
| *** libsysguy has joined #openstack-chef | 13:43 | |
| *** libsysguy has quit IRC | 13:43 | |
| *** smccully has quit IRC | 13:48 | |
| *** smccully has joined #openstack-chef | 13:49 | |
| *** libsysguy has joined #openstack-chef | 13:55 | |
| *** stevemar has joined #openstack-chef | 13:57 | |
| *** otter768 has joined #openstack-chef | 13:59 | |
| *** emagana has quit IRC | 14:00 | |
| *** otter768 has quit IRC | 14:04 | |
| *** libsysguy has quit IRC | 14:05 | |
| *** libsysguy has joined #openstack-chef | 14:08 | |
| *** _joel has joined #openstack-chef | 14:10 | |
| openstackgerrit | Mark Vanderwiel proposed stackforge/openstack-chef-repo: Bump rabbitmq and erlang versions https://review.openstack.org/176792 | 14:46 |
| *** libsysguy has quit IRC | 14:55 | |
| *** libsysguy has joined #openstack-chef | 15:00 | |
| jklare | markvan_: any ideas about this error? i thought we fixed it with the libraries installed by the common bootstrap.sh http://logs.openstack.org/37/165737/2/experimental/gate-cookbook-openstack-integration-test-chef-rake/cb405a7/console.html | 15:20 |
| *** dvlwrk has joined #openstack-chef | 15:20 | |
| markvan_ | jklare: yeah, was just looking at that. I think since we are running the bundle command outside the ChefDK env, we are going to see issues like this. It was the reason to switch to the ChefDK as it has the gems already installed. | 15:22 |
| *** wenchma has quit IRC | 15:22 | |
| jklare | markvan_: so just skip the whole bundle step? | 15:22 |
| jklare | markvan_: i will try this | 15:23 |
| markvan_ | Seems like we should just skip the bundle install step, or detect that ChefDK is installed and issue the commands with "chef exec bundle" at least that way the it will pick up the already installed DK gems. | 15:23 |
| markvan_ | I think we want to move toward "requiring" ChefDK for our dev setup, and therefore change the Rake to use "chef exec" to run the commands. If there comes a time where we need to update or install a gem, we can do that explicitly within the ChefDK env. | 15:24 |
| markvan_ | If folks don't want to use ChefDK, then just do the old steps (as doc'd in the testing.md) to bundle install and berks vendor. | 15:25 |
| markvan_ | Seems like a big waste of time to reinstall gems that are already part of the ChefDK. | 15:25 |
| jklare | sure | 15:26 |
| jklare | but if we want to use chef exec bundle we probably need to bundle first | 15:26 |
| jklare | like chef exec bundle install or something | 15:26 |
| *** mattray has joined #openstack-chef | 15:26 | |
| *** ChanServ sets mode: +o mattray | 15:26 | |
| jklare | of we could just try the whole thing without a Gemfile | 15:26 |
| jklare | if we are positive that our gems are included in chefdk | 15:27 |
| markvan_ | yeah, I think at one point I had a clean setup and did everything with just the ChefDK gems, no bundle install, and therefore no bundle exec. I just used "chef exec" for everything. | 15:28 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-integration-test: bootstrap.sh for infra testing https://review.openstack.org/165737 | 15:29 |
| jklare | if this works it would save us a great deal of work | 15:29 |
| jklare | we could just throw the gemfiles | 15:29 |
| jklare | lets see | 15:29 |
| markvan_ | But I think at some point I found and issue, but don't recall which gem off hand, bit it was 1 gem. So, we can easily handle that in the common bootstrap and just install/update gems as necessary after we install the ChefDK, one stop shop. | 15:30 |
| j^2 | looks like we’ll need to drop the hangouts officially: https://review.openstack.org/#/c/175000/ | 15:31 |
| jklare | j^2: yeah i read that | 15:31 |
| jklare | i actually do not understand it completely | 15:31 |
| j^2 | The only bar to entry in the big tent is "are you one of us", (do you handle your community in the OpenStack way), and we don't use Google Hangouts for our project team official meetings. There are multiple reasons for that, such as non-free software being needed, difficulty to casually follow the meeting without "joining the hangout", absence of searchable | 15:32 |
| j^2 | logs, etc. You may find them more efficient, but they are arguably less inclusive, and definitely different from how we do things in OpenStack... | 15:32 |
| markvan_ | j^2: yeah, saw that....while I can't fully agree with their points, it seems lke an up hill battle to try to convince them of how it's working well for us | 15:32 |
| j^2 | yeah but it goes to do we want to be part of OpenStack name space, and offically not use Hangouts, or stay in the Stackforge and offically use the hangouts | 15:33 |
| markvan_ | so maybe a 2 stage approach, switch to irc 9or whatever they suggest) for all our meetings, and then after we're in, begin a discussion about how we think video chats are a good way to communicate. | 15:34 |
| jklare | i do not see the reasoning why a video chat solution should not be allowed | 15:34 |
| markvan_ | There has to be others that would agree that video chats are a good tool for (socail) community work | 15:34 |
| jklare | and in the end we are not bound to hangouts | 15:35 |
| markvan_ | but we need to fight that battle after we're invited to the party | 15:35 |
| jklare | i guess we all are completely open for any other video chat solution | 15:35 |
| jklare | sure | 15:35 |
| markvan_ | yup, there has to be something out there that other folks have tried and would be acceptable | 15:36 |
| jklare | markvan_: seems to work without bundling :D, just the specs are failing now | 15:37 |
| markvan_ | yeah, and of course it's something related to I change I made for rhel 7.1, never saw that one before: Permission denied @ rb_sysopen - /opt/chefdk/embedded/lib/ruby/gems/2.1.0/gems/fauxhai-2.2.0/lib/fauxhai/platforms/redhat/7.1.json | 15:38 |
| markvan_ | fauxhai is supposed to auto download the latest platforms, seems like it's there but wrong file or dir ACL? | 15:39 |
| jklare | i saw something like this before, but cant remember how i solved it | 15:39 |
| markvan_ | I'm going to try that spec locally | 15:41 |
| markvan_ | jklare: Maybe we should try this Rake change with Common and Compute, that would give be a better warm fuzzy about this working in the gate | 15:42 |
| jklare | markvan_: on it :D | 15:43 |
| sc` | this is why i asked for a new gem to be cut :( | 15:45 |
| markvan_ | @core http://paste.openstack.org/show/205733/ down to just the 2 ubuntu bugs for running aio_nova, both have workarounds, so we can at least run tests now. I'm going to try my first tempest run | 15:45 |
| os-chef-bot | @j^2 @markvan @mattray @wenchma @jklare @cmluciano @zhiwei http://paste.openstack.org/show/205733/ down to just the 2 ubuntu bugs for running aio_nova, both have workarounds, so we can at least run tests now. I'm going to try my first tempest run | 15:45 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-compute: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177321 | 15:45 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-compute: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177321 | 15:46 |
| markvan_ | sc`: yeah, I'm not sure this is related to the gem itself, it might be something else | 15:46 |
| sc` | likely so, but the easiest way around it would have been to bump the version by +0.0.1 | 15:47 |
| markvan_ | I just ran the tests locally (ChefDK) and it worked fine | 15:47 |
| jklare | maybe a new chefdk version? | 15:47 |
| sc` | what version of chefdk? | 15:48 |
| j^2 | 0.5.0 should be out soon | 15:48 |
| markvan_ | humm, I'm still using 0.4.0 | 15:48 |
| jklare | i was talking about the version mark tested with and the one we install with bootstrap.sh | 15:48 |
| markvan_ | but I know I have messed a bit with the gems in there. | 15:49 |
| j^2 | ah | 15:49 |
| markvan_ | jklare: yup, I'm going to dig to see which fauxhai I'm using and how it looks for rhel 7.1 | 15:49 |
| jklare | btw. did we agree on joining the big tent without our beloved hangouts now? | 15:50 |
| markvan_ | humm, found one big difference, the test shows "/opt/chefdk/embedded/lib/ruby/gems/2.1.0" I have 2.0.0. What do you guys have in this ChefDK path for this ruby version? | 15:51 |
| sc` | j^2: so re: big tent. would it be agreeable to just hold meetings over irc, and hangouts for ad hoc face time? or no video chat period? | 15:51 |
| markvan_ | jklare: +1 to using irc for a while to get into bigtent | 15:51 |
| sc` | the video bit is useful imho, as the distributedness dehumanizes some things to some extent | 15:52 |
| sc` | just my stark view on it :) | 15:52 |
| j^2 | sc`, markvan_, jklare: yeah to get to the big tent IRC is probably the best way. and like the ah-hoc conversations though too, but yeah offcially we should axe hangouts :) | 15:52 |
| markvan_ | yeah, after we're in, we can discuss with others about ways to use a video chat, maybe a topic to throughtout in Vancouver | 15:54 |
| j^2 | sounds good. i’ll figure out a wording for the review to say we’re axing it. i have a meeting to get to in about 20 and i don’t want to be late :) | 15:54 |
| markvan_ | jklare: sc` what do you have in this dir? /opt/chefdk/embedded/lib/ruby/gems/ | 15:55 |
| jklare | markvan_: i have to admit that i acutally do not use chefdk locally because it breaks my rvm stuff :( | 15:56 |
| markvan_ | jklare: ah, ok. I think they we're trying to fix that up a bit in 0.5 | 15:56 |
| sc` | markvan_: 2.1.0 | 15:57 |
| markvan_ | I'm running this box on windows 7 with ChefDK, and it's been running well. | 15:57 |
| sc` | i use both chefdk and rvm | 15:57 |
| markvan_ | sc`: thx, so why do I have 2.0.0. humm. whas there a 0.4.1 that came out? | 15:57 |
| markvan_ | sc`: do you have the /opt/chefdk/embedded/lib/ruby/gems/2.1.0/gems/fauxhai-2.2.0/lib/fauxhai/platforms/redhat/7.1.json file? | 15:58 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177327 | 15:58 |
| markvan_ | fauxhai is trying to auto download the latest platform files from git, this permission issue might be because it can't write that new file since we're running tests without sudo | 15:58 |
| sc` | using chefdk 0.4.0 here. the fauxhai json doesn't exist | 15:58 |
| *** libsysguy has quit IRC | 15:59 | |
| markvan_ | Can you give the failing spec a spin? | 15:59 |
| jklare | its 0.4.0-1 right? | 15:59 |
| *** otter768 has joined #openstack-chef | 16:00 | |
| markvan_ | my version just shows 0.4.0, but let me check the actual version files (different on windows...). And I can try this spin on my linux boxes | 16:00 |
| *** libsysguy has joined #openstack-chef | 16:04 | |
| *** otter768 has quit IRC | 16:04 | |
| sc` | just for grins i jumped in the pristine copy of compute and ran rspec. loads of deprecation warnings but no failures so far. what are the failing specs? i missed it | 16:05 |
| sc` | even after that rspec run, i just have up to 7.0.json in /opt/chefdk | 16:07 |
| jklare | sc`: also this was not the inital error, here is another spec that failed in the same way but more impressive :D http://logs.openstack.org/21/177321/2/experimental/gate-cookbook-openstack-compute-chef-rake/1f70d13/console.html | 16:07 |
| *** smccully has quit IRC | 16:08 | |
| sc` | how strange. it makes me wonder if whatever user is kicking off the test just can't write to /opt/chefdk | 16:11 |
| sc` | but 7.1.json doesn't even show up on my system | 16:11 |
| jklare | but why is this only happeing for redhat? | 16:12 |
| *** smccully has joined #openstack-chef | 16:12 | |
| jklare | or for the redhat specs to be more precise :D | 16:12 |
| *** thumpba has joined #openstack-chef | 16:12 | |
| sc` | ok, i'm wrong | 16:13 |
| sc` | E_NOCOFFEE | 16:13 |
| sc` | it's a problem | 16:13 |
| markvan_ | jklare: it's the new redhat 7.1 I recently added. as sc`stated, the fauxhai gem is unique. I added it to that project, but the gem did not get a newer version. | 16:13 |
| sc` | my specs failed | 16:13 |
| sc` | same difference. can't write 7.1.json | 16:13 |
| markvan_ | instead, fauxhai tries to auto download the latest when it run, but since the tests are not run with sudo, I think we get permission deinied trying to bring down that file. | 16:14 |
| sc` | so it needs to be running as root, or at least have the chefdk bits chowned to a user with write privs | 16:14 |
| markvan_ | sc`: running as root, so can acess that dir? | 16:14 |
| markvan_ | So, jklare this might be the first one we need to add to the Common bootstrap to force an update to the fauxhai gem after the ChefDK install (chef exec gem install fauxhai --no-user-install) | 16:15 |
| markvan_ | but, not sure that will solve this as the gem is not version updated, it relies on it's runtime auto download. So, that would be we would need to run a dummy spec in the Common bootstrap while to get that update to happen | 16:16 |
| markvan_ | Or we all push yet again to get fauxhai to put out a new version....would make life to easy | 16:17 |
| openstackgerrit | Merged stackforge/openstack-chef-repo: Bump rabbitmq and erlang versions https://review.openstack.org/176792 | 16:18 |
| markvan_ | I wondier if during the fauxhai gem install, does the download take place? I'll try to figure that out | 16:18 |
| sc` | markvan_: running sudo chef exec rspec on my mac works | 16:20 |
| sc` | and now i have 7.1.json | 16:20 |
| markvan_ | No, it looks like (https://github.com/customink/fauxhai/blob/master/lib/fauxhai/mocker.rb) its all down when the fauxhai is actually used in a spec for mocking a platform, then the download happens. | 16:21 |
| jklare | chef exec gem install fauxhai actually install fauxhai 2.3.0 | 16:22 |
| markvan_ | sc`: thx for the confirmation. So, looks like 2 choices to solve this. 1 Beg for a new fauxhai version (and include that in DK 0.5.0), 2 Add messy logic to Common bootstrap to run a single Common redhat spec (chef exec rspec spec/client-redhat_spec.rb) to "prime" the 7.1 json | 16:23 |
| markvan_ | /chef/ sudu chef/ | 16:24 |
| sc` | the most straightforward way, imho, is for a new fauxhai gem to be cut, but that seems to prove more difficult | 16:24 |
| markvan_ | yeah, seems like they are really proud of this "auto" download magic, but it only works in some cases. Should be no need for sudo running tests (to update test tools) | 16:25 |
| markvan_ | jklare: humm, I don't think 2.3 has the 7.1, but we should check that. | 16:26 |
| sc` | 2.3.0 from rubygems doesn't have it afaik | 16:26 |
| markvan_ | and 2.3 was called the "annual" release....bummer | 16:26 |
| markvan_ | thx sc` | 16:26 |
| jklare | markvan_: currently waiting for the specs to fail | 16:26 |
| sc` | i mentioned this in my issue, which was summarily closed with "here's the code where it downloads from github" | 16:27 |
| markvan_ | jklare: so, want to try add that sudo rspec run to Common after we install the ChefDK? | 16:27 |
| jklare | markvan_: not sure, sounds bit messy, but i guess if it does the trick | 16:28 |
| markvan_ | sc`: yeah, I recall that....they did not want to discuss our (imo) valid use case where a new version is necessary | 16:28 |
| sc` | annoying | 16:28 |
| sc` | but it's not the first time that's happened | 16:28 |
| jklare | markvan_: sc` : mhhh interesting, this gem install fauxhai --no-user-install did the trick | 16:29 |
| sc` | the plot thickens | 16:29 |
| markvan_ | jklare: yeah I thnk sometihing like even running the Common rake would do the trick, but we only need 1 redhat test to run. | 16:29 |
| openstackgerrit | Merged stackforge/cookbook-openstack-dashboard: Allow ssl ciphers to be optionally configured for horizon https://review.openstack.org/176863 | 16:29 |
| sc` | i mean, if it works it works, but it's an awful hacky approach to get around bumping the version by 0.0.1 | 16:30 |
| markvan_ | jklare: k, maybe w ecan try adding that to the Common bootstrap (sudo chef exec gem install fauxhai --no-user-install) | 16:30 |
| markvan_ | sc`: yeah this seems like an issue to have j^2 bring up with the Chef folks, fauxhai is a vital part of doing specs, and putting out a release for major platform updates seems reasonable. | 16:31 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177327 | 16:31 |
| sc` | fauxhai isn't under the chef umbrella though | 16:31 |
| sc` | is it? | 16:31 |
| markvan_ | sc`: ah, don't know that part, good point | 16:32 |
| markvan_ | maybe we send them free pizza coupons? | 16:32 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-compute: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177321 | 16:33 |
| j^2 | sc`: yeah fauxhai I think is still under Seth Vargo. I can ask around to see if I can find someone to commit/push rights | 16:36 |
| markvan_ | jklare: I tried the chef exec gem install fauxhai --no-user-install, but still don't see the 7.1 json, so not sure this is going to work, might need to actually run a spec as well. | 16:42 |
| jklare | markvan_: yeah not sure, but it seemed to work on my ubuntu instance | 16:42 |
| markvan_ | I wonder why this does not fail on the current gates. could it be that ChefDK has a bug | 16:43 |
| jklare | markvan_: but the current run does not look like its working ... https://jenkins02.openstack.org/job/gate-cookbook-openstack-compute-chef-rake/3/console | 16:43 |
| markvan_ | maybe the ChefDK gem tree is locked down a bit to tight? unlike the normal gem install path? just guessing here | 16:44 |
| jklare | we could just change that | 16:45 |
| markvan_ | jklare: yeah, chmod on that one dir might do the trick for us. Not sure what user we run under, but something like (sudo chmod u+rw,g+rw /opt/chefdk/embedded/lib/ruby/gems/2.1.0/gems/fauxhai-2.2.0/lib/fauxhai/platforms/redhat/) | 16:48 |
| markvan_ | or something like that? | 16:48 |
| markvan_ | what does your box show for those permissions? | 16:49 |
| jklare | 0744 for all sub directories | 16:50 |
| jklare | but its fauxhai-2.2.0 for whatever reason | 16:51 |
| markvan_ | yeah would need to open that up, or just download this file and put it there with a (sudo wget...) | 16:51 |
| markvan_ | yeah that the embedded install path, there's also a ~/.chefdk with installed gems in it | 16:52 |
| sc` | how about opening up umask? perhaps that's locked down too tightly for fauxhai | 16:52 |
| *** _joel has quit IRC | 16:53 | |
| jklare | the path in home has 0774 for all subdirs | 16:53 |
| jklare | but i do not understand how this could be an issue at all, because all of these files should be owned by jenkins (or ubuntu in my case) | 16:54 |
| markvan_ | jklare: but we're using sudo to install the ChefDK, which is right, but means root only change to what was installed | 16:56 |
| markvan_ | so ,unless we run our tests with sudo it will fail right now | 16:56 |
| markvan_ | and I don't know how big a deal it is to run our gate tests that way | 16:57 |
| markvan_ | to test that now, we can just change the rake to "sudo chef exec" right? | 16:59 |
| jklare | sure | 17:00 |
| jklare | unless we dropped the privileges | 17:00 |
| jklare | not sure about that right now, but i can check that in infra | 17:00 |
| markvan_ | or back to my other idea, just run a (sudo chef exec rspec spec\client-redhat_spec.rb) in the common bootstrap to prime this 7.1 json, then other tests will be fine without sudo. | 17:02 |
| markvan_ | won't matter the version of fauxhai, it should do the download with that spec run | 17:03 |
| jklare | sounds like a doable workaround until the next fauxhai release | 17:05 |
| jklare | but since we probaly do not have a common redhat spec in all cookbooks | 17:07 |
| markvan_ | yup, I just tried that on my ubuntu box and it worked | 17:07 |
| jklare | this one would need to be special for each project | 17:07 |
| jklare | or we try some crazy wildcard stuff | 17:07 |
| markvan_ | do this in the common bootstrap just after the DK is installed. | 17:07 |
| jklare | but the spec\client-redhat_spec.rb does not exist in all projects, does it? | 17:08 |
| markvan_ | jklare: ah, yup. so maybe we create a dummy test on the fly that just makes use of the 7.1 but does nothing and is not even a cookbook spec, just a standalone spec. | 17:10 |
| jklare | markvan_: although i like doing stuff like that, this sounds a bit too workaroundy | 17:11 |
| *** _joel has joined #openstack-chef | 17:12 | |
| markvan_ | yeah, a bit ugly, but at least allows small mess to be in one place to easy remove later and other rakes/bootstrap don't care | 17:13 |
| jklare | i am still working on this permission change | 17:15 |
| jklare | maybe its easier to just make this folder accessible for everybody | 17:16 |
| *** GonZo2K has joined #openstack-chef | 17:18 | |
| *** GonZo2K has quit IRC | 17:18 | |
| *** GonZo2K has joined #openstack-chef | 17:18 | |
| markvan_ | jklare: hey check this out, how about a direct call into Fauxhai to do the work... | 17:24 |
| markvan_ | sudo chef exec ruby -e "require 'fauxhai'; Fauxhai.mock(platform:'redhat', version:'7.1')" | 17:24 |
| jklare | markvan_: that sound good | 17:24 |
| markvan_ | I think that will do the trick, and we can add more of these for other platforms as needed, it will do the auto download thingy | 17:25 |
| sc` | +1 to that if it works. it's the least hacky hack :) | 17:26 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: TESTING TESTS WITHOUT BUNDLING AND JUST CHEFDK https://review.openstack.org/177327 | 17:27 |
| jklare | ^^ i will ofc update the commit message and typos :D | 17:28 |
| markvan_ | And with this change to the Rakefile, is it time to then start updating the TESTING.md to make note of the ChefDK env requirement, and if you want manual, play with the Gemfile/bundles your self? | 17:31 |
| jklare | i think thats a good idea, i will update the TESTING.md as soon as this is working and we drop the gemfile | 17:33 |
| markvan_ | And this kinda also means making a statement about the Gemfiles, if they are not going to be used in the gates, do we want to keep them up to date? or maybe just have one up to date example in the Common cookbook for those that still want to do it that way | 17:33 |
| *** libsysguy has quit IRC | 17:33 | |
| jklare | i would just drop dem | 17:33 |
| jklare | since this this a cookbook and not a ruby project, using chefdk as a dependency should be sufficient | 17:34 |
| jklare | if it does the job ofc | 17:34 |
| markvan_ | yeah, I basically agree there. But I'm guessing there will be at least one gem out there in a ChefDK that we will want to update/change, that would mean just Common bootstrap updates. Just need to doc that somewhere. | 17:35 |
| markvan_ | and dropping also means one less file to dea with for stable branches, yeh! | 17:36 |
| jklare | ^^ | 17:36 |
| jklare | https://jenkins03.openstack.org/job/gate-cookbook-openstack-common-chef-rake/3/console | 17:37 |
| jklare | yay | 17:41 |
| jklare | success | 17:41 |
| markvan_ | :) | 17:41 |
| markvan_ | and about 4x faster... | 17:42 |
| jklare | ^^ | 17:42 |
| j^2 | Sweet! | 17:45 |
| markvan_ | j^2: cmluciano thx for getting that rabbitmq cookbook out...means I can escape a few status cages.... | 17:45 |
| cmluciano | that was all JJ | 17:46 |
| cmluciano | I just label things lol | 17:46 |
| j^2 | Yeah I need to put cheeseplus s change in today too | 17:46 |
| markvan_ | might help if EVERYONE could mark themselves as affected by this bug: https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1439280 still has not been fixed, seems like a simple version mistake for trusty. | 17:47 |
| openstack | Launchpad bug 1439280 in nova (Ubuntu) "Libvirt CPU affinity error" [Undecided,Confirmed] | 17:47 |
| openstackgerrit | Jan Klare proposed stackforge/cookbook-openstack-common: moving from bundler to chefdk https://review.openstack.org/177327 | 17:50 |
| *** harlowja_away is now known as harlowja | 17:51 | |
| jklare | need to go now, but if you leave comments for this patch i will try to get them in until monday | 17:51 |
| jklare | have a nice weekend | 17:51 |
| markvan_ | jklare: l8r, nice job moving this forward... | 17:59 |
| *** otter768 has joined #openstack-chef | 18:01 | |
| *** otter768 has quit IRC | 18:05 | |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-common: Move role and recipe include methods to Common https://review.openstack.org/177382 | 18:54 |
| *** mattray has quit IRC | 19:14 | |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-compute: Dont install db packages on compute nodes https://review.openstack.org/177387 | 19:14 |
| openstackgerrit | Merged stackforge/cookbook-openstack-client: Standardize cookbook name to openstack_client https://review.openstack.org/177209 | 19:24 |
| *** petertr7 has joined #openstack-chef | 19:34 | |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-common: Move role and recipe include methods to Common https://review.openstack.org/177382 | 19:43 |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-compute: Dont install db packages on compute nodes https://review.openstack.org/177387 | 19:47 |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-telemetry: Dont install db packages on compute nodes https://review.openstack.org/177399 | 19:48 |
| openstackgerrit | Samuel Cassiba proposed stackforge/cookbook-openstack-compute: Fix deprecation notices https://review.openstack.org/177400 | 19:50 |
| sc` | i expect that to be broken with fauxhai acting up | 19:50 |
| sc` | wanted to see what not-my-machine looked like :) | 19:51 |
| openstackgerrit | Mark Vanderwiel proposed stackforge/cookbook-openstack-bare-metal: Dont install db packages on compute nodes https://review.openstack.org/177420 | 19:58 |
| *** otter768 has joined #openstack-chef | 20:02 | |
| *** otter768 has quit IRC | 20:06 | |
| markvan_ | sc`: yeah those deprecation notices only come from the rake/chefdk side right now...we're still using chef 11 in the old gates | 20:19 |
| markvan_ | sc try running that with "check experimental" to see if that clears up the rake/chefdk side. | 20:20 |
| markvan_ | sc`: hang on, your doing that in compute....can only be done after the new common bootstrap goes in with the fauxhai fix | 20:21 |
| *** dvlwrk has quit IRC | 20:21 | |
| j^2 | wow, i got completely lost in the rabbitmq cookbook | 20:23 |
| j^2 | but it’s gonna have some of the largest longrunning PRs merged in abit | 20:23 |
| j^2 | :) :) :) | 20:23 |
| markvan_ | j^2: which ones, I can do a quick review if you want... | 20:24 |
| j^2 | no worries, I’ve already merged them locally, i’m running the integration tests now :) | 20:24 |
| j^2 | it’s just time consuming | 20:24 |
| j^2 | btw, i’m gonna put rabbitmq on openstack.chef.io whenever i finally make it just as _another_ ci/integration tester thingy | 20:25 |
| markvan_ | ah, ok...does that mean a 4.x release? | 20:25 |
| j^2 | i was debating that | 20:25 |
| sc` | markvan_: ah, wasn't aware it was still chef 11 | 20:26 |
| j^2 | when i push up my “i’m ready to push the cookbook” PR, i’ll ask around for the input on the 4.y.z or 3.Y.z release | 20:27 |
| markvan_ | ok, that would be good | 20:27 |
| *** dvlwrk has joined #openstack-chef | 20:30 | |
| j^2 | markvan_: https://github.com/jjasghar/rabbitmq/pull/257 | 20:43 |
| j^2 | cmluciano: https://github.com/jjasghar/rabbitmq/pull/257 | 20:43 |
| markvan_ | k | 20:43 |
| j^2 | anyone else interested in rabbitmq: https://github.com/jjasghar/rabbitmq/pull/257 | 20:43 |
| markvan_ | j^2: put my 0.02 in on that PR | 20:56 |
| j^2 | thanks | 20:56 |
| *** libsysguy has joined #openstack-chef | 21:23 | |
| *** _joel has quit IRC | 21:24 | |
| *** libsysguy has quit IRC | 21:26 | |
| *** libsysguy has joined #openstack-chef | 21:28 | |
| sc` | was just about to tell you kitchen/specs looked good | 21:34 |
| *** smccully has quit IRC | 21:52 | |
| *** stevemar has quit IRC | 21:53 | |
| *** neelashah has quit IRC | 22:01 | |
| *** otter768 has joined #openstack-chef | 22:03 | |
| *** otter768 has quit IRC | 22:07 | |
| *** harlowja_ has joined #openstack-chef | 22:41 | |
| *** harlowja has quit IRC | 22:41 | |
| *** petertr7 has quit IRC | 23:31 | |
| *** os-chef-bot has quit IRC | 23:47 | |
| *** os-chef-bot has joined #openstack-chef | 23:48 | |
| *** grealish has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!