| [10:40:19] | * josh_k has quit (Remote host closed the connection) |
| [10:40:48] | * josh_k has joined #aegir |
| [11:15:53] | * josh_k has quit () |
| [11:43:58] | * hestenet has joined #aegir |
| [11:56:18] | * gusaus has quit (Quit: gusaus) |
| [12:02:43] | * gandhiano has quit (Ping timeout: 246 seconds) |
| [12:04:50] | * gusaus has joined #aegir |
| [12:20:46] | * liammcdermott has quit (Quit: Leaving) |
| [12:27:36] | * Topplestack has joined #aegir |
| [12:38:52] | * Egyptian[Home] has quit (Quit: Leaving.) |
| [13:03:03] | * Yaazkal has quit () |
| [13:04:11] | * Yaazkal has joined #aegir |
| [13:16:41] | * Yaazkal has quit () |
| [13:25:53] | * gusaus has quit (Ping timeout: 252 seconds) |
| [14:49:06] | * hestenet has quit (Remote host closed the connection) |
| [15:34:20] | * boshtian has joined #aegir |
| [15:57:24] | * Topplestack has quit (Quit: My Mac has gone to sleep. ZZZzzz…) |
| [16:02:45] | * ivanjaros has joined #aegir |
| [16:03:04] | * Topplestack has joined #aegir |
| [16:33:07] | * stijnvbrande has joined #aegir |
| [16:35:40] | * DecipheredAFK has quit (Quit: ZNC - http://znc.in) |
| [16:36:04] | * DecipheredAFK has joined #aegir |
| [16:54:31] | * Topplestack has quit (Quit: My Mac has gone to sleep. ZZZzzz…) |
| [18:16:46] | * sdrycroft has joined #aegir |
| [18:21:39] | * e-anima has joined #aegir |
| [18:35:57] | * boshtian has quit (Read error: Connection reset by peer) |
| [18:37:33] | * boshtian has joined #aegir |
| [18:47:36] | <drupol> | helmo: why did you removed the aegir hosting queued service for systemd in 7.x-3.1 ? |
| [18:47:47] | <drupol> | I actually found it very handy |
| [18:59:35] | * ratioweb has joined #aegir |
| [19:21:30] | * boshtian has quit (Ping timeout: 240 seconds) |
| [19:43:14] | * gandhiano has joined #aegir |
| [19:54:14] | * boshtian has joined #aegir |
| [20:09:13] | * gandhiano has quit (Ping timeout: 246 seconds) |
| [20:15:27] | * dean has quit (Ping timeout: 260 seconds) |
| [20:21:40] | * dean has joined #aegir |
| [20:24:07] | * gandhiano has joined #aegir |
| [20:41:13] | * Egyptian[Home] has joined #aegir |
| [20:45:42] | * Egyptian[Home] has quit (Remote host closed the connection) |
| [20:47:15] | * Egyptian[Home] has joined #aegir |
| [21:20:04] | * Egyptian[Home] has quit (Quit: Leaving.) |
| [21:23:35] | * vantage|work has joined #aegir |
| [21:39:10] | * gandhiano has quit (Ping timeout: 245 seconds) |
| [21:42:38] | * gandhiano has joined #aegir |
| [22:09:31] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [22:17:21] | * ivanjaros has joined #aegir |
| [22:19:03] | * zombiebeard has joined #aegir |
| [22:35:43] | * shaneonabike1 has joined #aegir |
| [23:01:11] | * zz_drakythe is now known as drakythe |
| [23:03:25] | * Yaazkal has joined #aegir |
| [23:15:40] | * ratioweb has quit (Ping timeout: 272 seconds) |
| [23:36:29] | * ratioweb has joined #aegir |
| [23:37:51] | <helmo> | drupol: it was causing build issues |
| [23:38:27] | <drupol> | helmo: okay got it, do you plan to put it back ? We are using it here on a Fedora and it works pretty good actually. |
| [23:38:36] | <drupol> | Better than using regular cron to do the tasks |
| [23:38:41] | <drupol> | We're very happy of it. |
| [23:38:51] | <drupol> | I didn't know there were such drush command to process the queue |
| [23:39:34] | <helmo> | It's still open in https://www.drupal.org/node/2347557 and I hope to re-commit it later |
| [23:39:36] | <hefring> | https://www.drupal.org/node/2347557 => No systemd service file for queue runner [#2347557] => 21 comments, 12 IRC mentions |
| [23:59:23] | * anarcat has quit (Quit: going offline is good for you) |
| [00:02:32] | <gboudrias> | Hi all! |
| [00:04:17] | <drupol> | o/ |
| [00:06:09] | <gboudrias> | Just starting my day, we should probably release 6.25 today if possible |
| [00:06:21] | <gboudrias> | Um, 2.25 I mean |
| [00:07:19] | <gboudrias> | By which I mean 2.5 of course... Rough start :) |
| [00:07:54] | <drupol> | gboudrias: 2.5 of ? |
| [00:08:09] | <gboudrias> | drupol: Aegir |
| [00:08:24] | <drupol> | Oh... latest is not 7.x-3.1 ? |
| [00:08:29] | <gboudrias> | We released another Aegir 3 version last Thursday but we have to do the same for Aegir2 |
| [00:08:36] | <drupol> | Oooh ok |
| [00:08:39] | <drupol> | sorry :) |
| [00:08:42] | <drupol> | I'm out of sync |
| [00:08:45] | <gboudrias> | Np :) |
| [00:14:02] | <gboudrias> | In theory this is also the weekly chat |
| [00:14:30] | <gboudrias> | ergonlogic helmo cweagans_afk anything cool to report? |
| [00:14:57] | <ergonlogic> | sorry, running late |
| [00:15:03] | <ergonlogic> | I'll be back in ~10 mins |
| [00:16:57] | * notzach has quit (Quit: leaving) |
| [00:29:10] | * cweagans_afk is now known as cweagans |
| [00:29:23] | <ergonlogic> | ok, back |
| [00:29:33] | <cweagans> | Sorry for the lateness - it's been a weird morning. |
| [00:30:08] | <ergonlogic> | first day of school |
| [00:30:12] | <ergonlogic> | anyway |
| [00:30:44] | <ergonlogic> | I've mostly bee working on rearchiteting hosting_s3 somewhat, so as to limit the number of buckets created and used |
| [00:31:01] | <ergonlogic> | also, hosting_git |
| [00:31:35] | <ergonlogic> | and an initial draft of the support services offerings, and the underlying business plan |
| [00:31:36] | <cweagans> | I talked with realityloop and he said he was going to look into getting the local development stuff set up for aegir 4 - so kubernetes + openshift in a vm + a d8 install that can talk to the apis of both. He also mentioned that he might have some technical resources that can work on the project a bit, but nothing set it stone yet. |
| [00:32:18] | <ergonlogic> | yeah, I walke him through Valkyrie, and helped get him bootstrapped there |
| [00:32:52] | <cweagans> | DanielP has been in here a couple times. He seems eager to help with things, so next time he's in here, I'll reach out to him and see if there's anything he'd like to work on. |
| [00:33:31] | <ergonlogic> | short of Terra changing direction, I'd be happy for that local kube/osc to become V2 of Valkyrie |
| [00:33:47] | <helmo> | sorry for not attending, I'm currently busy with a few issues ... will read up later tonight |
| [00:34:13] | <ergonlogic> | I'm planning on doing some work on hosting_civicrm over the coming weeks for a clients |
| [00:34:30] | <ergonlogic> | and probably merging in provision_civicrm, as is the current practice, etc. |
| [00:35:22] | <cweagans> | ergonlogic: I'm not sure that's really a good fit, honestly. Local dev doesn't require a full kubernetes/openshift install to make it work. If we're going to rely on source2image to build the images to deploy, they won't even need docker (though it might be a good idea to do so anyway). We do, however, need those things locally to be able to develop the actual d8 frontend for Aegir (because the APIs need to be available). |
| [00:36:07] | <cweagans> | On the website front, still not much progress. I still really need some help with user stories for the site, particularly around the personas that we want to be targeting. Given that this is the third week I've asked for assistance, I'd like to propose that we just have a 1 hour call at some point and hammer them out as a group. Would anyone be particularly opposed to that? |
| [00:36:11] | <ergonlogic> | ok, so you're thinking just docker compose or something similar for local dev? |
| [00:36:39] | <ergonlogic> | I'm in |
| [00:36:44] | <cweagans> | Yeah, something lighter. Deploying code to a local kubernetes instance seems like way overkill |
| [00:36:58] | <ergonlogic> | sure |
| [00:37:00] | <cweagans> | docker compose might be a good fit, esp if we can provide images designed for local dev |
| [00:37:29] | <ergonlogic> | I guess I'd misread your comment from yesterday, because I thought it was just the docker end realityloop was going to work on |
| [00:37:37] | * fatguylaughing has joined #aegir |
| [00:38:22] | <cweagans> | Also of note - I use Dokku for my personal server, and they're rearchitecting things a bit. If Openshift's source2image doesn't fit our needs for whatever reason, the buildpack part of Dokku is going to be split out into a separate container that can just receive a git push and do something with it, so that might be a good thing to have on our radar. |
| [00:38:27] | <ergonlogic> | that said, I think there'll likely be quite a bit in common b/w developing sites locally, and aegir locally |
| [00:38:52] | <cweagans> | That could certainly be the case |
| [00:38:56] | <ergonlogic> | so, I don't see why we could have both a kube/osc backend, and something lighter |
| [00:39:13] | <ergonlogic> | I mostly use valkyrie for Aegir dev, at this point |
| [00:39:25] | <ergonlogic> | but even at that, Aegir is pretty heavy |
| [00:39:34] | <ergonlogic> | from a site dev standpoint |
| [00:39:38] | <ergonlogic> | so I see your point |
| [00:40:08] | <cweagans> | We totally could. I think we can punt on this for now, though - the prod deployment parts are much more complicated, so we should do those right and then expand/modify from there to get the local dev story taken care of. |
| [00:40:20] | <cweagans> | could have kube/osc backend + something lighter, that is |
| [00:40:38] | <ergonlogic> | right |
| [00:42:06] | <ergonlogic> | for that matter, local dev might not be something we have to worry about ourselves anyway |
| [00:42:30] | <ergonlogic> | there's Kalabox, for example |
| [00:42:56] | <ergonlogic> | acquia dev desktop |
| [00:43:19] | <ergonlogic> | geerlingguy's drupal-dev-vm |
| [00:43:31] | * hestenet has joined #aegir |
| [00:43:55] | <ergonlogic> | if deployment is mostly just a git push to our slug builder, we may not need much more than one of those |
| [00:46:51] | <ergonlogic> | we should plan for an Aegir 2.5 release |
| [00:46:55] | <ergonlogic> | since the ctools SA |
| [00:47:50] | * formatC_vt has quit (Ping timeout: 240 seconds) |
| [00:48:02] | <cweagans> | Probably :( |
| [00:48:14] | * formatC_vt has joined #aegir |
| [00:48:14] | * formatC_vt has quit (Changing host) |
| [00:48:14] | * formatC_vt has joined #aegir |
| [00:48:18] | <cweagans> | Really wish we'd have some time to get OBS running smoothly for things like this. |
| [00:48:26] | <ergonlogic> | oh, I almost forgot, I'd started looking at OBS |
| [00:49:03] | <ergonlogic> | yeah, but that's not actually where the majority of our time is spent on releases |
| [00:49:39] | <ergonlogic> | a lot of it is creating the release notes and nodes |
| [00:49:58] | <ergonlogic> | ensuring tests are passing, etc. |
| [00:51:39] | <ergonlogic> | I haven't found anything in OBS (yet) that indicates that they can trigger package builds automatically from git commits/tags |
| [00:51:49] | <ergonlogic> | like, no webhooks or whatnot |
| [00:52:01] | <ergonlogic> | they may be there, but they aren't evident, if they are |
| [00:52:43] | <ergonlogic> | without that, we're actually looking at a *more* manual process if we move to OBS |
| [00:54:54] | <ergonlogic> | oh... got the paypal acct switched over to me fully now. |
| [00:57:26] | * formatC_vt has quit (Ping timeout: 260 seconds) |
| [00:58:10] | * formatC_vt has joined #aegir |
| [00:58:23] | <bgm> | hey, ot question: during install, using the debian package on aegir3/jessie, I got: '@hostmaster pm-enable hosting_queued' could not be found. Is it a known/FAQ issue? |
| [01:00:13] | <ergonlogic> | hmm |
| [01:00:35] | <ergonlogic> | probably needs an extra 'drush cc drush' in there |
| [01:01:12] | <bgm> | I tried : su -s /bin/sh aegir -c 'drush cc drush && drush @hostmaster pm-enable -y hosting_queued' (in the debian postint), but doesn't seem to fix it |
| [01:01:16] | * sdrycroft has quit (Quit: Leaving.) |
| [01:02:50] | <gboudrias> | bgm: I just re-run hostmaster-install usually |
| [01:03:17] | <gboudrias> | From what I've seen this actually means hostmaster has failed to install somewhere |
| [01:03:43] | <gboudrias> | An issue I've noticed (but only sometimes) is that there's a file in /etc that can be unreadable if you installed postfix |
| [01:03:43] | <bgm> | hmm ok, how do I run that manually? as aegir? |
| [01:04:00] | <gboudrias> | bgm: Yeah just sudo into aegir and run hostmaster-install |
| [01:04:08] | <gboudrias> | (no arguments needed) |
| [01:04:27] | <gboudrias> | /etc/mailname is the file |
| [01:04:58] | <bgm> | gboudrias: thanks |
| [01:05:08] | <gboudrias> | You'll see it in the install if it's a problem, but sometimes /etc/mailname is readable only to root, which doesn't seem to make sense at all |
| [01:05:14] | <gboudrias> | np |
| [01:05:34] | <bgm> | aha, yep, seems to be the case under jessie? (or koumbit's puppet?) |
| [01:06:50] | <bgm> | reminds me that my postfix ansible rules were also missing the mailname :) |
| [01:06:50] | * boshtian has quit (Ping timeout: 240 seconds) |
| [01:08:49] | <gboudrias> | Yeah but I don't know if the issue is with another package, seems logical that the install would fail and the mailname permissions doesn't really seem like our "domain" |
| [01:14:16] | <bgm> | seems weird that aegir needs the mailname |
| [01:21:06] | <bgm> | running "drush hostmaster-install" worked fine |
| [01:21:18] | <bgm> | even after changing the permissions on mailname, didn't seem to work |
| [01:21:31] | <bgm> | and running "drush @hm en hosting_queued" worked after the manual install |
| [01:22:30] | <bgm> | for nginx, also had to: /etc/nginx/conf.d# ln -s /var/aegir/config/nginx.conf |
| [01:25:17] | <gboudrias> | hmm yeah, I think there's an issue for nginx but if not can you add it? |
| [01:25:19] | <bgm> | erm, scratch that, it was because of the manual install. |
| [01:25:28] | <gboudrias> | oh okay |
| [01:37:21] | * gandhiano has quit (Ping timeout: 250 seconds) |
| [01:38:00] | * stijnvbrande has quit (Quit: Connection closed for inactivity) |
| [01:45:08] | * vantage|work has quit (Quit: ChatZilla 0.9.92 [Firefox 40.0.2/20150812163655]) |
| [01:45:27] | * freiheit has quit (Quit: Leaving.) |
| [01:45:47] | * freiheit has joined #aegir |
| [02:00:37] | * zombiebeard has quit (Quit: zombiebeard) |
| [02:13:51] | * zombiebeard has joined #aegir |
| [02:33:07] | * formatC_vt has quit (Ping timeout: 252 seconds) |
| [02:34:07] | * formatC_vt has joined #aegir |
| [02:45:10] | * formatC_vt has quit (Ping timeout: 244 seconds) |
| [03:37:58] | <gboudrias> | ergonlogic cweagans helmo I'm initiating the release process for the 2.5 release |
| [03:38:11] | <ergonlogic> | gboudrias: thanks for stepping up :) |
| [03:38:31] | <gboudrias> | Np, I need the practice anyway :p |
| [03:40:04] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [03:40:56] | <gboudrias> | Hmm. the puppet tests seem broken: http://ci.aegirproject.org/ |
| [03:43:49] | <gboudrias> | Package authentication problems? http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... |
| [03:44:28] | <cweagans> | ergonlogic: have you ever looked into Omnibus? |
| [03:44:48] | <cweagans> | ergonlogic: https://github.com/chef/omnibus |
| [03:45:50] | <ergonlogic> | I tried using a package built with it for gitlab, and it didn't work too well |
| [03:46:01] | <ergonlogic> | but that might just have been their implementation |
| [03:46:19] | <cweagans> | ergonlogic: it's how Gitlab creates their packages. Unfortunately, it's all Chef. I wonder if there's something like it built with Ansible. Could be an interesting thing to look at for Aegir < v3 |
| [03:46:27] | <cweagans> | er, Aegir <= 3 |
| [03:46:55] | <ergonlogic> | that'd be nice |
| [03:59:27] | * gusaus has joined #aegir |
| [04:06:26] | <gboudrias> | ergonlogic: Any idea what this test does or why it started failing? http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... |
| [04:07:36] | <gboudrias> | Unstable test doesn't work but dev seems to work, not sure how that's possible: http://ci.aegirproject.org/ |
| [04:08:17] | <gboudrias> | (I'll update the makefile anyhow) |
| [04:13:19] | <ergonlogic> | gboudrias: it should run our full test suite against aegir2 installed from the 'unstable' .deb repo |
| [04:13:33] | <ergonlogic> | so, the jenkins-generated .debs |
| [04:14:12] | <gboudrias> | Yeah I didn't expect there to be some much new stuff, there are a lot of changes apparently |
| [04:14:22] | <gboudrias> | (In the puppet code) |
| [04:19:09] | <hefring> | Git => Update ctools version => http://cgit.drupalcode.org/hostmaster/commit/?id=269a53430db03a8945cab95... |
| [04:21:00] | <gboudrias> | (I'm taking a look at puppet aegir but it's been a long while) |
| [04:25:14] | * ivanjaros has joined #aegir |
| [04:45:39] | * formatC_vt has joined #aegir |
| [04:45:39] | * formatC_vt has quit (Changing host) |
| [04:45:39] | * formatC_vt has joined #aegir |
| [05:06:12] | * gandhiano has joined #aegir |
| [05:14:41] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [05:15:13] | <gboudrias> | For anyone who's curious, we finally published my Aegir 3 SaaS article: http://praxis.coop/en/blog/automated-drupal-saas-aegir3 |
| [05:15:32] | <gboudrias> | I think I did a reasonable job of summarizing what you need to build a similar service |
| [05:15:44] | <gboudrias> | (Tests are still running!) |
| [05:27:40] | <ergonlogic> | gboudrias: nice article! |
| [05:27:48] | <gboudrias> | ergonlogic: Thanks :) |
| [05:27:54] | <ergonlogic> | the fonts are really hard to read, though, imo |
| [05:28:27] | <ergonlogic> | like, I gave up on reading the code at the bottom, as it started giving me a headache |
| [05:28:38] | <gboudrias> | Oh yeah the code in particular yeah |
| [05:28:50] | <gboudrias> | I guess the main font could be better too |
| [05:29:52] | <gboudrias> | Actually, I have the technical problem that I can't put whitespace in the code, it's really silly... I think can just add empty <code> lines though |
| [05:30:05] | <gboudrias> | That would probably help, although the font is still a problem |
| [05:30:08] | <gboudrias> | Thanks for the feedback! |
| [05:35:13] | * josh_k has joined #aegir |
| [05:35:17] | * Yaazkal has quit () |
| [05:40:17] | <gusaus> | gboudrias: great summary - i’m happy to be testing it out, rather than setting it all up :) |
| [05:41:06] | <gboudrias> | hahah thanks |
| [05:44:08] | * hestenet has quit (Remote host closed the connection) |
| [05:45:09] | * shaneonabike1 has quit (Remote host closed the connection) |
| [05:51:01] | * hestenet has joined #aegir |
| [05:52:22] | <gusaus> | gboudrias: regarding using git to manage/update sites, is there anything i need to setup (aside from knowing what to do) - do i need to setup something like https://www.drupal.org/project/hosting_git ? |
| [05:53:10] | <gboudrias> | gusaus: You already have hosting_git actually, you can just enable it in the features ("hosting" in the menu) |
| [05:54:08] | <gusaus> | gboudrias: yup, i saw that… just didn’t want to enable something that might blow up the platform :P |
| [05:54:38] | <gboudrias> | gusaus: Heh, it shouldn't affect current platforms/sites |
| [05:57:14] | <gusaus> | ok - i’ll enable and see if i can answer my next question |
| [06:00:55] | * josh_k has quit (Remote host closed the connection) |
| [06:02:40] | * josh_k has joined #aegir |
| [06:20:49] | * ratioweb has quit (Quit: Leaving) |
| [06:28:53] | <gboudrias> | ergonlogic: For http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... , it seems like it's using a basebox that no longer exists, is this possible? |
| [06:29:03] | <gboudrias> | Meaning, would Jenkins get rid of it? |
| [06:29:20] | <ergonlogic> | I don't think it'd be purged |
| [06:29:24] | <helmo> | gboudrias: I wouldn't worry about that jobs ... last success 1 yr 4 mo |
| [06:29:37] | <helmo> | I now disabled the job |
| [06:29:48] | <gboudrias> | helmo: Darn you're right |
| [06:29:49] | <ergonlogic> | helmo: without that we aren't testing new .debs for Aegir2.x |
| [06:30:00] | <ergonlogic> | I'd started work on an Aegir 3.x equivalent |
| [06:30:04] | <gboudrias> | I still think we can fix it by using the Wheezy basebox instead |
| [06:30:13] | <gboudrias> | Since I can run it without error locally |
| [06:30:41] | <gboudrias> | The wheezy basebox is already in the Vagrantfile (commented out), so it's just a two-line change |
| [06:31:03] | <helmo> | ergonlogic: the -unstable for 3.x has never been run successfull :( |
| [06:31:28] | <gboudrias> | In fact, should I just push it and see? This is just the unstable Vagrantfile so it shouldn't affect the other tets |
| [06:31:29] | <ergonlogic> | (not now but...) ideally we'd make it a matrix test that runs on stable and old-stabl of Debian and Ubuntu |
| [06:31:31] | <gboudrias> | tests* |
| [06:31:46] | <ergonlogic> | helmo: I don't mean the one that runs on the latest Drush |
| [06:32:11] | <ergonlogic> | I couldn't find one that tested against the builds the jenkins does on each commit |
| [06:32:24] | <ergonlogic> | for Aegir3, that is |
| [06:33:08] | <ergonlogic> | even http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... is failing, at this point |
| [06:33:22] | <ergonlogic> | which should be using the published .debs for aegir2 |
| [06:34:20] | <helmo> | I don't think we have such a test at the moment |
| [06:34:37] | <ergonlogic> | helmo: http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... |
| [06:34:54] | <ergonlogic> | helmo: sorry, I meant: http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... |
| [06:35:26] | <ergonlogic> | that ought to be running tests against the .debs in the unstable repo |
| [06:35:28] | <gboudrias> | ergonlogic: Is there a reason why we're preferring Raring? I would imagine more people host on Debian than Ubuntu |
| [06:35:37] | <ergonlogic> | which ought to be the ones generated by jenkins |
| [06:35:55] | <ergonlogic> | gboudrias: for what? |
| [06:36:09] | <gboudrias> | ergonlogic: The two tests that are failing |
| [06:36:21] | <gboudrias> | The Vagrantfile it uses specifies a Raring box |
| [06:36:28] | <gboudrias> | Rather than the Wheezy one |
| [06:36:28] | <helmo> | that jos has had exacly one successfull run. |
| [06:36:35] | <ergonlogic> | gboudrias: it should probably be Trusty |
| [06:37:02] | <ergonlogic> | gboudrias: I'm not overly worried about non-LTS Ubuntus |
| [06:37:24] | <ergonlogic> | helmo: I'm more worried that http://ci.aegirproject.org/job/P_Aegir_Puppet_Module_functional_test_Aeg... isn't passing |
| [06:38:54] | <ergonlogic> | but it'd be nice to know if/when a change breaks our .debs too |
| [06:39:15] | <helmo> | did we remove the drush package? |
| [06:39:18] | <helmo> | from our repo? |
| [06:39:30] | <ergonlogic> | 5.10 should still be there |
| [06:40:00] | <ergonlogic> | yep: http://debian.aegirproject.org/dists/stable/main/binary-amd64/Packages |
| [06:40:01] | <helmo> | Hmm I see it is... new repo key? |
| [06:40:19] | <helmo> | didn't it expire some time ago? |
| [06:40:32] | <ergonlogic> | ==> default: err: /Stage[main]/Drush/Package[drush]/ensure: change from purged to present failed: Execution of '/usr/bin/apt-get -q -y -o DPkg::Options::=--force-confold install drush' returned 100: Reading package lists... |
| [06:40:40] | <ergonlogic> | I think it's just Puppet being stupid |
| [06:41:43] | <gboudrias> | Well, I can't reproduce because the box is no longer available |
| [06:41:46] | <gboudrias> | Testing with Trusty |
| [06:42:58] | <ergonlogic> | don't worry what's in the broken tests too much, they're broken :p |
| [06:43:41] | <ergonlogic> | I've found that the workspace sometimes needs to be purged, in Jenkins |
| [06:43:54] | <ergonlogic> | or it might keep around old puppet code, etc. |
| [06:43:58] | <gboudrias> | Hmm yeah alright |
| [06:44:11] | <gboudrias> | I mean, I'd still like to fix the unstable test |
| [06:44:24] | <gboudrias> | But we can probably release without it, right? |
| [06:44:38] | <ergonlogic> | right, but Raring was probably just the most current Ubuntu available at the time |
| [06:45:01] | <ergonlogic> | as long as we can test that the .debs work manually, sure |
| [06:45:14] | <gboudrias> | alright |
| [06:45:37] | <gboudrias> | can't seem to find a vagrant basebox in the ubuntu repo, not sure what's up with that |
| [06:45:58] | <gboudrias> | http://cloud-images.ubuntu.com/trusty/current/ |
| [06:46:50] | <gboudrias> | oh well |
| [06:46:54] | <gboudrias> | I'll do the release notes now |
| [06:47:12] | <gboudrias> | (already updated the makefile) |
| [06:48:12] | <gboudrias> | As expected, not a lot of updates |
| [06:48:16] | <helmo> | I'll see if I can do some testing in the morning... I'm off now |
| [06:48:58] | * josh_k has quit (Remote host closed the connection) |
| [06:49:27] | * josh_k has joined #aegir |
| [06:50:01] | <gboudrias> | helmo: Cool, see ya :) |
| [06:52:55] | <gboudrias> | Temporary release pastebin: http://pastebin.com/ERML9CQL |
| [06:56:00] | <ergonlogic> | gboudrias: you should just specify 'ubuntu/trusty' as the box |
| [06:56:08] | <ergonlogic> | or 'debian/jessie' |
| [06:56:46] | <ergonlogic> | Vagrant will find them on hashicorp atlas |
| [06:56:51] | <ergonlogic> | no need to specify a url |
| [06:56:55] | <gboudrias> | Ok cool |
| [07:01:45] | <gboudrias> | pushed the tags with release.sh |
| [07:01:54] | <ergonlogic> | github has an 'Open Build Service' service on repos |
| [07:03:02] | <ergonlogic> | so, we should be able to move most of this stuff off of jenkins |
| [07:04:40] | <gboudrias> | Yeah I followed the discussions a bit, looks pretty neat |
| [07:05:46] | <gboudrias> | ergonlogic: https://aegir.readthedocs.org/en/3.x/community/release-process/#26-build... do we still need to build manually or is that outdated? |
| [07:06:14] | <ergonlogic> | yeah, follow the docs |
| [07:06:26] | <ergonlogic> | I've only experimented a bit with OBS |
| [07:06:42] | <ergonlogic> | the repos there were generated by pointing OBS at our repos |
| [07:06:44] | <gboudrias> | okay, I see there's no equivalent jenkins job for 2.x so that makes sense |
| [07:06:47] | <ergonlogic> | our .deb repo |
| [07:06:49] | <gboudrias> | cool |
| [07:08:00] | <ergonlogic> | I wonder if there's a way to mirror d.o git repos to github (without the extra user that Jon's service requires) |
| [07:08:24] | <ergonlogic> | I assume that we can't install a post receive hook on d.o |
| [07:09:26] | <ergonlogic> | nor can we run them on guthub |
| [07:09:33] | <ergonlogic> | hmm.... |
| [07:10:34] | <ergonlogic> | if we can get commits mirrored to github, we could use travisci to trigger OBS builds via the cli client |
| [07:11:24] | <ergonlogic> | we could also use travisci to trigger our tests |
| [07:11:45] | <ergonlogic> | though it doesn't support vagrant |
| [07:12:06] | <ergonlogic> | and is still running on ubuntu 12.04, iirc |
| [07:12:40] | <ergonlogic> | so, we could go back to spinning up temporary micro AWS instances for our tests, or something |
| [07:12:43] | <gboudrias> | Seems like the tests also fail on Trusty |
| [07:13:01] | <ergonlogic> | ==> default: err: /Stage[main]/Drush/Package[drush]/ensure: change from purged to present failed: Execution of '/usr/bin/apt-get -q -y -o DPkg::Options::=--force-confold install drush' returned 100: Reading package lists... |
| [07:13:10] | <ergonlogic> | gboudrias: something like ^^^ ? |
| [07:13:16] | <gboudrias> | ergonlogic: yeah |
| [07:13:49] | <ergonlogic> | maybe add an 'apt-get update' step in the puppet module |
| [07:15:06] | <gboudrias> | okay I'll try that later, I'm just testing the manually built packages now |
| [07:15:29] | <ergonlogic> | http://stackoverflow.com/a/13655214 |
| [07:15:43] | <ergonlogic> | ok, let me know if that resolves it |
| [07:19:18] | <hefring> | Git => change version information for release 2.5 => http://cgit.drupalcode.org/provision/commit/?id=72c5f4aa4d57d415cf8cd6f7... |
| [07:19:18] | <hefring> | Git => Revert "change version information for release 2.5" => http://cgit.drupalcode.org/provision/commit/?id=79a9a9a09d3d9ca15162cc2d... |
| [07:21:12] | <ergonlogic> | when did hefring start announcing git commits to the channel? |
| [07:21:25] | <ergonlogic> | and can we get it to monitor 7.x-3.x? |
| [07:25:26] | <gboudrias> | Heh, no idea |
| [07:42:31] | <gboudrias> | packages seem to work on Wheezy |
| [07:47:06] | * ivanjaros has joined #aegir |
| [08:00:31] | * Egyptian[Home] has joined #aegir |
| [08:01:36] | * josh_k_ has joined #aegir |
| [08:01:36] | * josh_k has quit (Disconnected by services) |
| [08:05:49] | * zombiebeard has quit (Quit: zombiebeard) |
| [08:07:05] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [08:36:26] | * josh_k_ has quit (Remote host closed the connection) |
| [08:36:51] | * drakythe is now known as zz_drakythe |
| [08:36:55] | * josh_k has joined #aegir |
| [08:42:38] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [08:42:47] | <gboudrias> | Fails on Jessie but I think they always did |
| [08:44:21] | <gboudrias> | bgm / ergonlogic right? |
| [08:45:02] | * e-anima has quit (Read error: Connection reset by peer) |
| [08:48:10] | <gboudrias> | Be back later, I'll wait until someone else is online, just in case, so this will probably go to tomorrow morning |
| [08:53:25] | * gusaus has quit (Quit: gusaus) |
| [08:57:42] | * hestenet has quit (Remote host closed the connection) |
| [09:00:34] | <bgm> | gboudrias: except for the hosting_queued issue, installed fine for me on jessie |
| [09:09:23] | * gusaus has joined #aegir |
| [09:31:30] | * josh_k has quit () |
| [09:36:13] | * Topplestack has joined #aegir |
