| [11:20:46] | * ybabel has quit (Ping timeout: 268 seconds) |
| [12:18:48] | * theMusician has quit (Quit: theMusician) |
| [12:19:52] | * theMusician has joined #aegir |
| [12:21:49] | <gusaus> | this is a pretty good example of an OSS community site, huh https://civicrm.org/ |
| [12:24:35] | <gusaus> | has anyone made that into a distribution and put on an Aegir platform? :) |
| [13:50:52] | <ergonlogic> | gusaus: well, there's https://github.com/civicrm/civicrm-website-org... but I don't know how easy it'd be to replicate it |
| [13:57:57] | <gusaus> | ergonlogic: interesting - I was just pondering about how difficult it would be to turn something like that into a Drupal distribution in the OpenCollective room. There are probably enough orgs and OSS projects in their community that would pay for something like that |
| [14:00:59] | <gusaus> | especially if it was available as part of hosted offering powered by Aegir :) |
| [15:39:58] | * gusaus has quit (Quit: gusaus) |
| [18:09:50] | * boshtian has joined #aegir |
| [18:34:22] | * ybabel has joined #aegir |
| [19:13:35] | <helmo> | bgm, ergonlogic: can I get an rtbc on https://www.drupal.org/node/2699211 ? The old makefile is just not functional :( |
| [19:13:35] | <hefring> | https://www.drupal.org/node/2699211 => Example makefile doesn't work [#2699211] => 8 comments, 1 IRC mention |
| [21:51:48] | * oluabbeys has joined #aegir |
| [21:52:37] | * oluabbeys has quit (Client Quit) |
| [22:24:40] | * ybabel1 has joined #aegir |
| [22:24:53] | * ybabel has quit (Read error: Connection reset by peer) |
| [22:24:53] | * ybabel1 is now known as ybabel |
| [01:10:46] | * noecc has joined #aegir |
| [01:21:53] | * mstenta has joined #aegir |
| [01:26:38] | * fatguylaughing has joined #aegir |
| [01:38:14] | <colan> | if anyone has d6 / aegir 2 sites around, please test the latest Remote Import dev for https://www.drupal.org/node/2828482 |
| [01:38:15] | <hefring> | https://www.drupal.org/node/2828482 => Allow importing of disabled sites as well [#2828482] => 3 comments, 1 IRC mention |
| [01:43:54] | * viashimo has quit (Ping timeout: 246 seconds) |
| [01:51:02] | * viashimo has joined #aegir |
| [02:29:22] | * fatguylaughing has quit (Read error: Connection reset by peer) |
| [02:31:09] | * fatguylaughing has joined #aegir |
| [02:37:15] | * boshtian has quit (Ping timeout: 256 seconds) |
| [03:10:43] | * boshtian has joined #aegir |
| [03:25:05] | * boshtian has quit (Quit: boshtian) |
| [04:05:23] | * christefano has joined #aegir |
| [04:40:52] | * shaneonabike has joined #aegir |
| [04:58:58] | * theMusician has quit (Ping timeout: 268 seconds) |
| [05:10:04] | * theMusician has joined #aegir |
| [05:23:07] | * theMusician has quit (Quit: theMusician) |
| [05:41:43] | * gusaus has joined #aegir |
| [05:44:26] | <helmo> | ergonlogic: Scrum in 15min? |
| [05:53:58] | <jonpugh> | sure |
| [05:54:10] | * cweagans has joined #aegir |
| [05:54:32] | * cweagans has quit (Changing host) |
| [05:54:32] | * cweagans has joined #aegir |
| [05:54:55] | * cweagans has quit (Client Quit) |
| [05:55:14] | * cweagans has joined #aegir |
| [05:56:22] | <cweagans> | . |
| [05:57:37] | * Einherjer has joined #aegir |
| [06:02:47] | * theMusician has joined #aegir |
| [06:03:49] | * cweagans has quit (Quit: cweagans) |
| [06:06:30] | <helmo> | Hi all, ergonlog1c bgm jonpugh colan gboudrias memtkmcc .. It's Scrum time. |
| [06:06:45] | <jonpugh> | helllo |
| [06:07:11] | <jonpugh> | Two important patches from me this week: https://www.drupal.org/node/2824731 |
| [06:07:13] | <hefring> | https://www.drupal.org/node/2824731 => API Only: Allow creation of site and new platform in one request. [#2824731] => 9 comments, 2 IRC mentions |
| [06:07:30] | <jonpugh> | and https://www.drupal.org/node/2754069 |
| [06:07:31] | <hefring> | https://www.drupal.org/node/2754069 => API Only: Allow site nodes to specify "install_method", allowing something other than automated Drupal profile. [#2754069] => 18 comments, 4 IRC mentions |
| [06:08:52] | <jonpugh> | Both are being included in devshop |
| [06:09:14] | <jonpugh> | the latter is great, allows me to interrupt drupal profile run and do an SQL sync into the empty database instead. |
| [06:09:41] | * cweagans has joined #aegir |
| [06:10:20] | <helmo> | jonpugh: yes that's nicer then installing a dummy site and directly overwriting it :) |
| [06:10:48] | <jonpugh> | and more manageable than a Clone (plus import) task |
| [06:11:13] | <jonpugh> | I'm going to add things like import SQL directly from a file. |
| [06:11:55] | <jonpugh> | then combine that with a "DB Dump queue" which will periodically write to that file from a production site, including database sanitization, so dev and test sites can be created without impacting production DB |
| [06:12:15] | <jonpugh> | and allow remote downloads of that file for developers |
| [06:12:16] | * cweagans has left #aegir () |
| [06:12:25] | <helmo> | nice |
| [06:12:38] | <jonpugh> | not this week on that, though :) |
| [06:13:51] | <helmo> | If only we had an infinite amount of time ;) ... I've been wanting to try your api only issues but have not had serious time for it |
| [06:16:32] | <helmo> | jonpugh: should I test the 2754069-decouple-install or 2754069-decouple-install-alternate branch? |
| [06:16:39] | <jonpugh> | alternate |
| [06:18:46] | <ergonlogic> | that all sounds quite promising |
| [06:20:31] | * colan waves |
| [06:20:52] | <colan> | who's next? |
| [06:21:30] | <colan> | i'll go i guess. |
| [06:21:42] | <helmo> | ok, I've added it to my feature/quick-review branches |
| [06:21:42] | <ergonlogic> | go for it |
| [06:21:50] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [06:21:56] | <colan> | i started working with Remote Import. |
| [06:22:08] | <helmo> | I've also just now updated the test suite to replace 7.51 with 7.52 :) |
| [06:22:19] | <colan> | thumbs up on that. |
| [06:23:26] | <colan> | first problem i ran into was https://www.drupal.org/node/2826739 |
| [06:23:27] | <hefring> | https://www.drupal.org/node/2826739 => Overriden Provision_Service_remote_import_hostmaster methods incompatible with parent [#2826739] => 2 comments, 1 IRC mention |
| [06:23:58] | <colan> | ...which could be a php7 warning if nobody else has seen that yet. :) |
| [06:24:22] | * fatguylaughing has joined #aegir |
| [06:24:43] | <colan> | anyway, turns out that warning prevent the module from working which took me a while to figure out. so i fixed with https://www.drupal.org/node/2827791 |
| [06:24:44] | <hefring> | https://www.drupal.org/node/2827791 => List Remote Sites task produces no results with warnings [#2827791] => 4 comments, 2 IRC mentions |
| [06:25:15] | <ergonlogic> | colan: have you checked out the dev/2538788 branch? |
| [06:25:33] | <colan> | i added another commit to allow for importing disabled sites as that's useful. |
| [06:25:35] | <helmo> | Updateing the methods paramaters to match shoudn't be to hard ... ? |
| [06:25:48] | <colan> | ergonlogic: no, hadn't noticed it. |
| [06:26:39] | <colan> | i usually assume that branches like that are dead, sorry. |
| [06:27:12] | <ergonlogic> | it may have been merged... |
| [06:27:21] | <ergonlogic> | it's been a while since I've had to use it |
| [06:27:39] | <ergonlogic> | but I know we'd worked fixed it up at the time |
| [06:27:55] | <colan> | ergonlogic: yes it was! so no problem there. |
| [06:28:24] | <colan> | helmo: i looked, and it appears non-trivial, but i'm not that familiar with the code. |
| [06:28:52] | <colan> | but now it'll run with warnings so i'm not too concerned. though it would still be good to fix. |
| [06:30:03] | <colan> | if anyone has aegir 2 sites, please test https://www.drupal.org/node/2828482 |
| [06:30:03] | <hefring> | https://www.drupal.org/node/2828482 => Allow importing of disabled sites as well [#2828482] => 3 comments, 2 IRC mentions |
| [06:30:27] | <colan> | i think i got the code right for aegir2; i just have no way to test it. ;) |
| [06:31:08] | <colan> | the issue that's really wacky is https://www.drupal.org/node/2822454 |
| [06:31:09] | <hefring> | https://www.drupal.org/node/2822454 => Aegir 3.x does not supply an aegir_api variable [#2822454] => 4 comments, 1 IRC mention |
| [06:31:37] | <colan> | so i fixed in provision at https://www.drupal.org/node/2826770 |
| [06:31:38] | <hefring> | https://www.drupal.org/node/2826770 => Version setting "aegir_api" getting zeroed during backup task [#2826770] => 4 comments, 2 IRC mentions |
| [06:31:39] | <ergonlogic> | that is odd, since it's in settings.php |
| [06:32:10] | <colan> | ...but maybe i'm missing something, as they intentionally set that up to clear the variable on backups. |
| [06:32:36] | <colan> | it makes no sense to me. if there's no good reason, let's get my patch in there. :) |
| [06:33:52] | <ergonlogic> | so... you're trying to import the hostmaster site itself there? |
| [06:33:55] | <helmo> | I think the though was to let the backups be more 'neutral' |
| [06:34:14] | <colan> | what's funny is omega8cc raised this way back, but was ignored. |
| [06:34:53] | <colan> | ergonlogic: no, but it's the target server. |
| [06:34:56] | <helmo> | but I don't depend on it |
| [06:35:27] | <colan> | ergonlogic: the target server needs an api value, or Remote Import can't talk to it. |
| [06:36:57] | <colan> | helmo: but then the value should be in the backup, not change the api value of the server. |
| [06:37:04] | <ergonlogic> | ok, so aegir_api gets set to 0 in the settings.php that's bundled into the backup, or remains so afterwards? |
| [06:37:38] | <colan> | ergonlogic: remains so afterwards. if you drush vget the variable after a backup task runs, it's 0. |
| [06:38:02] | <colan> | so then you have to run a verify on hostmaster to get it back to 3. |
| [06:38:38] | <colan> | Remote Import really talks to hostmaster, i think, not the server, so maybe that's what's confusing. |
| [06:39:16] | <colan> | hostmaster verify sets it 3. hostmaster backups sets to 0. |
| [06:40:17] | <ergonlogic> | right, just confirmed that |
| [06:40:38] | <helmo> | So if we really want to leave it out of the backup it should be restored earlier |
| [06:41:18] | <ergonlogic> | we re-write the settings.php again following the backup, so it should get reset there, I'd have thought |
| [06:41:55] | <helmo> | but as it's just one line of config ... I don't really see the problem of having that in the backup. It could even have the bennefit of external platform X using it's value to determin the best import method for the Aegir version where this backup came from. |
| [06:42:14] | * colan nods |
| [06:42:19] | <ergonlogic> | agreed |
| [06:42:29] | * spyd has quit (Ping timeout: 252 seconds) |
| [06:42:51] | <colan> | maybe they had a good reason; i just don't know what it would be. |
| [06:43:02] | <ergonlogic> | do we check that variable on imports or something? |
| [06:44:25] | <colan> | if you want to zero it in the backup, at least set it back to what it was before. |
| [06:44:31] | <colan> | ...when done. |
| [06:44:53] | <ergonlogic> | it looks like, in Provision, it's never even checked |
| [06:45:30] | <colan> | i've grepped though all of core projects, and yeah, nobody checks it. |
| [06:45:42] | <colan> | it's only set in those 3 template files. |
| [06:46:53] | <colan> | in golden it's only Remote Import. though i may be missing a project or two. |
| [06:47:17] | <colan> | so if remote import is the only one using it... |
| [06:47:31] | <helmo> | I'll rtbc |
| [06:47:35] | <colan> | thanks. |
| [06:47:36] | <ergonlogic> | that's all I see too |
| [06:47:38] | <helmo> | hefring: logpointer |
| [06:47:44] | <helmo> | hefring: log pointer |
| [06:47:44] | <hefring> | http://hefring.mig5.net/bot/log/aegir/2016-11-17#T643176 |
| [06:48:03] | * theMusician has quit (Quit: theMusician) |
| [06:49:27] | <colan> | i think that's it for me. oh, i know i keep saying this, but i'll be getting back to hosting_https soon as that's become a priority for something. |
| [06:49:38] | * spyd has joined #aegir |
| [06:49:38] | <colan> | now i'm done. :) |
| [06:49:55] | <ergonlogic> | I think the real problem though is that (presumably) backup_in_progress isn't being unset/reset by the time we're re-generating the setting.php, post-backup |
| [06:51:02] | <colan> | ah, race condition. well, it seems we don't need to check it anyway. |
| [06:52:25] | <colan> | who's next? |
| [06:53:33] | <ergonlogic> | I mean that backup_in_progress isn't used anywhere either |
| [06:53:43] | <ergonlogic> | anyway... all just reason for a re-write |
| [06:54:07] | <ergonlogic> | un-explainable cruft |
| [06:54:15] | <colan> | ha! |
| [06:54:19] | <helmo> | I don't have much more to report then that I've been reviewing a bit. Not much new stuff. |
| [06:54:30] | <helmo> | Jon: The oldest needs-review could use your attention... https://www.drupal.org/node/2413023 |
| [06:54:31] | <hefring> | https://www.drupal.org/node/2413023 => Dupal 8 support for watchdog access [#2413023] => 2 comments, 1 IRC mention |
| [06:54:45] | <helmo> | jonpugh: ^ |
| [06:56:10] | <helmo> | I'd like to make it a weekly topic to get a new comment on the oldest needs-review issue in the combined queue. |
| [06:56:27] | <ergonlogic> | good idea |
| [06:56:47] | <helmo> | This one is for hosting_logs and exposed watchdog messages from a hosted site in the hostmaster site |
| [06:59:35] | <helmo> | Anyone else? |
| [07:00:06] | <helmo> | Otherwise it's probably time to jump to the #aegir-coop channel for an update there |
| [07:00:07] | <ergonlogic> | yep, I've made some further progress on code-coverage and such on the aegirNG |
| [07:00:46] | <ergonlogic> | and added a feature template, that will prompt to add a gherkin-like user story, etc. |
| [07:01:16] | <ergonlogic> | and have options to build/push docker images as part of the CI process |
| [07:01:54] | <colan> | is this via gitlab ci? |
| [07:02:09] | <ergonlogic> | so, I've begun working through some of these processes in the issue queue on gitlab |
| [07:02:13] | <ergonlogic> | yes |
| [07:02:20] | <helmo> | The GitLab pipelines look awsome |
| [07:02:24] | <ergonlogic> | https://gitlab.com/aegir/aegir |
| [07:02:32] | <ergonlogic> | I made the project public |
| [07:02:50] | <ergonlogic> | there's like 115 commits that basically come down to "fix CI" |
| [07:03:01] | <ergonlogic> | so, we might want to squash those... |
| [07:03:47] | <ergonlogic> | anyway, it'll update http://docs.aegir.hosting/ on successful commits |
| [07:03:58] | <ergonlogic> | and generate http://docs.aegir.hosting/coverage/index.html |
| [07:04:11] | <colan> | wow, nice. |
| [07:04:12] | <ergonlogic> | which is now merging results from simpletest and behat |
| [07:05:10] | <ergonlogic> | there's some optimization we could presumably achieve by caching the platform build, etc. |
| [07:05:27] | <ergonlogic> | but I ran into issues with Ansible and AUFS |
| [07:05:47] | <ergonlogic> | those may have been from a broken image though, so it might be worth re-visiting |
| [07:06:14] | <ergonlogic> | it's announcing successes/failures in #aegir-dev, btw |
| [07:06:34] | <helmo> | cool ... hope we can keep those numbers on 100% :) |
| [07:06:43] | <ergonlogic> | yeah |
| [07:07:04] | <ergonlogic> | one of the nice things about templates in gitlabci is that you can add a checklist |
| [07:07:16] | <ergonlogic> | keeping coverage at 100% is one of the items there |
| [07:07:25] | <ergonlogic> | along with review by a core team member |
| [07:07:44] | <ergonlogic> | I added a new Eldir that's a basic sub-theme of bootstrap |
| [07:07:51] | <ergonlogic> | as a starting point |
| [07:08:12] | <ergonlogic> | https://gitlab.com/aegir/aegir/issues/1 |
| [07:08:44] | <ergonlogic> | of course, it appears to have broken simpletest, so I'll work that stuff out |
| [07:09:14] | <anarcat> | ergonlogic: that looks like an empty aegir project - is this aegir4? :) |
| [07:09:27] | <ergonlogic> | but I'd like you all to review the process as documented, etc. |
| [07:09:40] | <ergonlogic> | anarcat: yep, just an empty profile, atm |
| [07:10:06] | <anarcat> | ergonlogic: i see, maybe it should be made clear in the index page, in case newcomers find this by mistake |
| [07:10:11] | <anarcat> | they will think aegir is vaporware ;) |
| [07:10:25] | <ergonlogic> | of the docs? |
| [07:10:34] | <ergonlogic> | the gitlab project mentions that it's under dev |
| [07:10:39] | <ergonlogic> | nd a re-write |
| [07:10:53] | <anarcat> | yeah, i understand |
| [07:11:07] | <anarcat> | maybe just a pointer to the normal aegir3 docs and a warning that it's under dev? |
| [07:11:12] | <ergonlogic> | sure |
| [07:11:15] | <anarcat> | like not for prod |
| [07:12:31] | <colan> | where do you add checklist items? |
| [07:12:41] | <colan> | great idea, btw. |
| [07:12:50] | <ergonlogic> | The version documented here is a full re-write using Drupal 8. It is not yet functional, and |
| [07:12:53] | <ergonlogic> | Aegir 3.x remains the supported release for the foreseeable future. |
| [07:13:11] | <ergonlogic> | and I'll add a link |
| [07:13:21] | <ergonlogic> | colan: in the issue description |
| [07:13:36] | <ergonlogic> | or in an issue template in .gitlab/issue_templates |
| [07:14:24] | <anarcat> | ergonlogic: awesome |
| [07:14:30] | <anarcat> | i really enjoy working on gitlab btw |
| [07:14:43] | <anarcat> | i know there'S less people there, but there are lots of interesting features not present on github |
| [07:15:00] | <anarcat> | like external wiki and integrated (e.g. not external like travis) CI |
| [07:15:00] | <ergonlogic> | yeah, it's phenomenal |
| [07:15:12] | <anarcat> | i hope people eventually all switch |
| [07:15:22] | <anarcat> | it's kind of awesome that we can just fix whatever is not like we want to |
| [07:15:29] | <anarcat> | and that we can'T do that on github |
| [07:15:45] | <ergonlogic> | they've really boosted up the CI runners too, so there isn't as much delay as previously |
| [07:16:05] | <colan> | they've completely opened their dev process too. it's a public project. |
| [07:16:27] | <anarcat> | yeah it's just awesome |
| [07:16:37] | <anarcat> | they have improved the response time on gitlab.com too |
| [07:16:40] | <anarcat> | it used to be too slow |
| [07:16:43] | <anarcat> | but now... |
| [07:16:53] | <anarcat> | i saw the unicorn on github the other day ;) |
| [07:17:06] | <anarcat> | they probably don't have the level of scalability issues that github has though |
| [07:17:43] | <ergonlogic> | ok, update the description of the project to point to Aegir 3 docs |
| [07:17:52] | <helmo> | sure Github+travins must have a few orders of magnatude more data and jobs running |
| [07:18:01] | <colan> | if you follow their blog, you can see updates. for example, https://about.gitlab.com/2016/11/10/why-choose-bare-metal/ |
| [07:18:04] | <ergonlogic> | and will commit a docs fix that'll be in the new docs as soon as I get a passing build |
| [07:18:35] | <anarcat> | helmo: yeah, it's crazy the amount of data in github |
| [07:18:36] | * hestenet has joined #aegir |
| [07:18:57] | <colan> | the only issue i've found is that gitlab.com is slow, but they're working on it :) |
| [07:19:56] | <ergonlogic> | anyway, so far the NG project is mostly such bootstrapping |
| [07:20:18] | <ergonlogic> | but I'd like us to adopt a bit stricter of a dev process moving forward |
| [07:20:25] | <ergonlogic> | so as to keep us on the right track |
| [07:20:50] | <anarcat> | you guys know about https://rfc.zeromq.org/spec:42/C4/? |
| [07:21:07] | <ergonlogic> | ensuring test coverage, primarily |
| [07:21:19] | <colan> | ergonlogic: this would help with the "This issue is caused by commit blah in issue blah", which i've been seeing a lot of lately. |
| [07:21:55] | * theMusician has joined #aegir |
| [07:22:19] | <ergonlogic> | anarcat: I had seen it previously, yes... we should adopt something similar |
| [07:22:46] | <ergonlogic> | colan: ?? |
| [07:23:02] | <ergonlogic> | oh, testing, etc.? yeah |
| [07:23:38] | <ergonlogic> | ok, that's it for me |
| [07:23:42] | <anarcat> | ergonlogic: i find it annoying they use "problem: " everywhere |
| [07:23:51] | <anarcat> | ergonlogic: i worked on a simpler version for monkeysign |
| [07:24:21] | <anarcat> | http://monkeysign.readthedocs.io/en/2.x/contributing.html#patches |
| [07:24:28] | <anarcat> | i also added a code of conduct bit |
| [07:24:49] | <anarcat> | interesting |
| [07:24:54] | <anarcat> | monkeysign'S docs don'T show the license |
| [07:25:10] | <anarcat> | anyways, it'S GPLv3 |
| [07:25:34] | <ergonlogic> | anarcat: feel free to create an MR on https://gitlab.com/aegir/aegir :) |
| [07:25:43] | <anarcat> | no time sory :) |
| [07:25:48] | <anarcat> | writing another article for LWN |
| [07:25:54] | <anarcat> | and getting paid for it too! :p |
| [07:26:03] | <ergonlogic> | I would like to briefly discuss some co-op stuff though |
| [07:26:11] | <ergonlogic> | over in #aegir-coop |
| [07:26:24] | <anarcat> | with me? |
| [07:26:58] | <colan> | anarcat: if you become a member ;) |
| [07:27:34] | <anarcat> | not yet :) |
| [07:27:48] | <ergonlogic> | anarcat: I meant w/ existing members, but you're welcome to lurk |
| [07:28:27] | <anarcat> | ok |
| [07:30:21] | <colan> | anarcat: where's the code for that cccc thing? can find a link to repo. |
| [07:30:36] | <colan> | i'd like to star/watch/whatever. |
| [07:30:39] | <anarcat> | on github somewhere, hold on |
| [07:30:47] | <anarcat> | https://github.com/zeromq/rfc/ |
| [07:30:55] | <colan> | merci. |
| [07:31:13] | <anarcat> | https://github.com/zeromq/rfc/issues?utf8=%E2%9C%93&q=is%3Aissue%20autho... |
| [07:31:40] | <anarcat> | see also https://anarc.at/blog/2016-10-14-bug-reporting/ |
| [07:35:03] | * cweagans has joined #aegir |
| [07:35:35] | <anarcat> | and https://bestpractices.coreinfrastructure.org/ |
| [07:37:55] | <ergonlogic> | we should look at adding issue templates for bug reports and such too |
| [07:39:50] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [07:47:18] | * fatguylaughing has joined #aegir |
| [07:47:34] | * theMusician has quit (Quit: theMusician) |
| [08:09:36] | * noecc has left #aegir ("pax") |
| [08:10:30] | * shaneonabike has quit (Quit: Leaving.) |
| [09:45:21] | * christefano has quit (Quit: christefano) |
| [10:00:39] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [10:35:20] | * mstenta has quit (Ping timeout: 244 seconds) |
