IRC logs for #aegir, 2017-04-06 (GMT)

2017-04-05
2017-04-07
TimeNickMessage
[10:10:29]* joshr_ has quit (Ping timeout: 260 seconds)
[10:18:02]* realityloop has joined #aegir
[10:24:58]* realityloop has quit (Quit: Leaving..)
[12:10:29]* hestenet has joined #aegir
[12:10:35]* mutin-s has joined #aegir
[12:13:58]* s-mutin has quit (Ping timeout: 256 seconds)
[12:40:21]* hestenet has quit (Remote host closed the connection)
[12:41:03]* hestenet has joined #aegir
[12:45:42]* hestenet has quit (Ping timeout: 256 seconds)
[14:00:18]* hestenet has joined #aegir
[14:08:23]* hestenet has quit (Remote host closed the connection)
[14:16:58]* s-mutin has joined #aegir
[14:17:59]* mutin-s has quit (Ping timeout: 260 seconds)
[15:48:44]* realityloop has joined #aegir
[16:45:15]* realityloop has quit (Remote host closed the connection)
[17:39:41]* reaper013 has joined #aegir
[18:13:50]* ybabel has joined #aegir
[18:56:16]* ybabel has quit (Remote host closed the connection)
[18:59:02]* ybabel has joined #aegir
[19:06:25]* realityloop has joined #aegir
[19:26:51]* realityloop has quit (Remote host closed the connection)
[19:39:11]* realityloop has joined #aegir
[19:44:00]* realityloop has quit (Ping timeout: 260 seconds)
[20:34:03]* realityloop has joined #aegir
[20:59:58]* realityloop has quit (Remote host closed the connection)
[22:27:02]* noecc has joined #aegir
[23:29:25]* realityloop has joined #aegir
[23:33:35]* realityloop has quit (Ping timeout: 240 seconds)
[00:00:35]* hestenet has joined #aegir
[00:12:44]* v20th has joined #aegir
[01:31:05]* realityloop has joined #aegir
[01:35:35]* realityloop has quit (Ping timeout: 240 seconds)
[01:44:03]* shaneonabike1 has joined #aegir
[02:03:43]* reaper013 has quit (Quit: Page closed)
[02:24:31]* shaneonabike1 has quit (Quit: Leaving.)
[02:24:39]* aegir080 has joined #aegir
[02:27:48]* realityloop has joined #aegir
[02:29:19]* aegir080 has quit (Ping timeout: 260 seconds)
[02:29:24]* vatican has joined #aegir
[02:32:40]* realityloop has quit (Ping timeout: 268 seconds)
[02:42:28]* v20th has quit (Ping timeout: 246 seconds)
[02:48:40]* realityloop has joined #aegir
[02:52:40]* vatican has quit (Read error: Connection reset by peer)
[02:53:03]* vatican has joined #aegir
[02:53:19]* realityloop has quit (Ping timeout: 246 seconds)
[03:07:32]* hestenet has quit (Remote host closed the connection)
[03:07:48]* v20th has joined #aegir
[03:08:01]* hestenet has joined #aegir
[03:09:37]* realityloop has joined #aegir
[03:12:58]* hestenet has quit (Ping timeout: 264 seconds)
[03:14:07]* shaneonabike1 has joined #aegir
[03:14:20]* realityloop has quit (Ping timeout: 260 seconds)
[03:28:43]* realityloop has joined #aegir
[03:32:57]* realityloop has quit (Ping timeout: 240 seconds)
[03:49:35]* realityloop has joined #aegir
[03:52:41]* gusaus has joined #aegir
[03:54:08]* realityloop has quit (Ping timeout: 240 seconds)
[03:59:51]* hestenet has joined #aegir
[04:06:59]<helmo>Hi all, ergonlog1c bgm jonpugh colan cweagans gboudrias memtkmcc .. Scrum time?.
[04:10:34]* realityloop has joined #aegir
[04:11:13]<helmo>I got a bit further with gitlab building Debian packages ... they are now uploaded to the Coop's server when reprepro will have to pick them up ... I think I now have to get them gpg signed to add them.
[04:11:41]<helmo>I also archived Jenkins to ci-jenkins-archive.aegirproject.org/ (GH pages) in preparation of it's shutdown.
[04:12:09]<bgm>cool
[04:15:29]* hefring has joined #aegir
[04:16:06]<helmo>I added a ci test for Debian stretch ... ergonlogic, you mention in https://www.drupal.org/node/2770819 that you got passed the mariadb password issue. Can you do the same in https://gitlab.com/aegir/provision/blob/feature/gitlab-testing/.gitlab-c...
[04:16:09]<hefring>https://www.drupal.org/node/2770819 => Install fails on MariaDB 10 without password (new default) [#2770819] => 11 comments, 6 IRC mentions
[04:17:05]<colan>helmo: assuming you're documenting all of this as you go so the rest of us will know where everything is / how it's set up?
[04:18:38]<helmo>Yes that's the idea :) I already added the GitLab build status in the table on https://github.com/aegir-project/tests
[04:19:07]<colan>Thanks for working on all of this, and to memtkmcc for providing incentive. :-)
[04:19:37]<helmo>I'll update the release process docs as we go luckily we still have ~6 weeks before the next release
[04:24:36]<colan>Fantastic.
[04:28:45]<colan>I ran into a memory issue with the backup manager, but I'm assuming it's an edge case.
[04:29:12]<colan>Should be the top one in the queue.
[04:30:09]<colan>Anyone have a good idea of how that's implemented?
[04:31:11]<colan>I just did things differently for now, but couldn't find a way to say "don't keep any backups after a certain time".
[04:31:24]<colan>1 year for example.
[04:32:21]<colan>Maybe this is a larger UI issue. :-)
[04:34:59]<jonpugh>colan: I'll say...
[04:35:01]<jonpugh>;)
[04:36:21]<helmo>colan: I just worked on another issue for that module ... https://www.drupal.org/node/2867233
[04:36:22]<hefring>https://www.drupal.org/node/2867233 => Can't disable site backup and multiple backups being spawned for same site that fail [#2867233] => 3 comments, 1 IRC mention
[04:37:46]<helmo>colan: so it's the example os dates to keep that get so huge?
[04:39:02]* shaneonabike1 has left #aegir ("PART #sovereign :PART #drupal-commerce :PING 1491503942")
[04:40:48]<colan>helmo: couldn't get that far as i ran into a 500 error. ;-)
[04:41:22]<colan>Didn't dig too deeply into it; just worked around.
[04:42:53]<helmo>colan: first suspect is _hosting_backup_gc_example() ... try adding a "return '';" in there next time you see this.
[04:43:37]* realityloop has joined #aegir
[04:45:05]<colan>Cool. In other news, https://www.drupal.org/project/quenforcer is ready for testing.
[04:47:09]<colan>Also, code reviews would be appreciated.
[04:48:58]* realityloop has quit (Ping timeout: 264 seconds)
[04:49:57]<jonpugh>That looks awesome, colan
[04:50:36]<jonpugh>So it will block users from creating content if the DB is full?
[04:52:09]<colan>jonpugh: well, not if it's full, but if the storage quota limit was reached.
[04:52:33]<colan>Unless that's what you mean.
[04:53:19]<jonpugh>Right, that's what I meant, I guess, "full" as far as the user knows
[04:53:33]<colan>Correct.
[04:54:22]<helmo>colan: nice
[04:55:31]<helmo>jonpugh: Can we get https://www.drupal.org/node/2851478 rolling? It looks like an easy on to just get out of the way ;)
[04:55:32]<hefring>https://www.drupal.org/node/2851478 => Move server's http service config symlinking out of Provision_Config class. [#2851478] => 8 comments, 3 IRC mentions
[05:04:41]<jonpugh>I'll take a look but might have to wait until next week...
[05:05:45]<jonpugh>I'm cleaning up work on devshop that led to all those patches... Install method, hosting git with makefiles and more
[05:06:24]<jonpugh>Reinstall ability...
[05:16:12]<jonpugh>helmo: added some comments, I think adding it to ProvisionService might be too broad
[05:17:07]<jonpugh>https://www.drupal.org/node/2851478#comment-12027012
[05:17:13]<hefring>https://www.drupal.org/node/2851478 => Move server's http service config symlinking out of Provision_Config class. [#2851478] => 9 comments, 4 IRC mentions
[05:26:38]<helmo>jonpugh: I think Neograph734 is correct in the comment just now ... the symlink_service is only defined in the base class, not called.
[05:26:56]<jonpugh>Ahh right
[05:27:48]<jonpugh>That's what I get for reviewing from my phone...
[05:28:10]<jonpugh>Maybe we should add a note in init_server() about using the method?
[05:29:19]* noecc has left #aegir ("pax")
[05:29:59]<jonpugh>Ok, commenting....
[05:31:58]<helmo>sure, that could help
[05:53:42]* ybabel has quit (Ping timeout: 268 seconds)
[05:56:44]* hestenet has quit (Remote host closed the connection)
[06:01:37]* ybabel has joined #aegir
[06:08:00]* roycroft has joined #aegir
[06:08:07]<roycroft>hello, folks
[06:08:22]<roycroft>i've been asked to import some drupal 7 sites into aegir
[06:08:44]<roycroft>it seems a fairly straightforward thing to do, with one potential issue
[06:09:00]<roycroft>the sites in question don't have their own hierarchy under sites
[06:09:37]<roycroft>there's just a settings.php in the sites/default
[06:10:34]<roycroft>aegir platforms prefer sites/sitename, and put the local modules/settings/files,/etc directories there
[06:10:54]<roycroft>is it as simple as creating sits/sitename and moving the stuff from sites/default to there?
[06:20:30]<viashimo>roycroft: there's a few other considerations: in the db for the site, uploaded content written by users can often refer to the path sites/default/... you will want ot check in dump to see if there's content to update
[06:21:16]<viashimo>roycroft: you'll probably want to run drush rr (maybe cc all is sufficient with newer versions of drush?) for the system table to pick up on any moved modules in the site-specific folder
[06:22:49]<roycroft>i got this when i attempted to do what i mentioned:
[06:22:51]<roycroft>Drush command terminated abnormally due to an unrecoverable error.
[06:22:54]<roycroft>when verifying the site
[06:23:07]* viashimo shrugs
[06:23:10]<roycroft>i'm going to try verifying again with the site named back to default
[06:23:20]<viashimo>the error message is pretty vague :)
[06:23:23]<roycroft>and then i'll look into the suggestions you just made
[06:23:25]<roycroft>yes it is
[06:23:45]<roycroft>it's like stabbing in the dark to fix the problem with a clue like that :)
[06:24:15]<roycroft>i'm verifying the platform again now
[06:24:49]<roycroft>and the verify was fine
[06:25:01]<viashimo>often when I do a first pass I take a long form approach: create a new stub site with the desired name, copy all the sites/default/* to sites/newsite/. Builders using the sites/default often chuck all their modules into sites/all, so I make sure those are in the platform or moved to sites/sitename/modules
[06:25:16]<viashimo>then I'll import the db (after a find/replace) with drush sqlc < db.sql
[06:25:44]<viashimo>oh, and as a note, don't overwrite the settings.php and drushrc.php in the stub site with the ones from default :)
[06:25:49]<roycroft>i manually copied the database to the aegir master
[06:25:52]<roycroft>the test aegir master
[06:26:05]<roycroft>what i did was go to the stand-alone machine
[06:26:18]<roycroft>made a tarball of the drupal install of the site
[06:26:23]<roycroft>did a mysqldump of the database
[06:26:30]<roycroft>scp'ed both to the aegir master
[06:26:43]<roycroft>untar'ed the drupal code in /var/aegir/platforms
[06:26:53]<roycroft>created an empty database and restored the db to it
[06:27:00]<roycroft>added the correct grant
[06:27:06]<roycroft>and then imported the new platform
[06:27:34]<roycroft>and now that i moved the site back to default the platform verifies fine
[06:27:38]<roycroft>but i have no site
[06:27:48]<roycroft>and i'm not sure how to generate a site
[06:28:02]<viashimo>roycroft: add site in the interface?
[06:28:23]<roycroft>but will it pick up the existing database?
[06:28:32]<viashimo>roycroft: no
[06:28:43]<roycroft>so what do i do?
[06:28:58]<viashimo>roycroft: that's why I import the database after making a stub site
[06:29:03]<roycroft>create the site and let aegir name a database, then restore my actual database to the aegir-generated one?
[06:29:09]<viashimo>roycroft: yup!
[06:29:11]<roycroft>ok
[06:29:13]<roycroft>i can do that
[06:29:18]<viashimo>roycroft: I've found that easier than mucking around with platform import
[06:29:51]* realityloop has joined #aegir
[06:30:24]<roycroft>grr
[06:30:35]<roycroft>it attempted to create the site before but could not verifyt it
[06:30:45]<roycroft>and now that i've moved the files back to default it can't delete it
[06:30:53]<roycroft>is there a drush command to force delete a site?
[06:31:36]<roycroft>i guess i can rename it back, delete it, and then start over
[06:33:18]<roycroft>sorry - the aegir way of doing things is different than what i'm used to, and sometimes it takes a few goes before something clicks with me
[06:33:23]<viashimo>roycroft: when I've got weirdness with an installed site I'll sometimes just hit up node/XYZ/delete (being the nid of the site in the hostmaster interface). It's not a recommended approach, but it'll clear out the references in the hostmaster db. You just have to clean up the related databases manually
[06:33:44]<viashimo>roycroft: np, like any tool it takes time to get used to the ins and outs!
[06:34:18]<viashimo>I stumbled over this same stuff alot when I started using aegir
[06:34:30]<roycroft>i'm pretty well covered from screwups though
[06:35:06]<roycroft>i've built three identical aegir3 masters - one is a scratch machine, and i do vmware snapshots before doing anything so i can always roll back when i hose stuff
[06:35:32]<roycroft>one is a testing machine that i move to once i'm sure that my install isn't going to break things, and is used for site development and testing
[06:35:46]<roycroft>and the other is the production aegir master which we use to deploy sites in production
[06:36:16]<roycroft>i got tired of getting up in the middle of the night to restore a bunch of stuff because my boss couldn't sleep and decided to do something fun and scary on a production server
[06:43:51]* ybabel has quit (Quit: ybabel)
[06:49:56]<roycroft>it looks like i have the site imported
[06:50:06]<roycroft>i did as you suggested, and created a stub site
[06:50:38]<roycroft>then i dropped its database, recreated it, and poplated it from the stand-alone site's database dump
[06:51:04]<roycroft>i then copied the stand-alone site's files hierarcy into the stub site's
[06:51:28]<roycroft>it looks like it's working, but my boss developed and maintains the site, so i handed it over to him for verification that everything is working
[06:52:21]<roycroft>fortunately there were no modules/libraries directories in default
[06:57:57]* realityloop has quit (Quit: Leaving..)
[07:07:55]* hestenet has joined #aegir
[07:27:44]* vatican has quit (Ping timeout: 260 seconds)
[07:30:56]<viashimo>roycroft: great!
[08:05:04]* v20th has quit (Ping timeout: 260 seconds)
[08:20:26]* v20th has joined #aegir
[08:22:35]* ybabel has joined #aegir
[09:19:40]* mstenta has quit (Ping timeout: 268 seconds)
[09:31:41]* mstenta has joined #aegir