IRC logs for #aegir, 2018-05-02 (GMT)

2018-05-01
2018-05-03
TimeNickMessage
[16:37:24]* ybabel has joined #aegir
[18:19:22]* testinr[m] has quit (Ping timeout: 245 seconds)
[18:20:37]* ergonlogic[m] has quit (Ping timeout: 245 seconds)
[18:23:18]* ergonlogic[m] has joined #aegir
[18:24:03]* testinr[m] has joined #aegir
[19:16:15]* JohnYang has quit (Ping timeout: 260 seconds)
[20:17:32]* calbasi has joined #aegir
[22:27:44]* anarchat is now known as anarcat
[22:49:19]<bgm>Anyone remember the issue for when installing a site, the 'encryption' setting must not be enabled? it can only be installed after the site install? (because the vhost doesn't have the well-known block yet)
[22:49:36]<bgm>I can't seem to find the issue, but pretty sure there was one at some point
[22:50:14]<helmo42[m]>yes ... could be left in the Gitlab issue queue?
[22:51:26]<helmo42[m]>bgm: archive is in https://www.drupal.org/project/hosting_https/issues/2934484
[22:51:29]<hefring>https://www.drupal.org/project/hosting_https/issues/2934484 => Move back issue management to D.o [#2934484] => 4 comments, 1 IRC mention
[22:51:37]<bgm>ah hmm, might be https://gitlab.com/aegir/hosting_https/issues/19 ? (which says it's fixed)
[22:52:59]<helmo42[m]>as far as I know it is fixed
[22:55:38]<bgm>I think it was fixed on Apache only, from what I see in the patch (https://gitlab.com/aegir/hosting_https/merge_requests/36/diffs)
[23:05:13]<bgm>if I understand correctly, it's because in Apache the "well-known" can be injected in 'hook_provision_apache_server_config', so it's global, but we can't do the same with nginx (as a hosting submodule)
[23:08:13]<helmo42[m]>From colon's last comment in #19 it seends like (that he thought) it was fixed
[23:08:18]<helmo42[m]>Can't you add it via ./submodules/nginx_https/drush/Provision/Config/Nginx/Https/Server.php ?
[23:09:50]* shaneonabike1 has joined #aegir
[23:11:15]* shaneonabike1 has left #aegir ()
[23:12:03]<bgm>helmo42[m]: I think that adding it there would conflict with the 'location' blocks inside vhosts. That's how it was initially, but it was moved to vhosts. (I'm not sure why)
[23:17:26]<colan1>bgm: helmo42 nginx was done before apache. if anything, it might not work on apache.
[23:18:09]<bgm>colan1: silly question, but have you tested it recently on nginx?
[23:19:27]<colan1>bgm: i did just test it, and it worked, but can't remember if it was apache or nginx (had to set up some apaches recently). it's possible there's an nginx regression.
[23:20:14]<colan1>bgm: just make sure to turn full logging back on to get the LE output.
[23:20:47]<bgm>hmm, ok, weird, I just tested another time, and this time it worked :)
[23:21:56]<bgm>is there a magic option to show LE output in the Aegir task log?
[23:24:41]<bgm>ok, my bad, sorry
[23:24:58]<bgm>my colleague has been running into this, but I can't reproduce it anymore
[23:26:19]<colan1>bgm: that going away is caused by https://www.drupal.org/project/hosting/issues/2912540; see the workaround there to bring the logs back.
[23:26:19]<hefring>https://www.drupal.org/project/hosting/issues/2912540 => Reduce the number of log messages shown to the user. [#2912540] => 11 comments, 2 IRC mentions
[23:28:04]<bgm>colan1: cool, thanks
[00:15:29]<colan1>helmo42: also seeing another php7.2 thing, though it fills up the logs a bit less than the other one: ` 8305 01/May 17:13 warning php Warning: count(): Parameter must be an array or an object that implements Countable in _form_validate() (line 1441 of /var/aegir/hostmaster-7.x-3.x-2018-04-25-1750/includes/form.inc). ` not sure if it's too late to add the core patch for that one as well?
[00:16:46]<helmo42[m]>well ... the 3.15.1 tag has been pushed... I was just about to create the release node on D.o
[00:17:11]<colan1>ok, next time then.
[00:49:08]* helmo has changed the topic to Aegir hosting system 3.15.1 released! http://docs.aegirproject.org/en/3.x/release-notes/3.15/ | Planning for AegirNG has begun: https://gitlab.com/aegir/aegir | IRC Logs: https://hefring.aegirproject.org/bot/log/aegir | Scrum/Coop: Thursday's 18:00 UTC
[02:55:31]<roycroft>hello, folks
[02:55:37]<roycroft>i am frustrated today
[02:55:50]<roycroft>we have a machine running an aegir master that is used for development
[02:56:07]<roycroft>we have another machine with an aegir master that is used for staging and deployment
[02:56:33]<roycroft>i took a site off an old aegir 2 machine and did a remote_import onto the development machine
[02:56:47]<roycroft>i updated modules and the like and got the site running fine
[02:56:59]<roycroft>then i copied the platform to the staging/deployment machine
[02:57:20]<roycroft>and did the same thing, this time importing from the development machine
[02:57:41]<roycroft>the site works fine, but i get an error when running drush updatedb
[02:58:53]<roycroft>The following module is missing from the file system: <em [warning]
[02:58:54]<roycroft>class="placeholder">commerce_discount_date</em>.
[02:59:09]<roycroft>the module is there though
[02:59:38]<roycroft>i attempt to run the pending updates, and of course get an error
[03:00:07]<roycroft>i also copied the staging platform to a deployment platform
[03:00:19]<roycroft>the two are absolutely identical, other than the webserver each uses
[03:00:27]<roycroft>i can import the site into the staging platform
[03:00:38]<roycroft>but i cannot import it into the deployment platform
[03:00:43]<roycroft>this makes no sense to me
[03:01:53]<roycroft>am i overlooking something obvious?
[09:16:02]* zz_drakythe has joined #aegir
[09:16:46]* Github[m] has quit (*.net *.split)
[09:16:46]* drakythe has quit (*.net *.split)
[09:16:46]* dsnopek has quit (*.net *.split)
[09:16:46]* zz_drakythe is now known as drakythe
[09:25:05]* Github[m] has joined #aegir
[09:46:10]* ybabel has quit (Ping timeout: 265 seconds)