| [17:42:09] | * ybabel has joined #aegir |
| [17:54:41] | * ybabel has quit (Quit: ybabel) |
| [19:15:15] | * Shwele has joined #aegir |
| [21:34:03] | * ybabel has joined #aegir |
| [21:42:34] | * ybabel has quit (Quit: ybabel) |
| [22:48:06] | * Shwele has quit (Ping timeout: 260 seconds) |
| [23:28:57] | * shaneonabike has joined #aegir |
| [23:48:15] | * Shwele has joined #aegir |
| [01:15:00] | * ybabel has joined #aegir |
| [01:15:08] | * ybabel has quit (Client Quit) |
| [02:18:04] | <helmo42[m]> | Swede: http://ispconfig.org/ has an APS installer option |
| [02:18:40] | * theMusician has joined #aegir |
| [02:23:06] | <Shwele> | helmo42: thanks, will use it as backup plan, since colleagues are pushing me to go with virtualmin for now... tho if I burn that bridge will invest time in ispconfig since it does have mailboxes as well, which is crucial for current project. |
| [02:41:06] | <jonpugh> | hey helmo42[m] colan[m] ergonlogic[m] any other DNS "experts"... have a question. |
| [02:41:28] | <jonpugh> | I'm working on enabling HTTPS with lets encrypt in the devmaster.profile, I've got it working, but' |
| [02:41:42] | <jonpugh> | of course it will only really work if the site is public |
| [02:41:49] | <Shwele> | certbot or? |
| [02:41:50] | <helmo42[m]> | true |
| [02:42:06] | <jonpugh> | hosting lets encrypt uses dehydrated |
| [02:42:17] | <jonpugh> | So I think, can I detect a public hostmaster? |
| [02:42:19] | <jonpugh> | dns_get_record() |
| [02:42:39] | <jonpugh> | is it safe to say if there is an "A" record or a "CNAME" record that this is a public site? |
| [02:43:14] | <Shwele> | A record, CNAME needs to be binded to some name and name needs to be to IP. So A record Id say |
| [02:43:25] | <helmo42[m]> | jonpugh: I guess not ... could still be a local DNS |
| [02:43:45] | <jonpugh> | yeah... |
| [02:43:54] | <jonpugh> | but those people know what they are doing :D |
| [02:44:29] | <helmo42[m]> | what does dns_get_record return when you have an address in your local /etc/hosts file? |
| [02:45:14] | <jonpugh> | https://www.irccloud.com/pastebin/f2r79PAD/ |
| [02:45:20] | <jonpugh> | local ip |
| [02:45:29] | <jonpugh> | doh |
| [02:45:50] | <jonpugh> | yeah, it's gonna always be 127 |
| [02:46:00] | * shaneonabike has left #aegir () |
| [02:47:28] | <jonpugh> | oh wait, interesting... |
| [02:47:43] | <jonpugh> | it works on a real server... |
| [02:50:04] | <jonpugh> | helmo42[m]: interesting comparison... |
| [02:50:10] | <jonpugh> | https://www.irccloud.com/pastebin/DQVDFVYV/ |
| [02:50:31] | <jonpugh> | dns_get_record() works. gethostbyname() gets local etc hosts |
| [02:50:44] | <jonpugh> | I'll just double check that IP isn't 127 |
| [03:00:32] | * Shwele has quit (Quit: Leaving) |
| [03:01:21] | <colan[m]> | <freenode_jon "but those people know what they "> Does that mean I know what I'm doing? I do this all the time for private networking, say for DB servers. ;) |
| [03:02:22] | <colan[m]> | there's no point setting up your own DNS when you can just use the public one (even though not addressable). |
| [03:02:46] | <jonpugh> | that's what I mean |
| [03:02:57] | <colan[m]> | works very well with openvpn. |
| [03:02:59] | <jonpugh> | anyone using non-public dns for a devmaster instance would expect a little strange |
| [03:03:14] | <jonpugh> | but I think i got good logic |
| [03:03:26] | <jonpugh> | https://www.irccloud.com/pastebin/c7OSowVw/ |
| [03:03:47] | <jonpugh> | if there's an A record and a CNAME record and IP is not 127, enable HTTPS for the devmaster site |
| [03:04:26] | <jonpugh> | if they have special DNS, worst case is the site verify task fails because Letsencrypt won't work. That user in that situation could handle it. |
| [03:04:36] | <colan[m]> | yeah, but LE needs to be able to access the site. |
| [03:04:39] | <colan[m]> | It won't here. |
| [03:05:02] | <jonpugh> | We're just talking about defaults on install |
| [03:05:48] | <jonpugh> | users who block public access for their devmaster can just go in and change the server master settings to turn off letsencrypt |
| [03:06:00] | <jonpugh> | I'd rather ask those people to do an extra step than those using default setup :) |
| [04:44:59] | * CTassell has joined #aegir |
| [04:46:57] | <CTassell> | I'm having a problem with our Aegir setup. We have a web cluster (two backend servers) with a shared network drive. Every once in a while some of our sites will break with the error "The website encountered an unexpected error. Please try again later" Verifying the site or running "drush cc all" from the control node seems to fix it. This is a Drupal 8 site. Anyone know what's going on? |
| [04:54:13] | <colan[m]> | CTassell You need to look at the Web server logs. More error info will be in there. |
| [05:01:29] | <CTassell> | Hmm, there are tonnes of errors about uncaught exceptions, but this seems to be the one causing the issue: Serialization failure: 1213 Deadlock: wsrep aborted transaction: |
| [05:38:03] | <CTassell> | Can I add $databases['default']['default']['init_commands'] to the defaut.settings.php of the platform and then that will affect all the child sites? Or do I need to edit each individual settings.php? |
| [06:47:01] | <CTassell> | For reference, no, that doesn't work. YOu need to edit sites/<site>/local.settings.php for each site |
| [06:47:06] | * CTassell has quit (Quit: Ex-Chat) |
| [07:39:38] | <colan[m]> | there's actually global.inc, & a similar file for each platform.... |
| [07:40:34] | * Yaazkal has joined #aegir |
| [10:51:49] | * theMusician has quit (Quit: theMusician) |
