| [23:33:44] | <imb> | viashimo: Now I have Drush 8.4.5 and Drupal9 sites is working for me. I was able to succesfully migrate sites form D8 platforms to D9 platforms. |
| [23:37:10] | <imb> | I tried to migrate a D7 to D8 platform and I get this error "Drupal\Core\DependencyInjection\ContainerNotInitializedException: \Drupal::$container is not initialized yet. \Drupal::setContainer() must be called with a real container. in /var/aegir/platforms/d8dev/web/core/lib/Drupal.php:130", but I suppose to upgrade to D8 from D7 sites must be made with other procedures. |
| [23:39:38] | <imb> | Deleting sites seems to be failing (in 3.191). I get this error "Drush command terminated abnormally due to an unrecoverable error." and the site is marked with status "Failed". |
| [01:53:09] | <viashimo> | imb: afaik, aegir does not handle d7 -> d8, or d8 -> d9 migrations by itself. the d8 -> d9 might work because the code bases are close and the data doesn't require a migration, but the d7 is very different from d8 |
| [04:28:44] | <imb> | I'm confused about the best way to use certificates. Is it posible to have a server with Let's Encrypt sites and commercial certificates at the same time? The Legacy SSL Support alows me to create autosigned certificates wich I can replace with commercial certificates, but the "Legacy SSL support" and the new "Aegir HTTPS" can't be selected at the same time for one server. |
| [04:57:59] | <colan[m]> | imb: You're correct about the conflict. The legacy stuff shouldn't be needed *except* if you've got the situation you're describing (as that's the only thing missing from the new HTTPS stuff) If I were you, I'd either either work on the feature myself in the new system, or pay someone else to do it. Details: https://www.drupal.org/project/hosting_https/issues/2936037 |
| [04:58:00] | <hefring> | https://www.drupal.org/project/hosting_https/issues/2936037 => Add a 'manual' Certificate implementation [#2936037] => 2 comments, 4 IRC mentions |
| [04:58:44] | <colan[m]> | You could try running both, but I wouldn't recommend it, and I doubt anyone would support it. If you do, please let us know what happens. :) |
| [05:15:36] | <imb> | Thank you colan, but even in the case that "Manual certifate service" were developed, when you configure a server you can select just one service. I have activated the experimental https "Self-signed service" but this one and the "Let's encrypt" service can't be selected at the same time. I'm right? |
| [05:16:23] | <colan[m]> | I don't think anybody's ever tried that. Try it. |
| [05:16:57] | <colan[m]> | Logically, it shouldn't work, but I have no idea. Maybe you'll get lucky. |
| [05:26:02] | <bgm[m]> | I have some hacks in provision_symbiotic so that it's possible to use a wildcard cert, if available (should also work for a single-site cert): https://github.com/coopsymbiotic/provision_symbiotic/blob/master/tpl/cus... |
| [05:27:14] | <bgm[m]> | although... looking at the code, it's kind of weird that it's looking in /var/aegir/config/letsencrypt.d, and probably would not work with both enabled, but that can be fixed by looking another directory for certs. |
| [06:02:58] | <imb> | At the moment and as workaround I think that the legacy "SSL suport" fit better for us. We can have a wilcard certificates that managers can select when creating new sites and if some site requires a "let's encrypt" certificate we could use symbolic links to the certs previously generated by our "let's encrypt" cron task. In that way managers could select his preferred certificate when they deploy new sites. I |
| [06:03:04] | <imb> | think it will work. |
| [06:41:57] | <imb> | Anybody with errors when deleting a site? Should I open an issue? |
| [06:57:31] | * shaneonabike has joined #aegir |
| [10:23:06] | * shaneonabike has left #aegir () |
