| [11:00:24] | * Egyptian[Home] has joined #aegir |
| [11:01:36] | * ergonlog2c has joined #aegir |
| [11:04:42] | * ergonlogic has quit (Ping timeout: 250 seconds) |
| [11:04:50] | * ergonlog2c is now known as ergonlogic |
| [11:22:41] | * keke420 has quit (Remote host closed the connection) |
| [11:28:53] | * anarcat has quit (Ping timeout: 265 seconds) |
| [12:19:15] | * gandhiano has quit (Ping timeout: 260 seconds) |
| [12:31:34] | * gandhiano has joined #aegir |
| [13:01:50] | * Egyptian[Home] has quit (Read error: Connection reset by peer) |
| [13:01:57] | * Egyptian[Home] has joined #aegir |
| [13:08:50] | * gandhiano has quit (Ping timeout: 260 seconds) |
| [14:06:50] | * Egyptian[Home] has quit (Quit: Leaving.) |
| [14:07:06] | * Egyptian[Home] has joined #aegir |
| [14:27:14] | * Egyptian[Home] has quit (Ping timeout: 245 seconds) |
| [14:40:20] | * Yaazkal has quit () |
| [15:04:29] | * Egyptian[Home] has joined #aegir |
| [15:05:25] | * Egyptian[Home] has quit (Client Quit) |
| [15:05:36] | * Egyptian[Home] has joined #aegir |
| [15:15:38] | * fathima has joined #aegir |
| [15:23:38] | * fathima has quit (Ping timeout: 246 seconds) |
| [15:39:52] | * Egyptian[Home] has quit (Ping timeout: 250 seconds) |
| [18:41:00] | * rominronin has joined #aegir |
| [18:58:20] | * zz_drakythe has quit (Ping timeout: 250 seconds) |
| [18:58:46] | * kvanderw has quit (Ping timeout: 250 seconds) |
| [19:41:15] | * zz_drakythe has joined #aegir |
| [19:41:39] | * zz_drakythe is now known as drakythe |
| [19:41:45] | * kvanderw has joined #aegir |
| [19:55:44] | * Captain_Haddock has joined #aegir |
| [19:55:48] | * Captain_Haddock has quit (Changing host) |
| [19:55:48] | * Captain_Haddock has joined #aegir |
| [19:57:27] | * rominronin has quit (Remote host closed the connection) |
| [20:09:48] | * rominronin has joined #aegir |
| [20:31:18] | * David_Hernandez has joined #aegir |
| [20:48:05] | * kvanderw has quit (Ping timeout: 245 seconds) |
| [20:48:50] | * drakythe has quit (Ping timeout: 250 seconds) |
| [20:49:00] | * rominronin has quit (Remote host closed the connection) |
| [20:49:52] | * zz_drakythe has joined #aegir |
| [20:50:05] | * zz_drakythe is now known as drakythe |
| [20:50:47] | * rominronin has joined #aegir |
| [20:51:22] | * kvanderw has joined #aegir |
| [21:54:56] | * gandhiano has joined #aegir |
| [22:17:25] | * daften has joined #aegir |
| [22:30:55] | * gandhiano has quit (Ping timeout: 240 seconds) |
| [22:36:58] | * daften has quit (Remote host closed the connection) |
| [22:37:23] | * daften has joined #aegir |
| [22:41:42] | * daften has quit (Ping timeout: 255 seconds) |
| [22:54:59] | * daften has joined #aegir |
| [22:59:42] | * daften has quit (Ping timeout: 250 seconds) |
| [23:18:52] | * daften has joined #aegir |
| [23:50:01] | * rominronin has quit (Remote host closed the connection) |
| [23:55:22] | * daften has quit () |
| [23:57:08] | * gandhiano has joined #aegir |
| [23:57:21] | * rominronin has joined #aegir |
| [23:57:53] | * rominronin has quit (Client Quit) |
| [00:03:12] | * noecc has joined #aegir |
| [00:10:26] | * gandhiano has quit (Read error: Connection reset by peer) |
| [00:14:56] | * Yaazkal has joined #aegir |
| [00:18:38] | * David_Hernandez has quit (Quit: :wq!) |
| [00:18:58] | * David_Hernandez has joined #aegir |
| [00:19:23] | * David_Hernandez has quit (Remote host closed the connection) |
| [00:19:47] | * David_Hernandez has joined #aegir |
| [00:21:34] | * David_Hernandez has quit (Client Quit) |
| [00:25:00] | * gandhiano has joined #aegir |
| [00:27:57] | * zombiebeard has joined #aegir |
| [00:48:29] | * keke420 has joined #aegir |
| [00:49:51] | * daften has joined #aegir |
| [01:26:00] | * Egyptian[Home] has joined #aegir |
| [01:31:18] | * gandhiano has quit (Ping timeout: 256 seconds) |
| [01:33:49] | * gandhiano has joined #aegir |
| [01:40:02] | * gandhiano has quit (Ping timeout: 250 seconds) |
| [01:44:09] | * gandhiano has joined #aegir |
| [02:11:28] | * ergonlogic has quit (Quit: leaving) |
| [02:11:54] | * ergonlog1c is now known as ergonlogic |
| [02:12:19] | <ergonlogic> | bgm: happy holidays :) |
| [02:12:45] | <ergonlogic> | I seem to recall you pinging me about something, but I've been swamped lately, and I don't recall whether I'd followed up with you |
| [02:13:34] | <ergonlogic> | also, I think I've narrowed down the CiviCRM temp files path issue that NDI has been seeing |
| [02:13:46] | <bgm> | ergonlogic: hmm, not that I recall. I updated the civicrm ticket you had opened for i18n |
| [02:14:12] | <ergonlogic> | ok, I'll need to catch up on that then |
| [02:14:39] | <ergonlogic> | as for the temp path thing, I documented the work-around in some depth in their Jira issue UACASE-51 |
| [02:15:25] | <bgm> | is that on i.c.o or an ndi jira? |
| [02:15:50] | <bgm> | I don't think i'm aware of that issue (and I don't have access to their jira) |
| [02:15:50] | <ergonlogic> | ndi jira |
| [02:15:55] | <ergonlogic> | ok |
| [02:16:01] | <ergonlogic> | it's the iddue here: https://www.drupal.org/node/2571101 |
| [02:16:02] | <hefring> | https://www.drupal.org/node/2571101 => Update some variables in Config.IDS.ini [#2571101] => 5 comments, 1 IRC mention |
| [02:16:11] | <ergonlogic> | I'll paste the relevant stuff there |
| [02:16:45] | <bgm> | oh ok |
| [02:18:29] | * Egyptian[Home] has quit (Remote host closed the connection) |
| [02:18:39] | <ergonlogic> | done |
| [02:19:08] | <ergonlogic> | I'm not sure if that's something that ought to be in hosting_civicrm or civicrm.drush.inc |
| [02:20:27] | <bgm> | ergonlogic: i'm just confused as to why you would edit the Config.IDS.ini, rather than just delete it. |
| [02:20:43] | <bgm> | "there's a setting that specifies the path for temporary files" => in Config.IDS.ini? |
| [02:21:35] | <bgm> | just the clarify, the IDS is a library that detects sql injections and other common attacks |
| [02:24:08] | <ergonlogic> | the site can be inaccessible due to this |
| [02:24:20] | <ergonlogic> | and the tmp_path is stored in the db |
| [02:24:38] | <ergonlogic> | so re-generating the Config.IDS.ini just puts back the same path |
| [02:25:23] | <ergonlogic> | I don't doubt that there's a better way to do this, btw |
| [02:25:32] | <ergonlogic> | but this works |
| [02:25:51] | <bgm> | hmm ok |
| [02:25:57] | <bgm> | the civi paths/settings have terribly confusing names |
| [02:26:02] | <bgm> | but it's possible to override them in PHP |
| [02:26:08] | <bgm> | similarly to how drupal does |
| [02:26:12] | <ergonlogic> | right |
| [02:26:53] | <ergonlogic> | I'd think we could just update the tmp_path var somewhere in a deploy hook |
| [02:27:03] | <bgm> | i've been thinking for a while to always declare them in PHP.. I often declared them in a local.settings.php in a way that's migration-friendly |
| [02:27:26] | <ergonlogic> | well, these are absolute paths, aren't they? |
| [02:27:35] | <ergonlogic> | at least in Config.IDS.ini, they are |
| [02:27:47] | <ergonlogic> | so changing platforms would break it |
| [02:27:52] | <ergonlogic> | as would clones |
| [02:28:32] | <ergonlogic> | at this point, is civicrm.drush.inc generating the civicrm.settings.php? |
| [02:28:35] | <bgm> | hmm, right, possibly.. I think there was a work-around, but then the variables change between drush vs web |
| [02:28:39] | <bgm> | yep |
| [02:28:47] | <ergonlogic> | ok |
| [02:29:09] | <ergonlogic> | we'd talked about using the Aegir templating system to generate it instead |
| [02:29:24] | <ergonlogic> | that'd be a convenient way to hardcode those vars |
| [02:29:34] | <bgm> | the civicrm.settings.php changes rather often :/ |
| [02:29:42] | <ergonlogic> | oh? |
| [02:29:52] | <bgm> | usually not in a way that breaks upgrades, but they add stuff regularly between versions |
| [02:29:57] | <ergonlogic> | outside of Aegir-like operations? |
| [02:30:04] | <ergonlogic> | ah |
| [02:30:17] | <bgm> | no no, just saying I don't want to spend time merging changes after each new civi release |
| [02:30:24] | <ergonlogic> | ok |
| [02:30:50] | <ergonlogic> | that was part of the reason to ship our own civicrm.drush.inc, at one point, right? |
| [02:31:21] | <bgm> | we had "forked" civicrm.drush.inc, because the core was wasn't very aegir-friendly |
| [02:31:31] | <bgm> | but it's not necessary anymore, and it makes it easier to support multiple versions of civi |
| [02:31:41] | <bgm> | without having "if version > 4.x.x" etc |
| [02:32:11] | <ergonlogic> | maybe we could inject an include somewhere |
| [02:32:30] | <ergonlogic> | right above the functional stuff at the bottom |
| [02:33:25] | <bgm> | similar to local.settings.php ? |
| [02:33:37] | <ergonlogic> | right, but generated by Aegir |
| [02:33:48] | <bgm> | would be good to have it at the beginning of civicrm.settings.php, this way we could easily override the constants |
| [02:34:27] | <ergonlogic> | at the beginning? |
| [02:34:40] | <ergonlogic> | let me refresh my memory on what it looks like |
| [02:35:26] | * gusaus has joined #aegir |
| [02:35:28] | <bgm> | i guess there are 3 types of configurations in that file.. |
| [02:36:02] | <bgm> | the constants, the $civicrm_root global, and settings overrides (which are all commented out by default) |
| [02:36:20] | <ergonlogic> | ok, I see |
| [02:36:41] | <ergonlogic> | does setting.php get parsed first? |
| [02:36:49] | <bgm> | yep |
| [02:37:06] | <ergonlogic> | so we could just add it via the settings hook |
| [02:37:20] | <bgm> | (I guess there can be some rare cases where civicrm will bootstrap drupal, but we usually call it using Drush, ex: crons) |
| [02:39:04] | <ergonlogic> | $civicrm_root is the only thing we can't override, atm, then |
| [02:39:18] | <ergonlogic> | oh, I see |
| [02:39:23] | <bgm> | that's generated using an aegir/drush variable, so not a big deal |
| [02:39:57] | <ergonlogic> | what would happen in an API call to Civi? |
| [02:40:14] | <ergonlogic> | is drupal/settings.php bootstrapped? |
| [02:40:32] | <ergonlogic> | I guess it must be, since it relies on Drupal for users, etc. right? |
| [02:40:47] | <bgm> | yep |
| [02:42:18] | <bgm> | although now i'm doubting about the URL tracking thing for civimail.. |
| [02:42:39] | <bgm> | things that don't get called from the CSM are usually in civicrm/extern/* |
| [02:43:13] | <ergonlogic> | // Override the Temporary Files directory. // $civicrm_setting['Directory Preferences']['customFileUploadDir'] = '/path/to/upload'; |
| [02:43:32] | <ergonlogic> | so, presumably setting that would resolve the specific issue here |
| [02:43:38] | <bgm> | yep |
| [02:43:53] | <bgm> | but .. I checked, and the URL tracker is indeed that one in extern/url.php |
| [02:43:58] | <ergonlogic> | I'll post a link to this discussion in the drupal.org issue, so I can follow-up with a fix |
| [02:44:01] | <ergonlogic> | ok |
| [02:44:05] | <bgm> | so I'd probably avoid putting things only in drupal confs, not called from civicrm.settings.php |
| [02:44:14] | <bgm> | k, thanks :) |
| [02:44:25] | <ergonlogic> | hefring: log pointer |
| [02:44:25] | <hefring> | http://hefring.mig5.net/bot/log/aegir/2015-12-23#T603423 |
| [02:45:44] | <ergonlogic> | hmm, ok |
| [02:45:57] | <ergonlogic> | so what would you suggest for exterm/*? |
| [02:46:04] | <ergonlogic> | also, https://www.drupal.org/node/2378175 |
| [02:46:05] | <hefring> | https://www.drupal.org/node/2378175 => Run file operations under sudo [#2378175] => 5 comments, 2 IRC mentions |
| [02:46:29] | * gandhiano has quit (Ping timeout: 255 seconds) |
| [02:46:30] | <ergonlogic> | the recurring file permissions issues with Civi under Aegir could be resolved by that ^^^ |
| [02:47:02] | <bgm> | yeah, that would be nice |
| [02:47:44] | <bgm> | I don't get the fuss about the risk of privilege escalation |
| [02:48:22] | * gandhiano has joined #aegir |
| [02:48:23] | <ergonlogic> | well, we can wrap is securely |
| [02:48:51] | <ergonlogic> | we don't want www-data to be able to chmod 777 just anything |
| [02:49:03] | <ergonlogic> | but still, shouldn't be too difficult |
| [02:49:21] | <bgm> | yeah, just make a shell script that can be run by the aegir user only, and without any arguments |
| [02:49:30] | <ergonlogic> | and itd be optional with a sudoers entry that'd have to be configured |
| [02:49:31] | <bgm> | well, erm, just the destination path |
| [02:49:34] | <bgm> | "fix this directory" |
| [02:50:12] | <ergonlogic> | oh, good idea |
| [02:50:13] | <bgm> | (and the script would validate that it's a valid site directory, etc) |
| [02:50:27] | <ergonlogic> | if the aegir user can't cd into the dir, they can't run those ops |
| [02:51:12] | <ergonlogic> | right, and we'd presumably call it using the site/platform alias |
| [02:51:35] | <ergonlogic> | where the site or platform root is specified anyway |
| [02:51:50] | <bgm> | makes sense |
| [02:53:09] | * bgm stuck on an icinga2 monitoring issue, grumble grubmle |
| [02:53:30] | <ergonlogic> | so, to inject settings that'll be accessible to extern/* scripts... |
| [02:53:35] | <ergonlogic> | any ideas there? |
| [02:54:08] | <bgm> | probably not the cleanest thing to do, but how about appending variables to the civicrm.settings.php directly? |
| [02:54:25] | * ergonlogic has to tackle a DKAN JS graphing issue next :-/ |
| [02:54:33] | <bgm> | :) |
| [02:54:35] | <ergonlogic> | prepending? |
| [02:55:24] | <ergonlogic> | wouldn't they have to be above the "Do not change anything below this line. Keep as is" |
| [02:55:29] | <ergonlogic> | maybe not |
| [02:55:51] | <bgm> | nah, it's not a big deal |
| [02:56:02] | <bgm> | civi anyway calls hook_civicrm_config() afterwards |
| [02:56:07] | <bgm> | which is another way to change settings |
| [02:56:36] | <ergonlogic> | ah, interesting |
| [02:57:29] | <ergonlogic> | that'd have to be in the site code-base though, right? |
| [02:58:11] | <ergonlogic> | so we'd have to ship that at the platform level or something |
| [02:58:20] | <bgm> | yeah |
| [02:58:30] | <bgm> | civi can have global extensions in ~/platform/civi-foo/vendor/ |
| [02:58:40] | <bgm> | but it would still need to be enabled, updated, etc |
| [02:58:46] | <ergonlogic> | right |
| [02:58:49] | <ergonlogic> | seems messy |
| [02:58:52] | <bgm> | yep |
| [02:59:59] | <bgm> | also, we could ask to add a smarty variable in the civicrm.settings.php to have a clean way to inject stuff in there |
| [03:00:06] | <bgm> | like a pre/post kind of variable |
| [03:00:25] | <bgm> | but the civicrm.drush.inc code is messy, and it sounds time-consuming.. |
| [03:01:35] | <ergonlogic> | that'd certainly make injecting custom stuff easier |
| [03:01:53] | <ergonlogic> | I'm looking in extern/widget.php |
| [03:02:48] | <ergonlogic> | which requires civicrm.config.php |
| [03:03:02] | <bgm> | civicrm.config.php finds civicrm.settings.php |
| [03:03:16] | <bgm> | it's messy, but it deals with detecting the CMS and all that |
| [03:03:53] | <ergonlogic> | right |
| [03:04:04] | <ergonlogic> | just checking what all it's doing there |
| [03:04:20] | <ergonlogic> | so there it'll only bootstrap the civi settings |
| [03:04:53] | <ergonlogic> | what if a widget needs something from the Drupal side? |
| [03:05:01] | <ergonlogic> | user perms or something |
| [03:05:04] | <bgm> | in some places, it's hard to say, but yeah, I think it's safer to assume that sometimes Drupal doesn't get bootstrapped |
| [03:05:26] | <bgm> | extern/widget.php is kind of crappy, and no one uses it :) |
| [03:05:33] | <ergonlogic> | hmm, ok |
| [03:05:33] | <bgm> | the only important thing there is extern/url.php |
| [03:05:43] | <ergonlogic> | ok, I'll take a look at that then |
| [03:05:58] | <bgm> | although some of the IPN code in there could be important for paypal ipns |
| [03:06:07] | <bgm> | normally that stuff should have been migrated to something more proper |
| [03:09:20] | <ergonlogic> | ok, so there might be some corner cases |
| [03:09:48] | <ergonlogic> | but re-setting temp path var during deploy would resolve the immediate issue |
| [03:10:22] | <ergonlogic> | also "$this->userFrameworkUsersTableName = $dbprefix . 'users';" |
| [03:10:42] | <ergonlogic> | that looks like how users table is bootstrapped from various CMSs |
| [03:11:17] | <bgm> | where do you see that? |
| [03:11:29] | <bgm> | but yes, I'm OK with fixing the immediate issue and keep it simple :) |
| [03:12:02] | <ergonlogic> | vi CRM/Core/Config.php +331 |
| [03:12:29] | <ergonlogic> | so, right... injecting into settings.php would work in most cases |
| [03:12:56] | <ergonlogic> | but some stuff in extern/* might break, if we aren't careful |
| [03:13:32] | <ergonlogic> | ideally the CMS would be bootstrapped in extern/* |
| [03:13:47] | <ergonlogic> | except that it'd presumably make things much slower |
| [03:13:57] | <ergonlogic> | as there'd be a lot more code loaded |
| [03:14:25] | <ergonlogic> | pre- and post- includes in civicrm.settings.php would help |
| [03:14:36] | <ergonlogic> | but then we'd need to generate those, as required |
| [03:15:20] | <ergonlogic> | bgm: does that sum it up well? |
| [03:15:24] | <bgm> | yep |
| [03:16:07] | <ergonlogic> | if time permits, would you mind posting an issue to civicrm for that? |
| [03:16:40] | <ergonlogic> | the pre- and post- hooks, that is |
| [03:17:23] | <ergonlogic> | I'm not sure I speak enough civicrm-ese to document the requirement properly :) |
| [03:18:37] | <bgm> | sure.. although I'll open the issue when ready to work on it |
| [03:19:00] | <bgm> | i'm probably going to a sprint in end-january where we'll work on the d8 & backport ports |
| [03:19:23] | <bgm> | so it'll be a good time to add those changes |
| [03:19:40] | <ergonlogic> | ok, great |
| [03:19:48] | <bgm> | (fwiw: https://civicrm.org/blogs/colemanw/sprint-for-civi-drupal8-backdrop-wp-j...) |
| [03:19:50] | <ergonlogic> | let me know, and I'll try to set aside some time to help |
| [03:20:17] | <ergonlogic> | d8 support should help with i18n some too, I'd think |
| [03:20:29] | <ergonlogic> | well, maybe not |
| [03:21:03] | <ergonlogic> | ok, I think I have enough to fix the tmp_path thing |
| [03:21:15] | <ergonlogic> | so I'll try to tackle that in early Jan |
| [03:21:41] | <bgm> | :) |
| [03:30:01] | <bgm> | ot: icinga2 is mind-blowingly simple and awesome, but.. you-just-need-to-unlearn-all-those-bad-nagios-habits |
| [03:52:19] | * daften has quit (Remote host closed the connection) |
| [03:52:46] | * daften has joined #aegir |
| [03:55:10] | * gandhiano has quit (Ping timeout: 245 seconds) |
| [03:57:08] | * daften has quit (Ping timeout: 255 seconds) |
| [03:58:11] | * gandhiano has joined #aegir |
| [04:09:53] | <ergonlogic> | neat |
| [04:10:19] | <ergonlogic> | I'd been working with shinken, which re-implements nagios in python |
| [04:10:33] | <ergonlogic> | allowing re-use of nagios plugins |
| [04:43:20] | * gandhiano has quit (Ping timeout: 260 seconds) |
| [05:22:15] | * Egyptian[Home] has joined #aegir |
| [05:42:40] | * Egyptian[Home] has quit (Ping timeout: 265 seconds) |
| [06:22:01] | * kvanderw is now known as zz_kvanderw |
| [06:22:47] | * zz_kvanderw is now known as kvanderw |
| [06:24:19] | * gusaus has quit (Quit: gusaus) |
| [06:24:37] | * daften has joined #aegir |
| [06:51:00] | * Egyptian[Home] has joined #aegir |
| [06:52:55] | * noecc has left #aegir ("pax") |
| [07:06:45] | * Captain_Haddock has quit () |
| [07:36:09] | * gusaus has joined #aegir |
| [08:14:41] | * daften has quit () |
| [08:15:36] | * sierkje has joined #aegir |
| [08:20:26] | * Egyptian[Home] has quit (Ping timeout: 240 seconds) |
| [08:22:27] | * sierkje has left #aegir () |
| [09:13:03] | * kvanderw is now known as zz_kvanderw |
| [09:13:19] | * zz_kvanderw is now known as kvanderw |
| [10:02:14] | * zombiebeard has quit (Quit: zombiebeard) |
| [10:37:19] | * gandhiano has joined #aegir |
| [10:38:41] | * drakythe is now known as zz_drakythe |
| [10:38:46] | * zz_drakythe is now known as drakythe |
| [10:39:35] | * drakythe is now known as zz_drakythe |
| [10:40:36] | * zz_drakythe is now known as drakythe |
| [10:41:21] | * drakythe is now known as zz_drakythe |
| [10:53:07] | * Egyptian[Home] has joined #aegir |
