| [11:06:31] | <adrinux> | jlkinsel: the difference is that one can be automated and the other requires a human to sit there clicking on things |
| [11:07:50] | <jlkinsel> | er...let me clarify. I'm talking about running drush provision-save, then drush @hostmaster hosting-import (for a platform or a site) |
| [11:11:59] | <adrinux> | ah. sorry, I'm tired :) |
| [11:12:36] | <jlkinsel> | tgif. :) |
| [11:12:58] | <adrinux> | it's Saturday already :( |
| [11:13:07] | <jlkinsel> | that's not my fault. ;) |
| [11:13:13] | <adrinux> | lol |
| [11:13:30] | <adrinux> | goodnight |
| [11:13:32] | * adrinux has left #aegir () |
| [11:19:59] | * mrchrisadams_ has joined #aegir |
| [11:19:59] | * mrchrisadams has quit (Read error: Connection reset by peer) |
| [11:20:00] | * mrchrisadams_ is now known as mrchrisadams |
| [11:21:30] | <mig5> | jlkinsel: there isn't, really |
| [11:21:43] | <mig5> | if you are referring to this http://lists.aegirproject.org/pipermail/aegir/2011-January/000026.html |
| [11:21:53] | <jlkinsel> | yea |
| [11:21:54] | <mig5> | i run a verify of the platform (that has already been imported into aegir) after i have installed a site |
| [11:22:00] | <mig5> | which kicks off an import of the *site* |
| [11:22:14] | <mig5> | you could do the install of the site before doing a hosting-import of the platform perhaps |
| [11:22:19] | <mig5> | and it'll probably take care of that step |
| [11:22:25] | <mig5> | i did it in a different order coz i'm a dumbarse |
| [11:22:58] | <mig5> | so do: provision-save *platform; provision-install *site*; hosting-import *platform |
| [11:22:59] | <eft> | can't even spell dumb arse |
| [11:23:04] | <mig5> | probably smarter way |
| [11:23:26] | <mig5> | er, probably need a provision-save *site* after provision-save *platform* |
| [11:23:33] | <mig5> | then install, then import |
| [11:23:35] | <eft> | Happy Saturday mig5 :) |
| [11:23:55] | <jlkinsel> | yeah |
| [11:24:15] | <jlkinsel> | I'm trying an import of the platform, then save the site, then import site, but same basic idea |
| [11:24:58] | <mig5> | yeah well since a verify of a platform will import any new sites, easier to save that step and have the site ready on the platform before you import it |
| [11:25:07] | <mig5> | i should've thought of that :) |
| [11:25:19] | <mig5> | but thanks for bringing it up - since i want to write some docs on the community site re: hosting-import |
| [11:25:54] | <mig5> | and also working on a git post-receive hook that auto-builds a new platform with the updated code and does a test site in stall |
| [11:27:21] | <jlkinsel> | I'm actually thinking for new sites...after provisioning/importing the platform, doing the save/import will let aegir go ahead and create the site (I'm looking to get virtualmin to do this...) |
| [11:27:40] | <mig5> | that's what i mean |
| [11:27:44] | <mig5> | i think we're talking about the same thing |
| [11:27:48] | <jlkinsel> | 'k. :) |
| [11:28:04] | <mig5> | the save/import of the platform will import the site if the site is created on the platform *before* the platform is imported into aegir |
| [11:28:25] | <mig5> | as you said because of the implied provision-verify of the platform |
| [11:28:48] | <jlkinsel> | btw, feature idea - option to notify client when a site is migrated to a new platform, in case they need to access files at a different directory |
| [11:28:58] | <mig5> | yeah i think we have a ticket about that |
| [11:29:04] | <mig5> | let me find it |
| [11:29:22] | * Artusamak_afk is now known as Artusamak |
| [11:29:27] | <mig5> | http://drupal.org/node/710834 |
| [11:29:40] | <mig5> | looks like i never finished it :) |
| [11:29:41] | <mig5> | oops |
| [11:29:52] | * mrchrisadams has quit (Read error: Connection reset by peer) |
| [11:29:56] | <jlkinsel> | ah, there ya go |
| [11:29:59] | * mrchrisadams has joined #aegir |
| [11:30:03] | <mig5> | like most aegir stuff, i think it was one of those things that sounded really easy until you start working on it |
| [11:30:16] | <mig5> | and then it turns into this gigantic clusterfuck that requires a refactoring of *everything* that was ever written |
| [11:30:19] | <mig5> | and so you give up :) |
| [11:30:32] | <mig5> | sigh |
| [11:30:49] | <jlkinsel> | well, you're still makin awesome progress. :) |
| [11:33:11] | <mig5> | speaking of client notifications, i want to tackle this one at some point too http://drupal.org/node/1014142 |
| [11:33:22] | <mig5> | i'm glad omega8cc mentioned the signup form (sort of :) ) |
| [11:33:28] | <mig5> | because i forgot about that, but yes that makes that hard |
| [11:34:37] | <jlkinsel> | it's a hack, but shortcut way to do that - set the initial email to an invalid value that'll bounce, then reset it after that |
| [11:38:35] | * realityloop has joined #aegir |
| [11:39:32] | <omega8cc> | mig5: yep, that could be probably some "multi-switch" but maybe it could be enough to introduce simple global switch for those interested and still use by default the previous behavior? |
| [11:39:46] | * Artusamak is now known as Artusamak_afk |
| [11:42:09] | <mig5> | mmm |
| [11:42:37] | <mig5> | trying to think if there's a way we can detect in the backend (provision) if the new site was requested via the signup form |
| [11:42:47] | <mig5> | wondering if i can set a drush option from the signup form, dunno |
| [11:43:57] | <mig5> | i think this is the first time i've ever looked at hosting_signup.module |
| [11:43:57] | <mig5> | wow |
| [11:44:06] | <mig5> | there are some parts of aegir i've just never had to go! |
| [11:44:25] | <jlkinsel> | lol |
| [11:45:45] | <mig5> | bugger, so we just save the site node |
| [11:45:50] | <mig5> | and the nodeapi in hosting_site kicks off install |
| [11:46:00] | * psynaptic|afk has quit (Read error: Connection reset by peer) |
| [11:46:28] | * psynaptic|afk has joined #aegir |
| [11:47:41] | <maria_zk> | omega8cc: is barracuda install boost - ready? could you please hint on nginx configuration? |
| [11:51:02] | <omega8cc> | mig5: also client is created, but not sure if that helps |
| [11:52:02] | <omega8cc> | maria_zk: yes, it is explained in the readme and in the welcome e-mail sent for every Octopus instance |
| [11:52:47] | <omega8cc> | maria_zk: it works out of the box, simply enable boost in any site |
| [11:52:53] | <mig5> | yeah it doesn't really help since we can't set parameters there if we are only creating nodes (there's no drush.inc stuff for the signup method, for example) |
| [11:53:03] | <mig5> | looks too hard to do it that way, oh well |
| [11:53:07] | <omega8cc> | ah, right |
| [11:53:23] | <mig5> | we'd have to try and detect if the request to sve the node came via the signup form but that feels hacky |
| [11:53:36] | <omega8cc> | mig5: so go with simple global switch maybe |
| [11:53:52] | <mig5> | yeah, or i wonder if we can set it per client |
| [11:53:54] | <omega8cc> | and default to on |
| [11:53:57] | <mig5> | yeah |
| [11:54:17] | <omega8cc> | not worth the trouble, I think |
| [11:54:52] | <mig5> | if there's a global swithc and default to on, and oyu want to suppress, you'd have to turn it off, install a site, then turn it back on so signup form method can work (and let's not even think about simultaneous signup vs adding a site as admin) |
| [11:55:08] | <mig5> | so that doesn't seem ideal either |
| [11:55:17] | <mig5> | so i wonder if it defaults to on, but on each client node you can check 'suppress' |
| [11:55:31] | <mig5> | and since it's on by default, and signup form adds a client node, that will always be on (unless you turn it off for that client later) |
| [11:55:34] | <mig5> | dunno |
| [11:55:56] | <omega8cc> | sounds already confusing :) |
| [11:56:00] | <mig5> | :) |
| [11:56:24] | <maria_zk> | omega8cc: it says that it is included, and that should be enabled... |
| [11:56:38] | <maria_zk> | I cannot find any instructions for the configuration though :( |
| [11:57:38] | <omega8cc> | maria_zk: it is a recommendation, not a requirement, also you only need to configure the time to purge the boost cache in the boost settings |
| [11:57:56] | <omega8cc> | the rest is already configured for you |
| [11:58:01] | <omega8cc> | including nginx |
| [11:58:36] | <maria_zk> | omega8cc: so it should play out of the box. |
| [11:58:47] | <omega8cc> | you can find configuration options as a sub-page of performance settings |
| [11:58:50] | <omega8cc> | yes |
| [11:58:54] | <maria_zk> | amazing |
| [11:59:12] | <maria_zk> | and having the cache module, there is no need for memcache / cacherouter, correct? |
| [12:00:51] | <omega8cc> | yes, that is all in one |
| [12:01:11] | <omega8cc> | this things started from the boost queue where we worked on the boost compatible nginx config, so it worked one year before Barracuda was born |
| [12:01:28] | * ryanarmstrong has quit (Quit: Leaving.) |
| [12:01:30] | <maria_zk> | omega8cc: it sounds like a *lot* of work was behind this |
| [12:02:06] | <omega8cc> | a bit, for sure, testing, testing, testing etc |
| [12:03:07] | * erutan has quit (Quit: Leaving...) |
| [12:03:09] | <maria_zk> | so i just enable the crawler, and I am good to go |
| [12:03:29] | <maria_zk> | nginx doesn't use htacees, so no copy paste is required there |
| [12:05:10] | <omega8cc> | crawler id denied, it is good only for small and/or not busy sites, otherwise it is too easy to kill the server, but you can turn it on by commenting out one line in the global.inc |
| [12:05:35] | <omega8cc> | nothing to copy/paste |
| [12:05:47] | <maria_zk> | thank you :) |
| [12:05:49] | <omega8cc> | it is already built in the nginx config |
| [12:06:34] | * mrchrisadams has quit (Quit: mrchrisadams) |
| [12:06:48] | <omega8cc> | some people are saying we made it too easy, lol |
| [12:07:00] | <maria_zk> | omega8cc: I still can't believe it is so easy |
| [12:07:12] | <maria_zk> | there *must* be something extremely complicated somewhere |
| [12:07:18] | <omega8cc> | yeah, that is the main "problem" |
| [12:07:32] | <omega8cc> | people expect hard things |
| [12:07:39] | <omega8cc> | and there is none |
| [12:08:09] | * kvanderw has quit (Quit: kvanderw) |
| [12:08:10] | <jlkinsel> | I have a few minutes - I might be able to make it more complicated for you if you'd like, maria_zk . :) |
| [12:08:23] | <omega8cc> | lol |
| [12:08:29] | <maria_zk> | jlkinsel: lol :) |
| [12:09:00] | <maria_zk> | that's fine, my memory leaks on my openVZ keep me just confused enough |
| [12:09:33] | <maria_zk> | ok, so.. This page is being served live to anonymous visitors, as it is not currently in the static page cache. |
| [12:09:39] | <maria_zk> | hmmm.... |
| [12:10:51] | * nicholasalipaz has left #aegir () |
| [12:11:38] | <omega8cc> | that is correct |
| [12:12:08] | <omega8cc> | you must be logged out to make that page cached |
| [12:13:02] | <omega8cc> | only anonymous visitors can fire up boost, fill cache and re-use it |
| [12:14:45] | <maria_zk> | aha! |
| [12:15:07] | <maria_zk> | so this is now served as static html |
| [12:16:10] | <omega8cc> | yes, but you can still enable page views counter |
| [12:16:26] | <omega8cc> | and it will use some ajax and extra php file |
| [12:16:49] | <maria_zk> | and it will also include the js |
| [12:16:52] | <maria_zk> | right? |
| [12:17:17] | <omega8cc> | yes, just activate correct block, as explained in the boost manual |
| [12:17:40] | <maria_zk> | omega8cc: your help is priceless |
| [12:18:34] | <omega8cc> | see: http://drupal.org/node/545908 and "Stats Block" at the bottom |
| [12:19:41] | <omega8cc> | and the boost_stats.php file is alraedy in place, in every 6.x platform created by Octopus |
| [12:21:22] | <omega8cc> | now trying to install latest Open Publish on 512 Mb linode, it takes ages |
| [12:21:45] | <maria_zk> | great. so nothing to do there as well |
| [12:22:01] | <maria_zk> | hehe! I am doing exactly the same thing :) |
| [12:22:10] | <maria_zk> | I wasn't sure if the installer got stuck or st |
| [12:22:20] | <maria_zk> | so it does take ages. good to know |
| [12:22:57] | <omega8cc> | you need at least 1-2 GB of RAM and it will still take 10 minutes or even can fail |
| [12:23:28] | <maria_zk> | why is it so greedy? |
| [12:24:07] | <maria_zk> | omega8cc: ok, confused: javascript is not working for frontpage... |
| [12:24:09] | <omega8cc> | *heavy* distro, it is like mega-elephant |
| [12:24:13] | <maria_zk> | all blocks are enabled |
| [12:25:28] | <omega8cc> | did you enable also Cache Statistics Block option in the block settings? |
| [12:26:34] | <omega8cc> | and the Boost: AJAX core statistics block must be displayed (invisible) on all pages |
| [12:26:56] | <maria_zk> | cache statistics block... |
| [12:27:18] | <maria_zk> | no such block :( |
| [12:27:41] | <maria_zk> | Boost: Pages cache status | Boost: Pages cache configuration | Boost: Expiration Relationship |Boost: AJAX core statistics |
| [12:27:43] | <omega8cc> | it is not a block, it is an option you need to enable |
| [12:27:44] | <maria_zk> | that's all |
| [12:28:12] | <omega8cc> | in the block Boost: AJAX core statistics - in its settings |
| [12:29:40] | <maria_zk> | ok i enabled it |
| [12:29:52] | <maria_zk> | so js was not working because of this? |
| [12:31:11] | <omega8cc> | probably, I hope you JS enabled in the browser (sorry!) :) |
| [12:31:25] | <omega8cc> | hou have* |
| [12:31:38] | <omega8cc> | it should work now |
| [12:32:24] | <omega8cc> | of course you must have enabled page views counter in the stats settings etc |
| [12:33:16] | <maria_zk> | mmm obvisouly i need to do some serious reading on boost :) |
| [12:34:35] | <omega8cc> | so, as expected, the OP site has been created, but the task failed to complete and report the result in the front end, so I have to delete the task first and re-verify OP platform to discover created site |
| [12:35:24] | * ezra-g has quit (Quit: Mastering Drupal Chapters: http://masteringdrupal.com/shop) |
| [12:35:30] | <maria_zk> | well... mine is still loading. but I see where this is going... |
| [12:35:31] | <omega8cc> | maria_zk: as with everything, once you have it done the first time, everything is easy |
| [12:35:47] | <omega8cc> | check the ps axf |
| [12:35:54] | <omega8cc> | if the process really runs |
| [12:35:58] | <omega8cc> | php |
| [12:36:07] | <maria_zk> | omega8cc: say that again... you could see the huge question mark over my head when i first heard about barracuda |
| [12:36:19] | <omega8cc> | haha |
| [12:36:42] | <omega8cc> | yeah, so check if the site is created and the php still runs or not |
| [12:36:50] | <maria_zk> | what process am i looking for? |
| [12:36:52] | <omega8cc> | if not, do the same (as above) |
| [12:37:16] | <omega8cc> | php with provision-blahblah |
| [12:37:47] | <omega8cc> | eating 100% of your CPU so easy to find |
| [12:38:06] | <omega8cc> | type: htop |
| [12:38:13] | <maria_zk> | my cpu is idle :( |
| [12:38:32] | <omega8cc> | ok, so you need to do the same |
| [12:38:55] | <omega8cc> | OP really needs loooong timeouts set for php and mysql |
| [12:39:28] | <maria_zk> | aha |
| [12:39:45] | <omega8cc> | in Aegir goto admin/content/node |
| [12:39:53] | <omega8cc> | and delete the install task node |
| [12:40:12] | <maria_zk> | so in aegir, not octopus install... |
| [12:40:29] | <omega8cc> | the same |
| [12:40:38] | <maria_zk> | is it? |
| [12:40:58] | <omega8cc> | octopus is the install script and Aegir is.. Aegir :) |
| [12:41:19] | <maria_zk> | :) |
| [12:41:27] | <omega8cc> | in your hostmaster site |
| [12:41:30] | <maria_zk> | i have aegir.example.com and octopus.exampl.com |
| [12:41:46] | <omega8cc> | so that one where you created OP site |
| [12:42:12] | <maria_zk> | ok |
| [12:42:28] | <omega8cc> | delete the install task node, the site node and then re-verify OP platform |
| [12:44:10] | <maria_zk> | done |
| [12:44:22] | <maria_zk> | ok, so now i see how to delete a task that causes trouble |
| [12:45:58] | <maria_zk> | platform verified :) |
| [12:47:07] | <omega8cc> | wow, I hope it is fully installed (and not a partially installed), but that worked for me |
| [12:47:39] | <maria_zk> | great |
| [12:47:46] | <maria_zk> | i guess i'll need more ram to do this |
| [12:48:34] | <maria_zk> | ok, i have enabled statistics |
| [12:48:37] | <maria_zk> | page views count |
| [12:48:50] | <maria_zk> | let's see... |
| [12:48:55] | <maria_zk> | what else... |
| [12:48:59] | <omega8cc> | it worked with 512 MB, well, "worked", but still |
| [12:49:49] | <omega8cc> | check permissions for roles, related to stats maybe |
| [12:50:12] | <maria_zk> | logged in as user 1 ? |
| [12:50:23] | <maria_zk> | javascript still disabled in front page - only? |
| [12:50:41] | <maria_zk> | ah no, it is dead in other pages as well |
| [12:51:12] | <maria_zk> | for /admin pages js works fine |
| [12:51:40] | <omega8cc> | maybe you enabled too much at once, like extra js gzipping or something |
| [12:51:44] | <maria_zk> | yes |
| [12:51:47] | <maria_zk> | that i did |
| [12:51:57] | <maria_zk> | and i don't need gzipping, right? |
| [12:52:19] | <omega8cc> | so disable this, nginx compresses everything already |
| [12:53:17] | <maria_zk> | I thought so... no need for JavaScript Aggregator module then |
| [12:53:19] | <omega8cc> | we should hardcode some gzip related setting also for those contrib modules |
| [12:54:00] | <maria_zk> | that would definitely work for me |
| [12:54:02] | <omega8cc> | it can help but it can also break incompatible stuff |
| [12:54:18] | <maria_zk> | when i change an option in Performance, changes are not saved. |
| [12:54:26] | <maria_zk> | this is a buf, right? |
| [12:54:29] | <maria_zk> | bug |
| [12:54:33] | <omega8cc> | no |
| [12:54:37] | <omega8cc> | by design |
| [12:55:00] | <maria_zk> | ah |
| [12:55:00] | <omega8cc> | js/css optimization is forced to on |
| [12:55:05] | <maria_zk> | aaa i see... |
| [12:55:07] | <omega8cc> | and gzip to off |
| [12:55:18] | <omega8cc> | but only for core modules |
| [12:55:18] | <maria_zk> | Gzip page compression (Boost & Core): ? |
| [12:55:34] | <omega8cc> | should be off already |
| [12:55:51] | <omega8cc> | by default |
| [12:56:03] | <maria_zk> | yeap |
| [12:56:06] | <maria_zk> | still no joy |
| [12:56:11] | <maria_zk> | js is dead |
| [12:56:30] | <omega8cc> | but the extra js aggregator has its own gzip settings and we don't hardcode it to off |
| [12:56:59] | <maria_zk> | i disabled that... should i turn it back on? |
| [12:57:15] | <omega8cc> | are you sure the block is set to be available for all roles? |
| [12:57:30] | <maria_zk> | the ajax stats? |
| [12:57:38] | <omega8cc> | yes |
| [12:58:09] | <maria_zk> | no restriction |
| [12:58:24] | <maria_zk> | under the block settings for roles. i selected all roles now |
| [12:59:02] | <maria_zk> | no joy. it's as if i had js turned off in my browser |
| [12:59:08] | <omega8cc> | clear all caches, clear browser cache and re-login, no idea what you have wrong there |
| [12:59:11] | <maria_zk> | (but is IS turned on) |
| [12:59:15] | <maria_zk> | omega8cc: ok |
| [13:00:27] | <maria_zk> | so for js to work, i need the ajax statistics block. |
| [13:02:31] | <maria_zk> | thank you so much for your help... |
| [13:07:27] | <omega8cc> | maria_zk: yes, as explained in the boost docs, it should work unless you have some unknown js conflicts etc, but I never had an issue with this, so let's hope it will work for you |
| [13:08:31] | <omega8cc> | anyway, 3 am here, so night all |
| [13:09:00] | <maria_zk> | good night! |
| [13:09:03] | <maria_zk> | and thanks again |
| [13:10:38] | * dipen has quit (Quit: Leaving) |
| [13:12:49] | * omega8cc_ has joined #aegir |
| [13:15:03] | * omega8cc_ has quit (Remote host closed the connection) |
| [13:15:10] | * omega8cc_ has joined #aegir |
| [13:16:58] | * omega8cc has quit (Ping timeout: 272 seconds) |
| [13:16:58] | * omega8cc_ is now known as omega8cc |
| [13:19:27] | * totten has quit (Ping timeout: 240 seconds) |
| [13:29:57] | * realityloop has quit (Remote host closed the connection) |
| [13:32:08] | <CIA-24> | aegir/hostmaster: mig dev-migrate_aliases * r8472785ed723 /modules/hosting/migrate/hosting_migrate.module: fix migrate so that it doesn't complain about the alias when it's not even changing |
| [13:32:40] | * totten has joined #aegir |
| [13:44:20] | * Vertice has joined #aegir |
| [13:52:15] | * dipen has joined #aegir |
| [13:54:40] | * arianek_afk has joined #aegir |
| [13:56:28] | * q0rban has joined #aegir |
| [13:58:03] | * dipen has quit (Remote host closed the connection) |
| [13:59:37] | * dipen has joined #aegir |
| [14:01:24] | * arianek_afk has quit (Remote host closed the connection) |
| [14:03:58] | * gezick has joined #aegir |
| [14:04:22] | <CIA-24> | aegir/hostmaster: mig dev-migrate_aliases * r16e71bd22f03 /modules/hosting/clone/hosting_clone.module: handle checking of aliases in the clone form, and except on any alias that is already in use (unlike migrate, in which the site node is allowed to retain its current aliases) |
| [14:05:42] | <gezick> | what's the preferred method for configuring aegir's system user to be able to login to github (for cloning repos from makefiles and the like) |
| [14:06:19] | <mig5> | well you can git clone with https:// for private repos i believe. |
| [14:06:26] | <mig5> | or http:// , can't remember |
| [14:06:28] | <mig5> | and specify the user/pass |
| [14:06:49] | <mig5> | alternatively, drop in a private ssh key of which the public key has access to that repo |
| [14:06:49] | <gezick> | is that better than using the git: url and setting a server key or similar? |
| [14:06:53] | <mig5> | well |
| [14:07:01] | <mig5> | the git: url is for anonymous cloning |
| [14:07:15] | <mig5> | so a) requires a public repo, and b) you can't push back to a repo via git:// (on github) |
| [14:07:17] | * kvanderw has joined #aegir |
| [14:07:30] | <mig5> | so it depends what you want to do |
| [14:07:55] | <gezick> | so, i've actually got one pub, one priv, now that i look at it. |
| [14:08:02] | <gezick> | aegir is passing this back "Unable to clone main from git@github.com:nagey/profile.git." |
| [14:08:07] | <mig5> | right |
| [14:08:19] | <mig5> | well, it is using the SSH protocol, and wants your private key |
| [14:08:35] | * joestewart|afk is now known as joestewart |
| [14:08:38] | <mig5> | so put one there that has access to the repo (you can add public keys to a private repo on github if i recall correctly) |
| [14:08:50] | <gezick> | can i just drop a .ssh dir into /var/aegir? |
| [14:09:04] | <mig5> | yes |
| [14:09:43] | <mig5> | just ensure the ownership and permissions are ok (owned by aegir:aegir, .ssh dir is chmod 700, and the id_rsa file inside it is ideally chmod 400 ) |
| [14:10:05] | <mig5> | you don't need the id_rsa.pub on the aegir server side |
| [14:10:13] | <gezick> | gotcha. |
| [14:10:16] | <mig5> | but it does no harm to leave it there either. |
| [14:10:54] | <gezick> | definitely liking the whole makefile/installprofile/aegir thing… finally feels like the "right way" to do drupal… |
| [14:12:08] | <mig5> | ;) |
| [14:12:26] | <mig5> | did you feel the *click* |
| [14:12:31] | <mig5> | when it suddenly made sense why |
| [14:12:55] | <mig5> | (or i should give you some credit, maybe it just made sense to you off the bat: it didn't for me :) ) |
| [14:13:45] | <joestewart> | mig5: thanks for committing http://drupal.org/node/984396 It has been a hassle for Pressflow and Varnish. |
| [14:13:47] | <mig5> | now if every node etc was exportable with a UUID, well now |
| [14:14:03] | * q0rban has quit (Quit: Computer has gone to sleep.) |
| [14:14:25] | <mig5> | joestewart: yeah i started to really feel strongly about it once i started using Pressflow and Varnish myself :) well, its easy to use the local.settings.php, but still, i want to keep aegir from enforcing 'policy' |
| [14:14:42] | <joestewart> | But it made me wonder why local.settings.php is included before global.inc |
| [14:14:50] | <mig5> | we will need an example in the community docs 'How to enable caching on all sites' and aa snippet of global.inc |
| [14:15:47] | <mig5> | i thought the include_once meant the values couldn't override when in global.inc |
| [14:15:49] | <mig5> | maybe that isn't the case |
| [14:15:53] | <mig5> | i remember wondering about that too |
| [14:16:45] | <joestewart> | ah. I just looked at the ordering of global after individual settings. |
| [14:17:38] | <joestewart> | no the include once is for each file. The array entries would still get clobbered. |
| [14:17:55] | <mig5> | hmm |
| [14:17:59] | <mig5> | so that seems back to front then doesn't it |
| [14:18:34] | <mig5> | should go default > global override > individual override |
| [14:18:42] | <joestewart> | yes, I looked back at Adrian's commit and I think local.settings.php originally had some db settings. |
| [14:18:54] | <mig5> | Vertice: what do you think of that |
| [14:19:00] | <Vertice> | cant talk now |
| [14:19:01] | <mig5> | shouldn't local.settings.php come last so as to override even global overrides |
| [14:19:02] | <Vertice> | vegas |
| [14:19:02] | <joestewart> | I would think that would be better |
| [14:19:03] | <mig5> | ok |
| [14:19:07] | <joestewart> | np |
| [14:19:15] | <Vertice> | i just saw meatloaf check into our hotel |
| [14:19:19] | <mig5> | wtf |
| [14:19:56] | <Vertice> | at a friend's wedding weekend in las vegas |
| [14:20:00] | <gezick> | mig5: actually fairly new to drupal, from a more "traditional" oop and software dev background. it was the first way to drupal that actually made any sense to me ;-) |
| [14:20:08] | <mig5> | gezick: haha |
| [14:20:14] | <mig5> | wow that makes for a nice change |
| [14:20:19] | <Vertice> | hahaha |
| [14:20:21] | <mig5> | i have a lot of trouble convincing people of this 'paradigm' |
| [14:20:25] | <mig5> | 'oh it all sounds too hard' |
| [14:20:31] | <mig5> | 'install profiles, yuck' |
| [14:20:32] | <Vertice> | it's actually one of the things steven liked about it too |
| [14:20:37] | <mig5> | 'too many platforms!' |
| [14:20:43] | <Vertice> | ciao |
| [14:20:47] | <mig5> | see ya :) |
| [14:20:52] | <mig5> | give meatloaf a kiss from me |
| [14:21:09] | <Vertice> | siss |
| [14:21:22] | <Vertice> | you dont know where he;s been |
| [14:21:38] | <mig5> | it's vegas, baby |
| [14:21:58] | <Vertice> | haha |
| [14:22:00] | <Vertice> | tm |
| [14:22:33] | * obrienmd has joined #aegir |
| [14:23:50] | <gezick> | mig5: that's actually one part of the process that i'm still having a bit of trouble with. doing builds on a aegir platform seems like it'll just lead to a nightmare of 'too many platforms!' as you put it. |
| [14:23:55] | * realityloop has joined #aegir |
| [14:24:21] | <mig5> | well |
| [14:24:34] | <mig5> | that's why i wrote the ability to run 'DElete' on platforms :) |
| [14:24:38] | <mig5> | as well as sites |
| [14:24:56] | <gezick> | heh. fair point. |
| [14:25:33] | <gezick> | so, i've got a makefile/profile that i can make and install interactively, but when i try running it on my aegir server, i get "Drush was not able to start (bootstrap) the Drupal database." |
| [14:26:02] | <gezick> | but I'm able to install with the standard profile no problem |
| [14:26:42] | <gezick> | how should i go about debugging the cause for that? |
| [14:26:59] | <gezick> | or, rather, the "right way" to do it |
| [14:27:35] | <mig5> | hmm |
| [14:27:41] | <mig5> | try installing it from cli |
| [14:27:46] | <mig5> | which is a combination of steps, one sec |
| [14:28:08] | <mig5> | example: |
| [14:28:27] | <mig5> | gah, my copy/paste is screwed |
| [14:28:58] | <mig5> | http://lists.aegirproject.org/pipermail/aegir/2011-January/000026.html |
| [14:29:26] | <mig5> | towards the end there i do a 'provision-save' command to save the site context |
| [14:29:32] | <mig5> | with the appropriate --profile |
| [14:29:40] | <mig5> | and then i can drush @thatsite.com provision-install --debug |
| [14:29:45] | <mig5> | that might give some clues |
| [14:30:06] | <mig5> | does your profile require human interaction? i.e can't be automated? |
| [14:30:34] | <gezick> | i didn't do anything that should make it require humans |
| [14:30:38] | <gezick> | profile is pretty simple |
| [14:30:40] | <mig5> | because if so, you've no hope in making aegir install it unless you write a custom feature that lets you enter that data in the site form and gives it to the profile via drush |
| [14:30:43] | <mig5> | ok |
| [14:31:04] | <mig5> | damn i have to go out |
| [14:31:05] | <gezick> | i mean, needs the standard, setup your admin account, setup your databse (i didn't automate those) |
| [14:31:07] | * jcapelo has quit (Quit: jcapelo) |
| [14:31:08] | <mig5> | back in a few hours |
| [14:31:14] | <gezick> | but mostly it just calls standard profile |
| [14:31:17] | <mig5> | yeah aegir should be automate those parts |
| [14:31:17] | <gezick> | will try tht tho thx |
| [14:34:30] | * obrienmd has quit (Quit: Leaving.) |
| [14:41:17] | * omega8cc has quit (Quit: zzzzzz...) |
| [14:41:24] | * dipen has quit (Quit: Leaving) |
| [15:00:31] | * arianek has joined #aegir |
| [15:01:42] | * troubador1 has left #aegir () |
| [15:07:54] | * ezra-g has joined #aegir |
| [15:08:09] | * ezra-g has quit (Client Quit) |
| [15:14:19] | * Met4physica has quit (Quit: No Ping reply in 180 seconds.) |
| [15:14:24] | * Met4physica has joined #aegir |
| [15:15:17] | * josh_k has quit (Ping timeout: 240 seconds) |
| [15:23:49] | * obrienmd has joined #aegir |
| [15:24:57] | * simesy has quit (Quit: Gordon Bennett) |
| [15:35:21] | * obrienmd has quit (Quit: Leaving.) |
| [15:36:39] | * bixgomez has joined #aegir |
| [15:43:30] | * kvanderw has quit (Ping timeout: 240 seconds) |
| [15:46:48] | * kvanderw has joined #aegir |
| [16:00:05] | <realityloop> | mig5: ping |
| [16:09:00] | * jcapelo has joined #aegir |
| [16:13:28] | <realityloop> | any one here thats really good with DNS? |
| [16:16:08] | * bixgomez has quit (Remote host closed the connection) |
| [16:23:25] | * bixgomez has joined #aegir |
| [16:27:55] | * bixgomez has quit (Remote host closed the connection) |
| [16:28:20] | * simesy has joined #aegir |
| [16:35:36] | * kvanderw has quit (Ping timeout: 265 seconds) |
| [16:36:13] | * obrienmd has joined #aegir |
| [16:37:04] | * kvanderw has joined #aegir |
| [16:55:07] | <mig5> | hey realityloop |
| [16:55:36] | <realityloop> | mig5: hey mate.. I figured it out myself.. though it may lead to a feature request in the future.. :P |
| [16:56:00] | <realityloop> | tho I can't quite figure out how it would sensibly be implemented |
| [16:56:30] | <mig5> | dns and aegir related? |
| [16:56:45] | <realityloop> | mig5: a way for aegir DNS implementation to be able to handle stuff like this http://www.google.com/support/a/bin/answer.py?answer=60227 |
| [16:56:56] | <mig5> | ah yeah |
| [16:57:03] | <mig5> | the SRV records |
| [16:57:03] | <realityloop> | it's probably not on the roadmap tho |
| [16:57:31] | <mig5> | maybe not, but i don't see any reason to not support SRV records, or rather, restrict what type of records can be created (maybe hardcoding what *is* a valid record) |
| [16:57:50] | <mig5> | i think koumbit use xmpp so good chance they want to implement it already |
| [16:59:46] | <realityloop> | cool |
| [16:59:59] | <mig5> | did you test that backup size patch yet ;) |
| [17:00:38] | <realityloop> | mig5: no sorry.. I got caught up with work stuff.. let me try now |
| [17:01:07] | <mig5> | no dramas, me too |
| [17:04:43] | <eft> | mig5: you manic guy, you |
| [17:05:16] | <eft> | so you pulled the plug on the cache setting :) |
| [17:05:25] | <mig5> | yep |
| [17:05:28] | <mig5> | makes sense |
| [17:05:55] | <eft> | site builders 1 - sysops not so much |
| [17:06:12] | <mig5> | no, it's policy, it shouldn't be in aegir |
| [17:06:19] | <mig5> | sysops can enforce caching any number of ways |
| [17:06:34] | <mig5> | i don't buy into the idea that people with VPS don't have access to a sysadmin to do it properly though |
| [17:06:41] | <mig5> | if you have a VPS, toughen up and learn that shit like I did |
| [17:06:48] | <mig5> | or go back to shared hosting :) |
| [17:06:54] | <mig5> | yes it *is* that black and white :) |
| [17:07:44] | <eft> | hear that simey? |
| [17:07:48] | <eft> | hear that simsey? |
| [17:07:57] | <mig5> | he knows what i'm like |
| [17:08:15] | <mig5> | if you put your client's site on a vps, you have a responsibility to them to manage it properly |
| [17:08:25] | <mig5> | however, i *also* agree aegir shouldn't enforce the cache in the settings.php |
| [17:08:36] | <mig5> | :) |
| [17:09:08] | <mig5> | and he has a point to turn it on and off in the db at least for testing/building purposes |
| [17:09:23] | <mig5> | i was only remarking on the irony of trying to improve performance by asking the database what to do |
| [17:09:46] | <eft> | I know bro - I' |
| [17:10:01] | <eft> | m on both your sides :) |
| [17:10:07] | <eft> | gonna have to tag another issue as aegirWTF - cos it makes me giggle |
| [17:10:20] | <mig5> | oh there's plenty to pick |
| [17:10:26] | <mig5> | in fact most of the issue queue makes me go WTF |
| [17:10:54] | <realityloop> | mig5: where does backup.provision.inc live? |
| [17:11:04] | <mig5> | realityloop: provision/platforms/ |
| [17:11:09] | <mig5> | patch from top of provision |
| [17:11:22] | <mig5> | er provision/platform |
| [17:11:26] | <mig5> | not platforms |
| [17:14:10] | <CIA-24> | aegir/provision: mig * rd7825722c747 /provision.inc: #888194 - http_port was being fetched from web_server attribute instead of server |
| [17:14:48] | <realityloop> | mig5: this only shows sizes for backups created after the patch applied? |
| [17:15:09] | <mig5> | yeah |
| [17:15:15] | <mig5> | oh. |
| [17:15:17] | * realityloop creating a backup |
| [17:15:32] | <mig5> | right, i suppose i should analyse all existing backups and get the size of them and add those into the database |
| [17:15:36] | <mig5> | ugh |
| [17:15:43] | <mig5> | didn't think of that, or my brain didn't want me to :) |
| [17:15:55] | <realityloop> | it would be nice.. :) |
| [17:16:15] | <mig5> | maybe you want to patch my patch ;) |
| [17:16:20] | <realityloop> | but I can live with current implementation |
| [17:16:22] | <realityloop> | :) |
| [17:16:44] | <mig5> | nah i should fix it properly |
| [17:16:55] | <mig5> | crap |
| [17:17:09] | <realityloop> | well I can confirm that it shows sizes for ones added after implenentation :) |
| [17:17:35] | <mig5> | gimme a couple min |
| [17:29:38] | * josh_k has joined #aegir |
| [17:29:43] | <mig5> | bah can't get it to work |
| [17:29:46] | <mig5> | will work on it later |
| [17:33:59] | <mig5> | cant get filesize of existing backups because www-data can't read that dir (that's by design) |
| [17:34:03] | <mig5> | so we may just have to live with it |
| [17:34:59] | <realityloop> | it's no problem for me.. |
| [17:35:06] | <realityloop> | we've been using beta |
| [17:35:25] | <realityloop> | if it's in by 1.0 then it doesn't really affect things if you ask me |
| [17:37:53] | <mig5> | good idea to rotate out old backups anyway |
| [17:37:56] | <mig5> | :) |
| [17:37:59] | <realityloop> | yep |
| [17:38:12] | <realityloop> | gotta split.. catchya later |
| [17:38:15] | <mig5> | seeya |
| [17:38:20] | * realityloop has quit (Remote host closed the connection) |
| [17:52:58] | * arianek has quit (Remote host closed the connection) |
| [17:57:27] | * josh_k has quit (Quit: Holla!) |
| [18:29:50] | * sepehr has joined #aegir |
| [18:48:15] | * totten has quit (Ping timeout: 240 seconds) |
| [18:56:02] | * eft has left #aegir () |
| [19:03:07] | * totten has joined #aegir |
| [19:04:22] | * sepehr has quit (Ping timeout: 255 seconds) |
| [19:16:00] | * simesy_ has joined #aegir |
| [19:16:00] | * simesy has quit (Read error: Connection reset by peer) |
| [19:16:01] | * simesy_ is now known as simesy |
| [19:23:49] | * sepehr has joined #aegir |
| [19:26:32] | <CIA-24> | aegir/hostmaster: mig * rd2a286f52fb8 /modules/hosting/platform/hosting_platform.drush.inc: switch to use hosting_domain_allowed hook when looking for sites to import. this prevents a site being imported that is already defined as an alias of another site |
| [19:27:03] | * Irishgringo has quit (Ping timeout: 240 seconds) |
| [19:34:15] | * Chipie has quit (Ping timeout: 240 seconds) |
| [19:38:58] | <CIA-24> | aegir/hostmaster: mig * rf4fd43d07647 /modules/hosting/site/hosting_site.module: #860582 by hadsie, mig5: Get site by alias as well as url |
| [19:39:52] | * psynaptic|afk has quit (Read error: Connection reset by peer) |
| [19:39:58] | * psynaptic|afk has joined #aegir |
| [19:45:36] | * sepehr has quit (Ping timeout: 240 seconds) |
| [19:59:49] | * sepehr has joined #aegir |
| [20:03:45] | * kylemathews has quit (Read error: Operation timed out) |
| [20:04:46] | * nick_santa has joined #aegir |
| [20:11:36] | * jcapelo has quit (Quit: jcapelo) |
| [20:12:14] | * ronia has joined #aegir |
| [20:33:49] | * Chipie has joined #aegir |
| [20:38:29] | * Chipie1 has joined #aegir |
| [20:51:15] | * NeoID has quit (Quit: Forlater kanalen) |
| [20:57:46] | * nick_santa has quit (Quit: nick_santa) |
| [21:04:16] | * sepehr1 has joined #aegir |
| [21:06:12] | * sepehr has quit (Ping timeout: 250 seconds) |
| [21:11:25] | * EugenMayer has joined #aegir |
| [21:16:09] | * Chipie1 has quit (Quit: Leaving.) |
| [21:18:16] | * sepehr1 has quit (Ping timeout: 240 seconds) |
| [21:21:09] | * nick_santa has joined #aegir |
| [21:26:06] | * nick_santa has quit (Quit: nick_santa) |
| [21:26:55] | * AquaticDisorder has joined #aegir |
| [21:34:09] | * sepehr has joined #aegir |
| [21:52:34] | * AquaticDisorder has left #aegir ("http://quassel-irc.org - Chat comfortably. Anywhere.") |
| [22:08:18] | * EugenMayer has quit (Quit: Leaving.) |
| [22:23:23] | * Chipie has quit (Ping timeout: 276 seconds) |
| [22:28:56] | * Chipie has joined #aegir |
| [22:37:53] | * psynaptic|afk has quit (Read error: Connection reset by peer) |
| [22:38:00] | * psynaptic|afk has joined #aegir |
| [22:48:28] | * sepehr has quit (Ping timeout: 255 seconds) |
| [23:11:00] | * sepehr has joined #aegir |
| [23:17:38] | * sepehr has quit (Ping timeout: 240 seconds) |
| [23:25:45] | * FransK_ has joined #aegir |
| [23:25:53] | * FransK has quit (Ping timeout: 264 seconds) |
| [23:28:45] | * NeoID has joined #aegir |
| [23:32:51] | * sepehr has joined #aegir |
| [23:37:51] | * psynaptic|afk has quit (Read error: Connection reset by peer) |
| [23:38:06] | * psynaptic|afk has joined #aegir |
| [23:41:01] | * psynapti_ has joined #aegir |
| [23:41:09] | * psynaptic|afk has quit (Read error: Connection reset by peer) |
| [00:24:11] | * Chipie has quit (Quit: Leaving.) |
| [00:27:58] | <skwashd> | anyone wanting to use drupal commons with aegir please test the patch at http://drupal.org/node/937864#comment-4020122 |
| [00:30:43] | * Chipie has joined #aegir |
| [00:35:05] | * NikLP has joined #aegir |
| [00:40:56] | * psynapti_ has quit (Read error: Connection reset by peer) |
| [00:41:11] | * psynaptic|afk has joined #aegir |
| [00:49:40] | * bejames has joined #aegir |
| [00:49:54] | <skwashd> | Vertice: hey |
| [00:55:36] | * MadTBone_ has joined #aegir |
| [01:00:10] | * Irishgringo has joined #aegir |
| [01:08:01] | * mlncn has quit (Remote host closed the connection) |
| [01:11:30] | * simesy has quit (Quit: simesy) |
| [01:11:32] | * scientist has joined #aegir |
| [01:14:35] | * Irishgringo has quit (Remote host closed the connection) |
| [01:40:51] | * anarcat has joined #aegir |
| [01:40:54] | * lukus has quit (Ping timeout: 240 seconds) |
| [01:43:03] | * psynaptic|afk has quit (Ping timeout: 240 seconds) |
| [01:50:15] | * omega8cc has joined #aegir |
| [01:50:19] | * omega8cc has quit (Remote host closed the connection) |
| [01:50:58] | * omega8cc has joined #aegir |
| [01:55:15] | * lukus has joined #aegir |
| [01:59:46] | * Chipie has left #aegir () |
| [02:04:50] | * scientist has quit (Ping timeout: 240 seconds) |
| [02:10:41] | * ezra-g has joined #aegir |
| [02:10:44] | * ezra-g has quit (Client Quit) |
| [02:16:18] | * anarcat looking at rabbitmq to resolve our concurrency and load problems |
| [02:17:00] | * kvanderw has quit (Quit: kvanderw) |
| [02:25:43] | * Irishgringo has joined #aegir |
| [02:47:22] | * milk has joined #aegir |
| [02:47:28] | * EugenMayer has joined #aegir |
| [03:02:45] | * obrienmd has quit (Quit: Leaving.) |
| [03:03:54] | * scientist has joined #aegir |
| [03:13:20] | * psynaptic has joined #aegir |
| [03:18:51] | <NeoID> | omega8cc, do you know why BARRACUDA writes "Aegir" to /etc/hostname? |
| [03:19:35] | <omega8cc> | mig5: the default > individual override > global override logic is correct imo, since it allows you to keep control in global.inc so any custom stuff in local can't do things the sysadmin doesn't allow, of course it makes sense only when there is > 1 of people with the write access to local |
| [03:20:58] | <omega8cc> | NeoID: the word "Aegir" in /etc/hostname ??? |
| [03:21:59] | <NeoID> | I've edited /etc/hostname to my servers domain like "server.domain.ltd", but each time I execute BARRACUDA.sh, it's changed to "Aegir". |
| [03:22:52] | <omega8cc> | NeoID: it should write fqdn there, to avoid issues on ubuntu, and this is what it should write there on the initial install only and never on upgrade |
| [03:23:24] | <NeoID> | I'm currently trying on a clean install :) |
| [03:24:07] | <omega8cc> | NeoID: so please define your expected fqdn hostname in the Barracuda config |
| [03:25:04] | <omega8cc> | or it will guess your hostname and go with any weird setup you have there |
| [03:25:10] | <NeoID> | ok |
| [03:30:56] | <omega8cc> | NeoID: I mean _MY_HOSTN, _MY_FRONT and _MY_OWNIP |
| [03:31:04] | <NeoID> | omega8cc, sure |
| [03:31:21] | <omega8cc> | all three must be defined, of course |
| [03:31:30] | <NeoID> | I know it may be a stupid question, but regarding running this script behind a router (NAT), I have a WAN-IP where the ports are forwarded to the server and a local IP. Which IP should I use in /etc/hosts and as _MY_OWNIP in BARRACUDA.sh? |
| [03:32:16] | <NeoID> | i've forwarded all related domain names to my WAN-IP |
| [03:34:14] | <omega8cc> | that can be tricky depending on how you have it configured, but there is a separate config section for servers running on local network and with traffic NAT'd there, see "LOCALHOST ON LOCAL NETWORK" |
| [03:37:40] | <omega8cc> | dinner time, bbl |
| [03:37:52] | <NeoID> | ok, thanks for the info, I'll give it a new try :) |
| [03:38:39] | * omega8cc has quit (Remote host closed the connection) |
| [03:38:48] | * omega8cc has joined #aegir |
| [03:39:51] | * totten has quit (Ping timeout: 240 seconds) |
| [03:40:10] | <EugenMayer> | is there any particular reason while site has "platform" as required during creating? |
| [03:40:11] | * omega8cc has quit (Remote host closed the connection) |
| [03:40:18] | * omega8cc has joined #aegir |
| [03:40:23] | * Irishgringo has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203074432]) |
| [03:44:06] | <NeoID> | EugenMayer, I don't know if I understand you corectly, but a plattform is what version/distro of Drupal your site should run on... Like "Drupal 6.20" is a platform.. |
| [03:44:53] | * scientist has quit (Read error: Connection reset by peer) |
| [03:47:23] | * obrienmd has joined #aegir |
| [03:48:03] | * scientist has joined #aegir |
| [03:54:25] | * obrienmd has quit (Quit: Leaving.) |
| [03:55:32] | * kylemathews has joined #aegir |
| [03:55:42] | * totten has joined #aegir |
| [03:58:00] | * Zelfje has quit (Ping timeout: 276 seconds) |
| [04:00:53] | * Zelfje has joined #aegir |
| [04:02:06] | <anarcat> | mig5: i have copied the aegirproject.org page in our hosting, once DNS switches (i just asked devseed to just change the NS records) the new page (without DCPH and with link to c.a.o instead of g.d.o) will be up |
| [04:03:24] | * bejames has quit (Quit: Core meltdown) |
| [04:05:09] | <hefring> | community => Resurrected two major pieces of documentation => http://community.aegirproject.org/node/286 |
| [04:06:05] | <EugenMayer> | NeoID: well, a site does not have to have a platform OOTB during creation |
| [04:06:12] | <EugenMayer> | this is close to a raise condition |
| [04:10:30] | * bixgomez has joined #aegir |
| [04:12:27] | * kvanderw has joined #aegir |
| [04:17:26] | * bixgomez has quit (Read error: Connection reset by peer) |
| [04:17:30] | * MadTBone_ has quit (Ping timeout: 240 seconds) |
| [04:20:23] | * bixgomez has joined #aegir |
| [04:24:41] | * milk has quit (Quit: baaaiiii) |
| [04:25:03] | * scientist has quit (Ping timeout: 240 seconds) |
| [04:35:25] | * jonhattan_ has joined #aegir |
| [04:45:13] | * killes has left #aegir () |
| [04:59:42] | * mlncn has joined #aegir |
| [05:00:49] | * jackinloadup has joined #aegir |
| [05:03:08] | * Zelfje has quit (Ping timeout: 276 seconds) |
| [05:05:18] | * Zelfje has joined #aegir |
| [05:06:20] | * bixgomez has quit (Remote host closed the connection) |
| [05:09:27] | * jonhattan has quit (Ping timeout: 240 seconds) |
| [05:10:06] | <EugenMayer> | could it be that aegir has issues with "libraries" ? |
| [05:22:26] | * jonhattan has joined #aegir |
| [05:25:24] | <omega8cc> | EugenMayer: every site in Aegir (including hostmaster site) does have to have platform (multisite) where it can be deployed, that is how standard Drupal multisite (not just Aegir) works |
| [05:25:50] | <EugenMayer> | omega8cc: i see. Its clear to me that a site needs a platform |
| [05:26:09] | <EugenMayer> | dont worry iam not new to aegir anymore. The thing is, why it is required during creating the site |
| [05:26:32] | <omega8cc> | it sounded it is not clear for you, for some reason |
| [05:26:32] | <EugenMayer> | a site CAN be a no platform fromt he start and then actually created on one |
| [05:26:51] | <omega8cc> | no |
| [05:27:06] | <EugenMayer> | A site is : server, domain, meta settings |
| [05:27:16] | <omega8cc> | sites != platform |
| [05:27:18] | <omega8cc> | no |
| [05:27:18] | <EugenMayer> | its like "where and how to access it", isnt it? |
| [05:27:19] | * jonhattan has quit (Read error: Connection reset by peer) |
| [05:27:22] | * jonhattan_ is now known as jonhattan |
| [05:27:35] | <omega8cc> | server is a site?? |
| [05:27:37] | <EugenMayer> | ye, a site is not "what"..exactly .. thats platform. |
| [05:28:12] | <omega8cc> | no, you are using terminology the wrong way probably, however I don't know why |
| [05:28:36] | <EugenMayer> | Yeah i might mix up where the server is relating to, makes sense it is the platform |
| [05:29:06] | <EugenMayer> | What is the site for you, thinking about its meta data.? |
| [05:29:27] | * bixgomez has joined #aegir |
| [05:30:07] | <EugenMayer> | Domain, Language (database server .. well ok, interesting..) Profile ( well that make sense when you have a platform ) |
| [05:30:55] | <EugenMayer> | omega8cc: for me a site is currenly like a mutant, it is probably design that way to "be easy". But i rather find it confusing having profile / platform / dbserver in there somehow, but maybe thats taste |
| [05:30:59] | <omega8cc> | we are talking about existing terminology, so please don't talk a new chinese to me ;) see: http://community.aegirproject.org/node/36 |
| [05:31:05] | <EugenMayer> | Hehe |
| [05:33:11] | <EugenMayer> | omega8cc: interesting doc. Thanks |
| [05:33:28] | <EugenMayer> | For me, it just probably has a different perspective |
| [05:33:39] | * kvanderw has quit (Quit: kvanderw) |
| [05:33:51] | <EugenMayer> | i create a "drupal project" which is : a site, the current platform it is on, makefile |
| [05:34:54] | <EugenMayer> | as the platform is auto-created from the makefile, my initial "steps are".Create a makefile, Create a site ( oh well..the platform ).. create a "Drupal Project" and add those references there. |
| [05:35:32] | <EugenMayer> | Click on install. The platform gets build, the site get intalled there and thats it. |
| [05:36:14] | <EugenMayer> | I just dont deal with platforms manually at all. I dont know about "platforms". I have a makefile version. And i deploy a project using a makefile |
| [05:36:52] | * jackinloadup has quit (Quit: jackinloadup) |
| [05:36:56] | <EugenMayer> | I think the initial question is here, does the platform has to be existing when a site gets created. I think reading your documentation it should, i undestand |
| [05:37:58] | <omega8cc> | EugenMayer: I think you missed a few important comments/blog posts and docs written by Vertice and mig5, that is already clearly explained and I don't see any reason to debate on it: what and why th e site *is* in Aegir - in short, it is an instance of app, and the app is an install profile (with related makefiles), and of course install profile lives in the platform context, that is it. |
| [05:38:07] | <EugenMayer> | I guess i have to create my own UI for creating a Drupal Project then( 1. select make file 2. create platform. 3. show site form prefilled + enter misssing fields 4. save and create project |
| [05:38:41] | <EugenMayer> | yeah i see |
| [05:39:18] | * edburton has joined #aegir |
| [05:40:22] | <EugenMayer> | omega8cc: thank you for clarifing. |
| [05:40:36] | <EugenMayer> | omega8cc: just a small issue : Platform "package listing" does not include libraries |
| [05:40:39] | * kvanderw_ has joined #aegir |
| [05:40:56] | <EugenMayer> | (while those are supported by make files / platforms ) |
| [05:41:02] | <EugenMayer> | is that by intention? |
| [05:41:14] | <omega8cc> | in short: you can't install a Drupal site without Drupal platform, no matter if that is multisite in Aegir or standalone, you are always starting with Drupal root, which is a "platform", in fact, since it always includes install profile (the app) |
| [05:42:01] | <EugenMayer> | yes i see |
| [05:42:28] | <EugenMayer> | omega8cc: in case of a makefile you just first "gather informations |
| [05:42:35] | <EugenMayer> | and then actually run "create it all" |
| [05:42:41] | <EugenMayer> | and thats where i run into issues |
| [05:43:26] | <EugenMayer> | i have to create the platform before i can continue gathering the informations of a site. But i see, due to the "profile selection", that order is a must |
| [05:43:27] | * TrevorT has joined #aegir |
| [05:43:51] | <omega8cc> | libraries are rather "includes" for modules, so they are not listed separately |
| [05:44:37] | * kvanderw_ is now known as kvanderw |
| [05:44:55] | <EugenMayer> | omega8cc: "includes"? |
| [05:45:05] | <omega8cc> | and they are often a third party tools, "embedded" by modules, like wysiwyg etc but not loaded as modules |
| [05:45:08] | <omega8cc> | yes |
| [05:45:20] | <EugenMayer> | yes they are no modules, there are libraries. |
| [05:46:17] | <EugenMayer> | But pretty strange not to list them, or? They can be packaged by feature server and have a normal version like all the other things. They are just deployed onder ./libraries/XXXX and not "activated" in general. But well i dont care to much |
| [05:47:10] | <EugenMayer> | i dont use the platform view anyway, as i have the make file view. If its by intention, iam fine ( eventhough it kind if is strange that you have to upgrade a platform which has no changes .. because e.g. the platform just changed the ckeditor version .. what is not visible ) |
| [05:47:20] | <EugenMayer> | ... in the "platform" perspective |
| [05:47:33] | * TrevorT is now known as Pls |
| [05:47:53] | <Pls> | hey omega8cc, I really like your BARRACUDA script, but however can't get it working (install doesnt start run). And reason of failing is DNS misconfiguration. I have similar issue as https://github.com/omega8cc/nginx-for-drupal/issues#issue/26 and already applied all commands from http://groups.drupal.org/node/89594#comment-283074 and it still stays its wrong. I have also made an A DNS record |
| [05:47:53] | <Pls> | for my server.domain.com and aegir.domain.com. Would appreciate if you can point me to the right direction. |
| [05:49:28] | * Vertice has quit (Quit: Vertice) |
| [05:50:27] | <edburton> | I wonder if anyone can help, I'm trying to install aegir on a linode using http://www.linode.com/stackscripts/view/?StackScriptID=203 but keep failing with "... waiting apache2: Syntax error on line 278 of /etc/apache2/apache2.conf: Could not open configur ation file /etc/apache2/conf.d/aegir.conf: No such file or directory", not sure where to go from here :-( |
| [05:51:01] | <EugenMayer> | edburton: your symlinks are broken or not accessable |
| [05:51:10] | <Pls> | edburton: It seems you have symlinks missing. |
| [05:51:16] | <omega8cc> | Pls: please open or continue existing issue in github queue, so we can follow it |
| [05:52:02] | <Pls> | omega8cc: Ok, right. Will describe more details there. |
| [05:52:31] | <omega8cc> | Pls: also, remember to include your server/system details, as explained in submission guidelines |
| [05:53:06] | <edburton> | thanks, I may be a little out of my depth, can you see how/where symlinks could end up missing during this StackScript? http://www.linode.com/stackscripts/view/?StackScriptID=203 , am trying to deploy StackScript as-is so that's on Debian 5.0 |
| [05:55:25] | <NeoID> | omega8cc, is there any reason for why you choose to use smtp relay insted of exim in your script? |
| [05:56:58] | <omega8cc> | NeoID: we use postfix, and the smtp relay is an *option* you can use when your host require/provide it with port 25 closed by default |
| [05:57:51] | <omega8cc> | no exim here |
| [06:04:24] | <NeoID> | so on linode (as I remember you use) I have to use a custom relay? |
| [06:07:46] | <omega8cc> | NeoID: no, they don't block smtp. We don't use Linode, I have my private 512 MB linode in Newark for testing only. |
| [06:08:10] | <NeoID> | ok :) |
| [06:08:57] | <omega8cc> | we use Voxel and they block (by default) outgoing smtp |
| [06:09:12] | * EugenMayer has quit (Quit: Leaving.) |
| [06:10:31] | * omega8cc_ has joined #aegir |
| [06:14:42] | * omega8cc has quit (Ping timeout: 246 seconds) |
| [06:14:43] | * omega8cc_ is now known as omega8cc |
| [06:17:56] | <omega8cc> | lol: FAQ: No, it's not april first. Yes, I am crazy. No, this is not a troll. - http://drupal.org/node/1044692 |
| [06:18:27] | <anarcat> | :) |
| [06:22:36] | <jlkinsel> | interesting. if you continue down that path, aegir could be a good replacement for virtualmin |
| [06:22:46] | <anarcat> | that is the idea |
| [06:23:22] | * troubador has joined #aegir |
| [06:23:54] | * kvanderw has quit (Quit: kvanderw) |
| [06:24:16] | <jlkinsel> | will be interesting to see how that progresses. |
| [06:26:18] | * omega8cc has quit (Remote host closed the connection) |
| [06:26:26] | * omega8cc has joined #aegir |
| [06:29:38] | <troubador> | Hi all! I'm excited to try Aegir, and I'm glad to see that the Ubuntu install script on the wiki (https://github.com/doka/install-aegir-on-ubuntu) seems to have almost all of the shell commands in the current INSTALL.txt. But the install script notes emphasize that it shouldn't be used on a production server, because it has no security hardening. I'm not new to Linux, but I am new to managing my own server, so can anyone suggest links to |
| [06:30:36] | <jlkinsel> | for hardening linux overall you mean? |
| [06:31:05] | <troubador> | Well, for hardening a linux installation which will be devoted exclusively to Aegir. |
| [06:31:38] | <troubador> | If I start with a bare Ubuntu 10.04, and run that Ubuntu install script, I want to know what's next to secure that installation. |
| [06:31:42] | * maria_zk has quit (Ping timeout: 240 seconds) |
| [06:32:03] | <troubador> | I know that things like SMTP and other services would be off-topic. |
| [06:32:40] | <troubador> | Although Aegir requires postfix, so I guess that needs hardening too. |
| [06:32:40] | <jlkinsel> | gotchya. I'd suggest two points - ubuntu has some pretty good security docs https://help.ubuntu.com/community/Security and then also look at securing the drupal install |
| [06:33:33] | <jlkinsel> | for drupal, I'd recommend http://crackingdrupal.com/ |
| [06:33:54] | <troubador> | Heh, yeah, cracking drupal was one of the first books I got. Good stuff. |
| [06:34:11] | * maria_zk has joined #aegir |
| [06:34:54] | <troubador> | Thanks for the Ubuntu security docs, they look good. |
| [06:35:09] | <jlkinsel> | sure. I'm not a fan of ubuntu personally, but that should do what ya need |
| [06:37:05] | <troubador> | Yeah. I mean, it will all be through ssh, so maybe Debian would be fairly similar. |
| [06:37:43] | <troubador> | I am basically just looking for whatever is considered "best practice" for Aegir people. |
| [06:38:17] | <troubador> | But probably there's not a consensus on a particular distro or particular hardening practices. |
| [06:38:23] | <jlkinsel> | i think most of the "aegir people" like debian, I use mostly centos myself |
| [06:38:37] | <jlkinsel> | (I'm not an "aegir people" - just another user:) ) |
| [06:38:40] | <troubador> | Heh. |
| [06:43:23] | * Vertice has joined #aegir |
| [06:44:59] | <mig5> | omega8cc: ega8cc> mig5: the default > individual override > global override logic is correct imo, since it allows you to keep control in global.inc so any custom stuff in local can't do things the sysadmin doesn't allow, of course it makes sense only when there is > 1 of people with the write access to local |
| [06:45:15] | <mig5> | but the problem is if you want something to override the global override |
| [06:45:19] | <mig5> | ie somethign site-specific |
| [06:45:27] | <mig5> | doesn't it make that impossible right now? |
| [06:45:32] | <mig5> | i would think that would be pretty serious |
| [06:45:44] | <mig5> | otherwise what's the point of a local override if a global one can override *that* |
| [06:47:26] | <mig5> | let's say i wanted to enable anonymous caching for all sites (in global.inc) |
| [06:47:37] | <mig5> | but one site i wanted to set it to aggressive, so i put that in local.settings.php |
| [06:47:47] | <mig5> | as i understand it, the global.inc would currently override that |
| [06:47:58] | <mig5> | i should of course actually test this to see if it works |
| [06:48:02] | <omega8cc> | mig5: the logic seems to follow the system logic, as the server admin (root) should be able to override anything, while site admin (non-root) should be allowed to use only what is not denied or forced in global |
| [06:49:32] | <mig5> | yeah |
| [06:49:35] | <mig5> | i think that's useless |
| [06:49:37] | <mig5> | :) |
| [06:50:08] | <omega8cc> | haha, I'm using it, anyway :p |
| [06:50:22] | <mig5> | the logic surely should be $default_configuration > overridden by $global_configuration > overridden by $local_configuration |
| [06:50:31] | <mig5> | i can't think of another project in existence that wouldn't do it that wya |
| [06:50:35] | <mig5> | hah |
| [06:50:38] | <mig5> | oh well |
| [06:51:06] | <mig5> | think of it like a package in debian or something |
| [06:51:12] | <mig5> | /usr/share/$program/config is default |
| [06:51:21] | <mig5> | /etc/$program/config is global |
| [06:51:29] | <mig5> | /home/miguel/.$program/config is override |
| [06:51:51] | * NikLP has quit (Ping timeout: 240 seconds) |
| [06:52:11] | <mig5> | anyway, maybe noone cares, so i will not bother to 'fix' it unless someone reports it as a nuisance |
| [06:53:01] | <omega8cc> | it is pure theory on my end since I don't allow to use local at all (unless for crazy table prefixes), so change it how you think makes more sense, yeah, I understand your poin t, that example with agressive caching for one site while forced standard for the rest is very good |
| [06:56:17] | <omega8cc> | mig5: go with it, it makes sense, after all no one opens access to local for end users probably, so it is safe |
| [06:58:40] | <mig5> | i think i'll go back to bed first |
| [06:58:47] | <mig5> | woken up at 6am sunday morning by Nagios :) |
| [07:01:21] | <omega8cc> | oh |
| [07:01:35] | <omega8cc> | mig5: have a better Sunday! |
| [07:07:26] | <NeoID> | omega8cc, thanks to you I think I managed to setup aegir on another server, but "run cron manually" doesn't do anything on sites I host with aegir... |
| [07:10:04] | * maria_zk has quit (Read error: Connection reset by peer) |
| [07:12:15] | * Pls has quit () |
| [07:16:35] | <omega8cc> | NeoID: correct, Barracuda and Octopus denies manual cron completely, you should rely on Aegir cron for sites to avoid issues (like duplicate content, rss etc) |
| [07:17:00] | <NeoID> | ok, again thanks alot, seems to be running fine here :) |
| [07:17:13] | <omega8cc> | it should |
| [07:17:47] | <NeoID> | just hope a later upgrade doesn't break stuff :p |
| [07:18:39] | <omega8cc> | no, upgrade path just work and is well tested, both for OS libs/services and Aegir instances |
| [07:18:48] | * MadTBone_ has joined #aegir |
| [07:19:08] | <omega8cc> | works* |
| [07:23:03] | * NikLP has joined #aegir |
| [07:28:20] | * obrienmd has joined #aegir |
| [07:33:54] | * realityloop has joined #aegir |
| [07:35:14] | * scientist has joined #aegir |
| [07:35:22] | * smoothify has joined #aegir |
| [07:37:26] | * NeoID has quit (Quit: Forlater kanalen) |
| [07:37:44] | * NeoID has joined #aegir |
| [07:43:06] | * omega8cc_ has joined #aegir |
| [07:46:26] | * bixgomez has quit (Remote host closed the connection) |
| [07:47:10] | * omega8cc has quit (Ping timeout: 265 seconds) |
| [07:47:10] | * omega8cc_ is now known as omega8cc |
| [07:49:56] | * jackinloadup has joined #aegir |
| [08:06:25] | * jackinloadup has quit (Quit: jackinloadup) |
| [08:07:49] | * NikLP has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]) |
| [08:09:30] | * simesy has joined #aegir |
| [08:10:22] | <NeoID> | omega8cc, just one last question.. what if I want to host a non-drupal site? Just place stuff in /var/www and use /etc/nginx and /etc/bind as usually without having to be afraid of breaking stuff when upgrading? |
| [08:11:51] | <NeoID> | like, creating a vhost-file in sites-available and create a zonefile in /etc/bind... |
| [08:19:23] | * bixgomez has joined #aegir |
| [08:21:08] | <omega8cc> | NeoID: yes, but I recommend to add your vhosts using as templates these created for buddy sql or collectd in /var/aegir/config/server_master/nginx/vhost.d/* |
| [08:21:14] | * jonhattan has quit (Quit: Ex-Chat) |
| [08:21:49] | <NeoID> | ok, I didn't install thoose :) |
| [08:23:21] | <omega8cc> | NeoID: add one of them to see the example as they are using php-fpm with more generic config, skipping all drupal related stuff |
| [08:23:30] | * realityloop has quit (Remote host closed the connection) |
| [08:24:09] | <NeoID> | oh. I see..., I'll do that, thought they where normal vhost files as I usually use |
| [08:24:23] | * Met4physica has quit (Ping timeout: 255 seconds) |
| [08:24:57] | * edburton has quit (Quit: edburton) |
| [08:25:04] | <omega8cc> | NeoID: you can try your own vhosts, sure |
| [08:25:09] | * bixgomez has quit (Remote host closed the connection) |
| [08:28:56] | * bixgomez has joined #aegir |
| [08:33:51] | * scientist has quit (Read error: Connection reset by peer) |
| [08:36:07] | * scientist has joined #aegir |
| [08:39:25] | * V1ntage has joined #aegir |
| [08:39:47] | <V1ntage> | hey, quick question, is it required to install Bind9 for Aegir to work properly? |
| [08:52:51] | * obrienmd1 has joined #aegir |
| [08:54:05] | * obrienmd has quit (Ping timeout: 260 seconds) |
| [08:56:59] | * simesy has quit (Quit: simesy) |
| [09:05:35] | * obrienmd1 has quit (Quit: Leaving.) |
| [09:12:57] | <omega8cc> | V1ntage: no |
| [09:13:26] | <V1ntage> | omega8cc: allright, thanks :) |
| [09:20:25] | * ronia has quit (Quit: ronia) |
| [09:26:57] | * adrinux has joined #aegir |
| [09:35:03] | * scientist has quit (Read error: Connection reset by peer) |
| [09:37:41] | * scientist has joined #aegir |
| [09:40:37] | * Irishgringo has joined #aegir |
| [09:43:03] | * scientist has quit (Ping timeout: 240 seconds) |
| [10:03:00] | * jackinloadup has joined #aegir |
| [10:06:37] | * nlisgo has joined #aegir |
| [10:13:02] | * NeoID2 has joined #aegir |
| [10:13:56] | <NeoID2> | omega8cc: after restarting the server, I get 403 Forbidden when visiting the site I just made... any ideas? http://client2.idmedia.no |
| [10:14:13] | <NeoID2> | everything seems to be running |
| [10:14:42] | <CIA-24> | aegir/provision: mig * r703a2354d9a5 /platform/backup.provision.inc: #1042758 - display filesize of backups |
| [10:14:49] | <CIA-24> | aegir/hostmaster: mig * r7a1af61cb1fc /modules/hosting/site/ (3 files): #1042758 - display filesize of backups |
| [10:16:58] | <omega8cc> | 403 is due to redirect to install.php which is a sign you have something wrong with permissions to settings.php file or site dir, or the path in the vhost is wrong, maybe try to reverify the site and make sure their domain is not used in any other vhost |
| [10:21:22] | <NeoID2> | oh.. reverifying did the job.. really strange this happens after restarting the server though... |
| [10:21:52] | * NeoID2 tries another restart |
| [10:23:42] | <NeoID2> | seems to work again, thanks :) |
| [10:24:02] | <omega8cc> | k |
| [10:24:05] | <NeoID2> | I'll make a documentation regarding all these questions, so I really thank you for your time |
| [10:24:24] | <omega8cc> | thanks! good idea |
| [10:24:51] | <NeoID2> | but need to go, see you later :) |
| [10:24:55] | * NeoID2 has quit (Quit: Page closed) |
| [10:25:49] | * mrchrisadams has joined #aegir |
| [10:30:23] | * MadTBone_ has quit (Read error: Operation timed out) |
| [10:43:02] | * Slydder1 has quit (Quit: Leaving.) |
| [10:46:02] | * smoothify has quit (Quit: Ex-Chat) |
| [10:48:25] | * Deezire has quit (Ping timeout: 260 seconds) |
| [10:55:16] | * Deezire has joined #aegir |
