| [10:00:08] | * gboudrias has quit (Quit: Leaving.) |
| [10:02:19] | * Egyptian[Laptop] has quit (Quit: Leaving.) |
| [10:02:27] | * Egyptian[Laptop] has joined #aegir |
| [10:06:38] | * Egyptian[Laptop] has quit (Ping timeout: 240 seconds) |
| [10:09:31] | * obicke has joined #aegir |
| [10:13:43] | * alberto56 has quit (Ping timeout: 265 seconds) |
| [10:17:49] | * berniecram has joined #aegir |
| [10:25:08] | * drakythe has joined #aegir |
| [10:26:43] | * obicke has quit (Ping timeout: 245 seconds) |
| [10:30:58] | * AquaticDisorder has quit (Remote host closed the connection) |
| [10:35:31] | * Egyptian[Laptop] has joined #aegir |
| [10:41:36] | * obicke has joined #aegir |
| [10:45:25] | * wamilton has joined #aegir |
| [10:48:53] | * boztek has joined #aegir |
| [10:49:33] | * samh has joined #aegir |
| [10:52:10] | * itomicsam has joined #aegir |
| [10:52:58] | * synesthete|away is now known as synesthete |
| [10:53:28] | * synesthete is now known as synesthete|away |
| [10:59:22] | * kylemathews has quit (Ping timeout: 252 seconds) |
| [11:00:06] | * josh_k has quit (Quit: Holla!) |
| [11:05:04] | * jhedstrom has quit (Remote host closed the connection) |
| [11:05:51] | * blurryaegir has quit (Ping timeout: 245 seconds) |
| [11:08:23] | * Blegh is now known as Phizes |
| [11:08:59] | * obicke has quit (Ping timeout: 260 seconds) |
| [11:09:45] | * Egyptian[Laptop] has quit (Ping timeout: 256 seconds) |
| [11:10:05] | * itomicsam has quit (Quit: Leaving) |
| [11:17:02] | * drakythe has quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/) |
| [11:17:35] | * blurryaegir has joined #aegir |
| [11:18:51] | * kepford has joined #aegir |
| [11:22:32] | * obicke has joined #aegir |
| [11:34:08] | * kylemathews has joined #aegir |
| [11:36:40] | * boztek has quit (Read error: Connection timed out) |
| [11:38:40] | * obicke has quit (Ping timeout: 244 seconds) |
| [11:46:31] | <darthsteven> | anarcat: Pong |
| [11:47:13] | <darthsteven> | hefring: Log pointer? |
| [11:47:13] | <hefring> | http://hefring.mig5.net/bot/log/aegir/2012-05-10#T205877 |
| [11:47:55] | <halcyonCorsair> | mig5: so we should feel free to ddos your site for fun now? :P |
| [11:50:33] | <darthsteven> | hefring: Tell anarcat sorry I missed your ping, it's 3am here so will be back online in a few hours |
| [11:50:33] | <hefring> | darthsteven: I'll pass that on when anarcat is around. |
| [11:51:58] | * obicke has joined #aegir |
| [11:52:23] | * Egyptian[Laptop] has joined #aegir |
| [11:56:58] | * alberto56 has joined #aegir |
| [12:01:53] | <anarcat> | darthsteven: ack |
| [12:01:53] | <hefring> | anarcat: 11 min 20 sec ago <darthsteven> Tell anarcat sorry I missed your ping, it's 3am here so will be back online in a few hours |
| [12:01:53] | * alberto56 has quit (Ping timeout: 256 seconds) |
| [12:11:12] | * synesthete|away is now known as synesthete |
| [12:11:42] | * synesthete is now known as synesthete|away |
| [12:18:32] | * blurryaegir has quit (Quit: Page closed) |
| [12:19:02] | * Harley has joined #aegir |
| [12:24:20] | * gusaus is now known as gusaus|lurk |
| [12:24:29] | * Harley has quit (Ping timeout: 260 seconds) |
| [12:29:38] | * Harley has joined #aegir |
| [12:36:13] | * patcon has joined #aegir |
| [12:37:03] | * q0rban has joined #aegir |
| [12:51:05] | * snlnz has quit (Quit: Pain is just weakness leaving the body... <k!15b8>) |
| [13:13:29] | * august1914 has quit (Quit: caio) |
| [13:23:50] | * blurryaegir has joined #aegir |
| [13:25:45] | <blurryaegir> | Hi guys. Is a site bound to its install profile in D7 since install hook_update_n's that are applied to the profile effect all sites built by that profile? Meaning, if I want to migrate that site to a new platform, does it matter if the profile exists on that platform? |
| [13:36:55] | <blurryaegir> | Google solved it, should have searched first, disregard my question |
| [13:43:36] | * gusaus|lurk has quit (Quit: gusaus|lurk) |
| [13:44:17] | * Harley has quit (Ping timeout: 260 seconds) |
| [13:54:45] | * Phizes has quit (Ping timeout: 252 seconds) |
| [13:56:29] | * Phizes has joined #aegir |
| [14:13:59] | * q0rban has quit (Quit: Computer has gone to sleep.) |
| [14:32:23] | * wamilton has quit (Ping timeout: 250 seconds) |
| [14:34:23] | * AquaticDisorder has joined #aegir |
| [14:44:00] | * AquaticDisorder has quit (Remote host closed the connection) |
| [15:18:20] | * lolmaus has joined #aegir |
| [15:20:00] | * boztek has joined #aegir |
| [15:35:51] | * penyaskito has joined #aegir |
| [15:38:30] | * boztek has quit (Read error: Connection timed out) |
| [16:06:33] | * simg has joined #aegir |
| [16:06:34] | * simg has left #aegir () |
| [16:07:24] | * EclipseGc has quit (Quit: EclipseGc) |
| [16:10:47] | * Phizes has quit (Ping timeout: 252 seconds) |
| [16:12:05] | <darthsteven> | anarcat: Ping? |
| [16:12:48] | * Phizes has joined #aegir |
| [16:16:12] | <mig5> | 2:16am there :) |
| [16:16:15] | <mig5> | what fun these timezones |
| [16:26:14] | * Tigeda has joined #aegir |
| [16:27:28] | <darthsteven> | Any idea what he wanted? |
| [16:27:28] | <darthsteven> | Drupalcon stuff? |
| [16:30:01] | * blurryaegir has quit (Ping timeout: 245 seconds) |
| [16:31:10] | * wroxbox has joined #aegir |
| [16:42:18] | * Harley has joined #aegir |
| [16:44:30] | * synesthete|away has quit (Ping timeout: 272 seconds) |
| [16:44:42] | * Harley_ has joined #aegir |
| [16:46:31] | * wroxbox has quit (Remote host closed the connection) |
| [16:47:41] | * Harley has quit (Ping timeout: 260 seconds) |
| [16:55:06] | * Harley has joined #aegir |
| [16:55:18] | * Deciphered1 has joined #aegir |
| [16:58:25] | * Harley_ has quit (Ping timeout: 260 seconds) |
| [16:58:38] | * Deciphered has quit (Ping timeout: 240 seconds) |
| [17:01:44] | * wroxbox has joined #aegir |
| [17:01:56] | * Pasqualle has joined #aegir |
| [17:04:29] | * Harley_ has joined #aegir |
| [17:04:56] | * Phizes has quit (Ping timeout: 252 seconds) |
| [17:07:45] | * Harley has quit (Ping timeout: 260 seconds) |
| [17:08:51] | * brando753 has quit (Quit: Client Quit, Bye Everyone!) |
| [17:08:53] | * Harley has joined #aegir |
| [17:11:57] | * Harley_ has quit (Ping timeout: 260 seconds) |
| [17:13:24] | * mrfelton has joined #aegir |
| [17:17:52] | * rv0 has joined #aegir |
| [17:23:37] | * dob_ has joined #aegir |
| [17:25:24] | * Phizes has joined #aegir |
| [17:28:14] | * wroxbox has quit (Ping timeout: 265 seconds) |
| [17:32:39] | * Phizes has quit (Ping timeout: 255 seconds) |
| [17:35:16] | * Phizes has joined #aegir |
| [17:41:38] | * kylemathews has quit (Ping timeout: 240 seconds) |
| [17:44:49] | * sdrycroft has joined #aegir |
| [17:47:55] | * wroxbox has joined #aegir |
| [17:51:43] | * a_c_m has joined #aegir |
| [17:52:09] | <a_c_m> | skwashd_: pm? |
| [17:54:51] | * David_Hernandez has joined #aegir |
| [17:56:21] | * penyaskito has quit (Ping timeout: 244 seconds) |
| [18:04:30] | * grobot has joined #aegir |
| [18:04:45] | * rv0 has quit (Quit: rv0) |
| [18:05:17] | * ratioweb has joined #aegir |
| [18:05:39] | * brando753 has joined #aegir |
| [18:05:42] | * Phizes has quit (Ping timeout: 256 seconds) |
| [18:07:27] | * Phizes has joined #aegir |
| [18:10:16] | * berniecram has quit (Quit: berniecram) |
| [18:14:54] | * wroxbox has quit (Remote host closed the connection) |
| [18:23:00] | * wroxbox has joined #aegir |
| [18:23:02] | * wroxbox has quit (Remote host closed the connection) |
| [18:25:13] | * ivanjaros has joined #aegir |
| [18:26:28] | * Pasqualle_ has joined #aegir |
| [18:29:25] | * Pasqualle has quit (Ping timeout: 244 seconds) |
| [18:33:29] | * Phizes has quit (Ping timeout: 265 seconds) |
| [18:33:53] | * obicke has quit (Quit: Leaving.) |
| [18:35:04] | * Phizes has joined #aegir |
| [18:44:55] | * patcon has quit (Quit: patcon) |
| [18:46:45] | * Phizes has quit (Ping timeout: 248 seconds) |
| [18:48:22] | * Phizes has joined #aegir |
| [18:53:54] | * Phizes has quit (Ping timeout: 260 seconds) |
| [18:54:31] | * Phizes has joined #aegir |
| [18:56:21] | * Deciphered1 has quit (Remote host closed the connection) |
| [19:01:43] | * Phizes has quit (Ping timeout: 245 seconds) |
| [19:01:45] | * Deciphered has joined #aegir |
| [19:03:44] | * Phizes has joined #aegir |
| [19:07:03] | * Deciphered has quit (Remote host closed the connection) |
| [19:07:06] | * danquah has joined #aegir |
| [19:09:29] | * Deciphered has joined #aegir |
| [19:13:02] | * Phizes has quit (Ping timeout: 255 seconds) |
| [19:13:54] | * Deciphered has quit (Ping timeout: 245 seconds) |
| [19:14:59] | * Phizes has joined #aegir |
| [19:20:00] | * Deciphered has joined #aegir |
| [19:28:38] | * Deciphered has quit (Remote host closed the connection) |
| [19:30:54] | * Deciphered has joined #aegir |
| [19:35:19] | * Deciphered has quit (Ping timeout: 260 seconds) |
| [19:37:54] | * adrinux has joined #aegir |
| [19:38:40] | * Deciphered has joined #aegir |
| [19:44:23] | * Phizes has quit (Read error: Operation timed out) |
| [19:46:29] | * Bidik is now known as bidik |
| [19:46:33] | * Phizes has joined #aegir |
| [19:49:08] | * bidik is now known as Bidik |
| [19:51:43] | * Phizes has quit (Ping timeout: 245 seconds) |
| [19:53:46] | * Phizes has joined #aegir |
| [19:56:45] | * j0nathan has joined #aegir |
| [20:01:56] | <mig5> | scrum time! |
| [20:02:04] | <mig5> | darthsteven: anarcat: ergonlogic |
| [20:02:09] | * Phizes has quit (Ping timeout: 252 seconds) |
| [20:02:24] | <darthsteven> | Hello! |
| [20:02:24] | <hefring> | salut |
| [20:02:27] | * brando753 has quit (Remote host closed the connection) |
| [20:03:05] | <mig5> | i've recently started using the pack feature, as a very large client of mine will probably use it soon, so i decided to eat the dogfood and transitioned my own infrastructure onto a 5-server high-availability environment :) |
| [20:03:12] | <mig5> | my main issue has been http://drupal.org/node/1555398 |
| [20:03:13] | <hefring> | http://drupal.org/node/1555398 => MySQL GRANTs aren't given for multiple servers in Web Pack => Hostmaster (Aegir), Code, normal, active, 18 comments, 1 IRC mention |
| [20:03:20] | <mig5> | we have some funny corner cases with mysql grants |
| [20:03:35] | <mig5> | not actually limited to pack module, but to a case where a webserver might have multiple ip addresses in the same LAN assigned to it |
| [20:03:50] | <mig5> | and mysql might pick the wrong *source* IP to make the mysql connection attempt in order to fetch those to use for grants |
| [20:04:07] | <mig5> | i've worked around it in my case there. but anyway that's basically all i'm thinking about at the moment |
| [20:07:14] | <mig5> | darthsteven: can we catch up after the scrum? |
| [20:07:18] | <mig5> | presuming the scrum isn't done already :) |
| [20:07:54] | <darthsteven> | (we can) |
| [20:08:08] | <darthsteven> | that's great |
| [20:08:16] | <darthsteven> | I saw lots of comments on that issue :) |
| [20:08:28] | <darthsteven> | shall I go next? |
| [20:08:29] | <mig5> | yeah i got a bit carried away - but it was really hard to find *what* was the problem |
| [20:08:33] | <mig5> | sure |
| [20:09:29] | <darthsteven> | didn't really do too much this week |
| [20:09:41] | <darthsteven> | got a fix for the Drush invoke issue in Drush 5 in: http://drupal.org/node/1554092 |
| [20:09:42] | <hefring> | http://drupal.org/node/1554092 => drush_invoke_process uses 100% CPU => Drush, Base system (internal API), critical, fixed, 13 comments, 3 IRC mentions |
| [20:10:01] | * Phizes_ has joined #aegir |
| [20:10:51] | <darthsteven> | and working with the Drush peeps to get http://drupal.org/node/1550574 sorted too |
| [20:10:52] | <hefring> | http://drupal.org/node/1550574 => drush_invoke_process doesn't pass through alias => Drush, Base system (internal API), normal, active, 4 comments, 1 IRC mention |
| [20:11:06] | <darthsteven> | we should be able to support Drush 5 easily when that one is done |
| [20:11:19] | <darthsteven> | at the moment the 6.x-2.x branch has a workaround |
| [20:11:30] | <darthsteven> | but 6.x-2.x still works with Drush 5 |
| [20:11:34] | <darthsteven> | that's it from me! |
| [20:11:37] | <darthsteven> | next? |
| [20:11:47] | <mig5> | nice! |
| [20:12:47] | <mig5> | i suspect that'll be it, it's only 6:12am in montreal |
| [20:12:51] | <mig5> | i'll wait a few min |
| [20:15:20] | * Phizes_ has quit (Ping timeout: 244 seconds) |
| [20:20:58] | <mig5> | ok i think that's it |
| [20:20:59] | <mig5> | thanks |
| [20:21:23] | * sdrycroft has quit (Remote host closed the connection) |
| [20:22:35] | <darthsteven> | thank you too |
| [20:23:31] | * Phizes_ has joined #aegir |
| [20:30:14] | * Egyptian[Laptop] has quit (Ping timeout: 248 seconds) |
| [20:30:38] | * Phizes_ has quit (Read error: Operation timed out) |
| [20:39:28] | * grobot has quit (Ping timeout: 260 seconds) |
| [20:39:42] | * Phizes_ has joined #aegir |
| [20:40:33] | * grobot has joined #aegir |
| [20:40:56] | <darthsteven> | did you want to chat about something after the scrum? |
| [20:41:14] | <mig5> | i thought I PM'd you |
| [20:49:08] | * Pasqualle_ has quit (Ping timeout: 240 seconds) |
| [21:06:16] | <danquah> | mig5: in the zero-configuration screencast you mention that you didn't find Aegirs remote-server features mature enough, do you thing that is still the case? |
| [21:12:44] | <darthsteven> | danquah: it's not 100% perfect |
| [21:13:02] | <darthsteven> | I wouldn't run a production site using it |
| [21:13:09] | <darthsteven> | not until we get it all sorted out |
| [21:13:22] | <darthsteven> | but, you can use the new 'pack' module |
| [21:13:25] | <danquah> | ah, will that is a good thing to know |
| [21:13:30] | <danquah> | "well" |
| [21:13:39] | <darthsteven> | which should allow you to do remote servers in better way for now |
| [21:14:29] | * Phizes_ has quit (Ping timeout: 248 seconds) |
| [21:14:42] | <danquah> | sounds good - is http://community.aegirproject.org/content/web-clusters the place to go? |
| [21:15:08] | * rv0 has joined #aegir |
| [21:16:21] | * Phizes has joined #aegir |
| [21:16:43] | <darthsteven> | maybe |
| [21:16:58] | <darthsteven> | mig5 or anarcat are the people to ask really |
| [21:17:05] | <danquah> | gotcha |
| [21:19:27] | <danquah> | we'll probably use the current remote server setup for a while during dev - do you know whether there is anything I should avoid doing, or is the functionality just unstable in general? |
| [21:20:35] | * Deciphered has quit (Remote host closed the connection) |
| [21:29:03] | <darthsteven> | basically you should be fine |
| [21:29:14] | <darthsteven> | but there is potential for some data loss of files |
| [21:29:28] | <darthsteven> | basically if you update a file on the remote, in the drupal files directory |
| [21:29:34] | <darthsteven> | and then verify the site |
| [21:29:43] | <darthsteven> | the file from the master will overwrite it |
| [21:30:04] | <darthsteven> | but if you backup first, the files will be synced back first, and the overwrite will happen the same way |
| [21:30:04] | <danquah> | ah, that would be bad... |
| [21:31:36] | <darthsteven> | *right way |
| [21:31:44] | <darthsteven> | sorry, if you backup first then it'll be fine |
| [21:31:51] | <danquah> | ok, and the reason the new pack module would handle this better is because it's rewritten from scratch? I mean, in the setup we're using we won't need more than one prod server - so the pack module seems superfluous |
| [21:32:28] | <darthsteven> | no, so the pack module skips rsync'ing files around, and relies on you setting up NFS shares between the machines |
| [21:32:29] | <danquah> | ok, so always backup prior to verifying |
| [21:33:28] | <danquah> | ok |
| [21:34:39] | <danquah> | but - using the setup from mig5's posts I will be creating a lot of platforms and migrating my site between them. In that case i guess the pack module would still have to do some rsync'in to move between the platforms right? |
| [21:35:10] | <darthsteven> | errmm....really not sure, guess so |
| [21:35:11] | <mig5> | no, there's no rsync done with pack module, other than for the configuration data |
| [21:35:25] | <mig5> | the entire /var/aegir/platforms is shared, just the one copy, over NFS across all the servers |
| [21:35:43] | * zorki has joined #aegir |
| [21:35:45] | <mig5> | that screencast on zero-touch, predates pack module. I am using pack module now (as of, erm 48 hours) |
| [21:35:58] | <mig5> | http://community.aegirproject.org/content/web-clusters is the correct place to read. i have just updated that page with a little more info |
| [21:35:59] | <danquah> | hehe, ok |
| [21:36:01] | <mig5> | and a diagram |
| [21:36:10] | <mig5> | See 'Caveats' too. |
| [21:39:46] | * penyaskito has joined #aegir |
| [21:39:53] | <danquah> | so … rsync is out, and all that is necessary to migrate a site is to do a local copy on the master server? (with regards to files that is) |
| [21:40:20] | <mig5> | well to migrate, you'd create a new platform, and that would go into /var/aegir/platforms so would be available to all servers in the pack. |
| [21:40:38] | <mig5> | then yes, the migrate process tarballs up the site and then 'unpacks' it on the target platform, and runs drush updatedb, then updates things like the vhost config etc. |
| [21:41:06] | * Phizes has quit (Ping timeout: 244 seconds) |
| [21:41:23] | <mig5> | you always create the actual platform on the *aegir* server (which, at the moment, seems to have best results if it is also the 'master' server in a pack), and then the verification process will ensure that platform is available wherever it needs to be |
| [21:42:42] | <danquah> | ok |
| [21:42:52] | * Phizes has joined #aegir |
| [21:46:29] | <danquah> | so in the new setup the aegis server only builds the platform, but it is then "owned" by the master server, whereas with the remote server setup aegis would build the platform and see it as the master-copy, and then rsync what it saw as a copy out to the master server? |
| [21:46:47] | <danquah> | "the aegir server" that is :) |
| [21:47:17] | <danquah> | don't know if that made any sense :) |
| [21:50:29] | <mig5> | that's basically it yep |
| [21:50:38] | * penyaskito has quit (Quit: Saliendo) |
| [21:50:39] | <a_c_m> | mig5: pm? |
| [21:50:46] | <mig5> | and i think if the 'master' server in the pack is *not* the aegir server, aegir will still rsync it out |
| [21:50:52] | <mig5> | a_c_m: yep |
| [21:51:03] | <danquah> | got it |
| [21:51:10] | * a_c_m has left #aegir () |
| [21:51:16] | <danquah> | great, thanks for the clearification! |
| [21:51:26] | <mig5> | no problem. still learning it myself :) |
| [21:51:36] | <danquah> | :) |
| [21:52:01] | <danquah> | I'll probably pop in with a couple of questions as the setup evolves |
| [22:00:50] | * grobot has quit (Ping timeout: 252 seconds) |
| [22:03:03] | * grobot has joined #aegir |
| [22:04:19] | * Phizes has quit (Ping timeout: 245 seconds) |
| [22:04:43] | <usaussie> | looking for a way to manage/restrict quota for sites managed by my aegir server. My google-fu skills are failing me ---- is there a way to limit the size a site can grow to via Aegir? |
| [22:06:18] | * Phizes has joined #aegir |
| [22:08:52] | <mig5> | usaussie: nothing for disk size at the moment I think. there's a Quota API (docs are a bit scarce!) but it's more on 'quantity of sites' vs 'size of sites'. |
| [22:09:20] | <usaussie> | gotcha. |
| [22:09:45] | <usaussie> | im happy to manage this outside of aegir if i had to, but i was hoping for a aegir method to simplify the configuration. |
| [22:11:03] | <mig5> | not sure if you'll be able to make use of the existing quota api: http://api.aegirproject.org/api/search/6.x-1.x/quota |
| [22:11:07] | <mig5> | but might help |
| [22:11:18] | <usaussie> | one of the things i have to figure out is what that quota will be per site. Most clients are used to a full cpanel environment with extra storage outside their website, but now they'll just have access via drupal only to their sites/default/files space. So im wondering how much is the right amount (im higher ed too, not commercial) |
| [22:11:31] | <usaussie> | ahh … nice---- thanks for the link |
| [22:12:11] | <mig5> | see if you can ping sfyn if/when he's around too, I think he wrote most of the quota stuff with the rest of Koumbit |
| [22:12:11] | * noecc has left #aegir ("pax") |
| [22:12:33] | <usaussie> | cool |
| [22:16:31] | <danquah> | has anyone had any luck in calling provision from a remote server via drush? I've set up a local alias that has a "remote-host" that points to the aegir-server, and I would now expect to be able to do a "drush @aegir @somesite.com status" |
| [22:17:25] | <mig5> | you should just be able to run |
| [22:17:29] | <mig5> | drush @somesite.com status |
| [22:17:32] | <mig5> | from the aegir server |
| [22:17:57] | <mig5> | at least, it works like that if it's an aegir-managed site, even a remote site |
| [22:18:16] | <danquah> | I'm running from another server, and want drush to connect to the aegir-server and basically run that command |
| [22:18:30] | <mig5> | you can't run drush provision commands from the remote webservers *back* to the aegir server, as it's not designed for that |
| [22:18:49] | <mig5> | the remote webservers don't even need drush i think, by design |
| [22:19:25] | <danquah> | ah well, I'm not running it from the webserver |
| [22:19:36] | <mig5> | what happens when you try what you've done? |
| [22:19:41] | <mig5> | drush @aegir @somesite.com status |
| [22:19:51] | <danquah> | my plan was to have a Jenkins job kick off a deployment |
| [22:21:03] | <danquah> | well - it executes, and amongst other things shows me a list of aliases on the aegir-server |
| [22:21:07] | <mig5> | if you use fabric, you could have jenkins ssh to the aegir server and just run('drush @somesite.com status') |
| [22:21:30] | <danquah> | hm, yeah, I'm considering going that way now |
| [22:22:32] | <danquah> | it would just be nice to be able to stick to Drush - especially now that it actually supports running drush commands remotely |
| [22:22:46] | <danquah> | but i guess the "double alias" parameter confuses it |
| [22:26:40] | * noecc has joined #aegir |
| [22:27:08] | * Phizes has quit (Ping timeout: 272 seconds) |
| [22:28:45] | * Phizes has joined #aegir |
| [22:33:48] | * Phizes has quit (Ping timeout: 255 seconds) |
| [22:34:38] | * Phizes has joined #aegir |
| [22:40:28] | * iribarne has joined #aegir |
| [22:42:04] | * ratioweb has quit (Ping timeout: 244 seconds) |
| [22:49:42] | * obicke has joined #aegir |
| [22:53:27] | * Harley has quit (Remote host closed the connection) |
| [22:54:29] | * ratioweb has joined #aegir |
| [23:13:56] | * EclipseGc has joined #aegir |
| [23:14:40] | * drakythe has joined #aegir |
| [23:15:02] | * jwester has joined #aegir |
| [23:18:34] | * cdracars has joined #aegir |
| [23:18:34] | * ratioweb has quit (Ping timeout: 252 seconds) |
| [23:39:55] | * greenscientist has quit (Ping timeout: 265 seconds) |
| [23:40:12] | * scientist has joined #aegir |
| [23:43:17] | * simg has joined #aegir |
| [23:47:17] | * jeremyr has joined #aegir |
| [23:49:44] | * wamilton has joined #aegir |
| [23:50:30] | * simg has left #aegir () |
| [00:06:01] | * lolmaus has quit () |
| [00:06:36] | * patcon has joined #aegir |
| [00:08:21] | * Phizes has quit (Ping timeout: 244 seconds) |
| [00:09:41] | * ivanjaros has quit (Read error: Connection reset by peer) |
| [00:10:44] | * fatguylaughing has joined #aegir |
| [00:11:48] | * anarcat waves hi |
| [00:12:02] | * zombiebeard has joined #aegir |
| [00:12:44] | * obicke has quit (Quit: Leaving.) |
| [00:13:57] | * Blegh has joined #aegir |
| [00:14:36] | * EclipseGc has quit (Read error: No route to host) |
| [00:14:54] | * EclipseGc has joined #aegir |
| [00:22:07] | * alberto56 has joined #aegir |
| [00:23:40] | <fatguylaughing> | Anyone using the amazon cloud to handle an Aegir server? |
| [00:24:25] | <fatguylaughing> | I'm trying to create a platform, but the platform will not verify. It throws an rsync error. Could not rsync from '/var/aegir/config/server_master/apache/platform.d/platform_Vendors7x16.conf' to 'aegir@ip-10-12-106-46:/' |
| [00:24:36] | <fatguylaughing> | Funny thing is I don't know that ip at the end of the error. |
| [00:24:44] | <fatguylaughing> | My server ip is different. |
| [00:25:26] | <fatguylaughing> | The next error is a little bit more detailed. |
| [00:25:48] | <fatguylaughing> | /var/aegir/config/server_master/apache/platform.d/platform_Vendors7x16.conf could not be synced to remote server ip-10-12-106-46. Changes might not be available until this has been done. (error: ssh: connect to host ip-10-12-106-46 port 22: Connection timed out rsync: connection unexpectedly closed (0 bytes received so far) [sender] rsync error: unexplained error (code 255) at io.c(601) [sender=3.0.7]) |
| [00:26:43] | * alberto56 has quit (Ping timeout: 245 seconds) |
| [00:29:13] | * VeggieMeat has quit (Quit: leaving) |
| [00:31:24] | * josh_k has joined #aegir |
| [00:31:24] | * josh_k has quit (Changing host) |
| [00:31:24] | * josh_k has joined #aegir |
| [00:32:07] | * VeggieMeat has joined #aegir |
| [00:42:47] | * ivanjaros has joined #aegir |
| [00:56:49] | * j0nathan has quit (Ping timeout: 245 seconds) |
| [00:56:51] | * ouelmart1 has joined #aegir |
| [00:57:14] | * j0nathan has joined #aegir |
| [00:58:04] | * ouelmart has quit (Ping timeout: 245 seconds) |
| [00:58:52] | <fatguylaughing> | Fixed it by setting my hostmaster's ip to 127.0.0.1 |
| [00:59:01] | <fatguylaughing> | Apparently I hadn't changed the defaults. |
| [01:01:38] | * ratioweb has joined #aegir |
| [01:06:59] | * obicke has joined #aegir |
| [01:21:19] | * gandhiano has joined #aegir |
| [01:29:16] | * ouelmart has joined #aegir |
| [01:30:34] | * ouelmart1 has quit (Ping timeout: 245 seconds) |
| [01:43:26] | * august1914 has joined #aegir |
| [01:49:02] | * AquaticDisorder has joined #aegir |
| [01:50:59] | * Blegh has quit (Ping timeout: 255 seconds) |
| [01:52:35] | * Blegh has joined #aegir |
| [01:58:55] | * dob_ has quit (Ping timeout: 252 seconds) |
| [01:59:53] | * jhedstrom has joined #aegir |
| [02:02:20] | <gandhiano> | I need to apply a patch, but don't have the folder /var/aegir/.drush/provision - any idea why? |
| [02:02:38] | <bgm> | gandhiano: are you using the Debian packages? |
| [02:02:57] | <gandhiano> | bgm, yes |
| [02:03:04] | <gandhiano> | is this at another location? |
| [02:03:12] | <bgm> | gandhiano: should be in /usr/share/drush/commands/provision |
| [02:03:24] | <bgm> | although, you'll lose your patch when you next upgrade |
| [02:03:36] | <gandhiano> | bgm, ok, great |
| [02:03:41] | <gandhiano> | tks |
| [02:03:58] | <darthsteven> | gandhiano: which patch are you applying out of interest? |
| [02:04:22] | <gandhiano> | darthsteven, here http://drupal.org/node/1083366#comment-5979790 |
| [02:04:23] | <hefring> | http://drupal.org/node/1083366 => Dont sync $site/files on verify of a site => Provision, Code, major, needs work, 41 comments, 3 IRC mentions |
| [02:04:35] | <darthsteven> | ah right, that one :) |
| [02:05:47] | * danquah has quit (Quit: danquah) |
| [02:06:10] | <gandhiano> | actually my issue is reported here - http://drupal.org/node/1561102 - but that one might provide the solution for it |
| [02:06:11] | <hefring> | http://drupal.org/node/1561102 => User enabled themes on remote (spoke) get disabled on verify => Provision, Code, normal, active, 0 comments, 1 IRC mention |
| [02:07:28] | * danquah has joined #aegir |
| [02:07:55] | * danquah has quit (Client Quit) |
| [02:08:30] | <gandhiano> | bgm, are there also different files on deb one? I don't have a provision_drupal.drush.inc (the one I'm supposed to patch) |
| [02:09:28] | <bgm> | $ dpkg -L aegir-provision | grep provision_drupal |
| [02:09:28] | <bgm> | /usr/share/drush/commands/provision/platform/provision_drupal.drush.inc |
| [02:09:43] | <bgm> | gandhiano: ^ |
| [02:09:56] | <gandhiano> | bgm, tks again :) |
| [02:22:26] | * zombiebeard has quit (Quit: zombiebeard) |
| [02:23:08] | * tommyent has joined #aegir |
| [02:24:55] | <fatguylaughing> | Alright so I'm moving my site from one server to another. I have created a platform and a new site. I have put the old sites database into the new site. |
| [02:25:09] | <fatguylaughing> | I've pulled over any custom additions like custom modules. |
| [02:25:40] | <fatguylaughing> | During my verify of the site I get an error file not found. |
| [02:25:56] | <fatguylaughing> | One of the modules is still looking for an include on the old server |
| [02:26:18] | <fatguylaughing> | I've cleared cache and I've looked for the old server name everywhere and have replaced it with the new. |
| [02:26:25] | <fatguylaughing> | Still getting the error... |
| [02:27:04] | * j0nathan has quit (Quit: ¡Hasta luego!) |
| [02:31:01] | * zombiebeard has joined #aegir |
| [02:33:33] | * danquah has joined #aegir |
| [02:36:40] | * ratioweb has quit (Ping timeout: 260 seconds) |
| [02:39:03] | <tommyent> | I installed Aegir via Barracuda and can't find the php error logs for the life of me. Can anyone point me in the right direction |
| [02:39:05] | <tommyent> | ? |
| [02:40:31] | * Pasqualle_ has joined #aegir |
| [02:41:23] | * simg has joined #aegir |
| [02:43:14] | * adrinux has quit (Quit: Leaving.) |
| [02:52:25] | * gusaus|lurk has joined #aegir |
| [02:53:56] | * simg has left #aegir () |
| [02:57:20] | * zombiebeard has quit (Quit: zombiebeard) |
| [03:00:21] | * ouelmart has left #aegir () |
| [03:05:11] | * Nickey has joined #aegir |
| [03:12:51] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [03:21:46] | * Pasqualle_ has quit (Ping timeout: 265 seconds) |
| [03:37:07] | * kylemathews has joined #aegir |
| [03:40:07] | <eft> | darthsteven: do you have any advice on creating mathematical equations with Drupal? |
| [03:40:15] | * gboudrias has joined #aegir |
| [03:40:28] | <eft> | I see you were involved with DruTeX module |
| [03:51:41] | * tommyent has quit (Quit: tommyent) |
| [04:11:19] | * DejaA has joined #aegir |
| [04:15:30] | <DejaA> | 'afternoon all. Quick question: is there an easy way to add a task to be run with cron on all aegir sites or do you have to have to do it via modules on each individual platform? |
| [04:17:00] | * wroxbox_ has quit (Ping timeout: 260 seconds) |
| [04:17:31] | * alberto56 has joined #aegir |
| [04:24:36] | * fatguylaughing has joined #aegir |
| [04:27:34] | <fatguylaughing> | Didn't realized I disconnected. Did anyone answer my earlier question? |
| [04:30:38] | * jwester has quit (Ping timeout: 240 seconds) |
| [04:40:21] | * zombiebeard has joined #aegir |
| [04:45:04] | * rv0 has quit (Quit: rv0) |
| [04:45:12] | * danquah has quit (Read error: Connection reset by peer) |
| [04:45:27] | * danquah has joined #aegir |
| [04:47:40] | * gusaus|lurk has quit (Quit: gusaus|lurk) |
| [04:51:47] | * gusaus has joined #aegir |
| [04:58:32] | <JohnF___> | Hi everyone...new to Aegir and have a question about active development of a site that's run from Aegir. When a new version is developed, does one just create a new platform for it and migrate the previous version to it? |
| [04:58:56] | <fatguylaughing> | require_once(/var/aegir/platforms/vendors-7.x-1.6/sites/vendors.aegir4.worxco.net/modules/media_youtube/includes/MediaYouTubeStreamWrapper.inc): failed to open stream: No such file or directory bootstrap.inc:3069 |
| [04:59:00] | <JohnF___> | Seems like having your site in Aegir during iterative development could be kind of tedious, no? |
| [04:59:12] | <fatguylaughing> | I've created a new site on a new server |
| [04:59:18] | <fatguylaughing> | and ported in the database |
| [04:59:36] | <fatguylaughing> | The site name is no longer vendor4. |
| [04:59:45] | <noecc> | fatguylaughing: Perhaps http://drupal.org/project/registry_rebuild |
| [04:59:51] | <fatguylaughing> | vendors.aegir4.worxco.net |
| [04:59:57] | <fatguylaughing> | noecc: Checking now |
| [05:01:50] | * dob_ has joined #aegir |
| [05:04:35] | <anarcat> | darthsteven: ping |
| [05:04:45] | <fatguylaughing> | noecc: Heh, can't even install the module from the command line because of the error |
| [05:04:54] | <fatguylaughing> | noecc: Also, can't make it to the site due to the error |
| [05:05:03] | <anarcat> | mig5: ping |
| [05:05:31] | * ratioweb has joined #aegir |
| [05:06:06] | <fatguylaughing> | noecc: Found the error, I can fix it manually in the db |
| [05:06:11] | <fatguylaughing> | noecc: Thanks so much |
| [05:07:01] | <noecc> | fatguylaughing: share your soultion |
| [05:07:16] | <fatguylaughing> | About to as long as it works ;) |
| [05:09:42] | <darthsteven> | anarcat: pong! |
| [05:10:43] | <anarcat> | darthsteven: pm |
| [05:14:32] | * danquah has quit (Quit: danquah) |
| [05:20:36] | * banghouse is now known as banghouseAFK |
| [05:29:41] | * EclipseGc has quit (Ping timeout: 256 seconds) |
| [05:43:43] | <fatguylaughing> | noecc: Ping |
| [05:50:39] | * Blegh has quit (Ping timeout: 252 seconds) |
| [05:52:38] | * Blegh has joined #aegir |
| [05:54:25] | * EclipseGc has joined #aegir |
| [05:55:32] | <gandhiano> | how does verify run the rsync? Only from the hostmaster to remote? |
| [05:57:31] | <gandhiano> | I tried applying the patch from here - http://drupal.org/node/1083366#comment-5979790 - to make rsync update rather than only create new files; but I still don't get a user created theme on the spoke to be sync'ed to the hostmaster (and, as a probable consequence, it is leading to theme deactivation - as reported here - http://drupal.org/node/1561102 |
| [05:57:33] | <hefring> | http://drupal.org/node/1083366 => Dont sync $site/files on verify of a site => Provision, Code, major, needs work, 41 comments, 4 IRC mentions |
| [05:57:33] | <hefring> | http://drupal.org/node/1561102 => User enabled themes on remote (spoke) get disabled on verify => Provision, Code, normal, active, 0 comments, 2 IRC mentions |
| [05:57:37] | * obicke has quit (Ping timeout: 244 seconds) |
| [05:58:17] | * DejaA has quit (Quit: DejaA) |
| [05:58:50] | * banghouseAFK is now known as banghouse |
| [06:04:14] | * David_Hernandez has quit (Quit: Saliendo) |
| [06:08:35] | * danquah has joined #aegir |
| [06:11:25] | * Blegh has quit (Ping timeout: 265 seconds) |
| [06:13:02] | * Blegh has joined #aegir |
| [06:14:08] | * kepford has quit (Ping timeout: 240 seconds) |
| [06:15:43] | * kepford has joined #aegir |
| [06:20:07] | * ratioweb has quit (Ping timeout: 256 seconds) |
| [06:29:06] | * obicke has joined #aegir |
| [06:30:34] | * Blegh has quit (Ping timeout: 245 seconds) |
| [06:32:30] | * Blegh has joined #aegir |
| [06:43:22] | * Mfawzy has joined #aegir |
| [06:44:04] | <Mfawzy> | hi, for some reason I can't verify a platform on my aegir site :( |
| [06:44:29] | <Mfawzy> | and the task popup doesn't appear :( |
| [06:44:41] | <Mfawzy> | it just loads nothing |
| [06:44:56] | * gboudrias has quit (Quit: Leaving.) |
| [07:02:10] | * Deciphered has joined #aegir |
| [07:15:09] | * mrfelton has quit (Quit: mrfelton) |
| [07:21:24] | * Deciphered has quit (Ping timeout: 255 seconds) |
| [07:25:58] | * eft has quit (Ping timeout: 244 seconds) |
| [07:26:55] | * iribarne has quit (Quit: iribarne) |
| [07:26:56] | * ratioweb has joined #aegir |
| [07:29:39] | * Deciphered has joined #aegir |
| [07:46:08] | * Deciphered has quit (Ping timeout: 240 seconds) |
| [07:47:00] | * Mfawzy has quit (Ping timeout: 260 seconds) |
| [07:48:21] | <gandhiano> | anarcat, is the fix from here - http://drupal.org/node/1079274 - commited on which version? |
| [07:48:23] | <hefring> | http://drupal.org/node/1079274 => Sites data get lost on migrate or clone / dont "spoke" the site => Provision, Code, normal, closed (fixed), 10 comments, 1 IRC mention |
| [07:49:21] | <anarcat> | i don't know |
| [07:50:14] | <gandhiano> | anarcat, I'm using 1.8 and it still seems that the supposed change of behavior (not "spoking" sites/domain folders) is not there |
| [07:50:59] | <gandhiano> | I'm asking you because you comment on the end of the thread that fix is committed (already in March 11 2011) |
| [07:51:44] | * eft has joined #aegir |
| [07:52:01] | <anarcat> | that is very possible |
| [07:52:02] | <anarcat> | i did? |
| [07:52:19] | <gandhiano> | http://drupal.org/node/1079274#comment-4198110 |
| [07:52:20] | <hefring> | http://drupal.org/node/1079274 => Sites data get lost on migrate or clone / dont "spoke" the site => Provision, Code, normal, closed (fixed), 10 comments, 2 IRC mentions |
| [07:52:31] | <anarcat> | hum |
| [07:52:31] | <gandhiano> | :) |
| [07:52:34] | <anarcat> | this is all quite confusing |
| [07:52:50] | <anarcat> | this may have changed again |
| [07:52:50] | <gandhiano> | yes, it is |
| [07:52:56] | <anarcat> | i don't really know |
| [07:53:01] | <anarcat> | what is the problem you are experiencing? |
| [07:53:15] | <anarcat> | i am more familiar with the 'pack' cluster module that i wrote than this cluster module |
| [07:53:21] | <gandhiano> | I'm fiddling with this for days to find a solution to my issue (user themes on remote host getting disabled on verify - http://drupal.org/node/1561102 ) |
| [07:53:21] | <hefring> | http://drupal.org/node/1561102 => User enabled themes on remote (spoke) get disabled on verify => Provision, Code, normal, active, 0 comments, 3 IRC mentions |
| [07:54:01] | <gandhiano> | but I still fail to completely understand what exactly is [supposed to be] done in the verify process |
| [07:54:27] | <anarcat> | interesting |
| [07:54:33] | <anarcat> | verify clears the cache, amongst other things |
| [07:55:37] | <gandhiano> | I managed to workaround the file sync permissions issue that arise (or any problem of overwritting user contributions) by symlinking to external folder |
| [07:56:14] | <gandhiano> | but then, on verify it outputs that it only finds 4 themes (the base ones) |
| [07:56:20] | <gandhiano> | not the user contrib |
| [07:56:37] | <gandhiano> | going to the site results in errors (no theme is loaded), after a while reverts to bartik |
| [07:56:45] | <gandhiano> | and user theme is disabled |
| [07:56:54] | <gandhiano> | can be re-enabled, works good - until next verify |
| [07:57:12] | <anarcat> | fun fun |
| [07:57:18] | <anarcat> | is the theme available on the hub? |
| [07:57:24] | <anarcat> | because that's probably where the verify runs |
| [07:57:50] | <gandhiano> | no - because if I put it there, he will try to sync to the remote, resulting in failure (permissions) |
| [07:58:16] | <gandhiano> | I did try to put it there as well |
| [07:58:54] | <gandhiano> | but did not result (it might however have to do with the incomplete verify, in case the rsync failure aborts the whole process) |
| [08:00:33] | * Mfawzy has joined #aegir |
| [08:00:43] | <anarcat> | gandhiano: i wouldn't allow sftp on the spokes, personnally |
| [08:00:54] | <gandhiano> | anarcat, where then? |
| [08:00:59] | <gandhiano> | hostmaster? |
| [08:01:01] | <anarcat> | yes |
| [08:01:11] | <anarcat> | gandhiano: exactly because #1079274 was committed - it means that those files get synced from the hub |
| [08:01:27] | <anarcat> | and anyways verify, migrate and so on are ran on the hub |
| [08:01:36] | <anarcat> | which will necessarily mean havoc if the spokes have different things |
| [08:01:40] | <gandhiano> | right |
| [08:01:47] | <anarcat> | now i admit this is weird and probably broken |
| [08:01:54] | <anarcat> | but this is the way it works unfortunately |
| [08:02:02] | <gandhiano> | originally I thought of having a spoke as a "safer harbour" for development |
| [08:02:07] | <anarcat> | yeah |
| [08:02:09] | <gandhiano> | leaving the hostmaster protected |
| [08:02:11] | <anarcat> | that's what i would think too :) |
| [08:02:22] | <anarcat> | the poblem is that spokes are too dumb |
| [08:02:37] | <anarcat> | drush (let alone provision) isn't installed over there, and commands run on the hub |
| [08:03:00] | <gandhiano> | well, that is a major advantage, from a security perspective |
| [08:03:02] | <anarcat> | you need to trust your sftp users unfortunately |
| [08:03:04] | <anarcat> | obviously |
| [08:03:19] | <anarcat> | it sucks |
| [08:03:43] | <anarcat> | i'd like 2.x spokes to be smarter |
| [08:03:55] | <anarcat> | and call provision and drush remotely |
| [08:04:05] | <anarcat> | so that there's a complete autonomous environment on remote servers |
| [08:04:16] | <gandhiano> | is there a release plan for 2.x? |
| [08:04:26] | <anarcat> | there's a roadmap, but no clearly defined dates |
| [08:04:43] | <gandhiano> | just trying to plan here how to move forward with our server architecture |
| [08:05:04] | <gandhiano> | because sftp on hostmaster would have to be something very temporary |
| [08:05:30] | <anarcat> | we (at koumbit.org) are using the pack module, which is more reliable because it doesn't have to rsync those files all over the place (but relies on NFS) |
| [08:05:46] | <anarcat> | and i'm looking at fixing the security issues related with giving access to modules to users |
| [08:05:54] | * ivanjaros has quit (Read error: Connection reset by peer) |
| [08:05:56] | <gandhiano> | yes, I'm following that one |
| [08:06:04] | <gandhiano> | what is the pack module? |
| [08:06:27] | <anarcat> | https://drupal.org/node/762138 |
| [08:06:45] | <anarcat> | gandhiano: the pack module is an alternate cluster implementation that doesn't sync anything in the platforms/ directory |
| [08:06:57] | <anarcat> | it only rsyncs the apache config, and assumes that you have NFS configured for the rest |
| [08:07:07] | <anarcat> | there's a debian package that does the NFS config for you |
| [08:07:37] | <gandhiano> | so platforms are a common folder on all spokes, shared over NFS? |
| [08:07:44] | <anarcat> | yep |
| [08:08:11] | <anarcat> | i don't really think that the current cluster model (apart from pack) makes much sense |
| [08:08:25] | <anarcat> | because say you have two slave servers - if someone uploads a file, where does it go? |
| [08:08:29] | <anarcat> | depends on the load balancer |
| [08:08:35] | <anarcat> | and then it won't be available on the other server |
| [08:08:42] | <anarcat> | you *need* to have some shared file storage |
| [08:08:52] | <gandhiano> | yes, this seems to make a lot of sense |
| [08:08:55] | <anarcat> | if you have that, why not share the whole platform and stop fucking around :) |
| [08:09:16] | <gandhiano> | at the moment they are all vservers of the same machine, so it even makes it more senseful |
| [08:10:00] | <anarcat> | yeah |
| [08:10:17] | <anarcat> | apparently, there's an issue with mysql grant with the pack module, but i haven't found any problems on my side so far |
| [08:10:40] | <anarcat> | https://drupal.org/node/1555398 |
| [08:12:07] | <gandhiano> | anarcat, would I be able to keep the same hostmaster and add pack over it, or is it a totally different package? |
| [08:13:36] | <anarcat> | you should be able to migrate easily |
| [08:13:43] | <anarcat> | i am not sure what the exact steps are, but it should be possible |
| [08:13:49] | <anarcat> | certainly you don't need to change the hostmaster |
| [08:13:49] | <gandhiano> | great |
| [08:14:04] | <anarcat> | if you can afford to rebuild the slave nodes, that may be easier, but i'm not sure it's necessary |
| [08:14:10] | <gandhiano> | I also don't have a large number of sites, everything is still being tested |
| [08:14:24] | <gandhiano> | I think I can, no sites in production yet |
| [08:15:16] | <anarcat> | see http://community.aegirproject.org/content/web-clusters |
| [08:15:23] | <gandhiano> | I'm there ;) |
| [08:15:46] | * brando753 has joined #aegir |
| [08:15:58] | <anarcat> | if you can contribute to that documentation, that would be very appreciated |
| [08:16:03] | <gandhiano> | getting an idea of my next days work - this absolutely makes more sense |
| [08:16:29] | * zombiebeard has quit (Quit: zombiebeard) |
| [08:16:41] | * patcon has quit (Quit: patcon) |
| [08:16:51] | <gandhiano> | ok, tks anarcat |
| [08:16:57] | <anarcat> | no problem |
| [08:17:04] | <gandhiano> | for putting me on the (hopefully) right path ;) |
| [08:17:09] | <anarcat> | i think the cluster module should be more explicitely deprecated |
| [08:18:01] | <gandhiano> | cluster module is the default thing allowing remote servers? |
| [08:18:12] | <anarcat> | well, it's the most obvious one :) |
| [08:18:17] | <anarcat> | because it's called "cluster" :P |
| [08:18:30] | * Mfawzy has quit (Ping timeout: 260 seconds) |
| [08:18:34] | <gandhiano> | actually I did not look for cluster, but somehow got to it |
| [08:20:41] | * cdracars has quit (Quit: cdracars) |
| [08:21:28] | <gandhiano> | anarcat, I came to it via this one: http://community.aegirproject.org/node/30 |
| [08:21:38] | <gandhiano> | no mention of pack, at all |
| [08:27:07] | <anarcat> | yeah, not sure how we should proceed here |
| [08:27:16] | <anarcat> | it's quite confusing because we juggle with 3 different concepts |
| [08:27:19] | <anarcat> | remote servers |
| [08:27:30] | <anarcat> | multiple redundant remote servers (aka cluster) |
| [08:27:40] | <anarcat> | multiple redundant remote servers, NFS-based (aka pack) |
| [08:27:44] | <anarcat> | i think cluster should die |
| [08:27:53] | <anarcat> | but remote servers should live |
| [08:27:59] | <anarcat> | they make sense on their own |
| [08:28:08] | <gandhiano> | well, we need at least some top-level documentation that makes clear that these 3 concepts exist |
| [08:28:35] | * ratioweb has quit (Remote host closed the connection) |
| [08:29:12] | * josh_k has quit (Quit: Holla!) |
| [08:29:50] | * wamilton has quit (Ping timeout: 255 seconds) |
| [08:33:45] | <anarcat> | yeah |
| [08:33:48] | <anarcat> | this may be the right page |
| [08:34:12] | <anarcat> | in fact, http://community.aegirproject.org/node/396 duplicates http://community.aegirproject.org/node/30 |
| [08:34:27] | <anarcat> | so #30 should be merged into #396 and #30 should be an index |
| [08:35:33] | <mig5> | i added some docs to the pack docs |
| [08:35:39] | <mig5> | based on my last 48 hours of experience :) |
| [08:35:56] | <anarcat> | awesome :) |
| [08:36:51] | <gandhiano> | mig5, where? |
| [08:37:05] | <mig5> | the clusters page |
| [08:37:05] | <mig5> | http://community.aegirproject.org/content/web-clusters |
| [08:37:18] | <mig5> | you won't really notice if you hadn't looked at it earlier than yesterday |
| [08:37:22] | <gandhiano> | ah, ok, just now or a while ago? |
| [08:37:23] | <gandhiano> | ok |
| [08:37:26] | <mig5> | but i dropped in a diagram of my setup, and added some more verbosity |
| [08:37:27] | <gandhiano> | that was my question :) |
| [08:37:51] | <mig5> | anarcat: so we have this neat corner case bug when we do mysql GRANTs (pack or otherwise) |
| [08:38:02] | <mig5> | we do the 'mysql -u intentnllyinvalid' or whatever |
| [08:38:19] | <mig5> | and based on the *expected* error bac k form the database server, we preg_match to fetch the IP address of the server requesting the grant |
| [08:38:30] | <mig5> | (why we don't just use the ip_addresses array in the context, i don't know) |
| [08:38:51] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [08:38:52] | <mig5> | but, turns out if you have multiple IP addresses on a server, on the same LAN, there's a chance that the 'mysql' command-line command, uses the wrong interface to 'bind' to |
| [08:39:04] | <mig5> | and so the IP that comes back is not the one that the webserver would later use to bootstrap drupal |
| [08:39:22] | <mig5> | and you can't do 'mysql --bind-address $proper-ip' unless you're using the special version of mysql, MySQL NDB cluster |
| [08:39:25] | <mig5> | epic fail |
| [08:39:55] | <anarcat> | wtf |
| [08:40:10] | <anarcat> | why would commandline mysql bind to a different IP than the webserver? |
| [08:40:21] | <anarcat> | this is the whole reason we are using this hack instead of guessing our ip |
| [08:40:36] | <mig5> | in my case, i am using mysql-mmm, and at the time. my webservers were also my two database servers |
| [08:40:51] | <mig5> | and mysql-mmm injects the floating IP onto the relevant 'writer' mysql server, directly with 'ip' |
| [08:41:00] | <mig5> | e.g it's not a configured interface on either server, but you can see it with 'ip addr sh' |
| [08:41:14] | <mig5> | and in my case, it was picking the floating IP to make the connection, instead of the webserver/database server's 'main' ip address |
| [08:41:30] | <mig5> | i worked around it by moving the webservers off so there's 2x web servers, 2x database servers. which is actually better design anyway |
| [08:41:33] | <anarcat> | but wait, the IP is insert on the mysql server |
| [08:41:39] | <anarcat> | inserted* |
| [08:41:49] | <anarcat> | oooh the webserver was *on* the db server |
| [08:41:50] | <mig5> | at the time, the mysql server was the same machine as the webserver |
| [08:41:52] | <mig5> | yes |
| [08:42:00] | <anarcat> | that is a good edge case indeed |
| [08:42:01] | <mig5> | and so the webserver was not getting the GRANT |
| [08:42:03] | <mig5> | yep |
| [08:42:13] | <anarcat> | we assume the mysql server doesn't move around too much |
| [08:42:17] | <anarcat> | wow, huge wtf |
| [08:42:34] | <anarcat> | will you write this in the summary? :) |
| [08:42:42] | <anarcat> | i have no idea how to fix this in the general case |
| [08:43:02] | <mig5> | yeah me neither. there's that ticket in the queue about it where i write my findings. and i put a 'caveats' section in the pack documentattion pointing to that ticket |
| [08:43:43] | <mig5> | it is of course better design to split off the database servers from webservers anyway probably. but i could see this occurring even for a single webserver + database server, on the same box or otherwise |
| [08:43:51] | <mig5> | basically, in any case where a webserver might have two IPs |
| [08:44:19] | <mig5> | i'm not sure the exact logic in mysql as to why it chose the floating IP.. it was on the same /24, not sure if it picked the 'lower' ip or what. |
| [08:45:01] | <anarcat> | it would be useful to write an issue summary for that issue |
| [08:45:05] | <anarcat> | i made a comment about this in the issue |
| [08:45:16] | * Tigeda has quit (Ping timeout: 272 seconds) |
| [08:45:46] | <mig5> | ok sure |
| [08:45:51] | <mig5> | just going through the morning's emails :) |
| [08:51:07] | * apintorso has joined #aegir |
| [08:51:07] | * apintorso has quit (Changing host) |
| [08:51:07] | * apintorso has joined #aegir |
| [08:53:08] | * drakythe has quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/) |
| [08:53:44] | * jeremyr has quit (Read error: Connection reset by peer) |
| [08:55:02] | * Deciphered has joined #aegir |
| [08:55:03] | * Deciphered has quit (Changing host) |
| [08:55:03] | * Deciphered has joined #aegir |
| [08:56:04] | <anarcat> | mig5: i am happy *you* stumbled upon this one instead of me ;) |
| [08:56:14] | <anarcat> | i wish i had more experience with mysql-mmmmmmm though |
| [08:56:15] | <mig5> | hah :) |
| [08:56:52] | <mig5> | yeah essentially it's a daemon that monitors two 'agent' daemons on each db server, and it just flicks the floating IP around between which ever should be the 'master' or 'writer' |
| [08:57:02] | <mig5> | and does things like arp cache flushing etc to handle that pixie dust |
| [08:57:31] | <anarcat> | fun |
| [08:59:15] | * AquaticDisorder has quit (Remote host closed the connection) |
| [09:01:24] | * dob_ has quit (Remote host closed the connection) |
| [09:01:48] | * dob_ has joined #aegir |
| [09:02:25] | * jhedstrom has quit (Read error: Operation timed out) |
| [09:03:46] | * gandhiano has quit (Remote host closed the connection) |
| [09:06:20] | * dob_ has quit (Ping timeout: 260 seconds) |
| [09:13:15] | * gboudrias has joined #aegir |
| [09:21:43] | * fatguylaughing has joined #aegir |
| [09:23:11] | * jhedstrom has joined #aegir |
| [09:30:14] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [09:30:22] | * obrienmd has joined #aegir |
| [09:34:50] | * Nickey has quit () |
| [09:38:54] | * blurryaegir has joined #aegir |
| [09:40:27] | <blurryaegir> | So I can set a custom settings.php for all sites on the server with global.inc, and individual specific sites with local.settings.php.. But how can I set a customs settings.php for all sites on a specific platform? |
| [09:40:28] | * danquah_ has joined #aegir |
| [09:40:28] | * danquah_ has quit (Client Quit) |
| [09:43:05] | * danquah has quit (Ping timeout: 260 seconds) |
| [09:44:13] | <Deciphered> | Anyone know of a module that replaces the Aegir 1.x tables with Views? I was tossing around the idea of doing a quick module for my own needs but don't want to duplicate efforts |
| [09:45:38] | <mig5> | i think we have views support written into Aegir 1.x already |
| [09:45:45] | <mig5> | but just no dependency on it |
| [09:46:27] | <Deciphered> | Cereal? I'll have to look into that. |
| [09:46:41] | <mig5> | we have a hosting.views.inc in modules/hosting/ and various views handlers in the various sub-features |
| [09:46:48] | <mig5> | i've never tried it |
| [09:47:04] | <Deciphered> | As an aside, and I will log this if it hasn't already been done, if you disable the 'Clients' feature it will break Aegir 1.8, can't create sites anymore :O |
| [09:47:52] | <mig5> | yeah i think clients is kind of baked into the system with or without that feature, it's messy. |
| [09:48:04] | <mig5> | can you open a ticket with the exact steps/symptoms if you get a chance? |
| [09:48:19] | <Deciphered> | Indeed... I didn't really need clients on my local machine so I figured I'd disable it.... |
| [09:48:31] | * rovo has joined #aegir |
| [09:48:32] | <Deciphered> | Will log it shortly |
| [09:49:40] | <rovo> | Hey Guys, I'm being really moronic about something, I just need a clue... I am trying to host a single page outside of my Aegir/Nginx setup, in the www directory... but I can't wrap my head around what it is I need to tweak to allow that. |
| [09:51:28] | * Tigeda has joined #aegir |
| [09:52:26] | * apintorso has left #aegir ("pax") |
| [09:52:47] | * berniecram has joined #aegir |
| [09:56:01] | * penyaskito has joined #aegir |
