| [10:00:45] | * Zelfje has joined #aegir |
| [10:02:04] | * gusaus has quit (Quit: gusaus) |
| [10:04:12] | * gusaus has joined #aegir |
| [10:04:18] | * SeanBannister has quit (Ping timeout: 255 seconds) |
| [10:04:19] | * Guest72069 has joined #aegir |
| [10:04:20] | * Guest72069 is now known as SeanBannister |
| [10:07:41] | * Pasqualle has quit (Ping timeout: 240 seconds) |
| [10:08:35] | * MadTBone has quit (Quit: Leaving) |
| [10:09:47] | * MadTBone has joined #aegir |
| [10:13:41] | * patcon|myplanet has quit (Ping timeout: 240 seconds) |
| [10:14:33] | * j0nathan has joined #aegir |
| [10:15:49] | * MadTBone has quit (Ping timeout: 250 seconds) |
| [10:23:08] | * sethvincent has quit (Ping timeout: 252 seconds) |
| [10:28:34] | * nomad411 has quit (Quit: Back to the unreal world...) |
| [10:29:53] | * sk33lz has joined #aegir |
| [10:34:22] | * j0nathan has quit (Ping timeout: 258 seconds) |
| [10:35:26] | * j0nathan has joined #aegir |
| [10:37:21] | * ryanarmstrong has quit (Quit: Leaving.) |
| [10:55:25] | * samhassell has joined #aegir |
| [11:01:11] | * gusaus has quit (Quit: gusaus) |
| [11:04:23] | * sk33lz has quit (Quit: Leaving.) |
| [11:09:45] | * obrienmd has quit (Quit: Leaving.) |
| [11:22:46] | * q-rban_ has joined #aegir |
| [11:25:18] | * q-rban has quit (Ping timeout: 276 seconds) |
| [11:25:59] | * obrienmd has joined #aegir |
| [11:26:42] | * obrienmd has quit (Client Quit) |
| [11:38:58] | * sk33lz has joined #aegir |
| [11:39:40] | * AquaticDisorder has quit (Remote host closed the connection) |
| [11:46:14] | * q-rban_ has quit (Ping timeout: 246 seconds) |
| [11:47:08] | * q-rban has joined #aegir |
| [12:00:22] | * EclipseGc has quit (Quit: EclipseGc) |
| [12:21:04] | * bdiddy has joined #aegir |
| [12:28:49] | * ezra-g has joined #aegir |
| [12:28:50] | * ezra-g has quit (Client Quit) |
| [12:41:48] | * obrienmd has joined #aegir |
| [12:53:30] | * samhassell has quit (Ping timeout: 252 seconds) |
| [20:49:56] | * hefring has joined #aegir |
| [20:50:35] | <jpstrikesback> | anyone seen an issue where the queue dies without any errors, but wont startup again? |
| [20:52:21] | <jpstrikesback> | on the CLI all I can get is a "Could not startup." |
| [21:27:34] | <jpstrikesback> | nm…silly mem setting done at 4am |
| [21:41:50] | * lukus has joined #aegir |
| [21:47:00] | * AquaticDisorder has joined #aegir |
| [21:55:13] | * that0n3guy has quit (Ping timeout: 246 seconds) |
| [21:55:26] | * that0n3guy has joined #aegir |
| [21:55:26] | * that0n3guy has quit (Read error: Connection reset by peer) |
| [22:03:41] | * jonhattan has quit (Quit: llevaré una chaquete de guns'n'roses) |
| [22:07:35] | * psynaptic is now known as psynaptic|lunch |
| [22:17:50] | * yareckon has joined #aegir |
| [22:19:26] | <ngnp> | Is this a useful feature http://drupal.org/node/1190326 ? |
| [22:19:26] | <hefring> | http://drupal.org/node/1190326 => Running hoster specific tasks like git pull for development => Hosting, Miscellaneous, normal, active, 0 comments, 1 IRC mention |
| [22:22:45] | <ngnp> | bot? |
| [22:23:12] | <darthsteven> | ngnp: If you're doing deployments using something like Jenkins to automate tasks is much better than cron imho |
| [22:24:01] | * psynaptic|lunch is now known as psynaptic |
| [22:24:19] | <ngnp> | darthsteven: we want to config stuff through the site node ... is that possibe then? |
| [22:24:54] | <darthsteven> | ngnp: probably possible, might not be the best way though. Can't say I've thought about it too much |
| [22:25:02] | * ngnp reading http://drupal.org/node/968454 |
| [22:29:28] | * MadTBone has joined #aegir |
| [22:29:51] | * Artusamak_afk is now known as Artusamak |
| [22:32:14] | * q-rban has quit (Remote host closed the connection) |
| [22:32:28] | * q0rban has joined #aegir |
| [22:36:19] | * kvanderw has joined #aegir |
| [22:38:51] | * Pasqualle has joined #aegir |
| [22:46:41] | * MadTBone has quit (Ping timeout: 240 seconds) |
| [22:50:37] | <shrop> | deetergp mcteague: no roberto :( |
| [22:50:52] | <shrop> | kind of wish I didn't have to worry about that |
| [22:51:14] | <shrop> | sorry folks.. wrong channel |
| [23:02:50] | * Artusamak is now known as Artusamak_afk |
| [23:10:53] | * SocialNicheGuru has joined #aegir |
| [23:15:06] | * SocialNicheGuru has quit (Ping timeout: 244 seconds) |
| [23:15:48] | * SocialNicheGuru has joined #aegir |
| [23:15:55] | * omega8cc has joined #aegir |
| [23:16:53] | * troubador has joined #aegir |
| [23:20:16] | * SocialNicheGuru has quit (Ping timeout: 244 seconds) |
| [23:21:47] | * SocialNicheGuru has joined #aegir |
| [23:28:28] | * obrienmd has joined #aegir |
| [23:34:25] | * Artusamak_afk is now known as Artusamak |
| [23:36:08] | * obrienmd has quit (Quit: Leaving.) |
| [23:52:53] | * beautifulmind has joined #aegir |
| [00:14:44] | * nick_santa has quit (Quit: nick_santa) |
| [00:20:26] | * j0nathan has quit (Quit: ¡Hasta luego!) |
| [00:25:08] | * beautifulmind has quit (Quit: Leaving.) |
| [00:28:33] | * obrienmd has joined #aegir |
| [00:41:53] | * mikl has quit (Ping timeout: 240 seconds) |
| [00:50:54] | * ezra-g has joined #aegir |
| [00:55:18] | * Brandonian has joined #aegir |
| [00:56:19] | * nlisgo has joined #aegir |
| [00:59:42] | * ezra-g has quit (Quit: COD: The Conference Organizing Distribution. http://usecod.com) |
| [01:00:25] | * scientist has joined #aegir |
| [01:00:45] | * nlisgo has quit (Client Quit) |
| [01:04:23] | * lolmaus has joined #aegir |
| [01:14:08] | * Met4physica has joined #aegir |
| [01:14:16] | <Met4physica> | How do I set barracuda to use aegir 1.1? |
| [01:18:14] | * j0nathan has joined #aegir |
| [01:24:53] | <omega8cc> | Met4physica: it installs Aegir 1.1 already since 1.0-boa-T-8.1 and it is 1.0-boa-T-8.7 now, see http://bit.ly/newboa |
| [01:25:03] | * ngnp has quit (Quit: ngnp) |
| [01:25:08] | <Met4physica> | thank you very much |
| [01:25:37] | <omega8cc> | np |
| [01:26:04] | <Met4physica> | is it possible to set it up on a linode without a real domain name as the FQDN? |
| [01:26:20] | * jonhattan has joined #aegir |
| [01:26:26] | * psynaptic is now known as psynaptic|away |
| [01:26:27] | <Met4physica> | using something like li333-83.members.linode.com |
| [01:26:48] | <omega8cc> | no |
| [01:27:23] | <omega8cc> | you need your own wildcard domain pointed there and used in the configuration for the hostname and Aegir url |
| [01:28:09] | <omega8cc> | the li333-83.members.linode.com is really a dummy reverse name you should never use in anything |
| [01:28:10] | <Met4physica> | i am attempting to move someone to aegir before i can actually xfer their domain |
| [01:28:17] | <Met4physica> | perhaps there is an easy way to accomplish this |
| [01:30:23] | <Met4physica> | I could enter in the FQDN for their name and just change my local /etc/hosts file to point to it? |
| [01:30:34] | <omega8cc> | no, it is not possible, or rather not recommended to try this as it will confuse Aegir and you so you will end up re-installing it from scratch, while domains are really cheap these days so you can point any domain there and later add as an alias the final domain to the hostmaster site |
| [01:33:48] | <omega8cc> | well, it could be done if you will set the database to listen on localhost and the use /etc/hosts trick to access the web url, but then database grants will use FQDN you entered and you will have to use the /etc/hosts trick also on the server to get it to work |
| [01:35:36] | <omega8cc> | so yes, it is possible, if you know what are you doing ;) |
| [01:36:14] | <Met4physica> | ok i am not that smart probably |
| [01:36:19] | <Met4physica> | i will just buy a cheapo domain |
| [01:36:25] | <Met4physica> | and do it the 'normal way' |
| [01:37:17] | <omega8cc> | that will save you a lot of headaches later, I think |
| [01:38:13] | * V4ntage_work has quit (Remote host closed the connection) |
| [01:39:29] | <Met4physica> | i have been successfully using aegir for a long time, its just been awhile since redoing it |
| [01:39:42] | <Met4physica> | my firewall script has caught about a million ssh login attempts |
| [01:39:55] | <Met4physica> | i know i should change the port,but i get a sick sense of humor out of it |
| [01:43:37] | <omega8cc> | lol |
| [01:45:46] | <Met4physica> | might i ask if the ssl stuff has been made easier ? i remember that is why i dumped all my clients off my server and moved them to crappier hosts |
| [01:46:20] | <Met4physica> | i made a bunch of sites that ran very fast, and then one got moved to www.advomatic.com and it was slow as can be, they charge extra $ for caching addons, etc. etc. |
| [01:46:52] | <omega8cc> | easier in what sense? |
| [01:47:35] | <omega8cc> | I think it is easy.. |
| [01:57:53] | * mikl has joined #aegir |
| [01:57:53] | * mikl has quit (Changing host) |
| [01:57:53] | * mikl has joined #aegir |
| [01:59:10] | * Zelfje has quit (Ping timeout: 246 seconds) |
| [02:16:04] | * recidive has joined #aegir |
| [02:31:24] | * siliconmeadow has quit (Ping timeout: 255 seconds) |
| [02:36:29] | * patcon|myplanet has joined #aegir |
| [02:37:58] | * sk33lz has joined #aegir |
| [02:38:36] | * Zelfje has joined #aegir |
| [02:39:07] | * sdrycroft has quit (Quit: Leaving.) |
| [02:42:19] | * josh_k has joined #aegir |
| [02:42:19] | * josh_k has quit (Changing host) |
| [02:42:19] | * josh_k has joined #aegir |
| [02:52:12] | * kepford has joined #aegir |
| [02:55:21] | <anarcat> | darthsteven: ping |
| [02:56:23] | <darthsteven> | anarcat: Pong! |
| [02:57:10] | <darthsteven> | Can I help? |
| [02:57:16] | <anarcat> | darthsteven: i have looked at your daemon, and it's great! |
| [02:57:26] | <anarcat> | darthsteven: i have submitted a patch for review that would remove the need for a supervisor |
| [02:57:37] | <anarcat> | i submitted another closely related patch, i'd like to get feedback |
| [02:57:42] | <anarcat> | i'll deploy this in production now |
| [02:58:35] | <anarcat> | darthsteven: also, i believe we should strive to merge this into 2.x and even 1.x, and make it default in 2.x |
| [03:00:28] | <darthsteven> | Indeed, I'm walking home at the moment, but thinking that a locked, cron initiated deamon would maintain the status quo but make it a whole lot faster |
| [03:01:08] | <anarcat> | locked? cron? |
| [03:01:21] | <anarcat> | also, i did have a question |
| [03:01:27] | <darthsteven> | I.e you call the drush command as many times as you like, as often as you like |
| [03:01:36] | <darthsteven> | But we keep one process going |
| [03:01:45] | <anarcat> | darthsteven: the queue runner seems to run only the tasks queue, and not the cron queue, is that correct? if so how does the cron queue get ran? |
| [03:01:51] | <darthsteven> | (restarts every hour) |
| [03:02:00] | <anarcat> | i see |
| [03:02:07] | <darthsteven> | Znc does this |
| [03:02:23] | <darthsteven> | Makes it exacty the same as now to set up |
| [03:02:46] | <anarcat> | ah |
| [03:02:46] | <darthsteven> | In fact hosting dispatch could ensure a deamon was running |
| [03:02:50] | <anarcat> | yeah |
| [03:02:55] | <anarcat> | well, this is where i was going with this |
| [03:02:58] | <anarcat> | this should be merged in hosting |
| [03:03:09] | <anarcat> | and it could then just be a setting in the frontend |
| [03:03:20] | <anarcat> | and hosting-dispatch would just keep the hosting-tasks queue daemon running |
| [03:03:22] | <darthsteven> | Queue runner only cares about tasks |
| [03:03:25] | <anarcat> | ok |
| [03:03:28] | <darthsteven> | Not cron or other queues |
| [03:03:48] | <anarcat> | shouldn't it be extended to do whatever hosting-dispatch does? |
| [03:04:14] | <darthsteven> | That would require major refactoring |
| [03:04:21] | <anarcat> | yeah |
| [03:04:23] | <darthsteven> | Ghetto queue 2.0? |
| [03:04:32] | <anarcat> | absolutely, that's what i mean - merge in 2.x |
| [03:04:38] | <anarcat> | we could merge it as is, then refactor stuff around |
| [03:04:41] | <anarcat> | i'd like to do that |
| [03:04:51] | <anarcat> | but in the meantime, i'll use it as is in production on 1.x |
| [03:04:54] | <darthsteven> | Me too |
| [03:04:59] | <anarcat> | and i'd like to merge those patches in :) |
| [03:05:25] | <darthsteven> | Well, when I get home... |
| [03:05:32] | * grugnog has joined #aegir |
| [03:05:39] | <anarcat> | darthsteven: cool |
| [03:07:30] | <darthsteven> | I'm not sure the other queues need this sort of latency though? I don't care if cron doesn't run for 60 seconds... |
| [03:07:34] | <jpstrikesback> | is this expected behaviour: When using remote webservers (which must 1st be setup on the hostmaster) - Upon reverification, it overwrites the remote servers files with the hostmaster orig copy? |
| [03:08:29] | <anarcat> | darthsteven: maybe you're right, but that's for "cron"-like queues, maybe there's going to be other batch queues where latency is important |
| [03:08:41] | <anarcat> | jpstrikesback: common problem, see issue queue :) |
| [03:09:06] | <anarcat> | darthsteven: one thing i was wondering - i didn't touch the crontab so hosting-dispatch is still running, does that mean i have two task dispatchers running at the same time now? |
| [03:09:10] | <jpstrikesback> | k, just thought maybe that was the desired behaviour (i.e. for a Stag->Prod setup) |
| [03:10:11] | <jpstrikesback> | tho I guess that doesn't make sense since you can only access one at a time |
| [03:10:29] | <anarcat> | darthsteven: btw, the patches i sent actually conflict with each other, i have a branch here that merges both, and if you give me access to your repo, i'd gladly push that branch for your perusal :) |
| [03:11:08] | <darthsteven> | anarcat: The deamon will almost always win, besides the install hook disabled the normal task queue |
| [03:11:25] | <anarcat> | darthsteven: awesome |
| [03:12:23] | <darthsteven> | If it doesn't win, then there's a safeguard, if that fails we need locking on running tasks |
| [03:12:32] | <darthsteven> | :) |
| [03:13:33] | <anarcat> | darthsteven: in the issue queue you said you merged the restart stuff, that wasn't on the master branch? |
| [03:13:53] | <anarcat> | darthsteven: if so, i recomment you delete the master branch to avoid confusion |
| [03:14:54] | <darthsteven> | There's a master branch? Urgh, there shouldn't be... |
| [03:15:46] | <anarcat> | ok |
| [03:16:13] | <anarcat> | ok, i see the code |
| [03:16:19] | <anarcat> | backend_fork() is not what you want |
| [03:16:29] | <anarcat> | that would be basically creating a slow-burning fork bomb |
| [03:16:35] | <anarcat> | forking new processes forever |
| [03:16:46] | <anarcat> | i think what you want is exec |
| [03:17:06] | <anarcat> | the good thing is the original process dies so you don't get duplicate processes |
| [03:17:14] | <anarcat> | but the problem is that you loose the controling terminal |
| [03:17:24] | <anarcat> | so say i run the dispatcher, i have the process in front of me |
| [03:17:39] | <anarcat> | after some time (apparently random for the newbie), the process will stop and fork in the background |
| [03:17:42] | <anarcat> | which could be confusing |
| [03:17:50] | <anarcat> | and will certainly wreck havoc with supervisord |
| [03:18:02] | <anarcat> | the original approach was still compatible with supervisord in fact |
| [03:18:34] | <anarcat> | darthsteven: but i'll let you come home alraedy :) |
| [03:18:56] | <darthsteven> | I am home now, and on it |
| [03:19:07] | <anarcat> | ah ok :) |
| [03:19:13] | <darthsteven> | hmm... |
| [03:19:30] | <darthsteven> | yeah, I don't like forking either |
| [03:19:32] | <anarcat> | so in your code, the daemon is unconditionnally restarted, right? |
| [03:19:39] | <darthsteven> | no |
| [03:19:39] | <anarcat> | so we can get rid of the variable in the frontend |
| [03:19:43] | <darthsteven> | you need an option |
| [03:20:06] | <anarcat> | oh right sry |
| [03:20:21] | <anarcat> | ok, so here's my (new) take on this |
| [03:20:26] | <darthsteven> | master is gone |
| [03:20:28] | <anarcat> | i think we should just try to restart anyways |
| [03:20:29] | <anarcat> | cool |
| [03:20:33] | <darthsteven> | 6.x-1.x now has my latest code |
| [03:20:38] | <darthsteven> | right |
| [03:20:40] | <darthsteven> | go! |
| [03:21:12] | <darthsteven> | (and you now have commit access) |
| [03:21:40] | <anarcat> | ok |
| [03:21:48] | <anarcat> | should i merge in the pcntl_exec() stuff? |
| [03:22:18] | <darthsteven> | no |
| [03:22:24] | <darthsteven> | not for the moment I think |
| [03:22:28] | <darthsteven> | so... |
| [03:22:34] | <darthsteven> | what's up with using fork |
| [03:22:44] | <darthsteven> | doesn't the process that did the forking 'go away'? |
| [03:22:51] | <anarcat> | yes, it goes away |
| [03:22:56] | <anarcat> | but that's exactly the issue here |
| [03:22:58] | <darthsteven> | ? |
| [03:23:02] | <anarcat> | you don't *want* it to go away |
| [03:23:05] | <anarcat> | you want to *replace* it |
| [03:23:06] | <darthsteven> | ah right |
| [03:23:11] | <anarcat> | so |
| [03:23:14] | <anarcat> | if you replace it |
| [03:23:14] | <darthsteven> | because you have no control over it |
| [03:23:20] | <anarcat> | you don't need --auto-restart |
| [03:23:26] | <darthsteven> | ah righto |
| [03:23:28] | <anarcat> | because then from supervisord's POV, nothing changed |
| [03:23:37] | <anarcat> | it's just this thing that need tobe restarted if it fails |
| [03:23:46] | <anarcat> | it just that supervisord becomes optional, all of a sudden :) |
| [03:23:51] | <anarcat> | because the daemon behaves |
| [03:24:12] | <anarcat> | i'm not sure i'm being very clear here :) |
| [03:24:17] | <darthsteven> | you are |
| [03:24:44] | * darthsteven is thinking... |
| [03:25:14] | <anarcat> | so there are two things here: |
| [03:25:18] | * FatGuyLaughing has joined #aegir |
| [03:25:23] | <anarcat> | 1. using pcntl_exec() instead of fork() |
| [03:25:28] | <anarcat> | 2. making restart the default |
| [03:25:35] | <anarcat> | my priority is 1, i can live with an option for now |
| [03:26:00] | <anarcat> | but i believe that the way it's setup, we don't even need the option |
| [03:31:09] | <darthsteven> | okay, but for people without supervisor that doesn't handle the case where PHP dies |
| [03:31:10] | <darthsteven> | but |
| [03:31:35] | <darthsteven> | If we add locking on the queue, so that you can only have one running at a time |
| [03:31:44] | <anarcat> | yeah |
| [03:31:47] | <darthsteven> | then we can run it from crontab |
| [03:31:51] | <anarcat> | right now, i don't care about the case where php dies |
| [03:31:56] | <darthsteven> | sure... |
| [03:32:07] | <anarcat> | i'll handle it when i stumble into it, and then, maybe not by restarting it automatically, but maybe by fixing the damn php bug :P |
| [03:32:11] | <anarcat> | or the drupal bug |
| [03:32:14] | <darthsteven> | heh |
| [03:32:15] | <anarcat> | failing that, crontab |
| [03:32:20] | <darthsteven> | okay |
| [03:32:36] | <darthsteven> | if you get pcntl working then I'll merge it in |
| [03:32:42] | <anarcat> | ok |
| [03:32:50] | <anarcat> | i can make it optional if it there's, try pcntl otherwise fork |
| [03:32:50] | <darthsteven> | apologies for my drush backend fork fork |
| [03:33:35] | * kepford_ has joined #aegir |
| [03:34:44] | <anarcat> | darthsteven: i pushed the autorestart branch for your perusal |
| [03:34:53] | <darthsteven> | cool |
| [03:35:23] | <anarcat> | actually, now it's pushed |
| [03:36:45] | * kepford has quit (Ping timeout: 264 seconds) |
| [03:36:45] | * kepford_ is now known as kepford |
| [03:38:10] | <anarcat> | Warning: The command callback function drush_hosting_queue_runner has a deprecated name. It [notice] |
| [03:38:13] | <anarcat> | must begin with drush_hosting_queue_runner_. Skipping hook functions. [0.18 sec, 28.02 MB] |
| [03:40:15] | <darthsteven> | yup |
| [03:40:30] | <darthsteven> | drush bug in 4.4 means that I need to cause that error message |
| [03:40:36] | <darthsteven> | < 4.4 |
| [03:40:48] | <anarcat> | ah |
| [03:40:58] | <darthsteven> | otherwise the drush command is called twice! |
| [03:41:22] | <anarcat> | aah that's the thing |
| [03:41:28] | <anarcat> | < 4.4 or <= 4.4? |
| [03:44:09] | <anarcat> | darthsteven: if i have small fixes like translation (t()) or similar, can i just push it? |
| [03:44:18] | <anarcat> | darthsteven: what liberty do i have to commit my own changes, basically? |
| [03:44:49] | <darthsteven> | enjoy your commit access is all I'll say ;) |
| [03:45:25] | <anarcat> | ok :) |
| [03:45:27] | <anarcat> | thx |
| [03:45:31] | <anarcat> | same for aegir btw :P |
| [03:46:21] | <darthsteven> | I'd rather we didn't need to use d()->name in there, we shouldn't need to I don't think... |
| [03:46:40] | <darthsteven> | hmm...maybe we do though... |
| [03:46:46] | <darthsteven> | ah I don't know |
| [03:47:00] | <anarcat> | you scared of d()->name? :) |
| [03:47:05] | <anarcat> | i think it should be okay |
| [03:49:50] | * ronia has joined #aegir |
| [03:54:01] | <noecc> | On a remote server verify what is the command being executed for rsync? |
| [03:54:32] | <anarcat> | rah |
| [03:54:43] | <anarcat> | darthsteven: in the refactoring, d()->name stopped working, all of a sudden |
| [03:55:24] | <anarcat> | wow, d() returns null, wtf |
| [03:56:02] | <anarcat> | hell |
| [03:56:08] | <anarcat> | it's because we're in a callback |
| [03:56:16] | <anarcat> | if i comment out the callback => line, it works again |
| [03:56:22] | * yareckon has quit (Quit: yareckon) |
| [03:58:32] | <anarcat> | darthsteven: // We don't need the callback defined for Drush > 4.4 |
| [03:58:37] | <anarcat> | darthsteven: don't you mean >= 4.4? |
| [03:58:46] | <darthsteven> | :no |
| [03:58:56] | <darthsteven> | I mean from 4.5 onwards |
| [03:59:06] | <darthsteven> | 4.5 hasn't been released yet though :) |
| [03:59:32] | <darthsteven> | without the callback the drush command will get executed twice |
| [03:59:41] | <darthsteven> | because of hosting.drush.inc |
| [03:59:51] | <darthsteven> | drush matches the function name twice |
| [03:59:55] | <darthsteven> | and calls it twice |
| [03:59:58] | <darthsteven> | it's madness! |
| [04:00:13] | * amateescu has quit (Quit: Leaving) |
| [04:01:52] | * ronia has left #aegir () |
| [04:06:13] | <anarcat> | i don't see that happening here |
| [04:07:50] | <darthsteven> | well, its a real bug! was fixed in drush this morning, there's a link in the code |
| [04:08:13] | <anarcat> | oh yes, i see the bug |
| [04:08:14] | <anarcat> | wow, weird |
| [04:09:21] | <anarcat> | alright, so i cleaned up the functions a lot |
| [04:09:23] | <anarcat> | but somehow |
| [04:09:29] | <anarcat> | the exec stopped working |
| [04:09:30] | <anarcat> | grr |
| [04:10:23] | <anarcat> | Restarting queue runner with /usr/share/drush/drush.php @hostmaster hosting-queue-runner [notice] |
| [04:10:26] | <anarcat> | --debug --auto-restart [2.19 sec, 28.03 MB] |
| [04:10:27] | <anarcat> | wtf |
| [04:10:29] | <anarcat> | then it just dies |
| [04:13:41] | <anarcat> | okay, i see what's gonig on |
| [04:13:52] | <anarcat> | the child drush process doesn't see --auto-restart (or --debug for that matter) |
| [04:13:57] | <anarcat> | that's ... weird.. |
| [04:16:49] | <darthsteven> | I've got some code almost working |
| [04:16:49] | <darthsteven> | give me 20... |
| [04:19:09] | * kepford has quit (Read error: Connection reset by peer) |
| [04:19:29] | * kepford has joined #aegir |
| [04:24:54] | <anarcat> | darthsteven: i've got it working here |
| [04:26:03] | <darthsteven> | cool! |
| [04:28:58] | <darthsteven> | mine doesn't work, but I've pushed it anyway |
| [04:29:36] | * mrconnerton has joined #aegir |
| [04:31:14] | <anarcat> | crap :) |
| [04:31:18] | <anarcat> | go conflicts :P |
| [04:32:57] | <darthsteven> | Well, don't merge mine! |
| [04:33:04] | <darthsteven> | It doesn't work |
| [04:33:12] | <darthsteven> | Anyway gotta run |
| [04:33:27] | <darthsteven> | Might e back online in a couple of hours... |
| [04:34:10] | <anarcat> | ok |
| [04:34:34] | <anarcat> | ooh okay, you have a branch |
| [04:34:44] | <anarcat> | i updated my branch |
| [04:35:43] | <anarcat> | darthsteven: i will test that branch in production now |
| [04:55:35] | * kvanderw is now known as kvanderw|afk |
| [05:04:43] | * gusaus has joined #aegir |
| [05:04:45] | <anarcat> | darthsteven: okay, i have another question: why do we pause before executing tasks? |
| [05:05:59] | * SocialNicheGuru has quit (Remote host closed the connection) |
| [05:11:01] | * josh_k has quit (Quit: This computer has gone to sleep) |
| [05:13:17] | * josh_k has joined #aegir |
| [05:19:22] | * josh_k has quit (Ping timeout: 250 seconds) |
| [05:21:41] | <anarcat> | darthsteven: the daemon is now running, and seems to be stable, with the latest code i pushed in the autostart branch, i'll let you merge it in if you're satisfied |
| [05:22:48] | <anarcat> | darthsteven: note that i have turned some drush_log() calls into watchdog() (which still show up in drush --debug) so that the frontend has an idea of what's going on with the dispatcher |
| [05:24:20] | * josh_k has joined #aegir |
| [05:24:33] | * noecc has left #aegir ("pax") |
| [05:28:33] | * gusaus_ has joined #aegir |
| [05:28:41] | * gusaus has quit (Ping timeout: 240 seconds) |
| [05:28:42] | * gusaus_ is now known as gusaus |
| [05:29:27] | * noecc has joined #aegir |
| [05:34:03] | * josh_k has quit (Ping timeout: 276 seconds) |
| [05:36:05] | * dnotes has joined #aegir |
| [05:37:03] | * kvanderw|afk has quit (Quit: Computer has gone to sleep) |
| [05:38:39] | * josh_k has joined #aegir |
| [05:38:44] | <dnotes> | where is the aegir_root password stored? |
| [05:38:59] | * josh_k has quit (Client Quit) |
| [05:41:18] | * scientist has quit (Ping timeout: 258 seconds) |
| [05:47:03] | * jonhattan has quit (Read error: Connection reset by peer) |
| [05:51:27] | * ezra-g has joined #aegir |
| [05:59:43] | * henk_ has joined #aegir |
| [06:00:30] | * ezra-g is now known as ezra-g|brb |
| [06:03:34] | * scientist has joined #aegir |
| [06:03:52] | * ryanarmstrong has joined #aegir |
| [06:04:25] | * dnotes has quit (Quit: Leaving.) |
| [06:05:09] | <darthsteven> | anarcat: we pause because we have no way to add a task with arguments and execute it immediately at the moment, hosting_backup_queue does this |
| [06:05:31] | * josh_k has joined #aegir |
| [06:05:31] | * josh_k has quit (Changing host) |
| [06:05:31] | * josh_k has joined #aegir |
| [06:05:37] | <darthsteven> | anarcat: so we pause for a second and let it execute it |
| [06:05:53] | * lolmaus has quit () |
| [06:14:38] | * kvanderw|afk has joined #aegir |
| [06:18:07] | * kvanderw|afk has quit (Client Quit) |
| [06:21:37] | * dob_ has joined #aegir |
| [06:22:16] | * kvanderw has joined #aegir |
| [06:22:32] | <dob_> | why is aegir always setting the s-bit when creating a site? |
| [06:23:02] | * kvanderw has quit (Client Quit) |
| [06:23:09] | <dob_> | drwxrws--- 13 aegir www-data 4096 2011-06-16 15:38 files |
| [06:23:20] | * scientist has quit (Ping timeout: 263 seconds) |
| [06:23:57] | * scientist has joined #aegir |
| [06:27:21] | * ezra-g|brb is now known as ezra-g |
| [06:31:33] | <joestewart> | dob_: that s helps file creation maintain group ownership |
| [06:32:15] | * recidive has quit (Ping timeout: 252 seconds) |
| [06:32:20] | <dob_> | okay, strange, if i have trouble when writing to some directorys, no problems if i remove that setgid bit. |
| [06:40:10] | * willj has quit (Read error: Connection reset by peer) |
| [06:41:39] | * willj has joined #aegir |
| [06:42:08] | * ezra-g has quit (Quit: COD: The Conference Organizing Distribution. http://usecod.com) |
| [06:42:22] | * kvanderw has joined #aegir |
| [06:44:33] | * recidive has joined #aegir |
| [06:45:42] | * kvanderw has quit (Client Quit) |
| [06:48:00] | * scientist has quit (Ping timeout: 258 seconds) |
| [06:48:17] | * vcardoso has joined #aegir |
| [06:50:11] | * wiifm has quit (Read error: Connection reset by peer) |
| [06:52:15] | * psynaptic|away is now known as psynaptic |
| [06:54:42] | * FatGuyLaughing has quit (Quit: FatGuyLaughing) |
| [06:54:58] | * wiifm has joined #aegir |
| [06:55:04] | * FatGuyLaughing has joined #aegir |
| [06:56:49] | * Irishgringo has joined #aegir |
| [07:01:13] | * kvanderw has joined #aegir |
| [07:13:29] | * FatGuyLaughing has quit (Quit: FatGuyLaughing) |
| [07:13:36] | * dob_ has quit (Quit: dob_) |
| [07:15:07] | * recidive has quit (Read error: Connection reset by peer) |
| [07:15:21] | * kvanderw_ has joined #aegir |
| [07:15:33] | * kvanderw has quit (Ping timeout: 246 seconds) |
| [07:17:03] | <hefring> | community => New Site Creation Issue => http://community.aegirproject.org/discuss/new-site-creation-issue |
| [07:17:46] | * recidive has joined #aegir |
| [07:30:39] | * kvanderw_ has quit (Quit: Computer has gone to sleep) |
| [07:30:55] | <anarcat> | darthsteven: i don't get it |
| [07:32:15] | * kvanderw has joined #aegir |
| [07:32:29] | * Slydder has joined #aegir |
| [07:38:24] | <anarcat> | darthsteven: i guess i'd need to read hosting_backup_queue's code |
| [07:39:08] | <anarcat> | darthsteven: also, i've been thinking, and i wonder i wouldn't add a X tasks maximum for the daemon, because that's really what eats the memory (guessing), not the main loop |
| [07:40:54] | * vcardoso has quit (Ping timeout: 258 seconds) |
| [07:46:22] | <anarcat> | darthsteven: that being said, i think we should just merge the autostart branch in the mainline now |
| [07:46:31] | <anarcat> | darthsteven: note, however, that this means it automatically restarts by default |
| [07:48:08] | * Pasqualle has quit (Ping timeout: 276 seconds) |
| [08:08:09] | * FransK has quit (Ping timeout: 260 seconds) |
| [08:10:53] | * Brandonian has quit (Quit: Brandonian) |
| [08:11:12] | * Met4physica has quit (Read error: Operation timed out) |
| [08:19:31] | * kvanderw has quit (Quit: Computer has gone to sleep) |
| [08:23:07] | * kvanderw has joined #aegir |
| [08:26:34] | * scientist has joined #aegir |
| [08:34:39] | * alex87 has joined #aegir |
| [08:35:17] | <alex87> | i have a number of sites running on the same platform, which means the module count is quite high. is this the wrong approach? |
| [08:35:28] | * Slydder has quit (Quit: Leaving.) |
| [08:38:22] | * henk_ has quit (Remote host closed the connection) |
| [08:43:11] | * boztek has joined #aegir |
| [08:59:54] | * patcon|myplanet has quit (Read error: Operation timed out) |
| [09:00:03] | * kvanderw has quit (Quit: Computer has gone to sleep) |
| [09:01:19] | * FatGuyLaughing has joined #aegir |
| [09:06:34] | * patcon|myplanet has joined #aegir |
| [09:08:00] | * MadTBone has joined #aegir |
| [09:18:26] | * psynaptic is now known as psynaptic|away |
| [09:24:23] | * j0nathan has quit (Quit: ¡Hasta luego!) |
| [09:30:05] | * omega8cc has quit (Quit: Cheers! It's Time for Offline Reality) |
| [09:30:50] | * ryanarmstrong has quit (Ping timeout: 260 seconds) |
| [09:37:45] | * ryanarmstrong has joined #aegir |
| [09:46:30] | * fysa has quit (Quit: fysa) |
| [09:48:21] | * scientist has quit (Ping timeout: 255 seconds) |
| [09:51:57] | * patcon|myplanet has quit (Ping timeout: 255 seconds) |
| [09:54:44] | * realityloop has joined #aegir |
| [09:55:59] | <realityloop> | is there a way to set access control on a features server that drush_make can still get through? |
