IRC logs for #aegir, 2011-02-20 (GMT)

‹ 2011-02-19
2011-02-21 ›
TimeNickMessage
[11:04:32]* AntiNSA has joined #aegir
[11:06:52]<jlkinsel>wish these logs had timestamps, too. I frequently get the feeling I'm reading them out-of-order
[11:13:54]* mlncn has quit (Remote host closed the connection)
[11:16:57]* nick_santa has joined #aegir
[11:20:31]<kannary>omega8cc: So I did what you suggested by testing a local (master server) install of OA: Created new platform, created new site = It still falied for the same reasons
[11:22:03]* AntiNSA has quit (Ping timeout: 250 seconds)
[11:22:47]<omega8cc>kannary: hmm, that is weird, because it works without issues even on my private test linode 512 with php-cli memory set to 256M
[11:23:02]<kannary>Yes, I know
[11:23:26]<kannary>So estrange as to why this is happening.....
[11:23:28]<omega8cc>so, it sounds something went wrong with your setup on the master
[11:23:56]<omega8cc>what version of php do you have there?
[11:24:19]<kannary>I have a downgraded version: php 5.2.10
[11:24:28]<kannary>Ubuntu 10 LTS
[11:24:57]<omega8cc>ubuntu? hmm, so it says 5.2.10 when you ask: php -v ?
[11:25:57]<kannary>PHP 5.2.10-2 Yeah. This was a downgraded from 5. 3
[11:26:31]<omega8cc>have you added some stuff in php.ini, like extra php modules or something? or maybe you downgraded only php and not all its modules? no idea.. weird
[11:27:03]<kannary>just GD....
[11:27:21]<omega8cc>what your system error logs say?
[11:27:27]* AntiNSA has joined #aegir
[11:27:30]<kannary>ok, so I am not crazy....at least I am getting somewhere LOL
[11:28:58]<kannary>No errors, just a lot of www CRON[31769]: (aegir) CMD (/var/aegir/drush/drush.php '@hostmaster' hosting-dispatch ) entries
[11:29:39]<omega8cc>how much ram do you have there?
[11:30:19]<kannary>I have gone to as much as I can give it in the 512 node...I tried up to 480 and everything in between
[11:31:13]<omega8cc>I mean the linode ram, not memory limit for php-cli
[11:31:30]<omega8cc>512
[11:31:35]<omega8cc>?
[11:31:48]<kannary>It is a Linode 512 node....so 512. That's what you are asking?
[11:31:50]<kannary>one thing that DOES hint at something estrange happening is the fact I set PHP error logs on, and yet none are being generated
[11:31:59]<omega8cc>aand it is Apache
[11:32:09]<kannary>yeap
[11:34:42]<omega8cc>I guess it is too low ram then, as it works for me, but I have nginx + php-fpm and still, with php memory set to 256, the load goes high and it swaps heavily on any heavy provision task, but breaks only when trying OpenPublish, while Atrium works
[11:36:26]<omega8cc>oh, I hope OpenPublish folks will not take over Atrium completely, it is enough they owns now most of DevSeed code on d.o
[11:36:41]<jlkinsel>so when I run mysql -u intntnllyInvalid -h 'hostname' -P '3306' -e "" from the command line, I get the error provisioning is looking for. When provisioning runs it, it gets ERROR 2003 (HY000): Can't connect to MySQL server on 'hostname' (110)
[11:37:02]<kannary>I see, well, I will breadown and try to temporarily up to the next node size see if that works, though I did see a post your replied to of someone running as much as 1G and still getting the same problem...I do thank you for the effort to help me....
[11:37:55]<LloydPearsonIV>omega8cc: u have openpublish? can you share it? i have been trying to find it
[11:37:58]* arianek is now known as arianek_afk
[11:38:16]* FransK has quit (Ping timeout: 240 seconds)
[11:38:36]<omega8cc>kannary: if the upgrade to 1G will not help then you will know there is something wrong in your system settings
[11:38:48]<jlkinsel>LloydPearsonIV: did you look at http://openpublishapp.com/download ?
[11:38:59]<jlkinsel>fg
[11:39:01]<jlkinsel>bah
[11:39:04]* yuracagran has quit (Ping timeout: 240 seconds)
[11:39:18]<kannary>omega8cc: Thanks. I will try and see what is up.
[11:39:45]<omega8cc>LloydPearsonIV: not sure what you mean, but latest OP version requires php memory set to 1 G to deploy the site probably, as it fails with 512
[11:39:56]<LloydPearsonIV>omega8cc: thanks for the direct link. the homepage forwards to the phase2tech page
[11:40:51]<omega8cc>LloydPearsonIV: jlkinsel posted the link above
[11:41:19]<LloydPearsonIV>omega8cc: i must have missed it
[11:41:26]<LloydPearsonIV>nevermind
[11:41:42]<LloydPearsonIV>i got ti then, i thought u meant ealier than that
[11:41:51]<LloydPearsonIV>i meant to thank jlkinsel
[11:41:52]<LloydPearsonIV>lol
[11:41:59]<LloydPearsonIV>thanks jlkinsel
[11:42:01]<LloydPearsonIV>lol
[11:42:08]<omega8cc>you need a server with 2 GB of ram minimum to run *one* OP site
[11:42:41]<LloydPearsonIV>smh
[11:42:55]<LloydPearsonIV>i meant openpublic, not open publish
[11:42:57]<omega8cc>and memory for php-cli 1G to provision it, while it will still take 10-15 minutes even on 8 G ram and full 4 CPU cores
[11:43:00]<omega8cc>ah
[11:43:10]<LloydPearsonIV>they need to do something about those names
[11:43:14]<LloydPearsonIV>lol
[11:43:15]<omega8cc>I guess it will require 16 GB
[11:43:32]<omega8cc>since they are making progress with every distro
[11:43:41]<omega8cc>I mean "progress"
[11:44:01]* MadTBone_ has quit (Read error: Operation timed out)
[11:44:08]<LloydPearsonIV>i wonder why it requires so much memory
[11:44:36]<omega8cc>this is madness if you require 2G ram to run one drupal site
[11:44:40]<omega8cc>bloated
[11:45:03]<LloydPearsonIV>i like openpubish but it isnt worth upgrading my instance
[11:45:11]<omega8cc>open buffet syndrome x10
[11:45:20]<LloydPearsonIV>and i still havent figured out how to get the install profile to work on aegir anyways
[11:45:26]<omega8cc>they should rename it to OpenBuffet
[11:45:34]<LloydPearsonIV>i hope openpublic isnt the same way
[11:45:53]<omega8cc>I told you, it will require 16GB, lol
[11:46:29]<omega8cc>but seriously, it will be interesting
[11:46:48]<kannary>LloydPearsonIV: Maybe a similar problem as with OA and namespace
[11:47:18]<LloydPearsonIV>kannary: it is that same problem, but i forgot how to fix it
[11:47:27]<LloydPearsonIV>i di it once before it times past, but i forget
[11:47:47]<LloydPearsonIV>i am not that interested though until they bring it to D7
[11:47:50]* Chipie has quit (Ping timeout: 240 seconds)
[11:47:59]<LloydPearsonIV>aegir is the only distro i use that is still d6
[11:48:14]<LloydPearsonIV>i just wish they would port the spaces module to d7 soon
[11:48:19]<LloydPearsonIV>i miss it alot
[11:48:21]<LloydPearsonIV>lol
[11:49:40]<kannary>You relabel the profile install folder to something like (using "openpublish" as an example, not sure what they use) openpublish_install, also rename the openpublish.profile to openpublish_install.profile, then edit the openpublish_install.profile and change the name of all the functions inside accordingly....
[11:50:03]<kannary>make sure you catch also the calls to the functions as well
[11:50:33]<LloydPearsonIV>hopefully the fix that when they bring it to d7
[11:51:26]<kannary>Not sure it is a fix thing as I think it is an Aegir thing not liking such set up and reacting to it because -as mig5 says- Aegir is a strict sort of fellow
[11:51:47]<kannary>Aegir FAQ has his entry on that
[11:51:50]* AntiNSA has quit (Ping timeout: 268 seconds)
[11:53:11]<LloydPearsonIV>kannary: well according to Phase2, they siad they learned their lesson with openpublish as far as the name space thing goes
[11:53:24]<kannary>ok, then I am wrong
[11:53:26]<LloydPearsonIV>thats what they said about people having issues installing tattler on aegir
[11:54:29]* Chipie has joined #aegir
[11:54:40]<LloydPearsonIV>openpublish was the first drupal distro i had ever tried
[11:55:46]* obrienmd has joined #aegir
[11:55:55]<LloydPearsonIV>my favorite distro is open atrium but it seems that a d7 port is a long ways off
[11:56:20]* AntiNSA has joined #aegir
[11:56:51]<kannary>right now I am just trying to figuer out why I cannot install it...deling with a bizarro memory issue
[11:57:20]<LloydPearsonIV>what os are using?
[11:57:37]<LloydPearsonIV>what os are you using***
[11:57:49]<kannary>Ubuntu 10 LTS - with downgraded php to 5.2.10
[11:59:03]* totten has joined #aegir
[11:59:31]<LloydPearsonIV>oh
[11:59:48]* smthomas has left #aegir ()
[11:59:54]<kannary><--- Holy crap....THUNDER IN LA!!
[11:59:54]<LloydPearsonIV>i am using ubuntu 10 lts but i am keeping my php 5.3
[12:00:14]<LloydPearsonIV>i need it for D& port of notifications module
[12:00:22]<LloydPearsonIV>it required 5.3
[12:00:29]<LloydPearsonIV>d7***
[12:00:53]<kannary>Well, at the time I was dealing with D6 only, now I am sort of abandoning D^ and building my distro on D7
[12:01:04]<kannary>though serously running out of time
[12:01:15]<LloydPearsonIV>well if you plan on using the notifications module
[12:01:21]<eft>you D7 guys are early adopters IMHO
[12:01:42]<jlkinsel>yep
[12:01:43]<LloydPearsonIV>thats the point
[12:02:56]<LloydPearsonIV>plus the functionality you get with d7 makes me never want to touch 6 again
[12:02:58]<LloydPearsonIV>so i dont
[12:03:02]<kannary>Yeah....I am in a bit of a clusterfuck right now....with the modules I need not being there, some new D7's being in dev, my own older distro being half finsihed in D6, my ubuntu on 5.2.....etc, etc
[12:03:23]<eft>mind you, I'm a hypocrite cos I got a Sandy Bridge CPU and now it's all gone SNAFU
[12:03:39]<LloydPearsonIV>lol
[12:04:01]<LloydPearsonIV>the only module i am waiting for is spaces
[12:04:01]<eft>D7 has lots going for it
[12:04:31]<LloydPearsonIV>of course, it will be great once they finish that merci port too
[12:04:40]<eft>but considering key contrib like Views is only just catching up, I see no rush in converting
[12:04:50]<kannary>Yeah....no dealing with the admin menu for starters
[12:05:45]<LloydPearsonIV>eft: the features port was a bigdeal for me
[12:06:06]<LloydPearsonIV>i am trying to get more familiar with panels
[12:06:13]<LloydPearsonIV>my flow got messed up
[12:06:26]<eft>ah spaces.... http://drupal.org/node/1056888
[12:06:38]<LloydPearsonIV>i was just getting into the whole spaces ,context 7 features method of doing things
[12:06:52]<kannary>Well, if you have your business running and suchl yeah, but in my case I threw myself into Drupal 100% a year+ ago and with learning curve, my ambition, etc....I sort of shot myself in the foot while I had the foot in my mouth...
[12:07:02]* simesy has joined #aegir
[12:07:53]<LloydPearsonIV>eft: that seems like a rather unique issue.
[12:08:08]<hefring>community => aegir hostmaster extremely slow after rc update => http://community.aegirproject.org/node/373
[12:08:44]<eft>LloydPearsonIV: you mean edge case or I'm the only one who noticed it?
[12:09:22]<LloydPearsonIV>kannary: yeh i first found drupal in summer of 2008 after getting fed up with joomla
[12:10:16]* totten has quit (Ping timeout: 240 seconds)
[12:10:26]<LloydPearsonIV>eft: spaces having an incompatibility with views is a unique issue. atrium uses views & spaces...no problem
[12:10:28]<kannary>I threw myself into for a paying gig back in April of 2009 = Only time of my life I have actually worked while crying....I
[12:11:15]<milk>any thoughts on why a site created with aegir would give an error when trying to access /user, /admin, etc? as far as i can tell, platform_FeatureServer.conf looks fine (the same as platform_hostmaster.conf) and mod rewrite is active (via a phpinfo() file in the sites folder)..
[12:11:38]<LloydPearsonIV>kannary: nevermind..it was 2009
[12:11:46]<LloydPearsonIV>the summer
[12:11:52]<kannary>milk: No URL found error?
[12:11:56]<LloydPearsonIV>because right after that..aegir came out
[12:12:09]* mlncn has joined #aegir
[12:13:02]<kannary>LloydPearsonIV: To show you how cluleless and naive I was: I actually thought I could somehow combined pantheon and aegir.
[12:13:27]<LloydPearsonIV>kannary: thats not anive
[12:13:35]<LloydPearsonIV>people actually did it
[12:13:45]<LloydPearsonIV>well project mercury
[12:14:13]<milk>kannary; yup :/
[12:14:48]<LloydPearsonIV>milk: what distro did u use?
[12:15:29]<LloydPearsonIV>kannary: i venturesd away from druapl twice in that time only to come back
[12:15:59]<LloydPearsonIV>most of it was just waiting on d7 to finally come out
[12:16:24]<LloydPearsonIV>i had no ideathat it was going to take so long
[12:16:33]<LloydPearsonIV>i expected it to come out in the summer
[12:16:36]<LloydPearsonIV>lol
[12:16:48]<LloydPearsonIV>i even tried plone
[12:17:15]<milk>a base ubuntu 10.4 (in virtualbox) using the install.sh
[12:19:23]* obrienmd has quit (Quit: Leaving.)
[12:21:07]<kannary>milk: Found the info for you: http://groups.drupal.org/node/95114#comment-313189 That is what did it for me when this happened
[12:21:35]<eft>LloydPearsonIV: yes, but OA uses views 6x-2.12
[12:21:35]* AntiNSA has quit (Read error: Connection reset by peer)
[12:21:50]<eft>i mean 2.11
[12:22:43]<LloydPearsonIV>so its not compatible with views 3?
[12:24:33]* jlkinsel finally sees what's going on, curses
[12:25:03]<kannary>HUge thunder in LA. Sorry posted twice, but it is unusual for this city.....
[12:25:18]<milk>kannary; just checked the vhost config and it's using <VirtualHost *:80> already (as are all the other.confs)
[12:25:28]<kannary>Uhmmm
[12:25:47]* totten has joined #aegir
[12:25:56]<mig5>08:18 < eft> "at least I cant mark it as a dupe :)" - is that in http://drupal.org/node/1063146#comment-4108810
[12:25:59]<mig5>hehe
[12:26:03]<kannary>Hello boss
[12:26:05]<mig5>that was in irc and is my sense of humour
[12:26:17]<mig5>i think there's a language barrier between EugenMayer and I and probably anarcat
[12:26:27]<mig5>i was making fun of myself and he missed the subtlety
[12:26:42]<kannary>when you mean you checked, you mean you checked all the vhost files generated for Aegir ONLY or also the ones for the server itself
[12:27:24]* ryanarmstrong has joined #aegir
[12:27:45]<ryanarmstrong>Hey does anyone here have any experience in setting up DNS in aegir?
[12:27:57]<milk>i checked both those in /vhost.d, is that right?
[12:28:29]<milk>the hostmaster and site ones that is (well, the only two so far)
[12:30:07]* AntiNSA has joined #aegir
[12:30:37]<kannary>milk: Is this for a site hosted in the same server as the hostmaster or a remote server?
[12:30:38]<LloydPearsonIV>mig5: i am trying to add comments to site, platform & server content types so that I can keep a log of changes but they wont show up. what can i do to get them to show up?
[12:31:17]<milk>on the same server (virtualbox)
[12:31:23]<mig5>LloydPearsonIV: should be possible to just edit the theme like any other
[12:31:29]<mig5>i'm not a themer and have no idea how it works
[12:31:41]<jlkinsel>I'm migrating a site from server A to server B. why is server_master sshing to server A and then attempting to connect via mysql to server B?
[12:31:42]<LloydPearsonIV>mig5: that makes 2 of us
[12:31:44]<mig5>isn't this just part of drupal theme system?
[12:31:46]<mig5>ah
[12:32:00]<mig5>maybe you can compare with some other themes to see how they do it?
[12:33:01]<LloydPearsonIV>mig5: well seeing as from past experiences with aegir, if i were to change the theme, the site would break, i figured that eldir has some unique things going on with it
[12:33:28]<LloydPearsonIV>when i asked devs in the drupal irc they seemd puzzled as to why it didt already work in the first place
[12:33:37]<ryanarmstrong>mig5: tried from home and your frigg script worked perfectly. I got Aegir up and running on my Rackspace, thanks for all the help yesterday
[12:33:39]<kannary>mig5: I did that as well, and tried adding the preprocess for comments and calls in the pages, but no luck. So I gather LloydPearsonIV we probably need to mess with it some more
[12:33:48]<LloydPearsonIV>i propose as a feature request that you guys allow comments to work
[12:34:00]<mig5>i think you opened a ticket already about that yes?
[12:34:08]<milk>i had a similar problem accessing the hostmaster just after install, but a 'chown -R aegir:aegir .' (iirc) fixed that. there isn't anything else like that i might have missed?
[12:34:09]<mig5>we should certainly support comments, yes
[12:34:15]<LloydPearsonIV>mig5: just today, yes
[12:34:27]* michaelmcandrew has quit (Quit: slaters!)
[12:34:34]<milk>well, that was for accessing any file..
[12:34:40]* mlncn has quit (Remote host closed the connection)
[12:34:44]<kannary>milk: Tell me which vhost directories you checked?
[12:34:45]<LloydPearsonIV>mig5: it would make keeping track of changes much easier
[12:34:59]<kannary>inside config
[12:37:35]<milk>/var/aegir/config/server_master/apache/
[12:38:09]<milk>server_localhost is empty
[12:38:22]<LloydPearsonIV>mig5: the lanuguage barrier between eugenemayer, and you & arnarcat is...
[12:38:31]<mig5>entertaining? :)
[12:38:32]<LloydPearsonIV>you & anarcat do not understand douche
[12:38:40]<LloydPearsonIV>lol
[12:38:45]<mig5>oh we do
[12:38:48]<mig5>don't worry
[12:38:49]<kannary>milk: And there is no other folders?
[12:39:00]<kannary>yeah, I guess not
[12:39:02]<kannary>uhmm
[12:39:05]<LloydPearsonIV>i just met the kat today and the kat was being a douche to me for no reason
[12:39:45]* mlncn has joined #aegir
[12:40:00]<LloydPearsonIV>ok yeh i meant to ask you something
[12:40:28]<LloydPearsonIV>mig5: is webcluster still broken?
[12:40:48]<milk>might there be a log i could check relating to this?
[12:41:43]<kannary>Try restarting apache from shell and see if it spits out any issues
[12:43:10]<mig5>LloydPearsonIV: yep
[12:43:19]<mig5>there was a patch that fixed it, but it broke everything else, we had to revert it
[12:43:28]<LloydPearsonIV>lol
[12:43:36]<mig5>we don't know how to fix it just yet.
[12:43:53]<LloydPearsonIV>well there is always 1.0
[12:43:55]<LloydPearsonIV>lol
[12:43:58]<mig5>oh yeah
[12:44:04]<mig5>there won't be any bugs in 1.0, it will be perfect ;)
[12:44:18]<LloydPearsonIV>thatswhat i am counting on
[12:44:18]* AquaticDisorder has quit (Read error: Connection reset by peer)
[12:44:23]<milk>a-ha... "[warn] NameVirtualHost *80 has no VirtualHosts" then the same msg again
[12:44:34]<LloydPearsonIV>i get that too
[12:44:38]<LloydPearsonIV>i think its normal
[12:44:55]<LloydPearsonIV>milk: thats defintiely not the problem
[12:45:10]<LloydPearsonIV>milk: did u use the special ubunut install script?
[12:45:14]<milk>oh, dang
[12:45:24]<LloydPearsonIV>if u didnt
[12:45:28]<LloydPearsonIV>u shoudl try it
[12:45:36]<milk>yup, i did
[12:45:37]<LloydPearsonIV>it worked for me when i had trouble
[12:45:50]<kannary>There is a special UBUNTU install script.....?
[12:46:04]<milk>learnt my lesson the hard way previously (VB Guest additions breaking things issue)
[12:46:49]<milk>sorry, i said i used install.sh earlier, but i did in fact use install-aegir-on-ubuntu.sh
[12:47:24]<mig5>i've never used a special UBUNTU install script, no idea why one is needed, the main one should Just Work (and does)
[12:47:27]<LloydPearsonIV>kannary: yup
[12:47:43]<milk>as in http://community.aegirproject.org/node/246
[12:48:01]<LloydPearsonIV>yup
[12:48:16]<LloydPearsonIV>specifically for ubuntu 10.4 LTS
[12:48:18]<LloydPearsonIV>lol
[12:49:14]<kannary>son of a....
[12:50:12]<mig5>wtf
[12:50:20]<mig5>i really don't see why we need OS specific install scripts
[12:50:28]<mig5>our install script is very deliberately designed to be platform agnostic
[12:50:29]<kannary>oh, never mind....I am good...though this memory issue is kicking my ass like a pimp
[12:50:39]<milk>i tried diving in at the deep end with Barracuda a while back but thought i'd best stick with what the main site advised
[12:50:52]<mig5>lunchtime, bbl
[12:51:18]<kannary>yeah, the script that comes with aegir has worked like butter every singe time....5 upgrades since alpha 12 and not a glitch
[12:55:13]<milk>i'll remember that if i need to install from fresh again :)
[12:55:31]<milk>(which i might do if i can't sort this dang quandary).
[12:56:16]* josh_k has quit (Quit: This computer has gone to sleep)
[12:56:35]<kannary>milk: Got to sort of split, but I think your problem is probably easier to figure out than you think. Just as last resort, try checking the same thing in your apache vhost files and try changing it just to see
[12:57:04]<LloydPearsonIV>well i dont know why but when i tried to install aegir the regular way, it failed but when i used ubuntu sh it worked
[12:58:17]<HaloFX>Aegir is very happy on Ubuntu. Installed several times, and just finished Beta 1 to Rc1 upgrades a couple minutes ago.
[12:59:15]<HaloFX>I haven't worked up the guts to try it on the RHEL VPS servers yet though.
[12:59:43]<jlkinsel>works fine for me on centos
[13:00:15]<jlkinsel>then again, I've using various repos for modern php/mysql
[13:00:17]<HaloFX>jlkinsel: did you have to alter the install?
[13:00:19]<milk>kannary; where might they be? can't see anything obviously relevant under /etc/apache2/..
[13:00:35]<jlkinsel>HaloFX: besides the repos, nah
[13:00:59]<jlkinsel>i think with the older aegir installs I had to, but now it's fairly smooth
[13:06:25]* samhassell has joined #aegir
[13:08:28]* kannary has quit (Quit: kannary)
[13:11:50]* MadTBone_ has joined #aegir
[13:11:51]<milk>might any other kindly souls have any ideas on my issue? :)
[13:12:23]<HaloFX>I just realized I haven't paid as much attention as I should have on these VPS servers. When we ordered them I thought they were RHEL 5, but the server is saying 4.
[13:13:15]<LloydPearsonIV>i use AWS
[13:13:26]<LloydPearsonIV>any server iwant...just an instance away
[13:13:26]* jlkinsel is moving off aws
[13:13:45]<HaloFX>They are running mySQL 5.0.51a and PHP 5.2.9
[13:14:07]<LloydPearsonIV>jlkinsel: to each their own...lol
[13:14:27]<LloydPearsonIV>milk: seems like something wnet wrong with your installation
[13:14:37]* recidive has joined #aegir
[13:14:37]<LloydPearsonIV>may u need to try it agin
[13:14:42]<LloydPearsonIV>thats all i can say
[13:16:20]<HaloFX>I was eyeing Rackspaces cloud, and move over to Ubuntu.
[13:16:34]<milk>LloydPearsonIV; fair play. will give me a chance to try the main installer with the new release. thanks though!
[13:16:52]<LloydPearsonIV>mig5 uses rackspace i think
[13:16:54]* jlkinsel is working on building his own secure cloud. :)
[13:17:30]<LloydPearsonIV>jlkinsel: more power to u
[13:17:48]<LloydPearsonIV>i am actuallly working on something a lil more simple
[13:17:53]<jlkinsel>hehheh
[13:18:04]<LloydPearsonIV>like my own distributions & installation profiles
[13:18:10]<jlkinsel>aws is nice, don't get me wrong, I used them for years
[13:18:26]<LloydPearsonIV>i like them alot
[13:18:40]<LloydPearsonIV>much better than leasing a server
[13:18:46]<jlkinsel>yea
[13:18:46]<LloydPearsonIV>that was expensive
[13:18:51]<LloydPearsonIV>especially if its managed
[13:19:11]<LloydPearsonIV>then it takes them 3 days to prep it
[13:19:13]* AntiNSA has quit (Ping timeout: 250 seconds)
[13:19:15]<LloydPearsonIV>i want my server now
[13:19:21]<LloydPearsonIV>not in 72 hrs
[13:19:30]<LloydPearsonIV>not when u just charged me today
[13:19:32]<LloydPearsonIV>lol
[13:20:59]<LloydPearsonIV>is anyone here using memecache with aegir?
[13:21:01]<jlkinsel>i was taklin to some people in the industry...they noticed their dedicated server sales were dropping once they started offering cloud...like that should be a shock
[13:21:11]* jlkinsel hasn't touched memchache yet
[13:21:24]<jlkinsel>memecache - is that for lolcats? ;)
[13:21:39]<LloydPearsonIV>jlkinsel: i am also trying to figure out what is the benefit of the apc module
[13:22:02]* psynaptic|afk has quit (Remote host closed the connection)
[13:22:11]<jlkinsel>i'm doing varnish + apc
[13:22:17]<LloydPearsonIV>i already have apc on my server
[13:22:23]<LloydPearsonIV>and its great
[13:22:28]* psynaptic|afk has joined #aegir
[13:22:38]<LloydPearsonIV>but i dont see what the added benefit the module is going to provide
[13:23:01]<LloydPearsonIV>jlkinsel: i want to use varnish
[13:23:16]<LloydPearsonIV>but i am worried about breaking something
[13:23:25]<jlkinsel>it's nice. I'm hoping to work on the esi module this weekend
[13:23:29]<LloydPearsonIV>i have both wordpress & drupal on this server
[13:23:49]<jlkinsel>yep I'm going to be hosting a wp site within the month for a client
[13:23:56]<jlkinsel>that's on my mind as well
[13:24:38]<LloydPearsonIV>the aegir part shouldnt be much of a problem i hope, but i want to make sure i dont break the ws sites
[13:24:42]<LloydPearsonIV>wp**
[13:25:00]<LloydPearsonIV>well if it offers any insight
[13:25:01]* AntiNSA has joined #aegir
[13:25:12]<jlkinsel>yeah i need to google around
[13:25:33]<LloydPearsonIV>i have wordpress multisite installed in the /var/www folder
[13:26:03]<LloydPearsonIV>and i make a vhost with a * alias and put it in the post.d folder in aegir
[13:26:15]<LloydPearsonIV>so any domian that is not for aegir goes to wordpress
[13:27:03]<LloydPearsonIV>my main site is currently on wordpress but i will be migrating it to drupal soon
[13:28:07]<LloydPearsonIV>my plan is to use varnish upfront, us apc for opcode & use memcache for the data
[13:28:35]<jlkinsel>yep
[13:29:00]<LloydPearsonIV>from what i have read
[13:29:01]<jlkinsel>I've got a few more layers in there as well...mod_security, mod_evasive, php ids...
[13:29:10]<LloydPearsonIV>i should only need memcache & apc
[13:29:18]<jlkinsel>yea
[13:29:24]<LloydPearsonIV>if those are configured correctly, my sites will be like lighning
[13:29:28]<jlkinsel>I'm taking a much more security-heavy focus
[13:29:33]<jlkinsel>yeah you see a very nice improvement
[13:29:36]<LloydPearsonIV>varnish would pretty much be overkill
[13:29:56]<jlkinsel>you'd be surprised
[13:30:00]<LloydPearsonIV>but i have a project where overkill may be necessary
[13:32:45]<LloydPearsonIV>check this out
[13:32:47]<LloydPearsonIV>http://coldfrontlabs.ca/blog/making-drupal-fly-apc-memcache-and-squid
[13:33:55]<jlkinsel>yea
[13:34:00]<LloydPearsonIV>i plan on a similar setup substituting squid for varnish
[13:34:11]<jlkinsel>there's a few good threads on the high performance drupal group
[13:34:18]<LloydPearsonIV>yeh
[13:34:20]<LloydPearsonIV>i am in it
[13:34:44]<LloydPearsonIV>been in it for quite some time
[13:35:12]<LloydPearsonIV>i dont comment muchjust ty & lear nsome thing from the peopel who really know their stuff
[13:35:29]<jlkinsel>yep same
[13:35:30]<LloydPearsonIV>there is so much to learn with druapl
[13:35:56]<jlkinsel>understatement. :)
[13:36:05]<LloydPearsonIV>true
[13:36:28]<LloydPearsonIV>features context 7 spaces made things easier for me
[13:36:29]* gezick has quit (Ping timeout: 240 seconds)
[13:36:53]<LloydPearsonIV>that really helped understand alot of the concepts
[13:36:58]* gezick has joined #aegir
[13:36:58]<LloydPearsonIV>that and working with atrium
[13:37:14]<LloydPearsonIV>oh yeh
[13:37:17]<LloydPearsonIV>and drush
[13:38:02]<jlkinsel>picking up Pro Drupal and doing some heavy drupal work with it was my sorta breakthrough point, but I've still got tons to learn
[13:41:30]<LloydPearsonIV>pro drupal, as in pro drupal development???
[13:41:49]<jlkinsel>yea sorry
[13:41:58]<LloydPearsonIV>just amkin sure
[13:42:04]<LloydPearsonIV>there is so much druapl stuff out there
[13:42:34]<LloydPearsonIV>i used to spend alot of time on http://learnbythedrop.org
[13:42:42]<HaloFX>Cab you guys with AWS experience help me understand their pricing? Their site makes my head spin!
[13:42:48]<jlkinsel>LloydPearsonIV: 404
[13:42:49]<LloydPearsonIV>and some other site
[13:43:05]<LloydPearsonIV>sure
[13:43:11]<LloydPearsonIV>what do u need to know?
[13:43:11]<jlkinsel>HaloFX: sure...take the price per hour, multiply by 720, that's the monthly price for that image size
[13:43:33]<jlkinsel>there's tons of other costs, but that's the basic
[13:43:56]<LloydPearsonIV>jlkinsel:
[13:43:59]<LloydPearsonIV>http://learnbythedrop.com/
[13:44:04]<LloydPearsonIV>lol
[13:44:15]<LloydPearsonIV>i have a medium instance
[13:44:26]<jlkinsel>(caveat on the *720 - if you buy a "reserved" image you pay a percentage up front, but the monthly and total is cheaper)
[13:44:28]<LloydPearsonIV>it cost me about 130/month
[13:44:42]<LloydPearsonIV>when i had a small instance it cost me 60/month
[13:45:07]<LloydPearsonIV>but the performance improvement is so vast that it made it work it for me
[13:45:18]<LloydPearsonIV>worth it***
[13:45:45]<LloydPearsonIV>i dont plan on needing a large instance for a while
[13:45:54]<LloydPearsonIV>because i am going to cache the hell out of this one
[13:46:57]<HaloFX>Did you upgrade for the compute cycles or memory?
[13:47:24]<LloydPearsonIV>compute cycles
[13:48:01]<LloydPearsonIV>i can always use virtual memory
[13:49:15]<LloydPearsonIV>though vm isnt as good as physical memory...there is no virtual compute cycles
[13:49:18]<LloydPearsonIV>lol
[13:49:26]<HaloFX>It says the micro has 613mb of memory and the small has 1.7gb of memory.
[13:49:49]<LloydPearsonIV>the medium also has 1.7gb of memory
[13:50:02]<LloydPearsonIV>i never used the micro
[13:50:05]<LloydPearsonIV>its somewhat new
[13:50:26]<jlkinsel>it's free if you're a new customer
[13:50:47]<HaloFX>They may have changed things, I see Micro, Small, Large and Extra Large.
[13:51:25]<HaloFX>I was thinking of trying the free, but if it doesn't work out, moving is always a pain.
[13:52:31]<HaloFX>The one biggy I do like from AWS is they support Turnkey Linux images, which I run on my Dev. and use the built in back up to S3.
[13:52:47]* jlkinsel makes note to look into turnkey
[13:53:08]<HaloFX>Turnkey has really been awesome for me!
[13:53:59]<HaloFX>I use their LAMP stack, they have Drupal , but I don't like how they installed Drupal.
[13:55:08]<HaloFX>I use it on several VMs, and it sets up in minutes. I use Hyper-V, but they have prebuilt Virtual box images also.
[13:55:50]<LloydPearsonIV>i dont like virtual box
[13:55:56]<LloydPearsonIV>its slow
[13:56:23]<HaloFX>There are ISOs you can install on anything.
[13:56:24]<LloydPearsonIV>parallels is much faster but its a resource hog
[13:56:58]* AntiNSA has quit (Ping timeout: 240 seconds)
[13:57:13]<LloydPearsonIV>maybe i should only run it on my desktop and run virtualbox on my laptop
[13:59:41]<HaloFX>Couple clicks and it backs up to S3. A couple more and you can restore or move to another location. It only backs up the changes from the base image. I have 2 servers backing up right now. Only using 255mb on S3, 4 cents a month!
[14:00:30]<LloydPearsonIV>jlkinsel: looks like they are going to work on that commenting issue. thats good
[14:00:39]<jlkinsel>ocol
[14:00:40]<jlkinsel>cool
[14:01:14]<LloydPearsonIV>they said they wont ever enable comments by default but they should atleast be usable
[14:01:19]<LloydPearsonIV>which is good enough for me
[14:01:23]<LloydPearsonIV>lol
[14:01:26]<jlkinsel>yea
[14:02:33]* samhassell has quit (Ping timeout: 250 seconds)
[14:03:01]<LloydPearsonIV>well hopefully they can provide a workaround while they implement a permanent fix
[14:03:10]<LloydPearsonIV>those aegir release cylces are slow
[14:03:15]<LloydPearsonIV>lol
[14:03:34]* AntiNSA has joined #aegir
[14:03:44]<LloydPearsonIV>but i respect the work
[14:03:49]<LloydPearsonIV>so no complaints
[14:04:59]<LloydPearsonIV>meanwhile i also need to take some time and learn how to create a sub theme of eldir so that my logo can quit disappearing every upgrade
[14:05:03]<LloydPearsonIV>lol
[14:11:42]* obrienmd has joined #aegir
[14:23:11]* arianek_afk is now known as arianek
[14:41:46]* obrienmd has quit (Quit: Leaving.)
[14:42:01]* AntiNSA has quit (Read error: Connection reset by peer)
[14:44:11]<ryanarmstrong>Anyone know how to get sendmail to work?
[14:44:43]<ryanarmstrong>I used mig5 frigg script and everythign is working great except sendmail, so I'm not getting any emails when I install a new site with the login and pass
[14:45:48]* AntiNSA has joined #aegir
[14:50:29]* recidive has quit (Quit: zzz)
[14:58:46]<LloydPearsonIV>ryanarmstrong: i had that issue too
[14:58:58]<LloydPearsonIV>u using ubuntu?
[14:59:08]<ryanarmstrong>Debian
[14:59:17]<LloydPearsonIV>oh ok
[14:59:25]<ryanarmstrong>I used Mig5's frigg script, so it installs on Rackspace using Debian
[14:59:32]<LloydPearsonIV>the way i fixed it is
[14:59:49]<LloydPearsonIV>i installed sendmail
[14:59:57]<LloydPearsonIV>in uninstalled postfix
[15:00:04]<LloydPearsonIV>and installed sendmail instead
[15:00:13]<LloydPearsonIV>then it worked
[15:00:37]<ryanarmstrong>are there shell commands to uninstal postfix and install sendmail?
[15:00:42]<ryanarmstrong>I'm really new to this server stuff
[15:00:54]<LloydPearsonIV>apt-get install sendmail
[15:01:00]<LloydPearsonIV>i believe that will do it
[15:01:12]<ryanarmstrong>you said you uninstalled postfix first?
[15:01:18]<ryanarmstrong>is that an alternative to sendmail?
[15:01:20]<LloydPearsonIV>it will uninstall postfix & install sendmail
[15:01:22]<ryanarmstrong>oh
[15:01:26]<ryanarmstrong>awesome thank you!
[15:01:30]<ryanarmstrong>I'll try that
[15:01:32]<LloydPearsonIV>no problem
[15:01:37]<LloydPearsonIV>let me know if it works
[15:02:37]<ryanarmstrong>So Aegir will just automatically know to use sendmail?
[15:02:40]<ryanarmstrong>No config I need to do?
[15:02:56]<LloydPearsonIV>whithout being too technical
[15:02:58]<LloydPearsonIV>yes
[15:03:02]<ryanarmstrong>lol ok cool
[15:03:04]<LloydPearsonIV>its a php function
[15:03:11]<ryanarmstrong>ah ok, I gotcha
[15:03:40]<LloydPearsonIV>aegir is pretty smart
[15:03:48]<LloydPearsonIV>it can figure alot of things
[15:06:02]<LloydPearsonIV>since you are rather new, it will benefit you quite well to do some research 7 become familiar with basic linux ommands
[15:06:15]<LloydPearsonIV>as well as chmod
[15:06:39]<LloydPearsonIV>and defintely become famiialr with drush
[15:06:51]<LloydPearsonIV>familiar***
[15:08:54]<ryanarmstrong>Hmm, e-mails don't seem to be sending.
[15:09:06]<ryanarmstrong>Install seemed to go ok, I did get one message:
[15:09:13]<ryanarmstrong>Warning: 3 database(s) sources
[15:09:13]<ryanarmstrong> were not found, (but were created)
[15:09:13]<ryanarmstrong> please investigate.
[15:09:13]<ryanarmstrong>Starting Mail Transport Agen
[15:10:20]<LloydPearsonIV>interesting
[15:10:42]<LloydPearsonIV>u know what
[15:10:45]<LloydPearsonIV>smh
[15:10:53]<LloydPearsonIV>maybe u need to open a port
[15:11:02]<LloydPearsonIV>lol
[15:11:40]<LloydPearsonIV>i dont remember what port it is off the top of my head but i am pretty sure you can google what port is needed for sendmail
[15:11:47]<ryanarmstrong>I think it's 25
[15:12:02]<LloydPearsonIV>is port 25 open?
[15:12:25]* mlncn has quit (Remote host closed the connection)
[15:16:36]<ryanarmstrong>Sory, how do I test for that
[15:16:46]<ryanarmstrong>ping?
[15:17:57]<mig5>ryanarmstrong: your frigg script should have installed postfix
[15:17:59]<mig5>reading backlog
[15:18:05]* mlncn has joined #aegir
[15:18:09]<mig5>postfix sends mail outbound out of the box
[15:18:55]<ryanarmstrong>yea when I ran frigg, I never got my login email. So I just generated a new MD5 hash and put it in my Aegir installs db to login
[15:19:06]<ryanarmstrong>but I don't get an email when I create a new site either
[15:19:11]<mig5>and frigg doesn't set a firewall, so if you didn't get the email maybe there was something with the receiving end.
[15:19:18]<mig5>tail your /var/log/mail.log to debug
[15:19:40]<LloydPearsonIV>maybe there was typo when he put in the email adress
[15:20:25]<mig5>most likely. check the aegir client user (node/1)
[15:20:40]<ryanarmstrong>ok one second
[15:20:49]<LloydPearsonIV>well goodnight guys
[15:21:08]<LloydPearsonIV>mig5: thanks for looking into that comment thing for me
[15:22:01]<ryanarmstrong>No typo in my email address
[15:22:18]<ryanarmstrong>Either for the original Aegir setup or when I created the new site
[15:22:52]<mig5>the /var/log/mail.log will give it away
[15:22:54]<mig5>np LloydPearsonIV
[15:24:15]* LloydPearsonIV has left #aegir ()
[15:27:46]<ryanarmstrong>hmm ok, I've got the log here, nothing is jumping out at me
[15:27:55]<ryanarmstrong>I can see where it was using postfix vs sendmail
[15:28:36]<ryanarmstrong>postfix attempt
[15:28:38]<ryanarmstrong>Feb 20 03:41:58 www postfix/qmgr[1037]: 90A7628FFD: removed
[15:28:38]<ryanarmstrong>Feb 20 04:02:12 www postfix/master[1029]: terminating on signal 15
[15:28:38]<ryanarmstrong>Feb 20 04:02:37 www sendmail[2605]: gethostbyaddr(10.181.91.72) failed: 1
[15:28:38]<ryanarmstrong>Feb 20 04:02:37 www sendmail[2605]: alias database /etc/mail/aliases rebuilt by root
[15:29:25]<mig5>will need to get more of the log, that looks like half a transaction
[15:29:44]<mig5>possibly a dns issue as well, but i don't know sendmail very well (postfix Just Works for me)
[15:30:01]<ryanarmstrong>if I apt-get postfix that will remove sendmail correct?
[15:30:05]<ryanarmstrong>I can switch it back
[15:30:08]<mig5>it should do yes
[15:32:09]<ryanarmstrong>ok I ran apt-get install postfix
[15:32:26]<ryanarmstrong>if I run /etc/init.d/postfix reload it says that the postfix service isnt running
[15:32:30]<ryanarmstrong>so I did /etc/init.d/postfix start
[15:32:57]<ryanarmstrong>And it says it's starting it. But if I try the reload again, it says its not running again
[15:34:29]<mig5>read the mail log again after you start it
[15:34:33]<mig5>it'll say why it's not running
[15:34:38]<ryanarmstrong>ok
[15:34:53]<mig5>or start it from command line (just 'postfix' probably)
[15:35:12]<mig5>maybe not
[15:36:09]<ryanarmstrong>Ah this might be a big clue. I can see in the log where I am trying to start postfix
[15:36:10]<ryanarmstrong>Feb 20 04:31:31 www postfix/master[1759]: fatal: bind 0.0.0.0 port 25: Address already in use
[15:36:11]<ryanarmstrong>Feb 20 04:31:43 www postfix/postfix-script[1764]: fatal: the Postfix mail system is not running
[15:36:11]<ryanarmstrong>Feb 20 04:31:53 www postfix/master[1827]: fatal: bind 0.0.0.0 port 25: Address already in use
[15:36:11]<ryanarmstrong>Feb 20 04:31:58 www postfix/postfix-script[1832]: fatal: the Postfix mail system is not running
[15:36:11]<ryanarmstrong>Feb 20 04:34:42 www postfix/master[1929]: fatal: bind 0.0.0.0 port 25: Address already in use
[15:37:24]<mig5>so something is already listening on port 25
[15:37:28]<mig5>maybe sendmail was not removed
[15:37:31]<mig5>do netstat -pant | grep 25
[15:37:34]* MadTBone_ has quit (Read error: Operation timed out)
[15:38:08]<ryanarmstrong>tcp 0 0 127.0.0.1:25 0.0.0.0:* LISTEN 979/sendmail: MTA:
[15:38:13]<mig5>right
[15:38:17]<mig5>so sendmail is still listening
[15:38:24]<mig5>stop that, and perhaps apt-get remove sendmail after that
[15:38:26]<ryanarmstrong>hmm, apt-get said it would uninstal sendmail, must not have removed that though
[15:38:40]<ryanarmstrong>Sorry, how do I stop that?
[15:38:50]<mig5>probably /etc/init.d/sendmail stop
[15:38:54]<mig5>if it doesn't work, kill 979
[15:39:22]<ryanarmstrong>ok now its:
[15:39:23]<ryanarmstrong>tcp6 0 0 50.56.73.154:80 76.115.186.104:62537 TIME_WAIT -
[15:39:58]<ryanarmstrong>rand the remove, and it said sendmail isn't there so we should be good
[15:40:37]<ryanarmstrong>ran*
[15:40:39]<mig5>your netstat there is for something on port 80
[15:40:50]<mig5>so we can ignoe that
[15:43:05]<ryanarmstrong>ok cool
[15:43:19]<ryanarmstrong>So do I need to edit a file to get postfix to use port 25?
[15:43:41]<ryanarmstrong>like the main.cf?
[15:43:52]<mig5>no, postfix will run on port 25 out of the box
[15:44:02]<mig5>now that sendmail is gone, we should start it and then check the log again
[15:45:15]<ryanarmstrong>ok, now I got:
[15:45:15]<ryanarmstrong>Feb 20 04:44:33 www postfix/master[2151]: daemon started -- version 2.5.5, configuration /etc/postfix
[15:45:21]<mig5>that's good
[15:45:23]<ryanarmstrong>let me try sending a mail
[15:48:49]<ryanarmstrong>mig5…. just fucking shoot me now
[15:48:51]<ryanarmstrong>I'm an idiot
[15:49:09]<mig5>welcome to the club!
[15:49:11]<ryanarmstrong>Gmail was sending the fucking emails into my spam folder lol
[15:49:15]<ryanarmstrong>*smacks head*
[15:49:21]<mig5>;)
[15:49:24]<ryanarmstrong>It was working the entire damn time
[15:49:29]<mig5>of course it was :)
[15:49:38]<ryanarmstrong>*long sigh*
[15:50:07]<ryanarmstrong>Well thank you for all of the help anyways, had there been an actual problem I'm sure you would have fixed it lol
[15:50:27]<ryanarmstrong>*hangs head in shame*
[15:50:28]<ryanarmstrong>lol
[15:50:51]<ryanarmstrong>I guess its better for it to be a stupid blunder on my part then an actual problem I suppose
[15:52:17]* jerryitt has joined #aegir
[15:57:52]* ryanarmstrong has left #aegir ()
[16:03:53]* AntiNSA has quit (Ping timeout: 246 seconds)
[16:09:46]* obrienmd has joined #aegir
[16:10:25]* EclipseGc has joined #aegir
[16:17:02]* obrienmd has quit (Quit: Leaving.)
[16:28:12]* scientist has joined #aegir
[16:51:04]* totten has quit (Ping timeout: 240 seconds)
[16:53:49]* AntiNSA has joined #aegir
[16:53:54]<AntiNSA>stupid question, I was locked out of my server because I accidently hammered with filezilla and my wrong account settings. I can use the aja terminal on my linode account, but I forgot which file I have to delete to lift the ban. I hate to restore a backup from yesterday and lose all my work. Any help would be appreciated...
[16:55:11]* arianek is now known as arianek_afk
[16:56:55]* bejames_ has joined #aegir
[17:03:14]* josh_k has joined #aegir
[17:06:42]* totten has joined #aegir
[17:09:16]* snlnz has quit (Ping timeout: 255 seconds)
[17:17:00]* snlnz has joined #aegir
[17:17:02]<eft>mig5: ping?
[17:17:23]<mig5>hey eft
[17:17:43]<eft>so you were enjoying my efforts at diplomacy?
[17:18:06]<mig5>hehe
[17:18:26]<mig5>the guy was begging for someone like you to speak up
[17:18:45]<mig5>for someone to tell that they don't understand him :)
[17:18:57]<mig5>and that he isn't a noob and has been doing this for hundreds of years!!!1111 :)
[17:19:50]<eft>well I think he is a smart guy and ultimately means well
[17:20:39]<eft>just needs to exercise more tact at times
[17:20:46]<mig5>i think he means well too.
[17:21:03]<mig5>but he has a severe case of Poor Me, and he contradicts himself
[17:21:03]<eft>I know I've come across pretty brash sometimes before
[17:21:07]<mig5>neither of which impress me
[17:21:49]<eft>well we know how easily you aren't impressed :)
[17:21:54]<mig5>:)
[17:22:12]<mig5>his whole thing relies on the idea that i might feel bad, so he seems baffled when it becomes clear i just don't care :)
[17:22:22]<mig5>anyway
[17:22:31]<eft>anyway....
[17:22:37]<eft>this channel was hopping today
[17:22:37]<mig5>he needs to let go of those incidents a bit: opensource is not a place for holding grudges, i've tried it
[17:22:57]<mig5>yeah the channel often gets busy when people try the new releases
[17:23:04]* eft quotes mig5 "Oh Really"
[17:23:12]<mig5>?
[17:23:38]<eft>you know, your one word sarcastic catch-all
[17:23:44]<eft>re the grudges
[17:23:55]<mig5>i don't get it
[17:24:00]<mig5>you love the cryptic shit don't you
[17:24:01]<mig5>:)
[17:24:03]<mig5>http://2.bp.blogspot.com/_1INH3fOI6pM/SWy6iyyspyI/AAAAAAAAAZM/Z_-dZmCyDN...
[17:24:09]<mig5>http://files.myopera.com/drlaunch/albums/37656/no-wai001.jpg
[17:25:22]<eft>trust me to be blind to the fact it's an effing phenomenon
[17:26:02]<mig5>wait til i tell you about the Chmod Meme
[17:26:19]<eft>but I like my current avatar
[17:26:38]<mig5>i still haven't seen that damn movie
[17:26:58]<eft>I saw a good one last night "The Escapist"
[17:28:03]<mig5>i like Brian Cox
[17:30:23]<eft>he reminded me of Albert Finney
[17:35:39]* skwashd has joined #aegir
[17:35:55]<eft>skwashd: yo
[17:36:40]<skwashd>hey eft
[17:36:52]<skwashd>sup?
[17:36:59]<eft>how are things in your world?
[17:40:46]* EclipseGc has quit (Quit: EclipseGc)
[17:44:52]* arianek_afk has quit (Remote host closed the connection)
[17:46:50]* EugenMayer has joined #aegir
[17:51:46]* jerryitt has left #aegir ()
[17:54:53]<skwashd>eft: pretty good
[17:55:02]<skwashd>sorry for the lag - dog ran away again!
[17:55:14]<skwashd>just moved house and the fencing is dodgy
[17:55:43]<eft>I've had that experience - the dog just wants to go back to the old home :)
[17:56:10]<skwashd>nah ... he wants to go exploring
[17:56:29]<eft>you getting any $$ to get to Chicago?
[17:57:34]<anarcat>hey skwashd
[17:57:43]<skwashd>hey anarcat
[17:57:44]<anarcat>skwashd: seems like you're planning an awesome party - count me in!
[17:57:50]<skwashd>eft yeah ... still need about 400USD
[17:58:03]<skwashd>anarcat: sign up for a ticket :)
[17:58:17]<anarcat>yeah... that seems complicated with registration and al ;)
[17:58:17]<skwashd>anarcat: it is more of a laid back training session, but you're welcome to come
[17:58:19]<anarcat>but i'll try!
[17:58:35]<anarcat>sure
[17:58:38]<skwashd>anarcat: you just go to eventbrite and put in some basic info
[17:58:39]<anarcat>i hope i can be of help
[17:58:47]<eft>anarcat: seems like ur already at an awesome party 0200 EST
[17:59:08]<anarcat>if i can't have a full aegir session, at least i can have a training session :)
[17:59:11]<anarcat>or give, rather
[17:59:15]<anarcat>eft: uh?
[17:59:27]<skwashd>anarcat: it should be fun
[17:59:32]<anarcat>sure
[17:59:34]<eft>isn't it past your bed time in MTL?
[17:59:35]<anarcat>cool
[17:59:42]<anarcat>eft: how do you know by bed time? :P
[18:00:02]<skwashd>anarcat: mortendk should be appointed chief marketing officer for aegir
[18:00:15]<anarcat>skwashd: that would be freakin awesome
[18:00:21]<skwashd>anarcat: :)
[18:00:23]<eft>anarcat: look out the window :)
[18:00:39]<anarcat>eft: windows?
[18:00:49]<anarcat>i run debian you insensitive clod
[18:01:14]<eft>:)
[18:01:49]<eft>anarcat: I was acting the peacemaker here earlier : a la bon cop, bad cop
[18:02:32]<anarcat>eft: good luck with that
[18:02:41]<anarcat>but thanks :)
[18:02:52]<anarcat>i'm trying my best to ignore the thing right now
[18:02:54]<anarcat>weekends.
[18:04:21]<eft>I don't blame you - you should be in the hot tub with a martini about now
[18:04:35]<anarcat>thats a god idea
[18:04:56]<eft>who's she (god)?
[18:05:41]<eft>or, as Ricky Gervais said, "I'd like to thank God for making me an atheist"
[18:10:48]* univate has quit (Quit: Leaving)
[18:11:35]<skwashd>i'm off
[18:11:37]<skwashd>ttyl
[18:11:44]<skwashd>anarcat: please sign up
[18:11:46]* skwashd has quit (Quit: Leaving)
[18:15:36]<EugenMayer>(08:01:49) eft: anarcat: I was acting the peacemaker here earlier : a la bon cop, bad cop
[18:15:40]<EugenMayer>True, True
[18:15:42]<EugenMayer>Morning
[18:19:38]* univate has joined #aegir
[18:25:15]<eft>evening here :)
[18:25:30]<eft>you in Germany?
[18:28:26]* EugenMayer has quit (Ping timeout: 276 seconds)
[18:28:44]<sdboyer>anarcat mig5: ping
[18:30:00]<sdboyer>anarcat mig5: i wanna get info re: http://groups.drupal.org/comment/reply/128479/417444 ASAP b/c we're coming down for rebuilds tomorrow, and if i'm gonna write the special cases then i want to do it before then so that they get a good test run
[18:31:32]<sdboyer>basically i just need clone uris for hostmaster and provision
[18:34:49]<sdboyer>as for moving the issues around - i'm not actually deleting any project nodes, so there's nothing to be concerned about wrt losing issues. as for moving them, it'd be nice if there's a batched way to do it, but i dunno that there is
[18:35:01]<sdboyer>basically you just need to move them from one queue to another
[18:36:09]<sdboyer>i do have d.o super cow powers, but it doesn't make much difference b/c there's a bunch of state change that happens when an issue gets moved from queue to queue, so i don't think there's a bulk interface for it. which leaves just manually moving things...which you can do with normal d.o powers. so i'll leave that to you :)
[18:41:56]<omega8cc>hefring: tell anarcat sdboyer needs urgent reply/info re: http://groups.drupal.org/node/128479
[18:41:56]<hefring>omega8cc: I'll pass that on when anarcat is around.
[18:42:14]<sdboyer>omega8cc: thanks :)
[18:43:04]<omega8cc>sdboyer: np, do you need urls to git repos we use now?
[18:43:11]<sdboyer>omega8cc: yep
[18:43:18]<sdboyer>omega8cc: that's really all it is
[18:43:34]<sdboyer>omega8cc: i can pop the logic into the scripts right away if i have that
[18:43:35]<omega8cc>sdboyer: http://git.aegirproject.org/
[18:44:07]<sdboyer>so git://git.aegirproject.org/hostmaster.git and git://git.aegirproject.org/provision.git then?
[18:44:10]<omega8cc>sdboyer: so git://git.aegirproject.org/hostmaster.git and git://git.aegirproject.org/provision.git
[18:44:13]<sdboyer>hehe
[18:44:13]<omega8cc>yep
[18:44:16]<sdboyer>perfect
[18:44:18]<sdboyer>i'll put them in tonight
[18:44:26]<omega8cc>thanks!
[18:44:30]<sdboyer>it would be great if y'all then check them on monday and see if they're correct
[18:44:31]<sdboyer>sure
[18:44:56]<omega8cc>ok, I will ping anarcat and mig5 about it
[18:45:10]<sdboyer>fyi we have a framework for doing push notifications already in place
[18:45:15]<sdboyer>no plugins written for it yet, but
[18:45:31]<sdboyer>it's one thing i'd like to add. i believe y'all have at least irc notifications, right?
[18:45:40]<omega8cc>nice!
[18:45:50]<omega8cc>yes, and this channel is logged
[18:45:56]* sdboyer seems to recall seeing a bot get spammy when people push
[18:45:58]<sdboyer>ok cool
[18:46:09]<sdboyer>yeah it won't be available right away, but like i said
[18:46:38]<sdboyer>the framework is in place, and if we were to have a quick sprint on it at drupalcon i'm sure we could a) have the notifications system set up and b) even have a plugin or two written to get info out
[18:46:45]<sdboyer>wow, am i tired
[18:46:49]<omega8cc>would be nice to have
[18:46:51]<sdboyer>replace a with:
[18:47:16]<sdboyer>a) take the hook we provide and make a framework for an arbitrary number of plugins that can fire on an incoming push
[18:47:30]<sdboyer>which would then be selectable in the d.o ui
[18:48:01]<omega8cc>sdboyer: wow, that sounds good! thanks for your assistance!
[18:48:02]<sdboyer>so, yeah :)
[18:48:04]<sdboyer>you bet
[18:48:27]* sdboyer is tired of lowest-common-denominator thinking keeping people from doing cool stuff on d.o
[18:48:47]<sdboyer>i'm doing everything i can to use git as an excuse to reverse some of that mentality :)
[18:49:31]<omega8cc>yeah, it will be nice to be back home :)
[18:49:38]* sepehr has joined #aegir
[18:51:17]<sdboyer>and hey, we might even be able to deal with that shite about the naming restrictions on branches/tags used for packaging...
[18:52:23]<omega8cc>sdboyer: yes?
[18:52:55]<sdboyer>omega8cc: yeah. we don't restrict them in any way in terms of what you can push into repositories, but we still tightly restrict what can be used for releases
[18:53:15]<sdboyer>which, after looking at the mess that is our 44000 release nodes, i appreciate the reasons for a bit more
[18:53:16]<sdboyer>however
[18:53:43]<sdboyer>the system that restricts the naming conventions is no longer hardcoded in project; it's been moved to a ctools plugin
[18:54:23]<mig5>hey sdboyer - sorry i was afk
[18:54:25]<sdboyer>http://drupalcode.org/viewvc/drupal/contributions/modules/drupalorg/drup...
[18:54:28]<mig5>i see you got the git repos already, cool
[18:54:31]<sdboyer>mig5: np, i believe i got what i need
[18:54:32]<omega8cc>ah, so there is workaround always possible
[18:54:34]<sdboyer>yup, you should see them on monday
[18:54:41]<sdboyer>well not even a workaround
[18:54:44]<sdboyer>more a structural solution
[18:55:04]<sdboyer>aegir & drush are the most prominent core-independent drupal projects, but by no means the only ones
[18:55:39]<omega8cc>cool
[18:55:46]<mig5>just the best ones! :)
[18:55:47]* mig5 runs
[18:55:56]<sdboyer>oh i wouldn't argue there :)
[18:55:56]<omega8cc>:)
[18:56:02]<sdboyer>there's been talk for a long time about having a separate category (themes, modules, translations), but with making the release restrictions pluggable, it really gives it teeth
[18:56:03]<mig5>hehe
[18:56:35]* sepehr has quit (Ping timeout: 246 seconds)
[18:56:47]<sdboyer>because we could tie a looser set of restrictions to that other type
[18:57:19]<sdboyer>so we could retain tight restrictions for normal, core-dependent things (which is really a benefit) and have core-independent stuff do its own thing. no harm, no foul
[18:57:48]<sdboyer>dww got on me a bit when i started talking to moshe about this, said he'd take care of it...which i don't quite trust to believe it'll be seen through to its conclusion, but
[18:58:12]<sdboyer>point is, y'all should know it's a real possibility now, that you should lobby for it if you care, and that i'll support efforts for it if you do :)
[18:58:16]<mig5>you sound like our hero
[18:58:20]<sdboyer>awww
[18:58:22]<mig5>that's great stuff, i am really out of touch
[18:58:33]<sdboyer>we've only announced some of this stuff, really
[18:58:38]<sdboyer>it's easy to miss it
[18:58:53]<sdboyer>actually, also, given aegir's interest in doing meta-drupal stuff
[18:58:59]<sdboyer>this issue might be interesting to y'all:
[18:59:23]<sdboyer>http://groups.drupal.org/node/126529
[18:59:46]* sdboyer REALLY wants to start publishing APIs like this
[18:59:48]<mig5>ooer
[18:59:50]<omega8cc>I imagine we don't want to became a drupal-github-anything, but maybe I could also move my bash Aegir installers github repo to drupal git (just dreaming)
[19:00:02]<sdboyer>no reason you couldn't
[19:00:08]<mig5>i can see this being a gold mine for drush_make possibly
[19:00:11]<mig5>as well
[19:00:13]<omega8cc>really?
[19:00:13]<sdboyer>yuuuup
[19:00:17]<omega8cc>wow!
[19:00:23]<sdboyer>omega8cc: there are _no_ restrictions on what you can push in
[19:00:31]* AntiNSA has quit (Ping timeout: 250 seconds)
[19:00:36]<sdboyer>some size restrictions, though those are still loose and unclear, and obviously the license agreement
[19:00:43]<sdboyer>the second you push it to d.o, it becomes GPLv2+
[19:00:51]<sdboyer>other than that, it's your party, have fun :)
[19:01:15]<omega8cc>barracuda and octopus users keep me asking when we will move to d.o since the github issue queue is a total sh*t
[19:01:23]<sdboyer>hehe, it sooo is
[19:01:30]<sdboyer>oh man, here's another little preview for you guys
[19:01:38]<mig5>omega8cc: i totally think you should move to d.o with us if/when we do it
[19:01:40]<omega8cc>it is all GPL v3
[19:01:44]<sdboyer>perfect
[19:01:47]<mig5>for the queues and the ease of working with us
[19:01:54]<sdboyer>so one of the two most exciting moments of the last nine months working on this for me
[19:01:58]<omega8cc>great! I am already happy :)
[19:02:26]<sdboyer>back in november, i pointed tizzo to 'git from the bottom up' after he kept asking me questions about git's internal mechanics
[19:03:01]<sdboyer>he read it, found glip, and three days later had coded a drupal-native repo browser
[19:03:12]<sdboyer>like, filesystem browser. similar to gitweb, github, etc.
[19:03:29]<sdboyer>~350 lines of code, and it firmly covered the basics
[19:03:31]<sdboyer>that, though, is not the cool part
[19:03:36]<mig5>wow
[19:03:54]<sdboyer>well, that's cool, but it's not the FO SHIZZLE HOLY CRAP part :)
[19:04:16]<sdboyer>it was basically built overtop of an API that will enable us to write input filters that you can use to directly reference repo data in your posts on d.o
[19:04:36]<mig5>oh my
[19:05:03]<omega8cc>that is awesome!
[19:05:04]<sdboyer>so, no more copy/paste - write some input filter-y goodness, and it'll pull the code right out of git for you. diffs, too.
[19:05:06]<sdboyer>yar :)
[19:05:30]<sdboyer>we aren't launching with it for a variety of reasons, but we'd like to see it implemented by mid-april
[19:05:38]<sdboyer>at least the repo browser, if not also the input filter
[19:05:54]<sdboyer>so yeah, at that point our issue queues really beat github's :)
[19:06:01]<sdboyer>he's gonna be demoing the browser at drupalcon
[19:06:36]<sdboyer>and somebody, we'll have our wonderful per-issue repos, which should bring us on par with github's pull requests...
[19:06:41]<sdboyer>s/somebody/someday/
[19:07:11]* AntiNSA has joined #aegir
[19:08:40]<mig5>that'll be great for omega8cc to send us pull requests
[19:08:46]<omega8cc>I believe this migration to git will open new sources of power/devs, that is really great move
[19:08:47]<mig5>i hope we never have to generate a patch ever again :)
[19:08:53]<omega8cc>yep
[19:08:56]<omega8cc>:)
[19:09:54]<sdboyer>[00:08:46] <omega8cc> I believe this migration to git will open new sources of power/devs, that is really great move <-- i really hope so
[19:10:05]<sdboyer>it's gonna be a pretty tumultuous time for a while as we sort it all out
[19:10:08]<sdboyer>i look forward to seeing how it pans out
[19:10:44]<sdboyer>and i _really_ hope that we can keep the git team momentum going...i really really want to get as many people grinding on improving/adding things to the d.o git experience as possible
[19:10:56]* sdboyer doesn't like the whole cloistered, secluded infra thing
[19:11:59]<mig5>you seem like you are on the right track anyway
[19:12:08]<mig5>so many ways for it to have gone wrong already but sounds great
[19:12:09]<sdboyer>that and, y'know, i really really don't want a situation where i'm the only one who knows all of our git infra, and so in practice i have to make all the changes :)
[19:12:15]<mig5>hehehe
[19:12:16]* totten has quit (Ping timeout: 240 seconds)
[19:12:17]<mig5>Bus factor
[19:12:21]<sdboyer>haha, exactly
[19:12:26]<sdboyer>yeah i was saying that to nnewton the other day
[19:12:36]<sdboyer>dude the migration scripts that actually do the whole build, moving us from CVS to git...
[19:12:48]<sdboyer>nobody understands those but me. they're twitchy and sensitive
[19:12:56]<sdboyer>they're all in jenkins, so it's all very structured on one level, but
[19:13:00]<mig5>Document! Document! :)
[19:13:06]<sdboyer>yeah i wish :/
[19:13:12]<omega8cc>there are many already using git so they will be the leaders of the tumultuous time and others will follow the smart move :)
[19:13:15]<mig5>yknow with all that spare time you have
[19:13:18]<sdboyer>exactly!
[19:13:20]<sdboyer>hehe
[19:13:35]<sdboyer>i've been avoiding documenting temporary, migration-specific stuff
[19:13:46]* FransK has joined #aegir
[19:14:16]<sdboyer>so there are aspects of the migration which are really just warrens that nobody knows but me. but as long as i live through thursday, that's ok :)
[19:14:36]<sdboyer>fortunately i'm barely leaving the house at this point, and not many buses pass through my living room :P
[19:14:44]<mig5>haha
[19:14:46]<omega8cc>lol
[19:15:09]<sdboyer>afterwards though, yeah. i really, really want to drop that bus factor as mucha s i can
[19:16:45]<omega8cc>there is also volcano factor etc, so yeah, we need git clone sdboyer.git
[19:17:27]<sdboyer>well i dunno about the clone, but i'm pretty sure i know how git push and git pull feel
[19:17:45]<sdboyer>git's been doin that to me for a while now :)
[19:18:32]<sdboyer>oh, actual real question
[19:18:35]<mig5>sometimes i git log
[19:18:38]<mig5>there we go, a poop joke
[19:18:42]<sdboyer>LOL
[19:18:45]<sdboyer>WOOT FOR POOP JOKES
[19:19:02]<mig5>\o/
[19:19:04]<mig5>sorry go on
[19:19:12]<sdboyer>we have scripts that strip translation dirs and id tags from repos
[19:19:32]<mig5>ah
[19:19:37]<sdboyer>can those safely be run on hostmaster & provision?
[19:19:50]<sdboyer>well the id tags definitely can, but translations are the question
[19:19:54]<mig5>by id tags you mean the $Id ?
[19:19:56]<sdboyer>yup
[19:20:03]<mig5>yeah we ignoe those these days, and we have no translations dirs
[19:20:05]<sdboyer>if you don't have any it'll just silently pass by
[19:20:07]<mig5>yep
[19:20:10]<sdboyer>ok cool
[19:21:08]<mig5>we still have a lot of // $Id probably from our migration to git from cvs but yeah they are obviously empty
[19:21:41]<mig5>hmm some may be set in old tags/branches, i'm not sure it's a big deal unless you're telling me it might
[19:22:39]* FransK has quit (Remote host closed the connection)
[19:22:49]<mig5>no they seem empty too, ignore me
[19:23:18]<sdboyer>doesn't matter
[19:23:32]<sdboyer>the id tag stripping hits all branches
[19:23:41]<sdboyer>although...actually you guys might have to run it manually afterwards
[19:23:43]<sdboyer>easy to do, just
[19:24:15]<sdboyer>i have to do clones after the tag/translation cleanup is done, and i dont' have time to refactor at this point
[19:24:46]<sdboyer>but it's no biggie, the script is available, y'all can just run it
[19:25:32]* totten has joined #aegir
[19:35:55]<eft>hefring: tell mig5 to explain the git log poop joke when I grow up
[19:35:55]<hefring>eft: I'll pass that on when mig5 is around.
[19:36:27]* AntiNSA has quit (Ping timeout: 240 seconds)
[19:39:21]<omega8cc>mig5: http://twitter.com/#!/m5ig/status/39237559397650432 :)
[19:41:56]* AntiNSA has joined #aegir
[19:46:47]<sdboyer>omega8cc mig5: there ya go, folks: https://github.com/sdboyer/drupalorg-git/commit/4c410d7c7e7f1c20b8586bf8...
[19:47:37]* HaloFX has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
[19:48:59]* sepehr has joined #aegir
[19:49:23]<omega8cc>sdboyer: thanks :)
[19:49:31]<sdboyer>fsho
[19:50:26]<sdboyer>so yeah, on monday you should check http://git-dev.drupal.org/project/{hostmaster,provision} and they ought to have the latest from your repos instead of the crap in CVS
[19:50:43]<sdboyer>though...it's worth noting that you may not get credit for the early commits in there
[19:50:49]<sdboyer>depending on how it was migrated originally
[19:51:08]<sdboyer>chances are very very good that those CVS commits will go unattributed
[19:51:22]* sdboyer considers a filter-branch for those...
[19:51:40]<sdboyer>well, check on monday in any case
[19:51:45]<sdboyer>that's also something we can fix post-launch
[19:51:55]<omega8cc>sure
[19:52:09]<mig5>nice one
[19:52:09]<hefring>mig5: 16 min 14 sec ago <eft> tell mig5 to explain the git log poop joke when I grow up
[19:52:24]<eft>:)
[19:53:06]<mig5>no wai
[19:53:57]<eft>s/grow/wake
[19:57:52]* totten has quit (Ping timeout: 240 seconds)
[20:02:33]* AntiNSA has quit (Read error: Connection reset by peer)
[20:08:43]* AntiNSA has joined #aegir
[20:13:45]* ronia has joined #aegir
[20:17:10]* mlncn has quit (Remote host closed the connection)
[20:20:11]* eft has left #aegir ()
[20:23:29]* r4gni has joined #aegir
[20:25:44]* Samar has joined #aegir
[20:25:55]* Samar has quit (Client Quit)
[20:26:30]* ronia has quit (Ping timeout: 272 seconds)
[20:26:30]* r4gni is now known as ronia
[20:29:36]* nick_santa has quit (Quit: nick_santa)
[20:38:28]* Slydder1 has quit (Quit: Leaving.)
[20:42:44]* Slydder has joined #aegir
[20:47:07]* jonhattan has joined #aegir
[20:54:28]* EugenMayer has joined #aegir
[20:59:14]* AntiNSA has quit (Ping timeout: 276 seconds)
[21:01:50]* EugenMayer has quit (Ping timeout: 276 seconds)
[21:05:32]* AntiNSA has joined #aegir
[21:05:57]* skwashd has joined #aegir
[21:08:27]* AquaticDisorder has joined #aegir
[21:15:46]* snlnz has quit (Quit: Pain is just weakness leaving the body... <k!15b8>)
[22:39:56]* mlncn has joined #aegir
[22:40:38]* psynaptic|afk is now known as psynaptic
[22:40:39]* psynaptic has quit (Changing host)
[22:40:39]* psynaptic has joined #aegir
[22:42:30]* skwashd has quit (Quit: Leaving)
[22:49:43]* EclipseGc has joined #aegir
[22:58:52]* ronia has quit (Quit: ronia)
[23:09:50]* AntiNSA has quit (Ping timeout: 246 seconds)
[23:12:07]* totten has joined #aegir
[23:12:24]<mig5>hefring: tell EugenMayer this guy seemed to have a slow Dashboard after upgrading too, sounds like you reproduced it. if you troubleshoot it let us know what you find http://community.aegirproject.org/node/373
[23:12:24]<hefring>mig5: I'll pass that on when EugenMayer is around.
[23:14:09]* AntiNSA has joined #aegir
[23:21:48]* samhassell has joined #aegir
[23:34:03]* SqyD has quit (Quit: Leaving.)
[23:35:56]* josh_k has quit (Quit: This computer has gone to sleep)
[23:46:18]* V1ntage has joined #aegir
[23:54:25]* milk has quit (Quit: baaaiiii)
[00:00:07]* Lloyd_Pearson_IV has joined #aegir
[00:08:46]* psynaptic is now known as psynaptic|food
[00:09:49]* EclipseGc has quit (Quit: EclipseGc)
[00:18:16]* bejames_ has quit (Quit: Core meltdown)
[00:39:18]* psynaptic|food is now known as psynaptic
[01:09:29]* samhassell has quit (Ping timeout: 276 seconds)
[01:19:21]* mlncn has quit (Remote host closed the connection)
[01:25:26]* AntiNSA has quit (Ping timeout: 272 seconds)
[01:26:19]* AntiNSA has joined #aegir
[01:36:18]* mlncn has joined #aegir
[01:38:28]* jonhattan has quit (Quit: Ex-Chat)
[01:49:48]* EugenMayer has joined #aegir
[01:57:37]* boztek has joined #aegir
[02:00:56]* psynaptic is now known as psynaptic|break
[02:04:11]<boztek>anyone had any problems updating to rc1 where "/var/aegir/drush/drush.php @platform_hostmaster04rc1 provision-verify --backend" seems to hang?
[02:05:11]<Lloyd_Pearson_IV>boztek: my upgrade went pretty smooth despite mistakes
[02:05:19]<Lloyd_Pearson_IV>but give it some time
[02:05:28]<boztek>this is the first one in a long time to give me trouble
[02:05:30]<Lloyd_Pearson_IV>some parts take a while
[02:10:02]* EugenMayer has quit (Quit: Leaving.)
[02:14:45]<boztek>upgrade went smooth - looks like it just tooke longer to do that verify than it has in the past
[02:16:47]<Lloyd_Pearson_IV>yeh
[02:16:52]<Lloyd_Pearson_IV>thats what i figured
[02:17:03]<Lloyd_Pearson_IV>glad it wnet well
[02:38:07]* sepehr has quit (Ping timeout: 250 seconds)
[02:42:07]* EclipseGc has joined #aegir
[02:48:12]* simesy has quit (Quit: Gordon Bennett)
[02:57:00]* EclipseGc has quit (Quit: EclipseGc)
[03:01:34]* boztek has quit (Quit: Computer has gone to sleep.)
[03:14:16]* totten has quit (Ping timeout: 240 seconds)
[03:15:00]* AntiNSA has quit (Ping timeout: 272 seconds)
[03:19:11]* AntiNSA has joined #aegir
[03:25:57]* AntiNSA has quit (Ping timeout: 240 seconds)
[03:26:02]* psynaptic|break is now known as psynaptic
[03:29:59]* obrienmd has joined #aegir
[03:30:24]* obrienmd has quit (Read error: Connection reset by peer)
[03:30:31]* recidive has joined #aegir
[03:32:17]* AntiNSA has joined #aegir
[04:12:22]* totten has joined #aegir
[04:14:05]* jerryitt has joined #aegir
[04:14:49]* mikl has joined #aegir
[04:14:50]* mikl has quit (Changing host)
[04:14:50]* mikl has joined #aegir
[04:24:43]* mikl has quit (Quit: Computer has gone to sleep.)
[04:42:03]* recidive has quit (Quit: recidive)
[04:42:58]* mikl has joined #aegir
[04:42:59]* mikl has quit (Changing host)
[04:42:59]* mikl has joined #aegir
[04:43:53]* mikl has quit (Client Quit)
[04:45:03]* mikl has joined #aegir
[04:45:03]* mikl has quit (Changing host)
[04:45:03]* mikl has joined #aegir
[04:47:17]* jonhattan has joined #aegir
[04:57:17]* jerryitt has left #aegir ()
[04:59:05]* eft has joined #aegir
[05:03:41]* eft has quit (Ping timeout: 264 seconds)
[05:16:14]* psynaptic has quit (Remote host closed the connection)
[05:18:07]* psynaptic has joined #aegir
[05:18:12]* recidive has joined #aegir
[05:18:34]<anarcat>sdboyer: you still there?
[05:18:34]<hefring>anarcat: 10 hours 36 min ago <omega8cc> tell anarcat sdboyer needs urgent reply/info re: http://groups.drupal.org/node/128479
[05:18:42]<anarcat>sdboyer: i need to catchup on that thread
[05:19:24]<sdboyer>anarcat: it's all good, taken care of
[05:19:45]<sdboyer>anarcat: you guys are gettin cloned in, though you can verify it after rebuild
[05:20:07]<sdboyer>the only potential issue is that you probably won't get credit for your old commits
[05:20:23]<sdboyer>old as in imported-from-cvs-when-y'all-switched-over commits
[05:20:31]<sdboyer>i can work with y'all on that if you want, but it will mean invalidating all your hashes
[05:21:06]<sdboyer>anarcat: but yeah, you're all fixed up as of this commit: https://github.com/sdboyer/drupalorg-git/commit/4c410d7c7e7f1c20b8586bf8...
[05:24:08]* anarcat reading the backlog
[05:25:42]<anarcat>you sir, are amazing
[05:26:34]* V1ntage has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
[05:27:05]<anarcat>i'm looking at the git browser on git-dev.d.o and it doesn't seem to have cloned from our repo (yet?) http://git-dev.drupal.org/node/196005/commits
[05:27:59]<anarcat>i'm really happy by the response from the git team, this is quite amazing, and i don't see any "cons" in my list to migrate to drupal.org
[05:32:07]<anarcat>sdboyer: do you think it would be worth it / easy enough to git-filter our history to fix the authors to map correctly?
[05:32:14]<anarcat>i don't feel like it's worth it...
[05:32:16]<anarcat>mig5: ^
[05:33:05]<sdboyer>anarcat: aww, thanks :)
[05:33:22]<sdboyer>anarcat: yeah it won't do the clone until we rebuild. i'm taking the server down in 4.5 hours for that rebuild
[05:33:35]<sdboyer>so you'll see it when we reopen, definitely by tomorrow morning
[05:33:51]<anarcat>alright
[05:33:59]<sdboyer>anarcat: as for rewriting the history, yep, it should be doable. i've got a free second, let me have a look at your old stuff...
[05:34:27]* sdboyer uses this as an excuse to dodge out of the slog of the migration work for a minute :)
[05:34:29]<anarcat>if it blows up the commit ids though, that means all existing checkouts will need to be redone?
[05:34:32]<anarcat>hehe
[05:35:05]* mlncn has quit (Remote host closed the connection)
[05:36:30]<sdboyer>yep, everythign will need to be reconed
[05:36:35]<sdboyer>err, recloned
[05:36:47]<sdboyer>if the dev team coordinates, you all will be ok
[05:37:00]<sdboyer>make sure that you have everything pushed before launch day and all your dev branches will survive just fine
[05:37:07]<sdboyer>you'll just need to blow away your local copies and reclone
[05:37:25]<sdboyer>since it's just metadata changes it also might be possible to rebase through it
[05:37:29]<sdboyer>git's pretty good at that
[05:38:10]<sdboyer>(if you rebase the now-invalid branch onto its equivalent, it'll go through all the commits, realize they're content-identical and just throw away the commits with the old metadata)
[05:38:30]<anarcat>nice
[05:39:12]<sdboyer>yeah, can't promise anything there, but it's possible
[05:39:36]<sdboyer> commit d962cee2bdeaa8ddc88be1386280978ce8ef45aa
[05:39:36]<sdboyer> Author: adrian <adrian>
[05:39:36]<sdboyer> Date: Mon Jan 7 16:57:43 2008 +0000
[05:39:42]<sdboyer>yeah, that's what i expected to see
[05:39:55]* sdboyer considers
[05:40:12]<sdboyer>i may actually have all the data i need to do this without any intervention from y'all
[05:40:18]<anarcat>there's worse than that
[05:40:30]<sdboyer>do tell
[05:40:46]<anarcat>well, we were (are?) all junior gits back then...
[05:41:10]<anarcat>so the commits that have actual emails as author ids... well, some have varieties of emails :P
[05:42:11]* anarcat wonders what's the difference between "author" and "committer" in git, after reading git-log(1)
[05:42:11]<sdboyer>not a problem
[05:42:30]<sdboyer>well, some are potentially a problem, if they're not real email addresses
[05:42:37]<sdboyer>but that's why we deployed multiple_email on d.o
[05:42:41]<anarcat>$ git log --format=%ae | sort -u | wc -l
[05:42:42]<anarcat>14
[05:42:55]<anarcat>but there's 9 people in there
[05:43:03]<anarcat>so lots of dupes
[05:43:05]<sdboyer>anarcat: http://drupal.org/user/1274/edit/email-addresses
[05:43:10]<anarcat>and ugly shit like this: drumm@drumm-x200.(none)
[05:43:13]<sdboyer>anarcat: man, look how low your uid is, i never realized that :P
[05:43:20]<anarcat>yeah, i know :)
[05:45:00]<sdboyer>anarcat: slightly more complete: git rev-list --all --format=%ae | grep -v '^commit' | sort -u | wc -l
[05:45:03]<anarcat>rah, hostmaster is even worse, there's shit like aegir <aegir@gambit.greenbeedigital.com.au>
[05:45:13]* sdboyer really hates how rev-list emits extra data when you use --format
[05:45:27]<sdboyer>gotta be a flag i'm missing somewhere
[05:45:31]<anarcat>git rev-list --all --format='%an <%ae>' | grep -v '^commit'| sort -u
[05:46:13]<sdboyer>really though, it's not a problem
[05:46:25]<sdboyer>i just need uids for each of those emails
[05:46:45]<sdboyer>and i'll remap the commit to whatever you've chosen on your CVS account page
[05:46:49]<anarcat>well, i guess the problem is: i have no freaking idea who those guys are :P
[05:47:05]<sdboyer>(you have picked the full name and email you want for your migrated commits, right? :P)
[05:47:11]<sdboyer>ah, that's a very different sort of problem :)
[05:47:27]<anarcat>eh
[05:47:29]<sdboyer>well, i guess tweet around, ask around, see if you can find out...if so, i'll put em in. if not, tough tiddlywinks for them
[05:47:39]<anarcat>okay
[05:47:53]<sdboyer>let's move the discussion to a d.o issue, though
[05:47:54]<anarcat>(what about my full name and email?)
[05:48:00]<anarcat>okay
[05:48:01]* sdboyer doesn't really watch g.d.o closely
[05:48:16]<sdboyer>put it in http://drupal.org/project/great_git_migration and tag it 'git phase 2'
[05:48:29]<sdboyer>and note in the issue that i said this is ok or you might get people bitching at you :)
[05:48:31]<anarcat>this is about the commit authors right?
[05:48:42]<anarcat>feature request?
[05:48:44]<sdboyer>commit authors and committers
[05:48:47]<sdboyer>sure
[05:48:50]<sdboyer>migration scripts component
[05:49:01]<anarcat>gotcha
[05:53:16]<anarcat>sdboyer: i don't think i understand what that comment implied: <sdboyer> (you have picked the full name and email you want for your migrated commits, right? :)
[05:53:34]<sdboyer>anarcat: oh right sorry
[05:53:43]<sdboyer>go to the bottom of http://drupal.org/user/1274/edit/cvs
[05:54:00]<sdboyer>looks like you have picked, i see your koumbit email there
[05:54:24]<anarcat>right, i picked anarcat@koumbit.org
[05:54:35]<anarcat>i was wondering if i should add a +git in there
[05:54:39]<anarcat>anarcat+git@koumbit.org
[05:54:43]<anarcat>to ease filtering in the future
[05:54:53]<anarcat>as this email then ends up in all spam lists in the world
[05:56:28]<sdboyer>meh, possibly
[05:56:35]<sdboyer>we don't show it anywhere on d.o
[05:56:44]<sdboyer>i'm using a real one, as well
[05:56:51]<sdboyer>the only way to see the email addy is to clone the repo
[05:57:07]<sdboyer>frankly if spambots are that good, they deserve my email. lol
[05:57:37]<anarcat>eh
[05:57:48]<sdboyer>but, whatever you want to do - that's the benefit of the multiple_email module :)
[06:00:04]<sdboyer>heh, this'll be interesting...need to pass a git command for php to exec through serialization in beanstalk, and have it retain the escaped sh script string within that git command all correctly
[06:00:26]<sdboyer>anarcat: yeah it...may be easier if i compose these commands for you, then y'all run them just before launch so that the transforms are already done :P
[06:01:09]<anarcat>i'm writing this long issue with lots of details, i hope i won't piss people off :)
[06:03:10]<anarcat>sdboyer: should i open a separate issue for branch/tag naming conventions?
[06:04:21]<sdboyer>anarcat: haha...yeah, separate issue for branch/tag naming conventions. actually let me see if i can find one in the backscroll elsewhere, i know there is one
[06:04:56]<sdboyer>anarcat: so it kinda ends up being this one, probably - http://drupal.org/node/322626
[06:05:28]<sdboyer>anarcat: and just to make sure we're on the same page, there are zero restrictions on what branches/tags you can push in - just the branches/tags you can create releases from
[06:05:36]<anarcat>oooh that issue
[06:05:39]<sdboyer>anarcat: hehe, yeah
[06:05:49]<anarcat>i opened the first issue http://drupal.org/node/1067500
[06:05:53]<sdboyer>that's the one dww pointed us to when i brought this up with moshe the other day
[06:05:57]<anarcat>okay, i understand
[06:06:01]<anarcat>yeah, i remember that issue
[06:06:05]<sdboyer>it's a little more his territory than mine
[06:06:15]<sdboyer>but the fact is it is WAY more feasible now
[06:06:32]<sdboyer>since instead of rejecting the branches/tags outright, we accept them and just filter them out
[06:06:41]<sdboyer>so i think with some pressure we could get it resolved fairly quickly
[06:07:20]<sdboyer>the other question will be whether it gets resolved by specifying the branch/tag restrictor to use on a per-project or per-project-type basis (or both)
[06:09:05]<anarcat>i don't really understand why the restrictions are there in the first place :P
[06:09:14]* fysa has quit (Ping timeout: 260 seconds)
[06:11:21]<anarcat>sdboyer: problem with that issue is that it assumes that the proper fix is B) (per project type)
[06:11:24]<anarcat>http://drupal.org/node/322626#comment-4116688
[06:13:18]<sdboyer>anarcat: i tend to agree
[06:13:36]<sdboyer>anarcat: i think there's an easy way to win that argument, or at least change it, though
[06:13:42]<sdboyer>anarcat: core is a 'Drupal project'
[06:14:22]<sdboyer>anarcat: as are a dozen or so others. but the release restrictions those projects should have (if they "should" even have repos at all) are nothing like core's
[06:14:37]<sdboyer>ipso facto, those categories aren't enough
[06:14:42]<anarcat>right
[06:14:44]<sdboyer>trying to reuse them for this purpose is a mistake, i think
[06:15:20]<sdboyer>i mean they're a good starting point. but it should be global default < project type default < individual project setting
[06:15:31]<anarcat>well, that would be ideal
[06:15:36]<anarcat>but it's harder to implement
[06:15:38]* fysa has joined #aegir
[06:15:39]<sdboyer>i know this is something we could pretty easily put into vc_project
[06:15:55]<sdboyer>well, fairly. the fact that we use taxo for those types is awkward. but still
[06:16:36]<sdboyer>not too hard :) we do it all with ctools plugins, and the VersioncontrolRepository classes can take as many ofthose as we want. and those are per-project, so that gives us the ultimate end-goal of a per-project setting, if we locate them there
[06:16:47]<sdboyer>so per-project would actually be easier than per-type
[06:17:10]<sdboyer>global default < individual repo is a pattern we already use for half a dozen plugin types all across gd.d.o
[06:17:25]<anarcat>nice, i see
[06:17:30]<anarcat>so maybe we should just do that :)
[06:20:12]<anarcat>okay, i think i have done what i can here, not sure to go further
[06:20:25]<anarcat>sdboyer: thank you very much for the support, I hope this all goes well and we're able to switch back
[06:20:46]<sdboyer>anarcat: you bet. i hope so too
[06:20:48]<anarcat>sdboyer: i think that whatever happens now anyways, we'll go back to drupal.org - even with screwed up history or silly naming convention
[06:21:19]<sdboyer>anarcat: responded btw: http://drupal.org/node/1067500#comment-4116702
[06:21:20]<anarcat>and actually, thinking about it...
[06:21:22]<sdboyer>anarcat: yay! :)
[06:21:54]<anarcat>since we have all those tags already in here, it would be quite annoying to have to skip all those older tags that don't have release nodes associated to them on drupal.org everytime we need to do a release
[06:22:10]<anarcat>so in a way, if the convention is enforced, that could actually make things easier for us, because it will ignore older tags
[06:22:18]<sdboyer>there ya go :)
[06:22:44]<sdboyer>honestly i think that the meta-drupal work people do gets more important as drupal gets bigger and bigger
[06:23:08]<sdboyer>it's the stuff that 'wraps' drupal that lets us do a whole different level of awesome stuff to it
[06:23:23]<sdboyer>so, not having a friendly home for things like that on d.o is just...insulting, really
[06:23:27]* EclipseGc has joined #aegir
[06:23:41]<anarcat>yeah
[06:23:55]<anarcat>but then we may be trying to do everything at once... ;)
[06:23:58]<anarcat>let's just get git working :)
[06:24:11]<anarcat>okay, so i hear it would be better for me to try to cleanup my history first :)
[06:24:17]<anarcat>i can try!
[06:24:27]<anarcat>array("filter-branch -f --prune-empty --tree-filter " . escapeshellarg("'rm -rf branches'") . " -- --all", TRUE)
[06:24:31]<anarcat>why isn't this just:
[06:24:36]<anarcat>array("filter-branch -f --prune-empty --tree-filter 'rm -rf branches'") . " -- --all", TRUE)
[06:25:27]<anarcat>http://help.github.com/changing-author-info/ <- ouch. that's ugly.
[06:29:05]* josh_k has joined #aegir
[06:30:16]<anarcat>sdboyer: seems like derek is awake :)
[06:31:08]<sdboyer>anarcat: because when i tried it the way you initially wrote, it failed. i haven't tested the escapeshellarg() version yet, that'll be tested in the afternoon's builds. i'll have it sorted before the servers coem back up
[06:31:43]<sdboyer>anarcat: yeah, i don't know if there's a more efficient way to do the author renaming, but the fact that that's what github has posted suggests to me there really isn't
[06:34:47]<anarcat>here we go, i entered the ring with derek :)
[06:39:25]<anarcat>hahaha those guys are insane http://blog.l4rk.com/2009/05/pair-programming-git-github-gravatar.html
[06:43:05]* jonhattan has quit (Quit: Ex-Chat)
[06:44:51]* EclipseGc has a new debian install going to get a totally fresh rc1 of aegir up
[06:44:52]<EclipseGc>:-D
[06:49:46]* snlnz has joined #aegir
[06:50:29]* eft has joined #aegir
[07:00:20]* snlnz is gone, autoaway/10m (l!on)
[07:06:25]* SqyD has joined #aegir
[07:21:33]* Slydder1 has joined #aegir
[07:22:53]* Slydder has quit (Ping timeout: 264 seconds)
[07:22:55]* snlnz has quit (Ping timeout: 255 seconds)
[07:30:10]* snlnz has joined #aegir
[07:31:07]* nomad411 has joined #aegir
[07:42:32]* recidive has quit (Quit: zzz)
[07:48:28]* recidive has joined #aegir
[07:49:06]* EclipseGc loves how effortless it is to get a new aegir system up and running these days
[07:49:29]* recidive has quit (Client Quit)
[07:50:14]* anarcat is happy to hear that :)
[07:51:46]* recidive has joined #aegir
[08:03:27]* mlncn has joined #aegir
[08:04:27]* recidive has quit (Quit: zzz)
[08:11:02]* FatGuyLaughing has joined #aegir
[08:11:52]* FatGuyLaughing has quit (Client Quit)
[08:19:31]* cafuego_ is now known as cafuego
[08:19:46]* cafuego has quit (Changing host)
[08:19:46]* cafuego has joined #aegir
[08:34:41]* V1ntage has joined #aegir
[08:37:22]* mikl has quit (Quit: Computer has gone to sleep.)
[08:50:54]* realityloop has joined #aegir
[09:04:03]* Plsss has joined #aegir
[09:07:14]* TrevorT has quit (Ping timeout: 276 seconds)
[09:11:47]<mig5>anarcat: all good with the git migration stuff?
[09:11:52]<mig5>i don't care about credit for commits
[09:12:09]<mig5>and i don't care about re-checking out stuff: i check out every time i work on aegir anyway, then blow it away (cloud, maaaaan)
[09:12:29]<mig5>plus the upgrade.sh for the bleeding edge people on HEAD, will fetch new components each time anyway
[09:13:43]* anarcat nods
[09:14:22]<anarcat>sdboyer: i was thinking... the way we do releases on aegirproject.org right now is that we push a specially-formatted tag and then git takes care of the rest (ie. it generates a tarball and md5 from that)
[09:14:50]<anarcat>sdboyer: would it be possible to automatically generate the release nodes (from the git tag annotation) and subsequent tarballs when we push tags to drupal.org too? :)
[09:15:04]<anarcat>mig5: pretty amazing that feedback we got, didn't expect as much!
[09:16:58]* nick_santa has joined #aegir
[09:18:30]* skwashd has joined #aegir
[09:22:11]* Plsss has quit (Ping timeout: 276 seconds)
[09:22:16]* EugenMayer has joined #aegir
[09:22:30]<EugenMayer>Hello
[09:22:30]<hefring>ni hao
[09:22:30]<hefring>EugenMayer: 10 hours 10 min ago <mig5> tell EugenMayer this guy seemed to have a slow Dashboard after upgrading too, sounds like you reproduced it. if you troubleshoot it let us know what you find http://community.aegirproject.org/node/373
[09:23:22]<anarcat>maybe iterating over the task forms was a bad idea after all
[09:23:32]<mig5>i couldn't reproduce it on a fresh install
[09:23:34]<EugenMayer>Hello guys
[09:23:42]<mig5>and i can't imagine why it only affects upgrades
[09:23:56]<EugenMayer>mig5: well i started logging slow queries. Doesnt seem to look to good
[09:24:34]<EugenMayer>rc1 is extremly more CPU intensive an mysql eating. I was working on other issues this WE but i will find the days to check what it is
[09:24:44]<mig5>weird
[09:24:56]<EugenMayer>mig5: e.g. we miss an index on package type
[09:25:13]<EugenMayer>it will not boost it light years, but it wont harm it
[09:25:37]<mig5>yeah, but it's not like a missing index will suddenly make things *slower*
[09:25:41]<mig5>after we have had no indexes all this time
[09:25:47]<mig5>so it's something else
[09:25:49]<EugenMayer>i think we should cashe the most intense queries, and that are most like the one joining with the hosting_package table
[09:25:55]<mig5>and if it only affects the Dashboard, to me that suggests the bulk operations stuff
[09:26:09]<EugenMayer>Yeah, its not that index, i tried it. Its also not any other index
[09:26:22]<EugenMayer>its _very_ slow.
[09:26:29]<mig5>was it a fresh install or an upgrade from 0.4-beta2
[09:26:36]<EugenMayer>upgrade
[09:26:47]<EugenMayer>8 sites, 6 complex platforms, 6 servers
[09:27:45]<EugenMayer>but: i have a fserver running on aegir + drush_make_ui. Eventhough those things are slowing it down, it is def. related to the upgrade. I did not change anything on the other parts, just the upgrade was slowing down things
[09:27:46]<mig5>mmm
[09:27:57]<EugenMayer>most of the patches coming with rc1 were already in my b2
[09:28:05]<mig5>yep, well it sounds like the same issue that this 'kevin' guy had
[09:28:13]<mig5>so we probably need this in the Hosting queue somewhere
[09:28:31]<EugenMayer>Yeah, i is def. a hosting issue, not provisioning
[09:28:33]<mig5>i wonder why it would only affect upgrades and not fresh installs
[09:28:43]<EugenMayer>hmm
[09:29:21]<EugenMayer>i hacve 600 packages in hosting_package, so not a huge number here
[09:29:29]<EugenMayer>URGS
[09:29:32]<EugenMayer>23k in instances
[09:29:33]<EugenMayer>!
[09:29:39]<EugenMayer>that sounds _very_ odd.
[09:29:57]<mig5>it's hard to say if this is a new thing though.. we need to compare it with a 0.4-beta2 instance
[09:30:00]<mig5>and it varies for everyone
[09:30:01]<EugenMayer>And is a huge number. What about some upgrade things freaked out and we have tons of ghost in the hostin_package_instances table?
[09:30:08]<mig5>it's possible
[09:30:31]<EugenMayer>well my older b2 has 1,5k in there
[09:30:36]<EugenMayer>so thats factor 20x
[09:30:43]<EugenMayer>old != before upgrade though
[09:31:19]<EugenMayer>ar/www/aegir/hostmaster-0.4-alpha14/modules/ping...
[09:31:26]<EugenMayer>well that is a ghost 100%
[09:31:29]<EugenMayer>( instances )
[09:31:42]<EugenMayer>part of rc1, platform deleted ages ago
[09:32:11]<EugenMayer>what happens if i truncate _instances? Does it get regenerated?
[09:32:32]<mig5>i think so, on Verify. correct anarcat?
[09:32:44]<mig5>take a backup first EugenMayer :)
[09:32:54]<EugenMayer>its not my live site, iam not crazy :)
[09:33:00]<mig5>hehe
[09:33:33]<EugenMayer>well ok, _instances is empty
[09:33:45]<EugenMayer>and the dashboard loads for 10s .. so thats not the one
[09:33:58]* nomad411 has quit (Quit: Back to the unreal world...)
[09:34:11]<mig5>ok
[09:34:12]<mig5>so
[09:34:17]* boztek has joined #aegir
[09:34:23]<mig5>can you try and revert the bulk operations commit
[09:34:27]<mig5>i know the kevin guy said it didn't make a difference
[09:34:29]<EugenMayer>jap
[09:34:39]<EugenMayer>as i have a testing aegir locally thats no issue
[09:34:51]<mig5>ah, there's two commits actually
[09:34:57]<mig5>http://git.aegirproject.org/?p=hostmaster.git;a=commitdiff;h=7ac2c5fe7c9...
[09:35:01]<mig5>http://git.aegirproject.org/?p=hostmaster.git;a=commitdiff;h=4f16634d5cf...
[09:35:07]<mig5>the guy reverted the first one (the older one)
[09:35:12]<mig5>i wonder if the latter one needs reverting too
[09:35:31]<EugenMayer>hmm my provisioning is not a git checkout - will need to do this first
[09:35:40]<mig5>oh it's in hostmaster
[09:35:46]<mig5>but probably the same case for you
[09:35:48]<EugenMayer>ah, anyway
[09:35:55]<mig5>do a separate git clone and copy the .git dir into place
[09:36:10]<anarcat>mig5: what's that?
[09:36:23]<anarcat>you guys generate too much data for me to parse :)
[09:36:28]<mig5>anarcat: ignore earlier
[09:36:34]<mig5>this is not related to the package instances i don't think
[09:36:39]<anarcat>got it
[09:36:41]<mig5>i think it's the two bulk oeprations commits
[09:36:45]<mig5>or one of them
[09:36:54]<mig5>maybe that === TRUE stuff is needed
[09:37:00]<anarcat>ah, maybe
[09:37:01]<mig5>no idea
[09:37:51]<EugenMayer>mig5: looks i have to merge those
[09:37:51]<mig5>strange it should only affect upgrades. my 0.4-rc fresh install was nice and snappy
[09:38:01]<mig5>EugenMayer: they were committed as part of 0.4-rc1
[09:38:06]<mig5>are you on an older version?
[09:38:51]<EugenMayer>no, i was wrong. Just did a mistake
[09:39:51]<EugenMayer>ok now my intallation bootstraps the checkedout rc1, lets seem what happens if a revert those. Any preference which i should start with? latter one?
[09:40:32]<mig5>yeah i guess revert them back to front, so revert the newer one first, then the older one
[09:40:42]<mig5>git revert 4f16634d5cf68baca7f5035241fa39ea7e6ac7fe
[09:40:50]<mig5>git revert 7ac2c5fe7c98935d14c4eb8d37ee13937ba62860
[09:41:22]<EugenMayer>4f1 had no effect
[09:41:36]<mig5>hmm
[09:42:13]<EugenMayer>7ac either. switching to b2 to verify its faster
[09:42:47]<sdboyer>anarcat: sorry, i temporarily released the ball and chain binding me to my laptop so i could go booze it up then throw a friend into rehab
[09:42:51]<sdboyer>back now!
[09:42:53]<mig5>well, i'm stumped
[09:43:18]* realityloop has quit (Read error: Connection reset by peer)
[09:43:21]<EugenMayer>mig5: well now iam confused. went back to b2 with the same prov / db. Still far to slow
[09:43:26]<mig5>wtf
[09:43:27]<EugenMayer>so its a db issue?
[09:43:43]<sdboyer>mig5: i think you may care more about commit credit as time goes on. i'm hoping that git will usher in an age of commits being much more meaningful
[09:43:47]* realityloop has joined #aegir
[09:43:55]<sdboyer>anarcat: so, re: autogenerating release nodes
[09:44:02]<mig5>sdboyer: oh i probably will, what i mean is i don't care if it breaks for now
[09:44:09]<anarcat>sdboyer: no problem! :)
[09:44:25]<mig5>often i wish my commits weren't by me anyway :)
[09:44:29]<sdboyer>mig5: once commits are in, though, they're inviolate
[09:44:32]<sdboyer>mig5: lol :P
[09:44:33]<mig5>#migression
[09:44:45]<sdboyer>anarcat: certainly initially autogenerating release nodes will not be possible
[09:44:57]<sdboyer>anarcat: however, and we were talkin about this last night a bit
[09:45:10]* recidive has joined #aegir
[09:45:37]<sdboyer>anarcat: we already have the basic data necessary for a rich push-response environment
[09:45:54]<sdboyer>and hooks, currently only used by pift, that act on those incoming pushes and can do stuff
[09:45:54]<EugenMayer>mig5: truncated task_log ( 2k ) still ~8s
[09:46:04]<anarcat>... pift?
[09:46:18]<sdboyer>sorry, project_issue_file_test, the thing that hooks up all the automated testing on d.o
[09:46:53]<sdboyer>anarcat: one of the things i would like to do is define a spec for plugins that can respond to pushes. people can then write plugins for whatever, infra can review them, and if approved they then become available for every maintainer to selectively activate on their project
[09:47:07]* EugenMayer doh!
[09:47:18]<sdboyer>anarcat: i'm hoping we can at least define that spec by the end of drupalcon, so that people can get started with those plugins asap
[09:47:18]<EugenMayer>i was clearing the wrong DB. i was working on the old b2
[09:47:40]<anarcat>sdboyer: sounds fun :)
[09:47:59]* mikl has joined #aegir
[09:48:00]* mikl has quit (Changing host)
[09:48:00]* mikl has joined #aegir
[09:48:08]<EugenMayer>mig5: aahh! 118k in task log. Thats 100% the issue.
[09:49:47]<sdboyer>anarcat: hopefully :) so once we get that together, you'll be able to write a plugin that generates the release node if this and that condition is present
[09:50:46]<anarcat>EugenMayer: let me guess - ereg is deprecated? :)
[09:50:57]<EugenMayer>yes :)
[09:51:09]<EugenMayer>decades.
[09:51:40]<EugenMayer>well truncated task_log, instances without any effects
[09:51:50]<EugenMayer>i have 1k nodes though
[09:54:06]* mikl has quit (Quit: Computer has gone to sleep.)
[09:55:10]<EugenMayer>:/ I was tricked by my VPN. I was testing the live aegir, working with db deletion localy. Thats a bad start :) let me redump the db
[09:56:06]<EugenMayer>I was traveling today, working localy on my laptop. Just came online and got pushed the dns server, overriding my local ip...so i landed online.
[09:56:22]<EugenMayer>Sorry for the confusing. taking back all results. Now trying it slowly
[09:56:29]* V1ntage has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
[10:01:19]<EugenMayer>anarcat: mig5: after that issue. I have a lot of code to share and i have 2-3 deep-code questions. I think i found numerous bugs with contextes which have a lot of side effects and also will be important in the feature when the provision.api get used more (as its affected most of it). do you have time for a 15 minutes chat to clear out those questions, so i can finalize the code? I have a working code of a UI for site-vhost settings ( sep
[10:03:32]* Vertice has joined #aegir
[10:03:33]<EugenMayer>i also added backup-slim which is basical "backup" without files. migrate-ng checks if its a "same server migration", if not, it invokes backup, otherwise backup-slim. Basically i did not hack provisioning, but rather introduced those new tasks which base call as much code of the original ones as possible.
[10:09:40]* mrfelton has joined #aegir
[10:10:33]<EugenMayer>mig5: ok finally. reverting both of those checkins fixed the slowlyness. going back to rc1 and reverting them one by one
[10:12:58]* HaloFX has joined #aegir
[10:14:17]* AquaticDisorder has quit (Remote host closed the connection)
[10:14:17]<mig5>i can't chat now as i'm at work
[10:14:30]<mig5>but if you mean reverting those bulk operations commits really fixed it after all, please open a ticket in Hosting about it
[10:14:33]<mig5>thanks
[10:14:40]<EugenMayer>ok
[10:18:00]<EugenMayer>anarcat: if you would have time, give me a note
[10:18:04]<omega8cc>anarcat: btw, I believe aegir@gambit.greenbeedigital.com.au is mig5 :)
[10:19:47]<mig5>oops
[10:19:55]<mig5>that must be a very old commit
[10:20:09]<mig5>bbl
[10:20:37]<omega8cc>so, we should really delete zombies from packages when we are deleting stuff like platform, as zombies slows down create site form (and probably many more)
[10:20:50]<EugenMayer>omega8cc: jap
[10:22:07]<EugenMayer>omega8cc: but as the deletiong of platforms is not robust right now, e.g. you cant do that if migration fails or you have other isseus ( we dont have purge yet ), so deleting by hand happens, i would rather tend to introduce a cleanup-job which truncates the table and regenerates it completely ( once a day, once a week, on demand what ever )
[10:22:42]<EugenMayer>That would also fix that we need to cover that issue for upgraders i guess
[10:22:51]<EugenMayer>"existing installations"
[10:22:59]* LloydPearsonIV has joined #aegir
[10:23:43]* jerryitt2 has joined #aegir
[10:24:37]<LloydPearsonIV>anarcat: mig5 & Vertice , any update on a workaround for that comments & theme situation? hope i am not bugging, i just dont know where to expect the update on the status to be posted
[10:25:00]<omega8cc>EugenMayer: we need some maintenance/purge/regenerate action to clean up any old mess, but when we are deleting empty platform in the front-end, it should delete also all its related stuff anyway
[10:25:13]<EugenMayer>omega8cc: signed
[10:25:52]* sepehr has joined #aegir
[10:26:01]<EugenMayer>omega8cc: after testing migrations up and down i really thinkg we have more then 5 unrecoverable cases in the normal migrate right now. There are also some during installations or cloning
[10:26:23]<EugenMayer>omega8cc: so cleanup / purge would be great
[10:28:37]<omega8cc>EugenMayer: yeah, we have them "by design" because Aegir is designed to panic & fail if *anything* goes wrong, instead of to force action and create even more mess, but still, we need the ability to clean up the mess, no matter what is it
[10:31:20]<omega8cc>LloydPearsonIV: I believe it is a weekend, no? :)
[10:32:01]<LloydPearsonIV>omega8cc: i was simply trying to get an estiamte, i know its not magic, it just seems that way :)
[10:33:31]<mig5>LloydPearsonIV: we are waiting on your patch! ;)
[10:33:40]<omega8cc>lol
[10:33:41]<mig5>if we are moving too slowly for you, go right ahead
[10:33:51]<mig5>it's a theme, i have no idea how that shit works to be honest
[10:34:32]<LloydPearsonIV>mig5: i am not a dev or a themer, just an integrator. I was simply wanted to get an idea of when, not rush anybody
[10:34:47]<LloydPearsonIV>i hope & dont mean to make it seem as if am pressing
[10:34:53]<LloydPearsonIV>its more of curiosity
[10:36:28]<mig5>no idea of when at this stage
[10:36:42]<mig5>it's a very minor issue, no offense meant by that
[10:36:45]<mig5>just in relation to other stuff
[10:36:47]<mig5>and i'm at work
[10:38:11]<EugenMayer>mig5: be aware iam flooding the isseu queue! So only look there if you are in a good mood :)
[10:38:25]* EugenMayer dodges any flying objects
[10:39:22]<EugenMayer>omega8cc: interested in testing site settings / site vhost UIs ? i have finshed those features
[10:40:02]<LloydPearsonIV>mig5: no biggy, i guess i will just have to figure out something
[10:41:11]<EugenMayer>i think both features should even move into core. I dont think its practical to set settings in hooks, as you simply lose the overview. Things like base_url for notifications, or chain certs for ssl sites. Or whatever. It is simply needed to group those things into the site in the UI to have an overview whats happening on the site and make it more practical for the non-programmer
[10:42:26]<EugenMayer>AFAIS mig5 was simply just inviting people to write those things. The provision API and the hooks provided there, the task implementation and the pre_hosting_task hook is just a complete set of tools here. Thanks for the APIs over there, very handy
[10:43:07]<omega8cc>EugenMayer: where can I find them?
[10:43:24]<EugenMayer>still on my private repos, but if interest exist i push them on github
[10:44:00]<omega8cc>EugenMayer: sure, please push to github if you can
[10:44:20]<EugenMayer>give me some minutes, just creating the patch for the Chain-SSL-Cert support
[10:47:09]* Slydder1 has quit (Quit: Leaving.)
[10:50:08]<EugenMayer>omega8cc: maybe one thing before i push those: is it possible to declare the hook_provision_apache_vhost_config methods in the mymodule.drush.inc file? If i do so, they seem to never got called
[10:51:01]<EugenMayer>the issue with that is, that both, sites settings / sites vhost are then needing a special file in .drush/ to implement that hook which makes both modules harder to install, when they dont become core ones.
[10:52:39]* jerryitt has joined #aegir
[10:53:50]* jerryitt2 has quit (Ping timeout: 240 seconds)
[10:58:44]<omega8cc>EugenMayer: hmm... you should be able to re-use existing hooks and maybe submit it as a patch for core instead of fighting with contrib space? I believe extra vhost settings should be in core
[10:59:48]<EugenMayer>omega8cc: well thats not for me to decide. I add that file to the repos that, you have to move them under .drush for now, most probably to change then