IRC logs for #aegir, 2011-04-14 (GMT)

‹ 2011-04-13
2011-04-15 ›
TimeNickMessage
[10:01:08]<anarcat>those could deserve our attention too :) https://drupal.org/project/issues/search?issue_tags=aegirWTF
[10:01:16]<anarcat>that one especially - https://drupal.org/node/908314
[10:01:18]<anarcat>low hanging fruit
[10:01:28]<anarcat>but we should refrain from doing a bunch of crazy coding now
[10:01:36]<anarcat>let's just take the patches that are there if they look reasonable
[10:01:55]* arianek has joined #aegir
[10:03:01]* scientist has joined #aegir
[10:03:53]* snlnz has joined #aegir
[10:05:26]<mig5>good
[10:05:32]<anarcat>good good
[10:05:36]<mig5>i haven't tested the cluster feature itself though
[10:05:41]<mig5>but it didn't break anything else
[10:07:35]* smoothify has quit (Quit: Ex-Chat)
[10:08:11]* JoeMurray has joined #aegir
[10:08:15]<anarcat>good.
[10:09:09]* clou has quit ()
[10:09:58]* clou has joined #aegir
[10:10:02]* gusaus has quit (Quit: gusaus)
[10:11:29]* clou has quit (Client Quit)
[10:12:15]* clou has joined #aegir
[10:12:49]<anarcat>hey, another easy rollback fix maybe? https://drupal.org/node/1081536
[10:18:06]* anarcat maliciously turning a bunch of bug reports into feature requests :P
[10:20:07]<mig5>:)
[10:20:30]<mig5>suits me, you know i am brutal in the queues
[10:20:34]<mig5>or at least Eugen would say so :)
[10:20:59]<mig5>that's a crapload of fixed bugs in under a fortnight http://drupal.org/project/issues?text=&projects=hosting%2C+provision%2C+...
[10:21:03]<mig5>nice
[10:25:34]* psynaptic|away is now known as psynaptic
[10:28:30]<JoeMurray>I have a client who has about 85 sites they would like to migrate to Drupal / CiviCRM fairly standard config, and expect to double that within 6 months. I heard from bgm that aegir is good for single server production provisioning now, but remote functionality/stability isn't quite there for non-developers. Comments? Timeline for when it would be safe to recommend to semi-technical clients...
[10:28:32]<JoeMurray>...for remote provisioning (e.g. one site per vps that is auto provisioned by linode api)
[10:28:58]<JoeMurray>say 6- 12 months?
[10:31:05]<anarcat>rah there are like 30 issues more than 2 months old
[10:33:56]<anarcat>JoeMurray: that sounds reasonable
[10:34:46]<anarcat>JoeMurray: koumbit has remote server refactoring on its roadmap for the end of the summer (or rather "summer", as we milestone it internally :), but we can up that priority with external funding and support
[10:35:18]<JoeMurray>anarcat: thx. I will propose use big servers initially, with plan to move to remote server support later
[10:35:49]<anarcat>pretty much what we're doing here :)
[10:35:58]<anarcat>if you need any help, let us know
[10:36:18]<JoeMurray>anarcat: I don't know how flush with cash this client is, or how insistent on 1 site / vps, with all of separation and other benefits that provides
[10:36:46]<JoeMurray>bgm said you have up to 100 civi/drupal sites on a webhead server
[10:36:57]<JoeMurray>with MySQL on separate backend
[10:37:30]<JoeMurray>how big is your frontend server? 32GB? 15k drives?
[10:38:01]<JoeMurray>dual quad processors?
[10:39:06]<anarcat>dual quad, 16G, SATA 7k drives (iirc)
[10:39:22]<JoeMurray>thx!
[10:39:31]<anarcat>we replaced it with a 8G because we had a BIOS corruption, and it still works well
[10:39:38]<anarcat>DB is 12G SAS, iirc
[10:39:43]* eft has joined #aegir
[10:40:18]<JoeMurray>wow, much smaller than I would have guessed. But I have 500k contacts in a couple of my CiviCRM db's
[10:40:40]<JoeMurray>these guys will be probably 1 - 5k contacts
[10:41:38]<JoeMurray>do you master / slave your DB for high avail fail-over?
[10:41:40]<anarcat>yeah, i am not sure we have that big contact lists yet
[10:41:42]<anarcat>yes, we do
[10:41:47]<anarcat>which is why it's not so big :P
[10:41:52]<anarcat>we pay twice for those puppies :)
[10:42:22]<JoeMurray>yeah. do you have it set up to read from slave and write to master, or just use slave as hot-swap
[10:42:45]* Aurorus has joined #aegir
[10:43:00]* Aurorus has quit (Client Quit)
[10:44:23]<anarcat>just as hotswap for now
[10:44:39]<anarcat>we do mostly shared hosting on that stuff so we don't control the applications well enough to do that kind of scaling
[10:44:46]<anarcat>besides, it's something only d7 does, iirc
[10:44:49]* patcon|myplanet has joined #aegir
[10:44:52]<anarcat>maybe pressflow
[10:45:13]<JoeMurray>and for the 8G / 16G webheads, do have a cluster of 2 or 3 or ??
[10:45:27]<anarcat>we only have one webhead for now, which is a big issue for us
[10:45:53]<anarcat>so we are committed to fixing this in the short (trimester as opposed to years :)) horizon
[10:46:13]<anarcat>mig5: can you test a rc7 to head upgrade after i push this?
[10:46:33]* lukus has quit (Quit: Leaving)
[10:46:34]<JoeMurray>I think I'll propose they use cluster of two 8G, which will make it easy to add more if necessary
[10:46:55]<JoeMurray>many thanks - this is very useful
[10:47:12]<anarcat>yeah, but keep in mind that some people had trouble with cluster mode
[10:47:24]<JoeMurray>with aegir?
[10:47:27]<anarcat>yes
[10:47:55]<anarcat>you will certainly have better luck "sharding" amongst the servers (X sites on server A, Y sites on server B) than "clustering: (X + Y sites on servers A and B)
[10:48:37]<JoeMurray>which deal with load but not reliability as much as a cluster. ok
[10:49:18]<anarcat>right
[10:49:33]<JoeMurray>So i'm imagining your trimester horizon is when you think aegir will be production-ready on that front?
[10:49:47]<anarcat>well
[10:49:54]<anarcat>i think that's actually optimistic right now
[10:49:55]* mapleleaf_ has joined #aegir
[10:49:59]<anarcat>but that's what i'd *like* to think :)
[10:50:03]<JoeMurray>:)
[10:50:29]<JoeMurray>our roll-out is probably 5 months or more away. i'll leave this question open for a while
[10:50:33]<anarcat>simply put - we'll start coding this in the summer, and since Koumbit is running "the stable branch" in production, all this really depends on the 2.x release schedule too
[10:50:39]<anarcat>now koumbit is running the rcs in production
[10:51:02]<anarcat>but maybe we'll start pushing out betas for 2.x and run that in production on our side, which would rush towards a 2.x release in autumn
[10:51:08]<anarcat>but would still mean it would be production ready
[10:51:11]<anarcat>because we would run it :P
[10:51:16]* mapleleaf has quit (Ping timeout: 250 seconds)
[10:51:30]<anarcat>aegir releases have been traditionnally *really* stable regardless of the naming (alpha/beta/rc)
[10:51:31]* mapleleaf_ is now known as mapleleaf
[10:51:33]<anarcat>because we release often
[10:51:36]<anarcat>mig5: ping
[10:51:45]<JoeMurray>but my client doesn't have the internal resources like you do to deal with issues, so we'll wait for a stable stable rel :)
[10:52:15]<anarcat>that's why we sell our soulds to client like yours ;)
[10:52:18]<JoeMurray>yes, aegir alpha <> normal alpha
[10:52:22]<anarcat>souls*
[10:52:30]<anarcat>yeah
[10:53:22]<JoeMurray>as you may have heard from bgm, i'm putting in for some dev $ for Provision CiviCRM. I'm also going to suggest they look to Koumbit for Aegir support @ $200 / hr
[10:54:07]<JoeMurray>(with perhaps an irc answer or two as well)
[10:54:17]<JoeMurray>ciao
[10:56:44]<anarcat>that's sweet
[10:56:46]<anarcat>take care
[10:57:42]* gusaus has joined #aegir
[10:57:43]<mig5>hey anarcat
[10:57:46]<mig5>sorry went out for coffee
[10:58:04]<anarcat>mig5: all good - i committed a release-breaking thing that should be pretty easy to test
[10:58:11]<anarcat>have a way to do a quick upgrade to head?
[10:58:21]<anarcat>i tried to filter through the bug queue, but i am tired of it now :)
[10:58:36]<mig5>sure
[10:58:53]<mig5>we have a rc3 > 6.x-1.x upgrade job there in jenkins
[10:58:56]<anarcat>i isolated 3 issues we could target soon
[10:58:57]<anarcat>https://drupal.org/node/983056
[10:58:59]<mig5>running it now
[10:59:01]<anarcat>https://drupal.org/node/1081536
[10:59:06]<anarcat>https://drupal.org/node/1111572
[10:59:08]<anarcat>okay
[10:59:14]<anarcat>i'll look at it to figure it out :)
[10:59:16]* j0nathan has quit (Ping timeout: 258 seconds)
[10:59:20]<anarcat>oh.
[10:59:24]<anarcat>it was *already* created
[10:59:27]<anarcat>silly me. :(
[10:59:28]<anarcat>sorry!
[10:59:29]<mig5>running it :)
[10:59:31]<mig5>no probs!
[10:59:36]<anarcat>man that thing rocks
[10:59:37]<mig5>but now you know, in case i am not here
[11:00:08]<anarcat>that's soo cool
[11:01:00]<mig5>http://jenkins.greenbeedigital.com.au:8080/job/aegir%206.x-1.0-rc3%20to%...
[11:01:05]<mig5>console output
[11:01:13]<mig5>in real time
[11:01:13]<anarcat>yeah, i'm looking at that :)
[11:01:16]<mig5>:)
[11:01:22]<anarcat>that fucking rocks
[11:02:57]<mig5>the usb debian repo is a little slow to rackspace in recent days
[11:03:00]<mig5>dunno why
[11:03:02]<mig5>er
[11:03:05]<mig5>us* debian repo
[11:04:19]<anarcat>maybe having this based on images would be faster ;)
[11:05:12]<mig5>yes
[11:05:18]<mig5>in fact it uses one alreadyu
[11:05:22]<mig5>a Squeeze image
[11:05:33]<mig5>i should just update it to have the LAMP stack ready
[11:05:52]* anarcat cries
[11:05:52]<anarcat>function drush_provision_verify() {
[11:05:52]<anarcat> d()->command_invoke('verify');
[11:05:53]<anarcat>}
[11:05:57]<anarcat>wtf.
[11:06:09]<mig5>#aegirwtf
[11:06:11]<anarcat>completely blows the rollback stuff out the window
[11:06:35]<anarcat>it's magic!
[11:06:40]<mig5>fuck
[11:06:44]<anarcat>i grepped through 3 layers of hooks there
[11:06:48]* arianek is now known as arianek_afk
[11:06:54]<anarcat>[173-203-102-223.static.cloud-ips.com] out: Fatal error: Cannot redeclare hosting_package_update_6005()
[11:06:57]<anarcat>hehe
[11:06:59]<anarcat>i *knew* i would fail
[11:07:03]<anarcat>effin moron
[11:07:06]<anarcat>lemme fix that
[11:07:06]<mig5>-203-102-223.static.cloud-ips.com] out: Fatal error: Cannot redeclare hosting_package_update_6005()
[11:07:09]<mig5>[173-203-102-223.static.cloud-ips.com] out: (previously declared in
[11:07:12]<mig5>[173-203-102-223.static.cloud-ips.com] out: /var/aegir/hostmaster-6.x-1.x/profiles/hostmaster/modules/hosting/client/hosting_client.install:275)
[11:07:15]<mig5>[173-203-102-223.static.cloud-ips.com] out: in
[11:07:17]<mig5>[173-203-102-223.static.cloud-ips.com] out: /var/aegir/hostmaster-6.x-1.x/profiles/hostmaster/modules/hosting/package/hosting_package.install
[11:07:21]<mig5>oh you beat me to it
[11:07:23]<mig5>sorry
[11:07:29]<mig5>well i am glad we have this tool :) yay upgrade path testing
[11:09:59]<anarcat>alright, i started a new build
[11:10:02]<anarcat>i *like* this :)
[11:10:45]<mig5>i thought you would :)
[11:10:54]<mig5>i will work on a new image to make it faster too
[11:11:17]<anarcat>so
[11:11:29]<anarcat>platform verification, at the d() level, is unrollbackable
[11:11:40]<anarcat>because of that fucking moroning command_invoke() crap
[11:11:50]<anarcat> d()->command_invoke('verify');
[11:12:19]<anarcat>that in turns call d()->verify_platform_cmd()
[11:12:30]<anarcat>why all the new conventions?
[11:12:36]<anarcat>why not just stick with what worked?
[11:12:42]<anarcat>i am utterly confused by this
[11:13:24]<mig5>i've no idea
[11:13:30]<mig5>can we hunt down where it was introduced in the commit log
[11:14:04]<mig5>the message sometimes gives away what adrian/neil were thinking
[11:15:11]<anarcat>i think i don't actually *want* to do that
[11:15:22]<anarcat>half the time i end up with a 3k line diff
[11:15:29]<anarcat>and one phrase of commit log
[11:15:34]<mig5>:)
[11:15:34]<anarcat>"fixing servers"
[11:15:41]<mig5>'fix for breakage'
[11:15:43]<mig5>:D
[11:15:57]<anarcat>je suis de mauvaise foi... je vais checker
[11:16:02]<mig5>hmm
[11:16:06]<anarcat>my god, i am retarded
[11:16:17]<mig5>6006 this time :)
[11:16:25]<mig5>that's progress!
[11:16:41]<anarcat>Cannot redeclare hosting_package_update_6006 in hosting_client.install
[11:16:43]<anarcat>bravo
[11:16:56]<anarcat>committed
[11:16:59]<anarcat>pushed
[11:17:14]<anarcat>scheduled
[11:18:45]<mig5>building a new squeeze LAMP image now
[11:25:16]<anarcat>i am wondering
[11:25:23]<anarcat>aren't we creating the site apache configuration twice?
[11:25:53]<mig5>where?
[11:26:07]<anarcat>no, we aren't
[11:26:08]<anarcat>well
[11:26:12]<anarcat>that command_invoke crap
[11:26:25]<anarcat>ah maybe on verify
[11:26:26]<anarcat> function verify_site_cmd() {
[11:26:26]<anarcat> $this->create_config($this->context->type);
[11:26:26]<anarcat> $this->parse_configs();
[11:26:26]<anarcat> }
[11:27:07]<anarcat>rah
[11:27:08]<anarcat>no
[11:27:15]<anarcat>because we do this magic only on verify
[11:27:16]<anarcat>not install
[11:27:18]<anarcat>just
[11:27:20]<anarcat>because.
[11:27:25]<anarcat>go figure
[11:27:27]<anarcat>alright
[11:27:37]<anarcat>so okay
[11:27:38]<anarcat>wow
[11:27:40]<anarcat>er
[11:27:48]<anarcat>well, i'll cross my fingers and try to fix that shit
[11:28:01]<mig5>we can just keep testing until we know it works :)
[11:28:18]<mig5>i just finished building a LAMP stack and have been changing those build jobs
[11:28:23]<mig5>i'll do a fresh test install now
[11:30:49]<mig5>boo
[11:30:58]<mig5>rackspace cloud api breaks occasionally
[11:31:09]<anarcat>hehe
[11:33:01]<anarcat>alright, i did my stupid commit that may destroy verify rollbacks forever
[11:33:42]<anarcat>now *that* was fucked up
[11:33:43]<anarcat>server/
[11:33:47]<anarcat>oops
[11:33:47]<anarcat>https://drupal.org/node/1081536#comment-4345274
[11:34:20]* jerryitt1 has quit (Quit: Leaving.)
[11:34:43]<mig5>yay my builds are gonna be much faster now
[11:36:25]* sdarnoc23 has joined #aegir
[11:36:39]<mig5>that fix of yours looks nice enough
[11:36:48]<mig5>and some code docs too, nice
[11:36:50]<anarcat>i'm scared of it
[11:37:23]<mig5>it looks sane
[11:37:54]* mapleleaf has quit (Read error: Connection reset by peer)
[11:38:34]* mapleleaf has joined #aegir
[11:38:37]<anarcat>rah, it's not right
[11:38:50]<anarcat>what if verify fails *after* the platform is actually running, with sites in it?
[11:39:17]<anarcat>then i remove the platform apache config, and all clean_urls fuckup
[11:39:23]<anarcat>crap
[11:39:34]<mig5>ah shit you're right
[11:40:05]<mig5>we have an existing function to check if sites are present on the platform
[11:40:07]<mig5>we use it for platform delete
[11:40:13]<mig5>not sure if you want to add additional conditionals
[11:40:15]<anarcat>hum hum
[11:40:16]<anarcat>yeah
[11:40:24]<anarcat>back to the drawing board, that's all
[11:40:31]<mig5>no probs
[11:40:32]<anarcat>i'll just revert this
[11:43:21]<anarcat>hum
[11:43:30]<sdarnoc23>Hello Aegir. Attempting install. drush hostmaster-install hangs at http://jquery-ui.googlecode.com/files/jquery.ui-1.6.zip. Running wget manually to same, retrieves the file. Any ideas?
[11:43:40]<anarcat>wasn't there an issue about an install task for platforms?
[11:44:44]<anarcat>ah yea got it
[11:44:44]<anarcat>https://drupal.org/node/1108152
[11:45:18]<mig5>sdarnoc23: are you sure it's hanging on wgetting that, or is that the last thing the output shows?
[11:45:28]<mig5>and that it might be hanging on the next step?
[11:45:46]<mig5>anarcat: ooh good idea..
[11:45:50]<mig5>this could be specific to platform install
[11:45:52]<sdarnoc23>Not sure. It is last thing on console.
[11:45:52]<mig5>if there was such a thing
[11:46:00]<mig5>sdarnoc23: so do a ps, see what it's currently running
[11:46:30]<mig5>anarcat: so much for 'no crazy big code changes' :)
[11:46:32]<mig5>but this is a good one
[11:46:39]<anarcat>i am *not* doing this now
[11:46:58]<anarcat>https://drupal.org/node/1081536 will just have to wait
[11:47:10]<anarcat>hefring: can you please help us with those non-descript nids?? https://drupal.org/node/1081536
[11:47:17]<anarcat>stupid bots
[11:47:26]<mig5>i'll see what i can do there, you want it to parse the node title?
[11:47:48]<mig5>i don't know why it doesn't, i wonder if that was a feature built into Druplicon only and not the Bot module
[11:47:55]<anarcat>that would be good
[11:48:05]<anarcat>i think it's because druplicon actually connects to the d.o database
[11:48:23]<mig5>gotcha
[11:48:41]<anarcat>also giving up on https://drupal.org/node/1111572 for the night
[11:48:43]<anarcat>i'm done
[11:48:47]<anarcat>i'm ready for 1.0
[11:48:51]<anarcat>let's give it a little longer
[11:48:56]<anarcat>we can start compiling the release notes
[11:48:57]<anarcat>i wonder
[11:49:15]<anarcat>do we make real release notes of every major feature we have added since 0.3?
[11:49:28]<anarcat>maybe the docs team can start working on that?
[11:49:29]<mig5>that would be a big job :)
[11:49:41]<sdarnoc23>mig5: Thanks, sendmail isn't working. I'll fix it. Goodby.
[11:49:53]* sdarnoc23 has quit (Quit: Ex-Chat)
[11:51:19]* simesy has joined #aegir
[12:02:18]<anarcat>"coordinating the final !aegir 1.0 release, it's going to be a huge list of features and a hell of a party! https://drupal.org/node/1126736"
[12:02:23]<anarcat>there we go
[12:02:29]<anarcat>i'm going to stop there for the day
[12:02:45]<anarcat>mig5: but feel free to fire up an etherpad somewhere so we can start writing this shit :)
[12:03:07]<anarcat>ergonlogic / darthsteven : we're heading towards 1.0 and need docs people help! http://drupal.org/node/1126736
[12:04:20]* anarcat has changed the topic to Aegir hosting system 1.0-rc7 released! http://community.aegirproject.org/1.0-rc7 | Issue queue: http://is.gd/edalY | read this before asking: http://community.aegirproject.org/bugs | scrums on tuesday 2200UTC | 1.0 getting cooked for you here http://drupal.org/node/1126736 - smell the sweets, add your sugar!
[12:05:08]<mig5>nice work anarcat
[12:05:26]<anarcat>you too buddy :)
[12:05:37]<anarcat>man, that jenkins guy is saving our butts
[12:05:53]<mig5>i just cleaned up some of that code
[12:06:40]<anarcat>which code?
[12:06:47]<anarcat>the jenkins jobs?
[12:07:19]<anarcat>i wonder if i could make a 0.3 to 1.x upgrade job :P
[12:07:44]<anarcat>bloody hell the python script is inline right in there
[12:07:45]<anarcat>i see
[12:07:51]<mig5>yeah
[12:07:55]<mig5>i will fix that eventually
[12:08:10]<mig5>i think now that there's proper try/except stuff (wasn't originally), i could put this into a script on the server itself
[12:08:13]<mig5>however
[12:08:14]<anarcat>still it's nice
[12:08:17]<mig5>then it's a bit opaque for you again
[12:08:24]<anarcat>yeah
[12:08:24]<mig5>at least there you can tweak version numbers etc to make new jobs
[12:08:28]<anarcat>it's nice to see it
[12:08:39]<mig5>although i could parameterise that, as args to a shell script or something
[12:08:40]<mig5>but meh
[12:08:41]<anarcat>i wonder - can you stack tasks?
[12:08:43]<mig5>yes
[12:08:46]<anarcat>like have an install rc7 task
[12:08:55]<anarcat>then have an upgrade from X to 1.x task?
[12:08:57]<mig5>yes you can have chained before/after jobs
[12:09:01]<mig5>that only occur if the previous one succeeded, etc
[12:09:08]<anarcat>so that you have a generic upgrade task and you can chain it with any install you want?
[12:09:14]<mig5>the reason i haven't done that yet
[12:09:24]<anarcat>so every time we make a release, we make an "install that release" task and testing the upgrade path is trivial
[12:09:25]<mig5>is because i'm not really persisting the server IP between tasks
[12:09:29]<mig5>so i'd have to record that somewhere
[12:09:36]<mig5>and you can't send args to other jobs in jenkins yet apparently
[12:09:47]* auth has quit (Ping timeout: 252 seconds)
[12:09:51]<anarcat>humm
[12:10:01]<anarcat>well, jobs run in a specific context?
[12:10:03]<anarcat>like in a server?
[12:10:23]<mig5>yes, in this case, the task is executed by the 'jenkins' user on my jenkins server
[12:10:30]<mig5>and jenkins supports environment variables
[12:10:34]<mig5>so we could export that perhaps
[12:10:40]<anarcat>yaay
[12:10:45]<anarcat>man i like that
[12:10:56]<anarcat>well
[12:11:05]<anarcat>what i mean is more like - hum
[12:11:11]<anarcat>you ran this install task right?
[12:11:16]<anarcat>so now the server has rc7 installed
[12:11:21]<anarcat>now you run the upgrade task
[12:11:23]<anarcat>on the same server
[12:11:25]<mig5>well it's 6.x-1.x yeah
[12:11:26]<mig5>yes
[12:11:36]<anarcat>why do you need to record the ip? it's just "that server i'm running that task on"
[12:11:41]<mig5>but that upgrade task has to know which server to connect to
[12:11:42]<anarcat>no?
[12:11:44]<mig5>it's fabric calls
[12:11:46]<mig5>to a brand new vps
[12:11:58]<anarcat>okay
[12:11:59]<mig5>understand what i mean?
[12:12:02]<anarcat>i'm getting this slowly :)
[12:12:03]<anarcat>yeah
[12:12:12]<mig5>so the install task builds a total fresh new vps
[12:12:12]* anarcat heads into the cloud
[12:12:16]<mig5>the next job would need to know about that new vps
[12:12:19]<mig5>to connect to it and upgrade
[12:12:32]<anarcat>and then have a tear down job in the end, i guess
[12:12:33]<anarcat>i see
[12:13:00]<mig5>not impossible, i just haven't though of a good way to do it
[12:13:23]<mig5>first i tried sending an arg to the next task with the IP, but jenkins' wiki says it's not possible (yet)
[12:13:50]<mig5>so that's why that upgrade job does an install of rc3 first, then an upgrade, all at once, which is ugly
[12:13:56]<mig5>so i will be happy to chain tasks if i find away
[12:13:56]<anarcat>je vois
[12:13:57]<anarcat>Actions à la suite du build
[12:13:59]<mig5>a way*
[12:14:01]<anarcat>Construire d'autres projets (projets en aval)
[12:14:19]<anarcat>i understand
[12:14:38]<anarcat>still, creating "install" tests for every new release should be something we do as part of the release process, at least on 1.x
[12:14:51]<mig5>that's also remarkable how jenkins translates the app based on your locale :)
[12:14:59]<mig5>i didn't activate any FR in there
[12:15:25]<mig5>yes, i have been creating such install tasks for reach release
[12:15:26]<anarcat>yeah it's nice
[12:15:27]<mig5>hence that rc7 job
[12:15:54]<anarcat>maybe splitting the job in multiple steps would help
[12:16:03]<anarcat>with each step being its own python script...
[12:16:46]<anarcat>it's already a lot like this, i guess, it's just one step...
[12:17:08]<mig5>yes i would like the release-specific job to be less duplication
[12:17:28]<mig5>i've just been taking advantage of the New Job > Copy from Existing feature in jenkins
[12:17:59]<anarcat>a lot of work would be saved by using debian packages too ;)
[12:17:59]* auth has joined #aegir
[12:18:02]<anarcat>yeah, that works too
[12:18:10]<anarcat>so screw that, duplication works :P
[12:18:32]<anarcat>this is miles beyond what we had before
[12:18:35]<anarcat>which was nothing
[12:18:36]<anarcat>:P
[12:18:39]<mig5>yes re: debian packages: i have not yet nutted out the preseeding required
[12:18:44]<mig5>you are free to fix that :)
[12:19:06]<anarcat>the only preseeding required is mysql, really
[12:19:10]<mig5>i took a quick look at your scripts in the .deb's but i was sick and not really committed
[12:19:14]<anarcat>... if your host is setup right of course :P
[12:19:33]<mig5>well i was preseeding mysql already, now i have an image where it's already running
[12:19:34]<anarcat>so if it's a fresh debian install, then the default preseeds probably work (ie. no mysql pass)
[12:19:39]<mig5>so it just needs to handle the aegir install itself
[12:19:56]<mig5>i am hardcoding the mysql pass now, to be what i know it is in my squeeze lamp image
[12:19:59]<anarcat>apt-get install aegir-provision
[12:20:00]* EclipseGc has quit (Quit: EclipseGc)
[12:20:02]<anarcat>apt-get install aegir-hostmaster
[12:20:03]<anarcat>really :)
[12:20:12]<mig5>yeah but isn't there a debconfig to prompt for URL etc?
[12:20:14]<mig5>debconf*
[12:20:15]<anarcat>yeah, i noticed that
[12:20:25]<mig5>and client email etc
[12:20:26]<anarcat>yeah, but debconf can be told to shut up
[12:20:34]<anarcat>< /dev/null
[12:21:00]<mig5>so what i mean is i didn't know the correct parameters yet to force those responses, like how i do for postfix/mysql etc
[12:21:25]<anarcat>i see
[12:21:32]<anarcat>one of them i do this:
[12:21:33]<anarcat>echo debconf aegir/makefile string /var/aegir/makefiles/aegir/aegir-koumbit.make | debconf-set-selections
[12:22:15]<mig5>oooer
[12:22:16]<mig5>i see
[12:22:16]<anarcat>all the questions are defined here: http://drupalcode.org/project/provision.git/blob_plain/refs/heads/debian...
[12:22:24]<anarcat>| grep Template:
[12:22:27]<mig5>damn that's easier to read
[12:22:33]<mig5>i was picking apart an actual .debv that i had wgetted
[12:22:34]<mig5>stupid
[12:22:38]<anarcat>that's really all there is
[12:22:40]<anarcat>hehe
[12:22:49]<anarcat>http://drupalcode.org/project/provision.git/tree/refs/heads/debian:/debian is your friend
[12:23:01]<mig5>i wonder if you could add a jenkins job that *builds* your debs :)
[12:23:08]<mig5>configure your koumbit server as a 'slave' of my jenkins
[12:23:15]<mig5>and it can connect over ssh and do the job :)
[12:23:20]<mig5>i do my server backups that way :)
[12:23:29]<mig5>then any of us can do a full release, no bus factor :)
[12:24:26]<anarcat>of course
[12:24:38]<anarcat>http://community.aegirproject.org/node/543
[12:24:42]<anarcat>that's how you build the packages
[12:25:09]<anarcat>there's some glue missing between some of those commands, but otherwise it's pretty much auto
[12:25:19]<anarcat>like:
[12:25:19]<anarcat>git describe
[12:25:19]<anarcat>dch -v 1.0~rc3+28-1
[12:25:41]<mig5>oh - have you got a post-receive hook that drops your debs into your koumbit repo?
[12:25:42]<anarcat>git describe actually gives you 1.0-rc3-28-garbage
[12:25:47]<mig5>i didn't realise i could do that myself
[12:26:09]<anarcat>so you need to replace the first dash with aa ~, the second with a + and drop the garbage... kinda annoying
[12:26:13]<anarcat>yeah
[12:26:17]<anarcat>we don't have a hook though :)
[12:26:25]<anarcat>i build and upload manually, following those precise steps
[12:26:34]<mig5>i see
[12:26:37]<anarcat>ah and some steps are interactive, like dch
[12:26:52]<anarcat>and git-buildpackage of course asks me for my PGP password :P
[12:27:01]<mig5>bugger: i was about to say, we could put it all into release.sh, and a post-receive hook to get it into your repo, and then make it a jenkins job :)
[12:27:02]<anarcat>which i am *not* going to put in a jenkins script thank you very much :P
[12:27:04]<mig5>:)
[12:27:05]<mig5>hahah
[12:27:06]<mig5>snap
[12:27:08]<mig5>one click 'build'
[12:27:43]<mig5>yes i'm getting carried away
[12:27:57]<anarcat>nightly builds they're called :)
[12:28:11]<anarcat>well, we could have a pgp key for jenkins
[12:28:23]<anarcat>password less
[12:28:30]<anarcat>and jenkins can watcha git repo for changes apparently
[12:28:48]<anarcat>one thing that would be nice would be to have a generic test suite to run post commit
[12:29:02]<anarcat>like installing rc7, but on 1.x
[12:29:12]<mig5>yeah
[12:29:14]<anarcat>we do have that
[12:29:33]<anarcat>Gestion de code source
[12:29:34]<anarcat>git
[12:30:09]<mig5>yeah there's plugins
[12:30:32]<mig5>i should give you full admin access, just promise not to break my other server backup jobs that you can't yet see :)
[12:31:14]<anarcat>the drush folks made it so the committer of the broken commit gets the mail :P
[12:31:21]<mig5>yeah :)
[12:31:46]<anarcat>i modified the 1.x install script to inspect the git repos and fire up when there's a change
[12:31:49]<anarcat>... i *think*
[12:32:36]<anarcat>Envoyer des emails séparés aux personnes qui ont cassé le build
[12:32:46]<anarcat>Si cette option est choisie, les notifications par email seront envoyées aux personnes qui ont fait des modifications dans un build en échec (on suppose que ces changements ont cassé ce build).
[12:32:54]<anarcat>god jenkins is *insane*
[12:33:07]<anarcat>okay, let's see if that thing worked
[12:34:10]<anarcat>i changed a bit in the debian branch and pushed it, let's see if jenkins picks it up
[12:35:42]<mig5>have you got jenkins watching a git repo now though?
[12:35:49]<mig5>i don't think i'm making use of the git plugin myself yet..
[12:36:04]<mig5>oh git polling log, did you do that?
[12:36:15]<mig5>i see
[12:36:58]<mig5>i wonder how that interacts with the actual build job, i guess it keeps it separate
[12:37:13]<anarcat>i went in http://jenkins.greenbeedigital.com.au:8080/job/aegir%206.x-1.x%20install...
[12:37:20]<mig5>yep i see now
[12:37:22]<anarcat>and clikety my way around the git stuff
[12:37:35]<anarcat>basically adding the repos
[12:37:42]<anarcat>then I went into "Ce qui déclenche le build
[12:37:43]<anarcat>"
[12:37:48]<anarcat>not sure what that is in english :)
[12:37:52]<mig5>so in other words if you push a commit, it should do an aegir install ?
[12:37:54]<anarcat>"Scruter l'outil de gestion de version"
[12:37:58]<anarcat>i guess so!
[12:38:03]<mig5>cool
[12:38:14]<anarcat>it checks the git repo only every hour though, so that shouldn't be too bad
[12:38:24]<mig5>that's fine
[12:38:27]* CraHan_ has joined #aegir
[12:38:35]<anarcat>but the real magic is the mail notification to the people that broke the build :)
[12:38:42]<anarcat>now i hope *that* will work :)
[12:38:54]<mig5>yeah that'll be interesting, i don't see how it will know from the python tasks where that affected in git
[12:39:00]<mig5>maybe it will surprise me :)
[12:39:33]<mig5>i'm gonna make it run at 45 minutes past the hour so it happens in the next 5 min :)
[12:39:59]<anarcat>ah cool
[12:40:01]<anarcat>better
[12:40:07]<anarcat>i'm too tired to think straight like that
[12:40:25]<anarcat>i should go to sleep but i forgot to have supper so i'm all ... fucked up :)
[12:40:46]<mig5>ahaha
[12:41:12]<mig5>il est fou l'anarcat
[12:41:16]<anarcat>fou raide
[12:41:31]<anarcat>this is weird Scruter l'outil de gestion de version
[12:41:33]<anarcat>er
[12:41:36]<anarcat>http://jenkins.greenbeedigital.com.au:8080/builds
[12:41:48]<anarcat>it says those tasks ran on april 15
[12:41:54]<mig5>yes it's a strange histogram
[12:41:54]<mig5>well
[12:42:00]<mig5>the server is in gmt +10 :)
[12:42:04]<anarcat>i'm on the 13th here, and i know you're on the 14... but you're not *that* ambitious :P
[12:42:04]<mig5>i am from the future remember
[12:42:06]<mig5>oh
[12:42:07]<mig5>wait
[12:42:08]* CraHan has quit (Ping timeout: 276 seconds)
[12:42:10]<mig5>you're right :)
[12:42:11]<mig5>haha
[12:42:15]<anarcat>hehe
[12:42:21]<mig5>it's SKYNET
[12:42:44]<mig5>probably a jenkins bug
[12:42:57]<mig5>i am using the jenkins debian repo, it's some what bleeding edge
[12:43:08]<mig5>you'd think they'd, er, test for that sort of thing
[12:43:10]<mig5>in jenkins..
[12:43:10]<mig5>:)
[12:43:11]<anarcat>haha
[12:43:31]<mig5>bbs
[12:46:04]<anarcat>i don't think i've released 1.0 of anything either
[12:46:10]* anarcat food time
[12:47:26]<anarcat>hefring: tell ergonlogic that this is a nice way to present install instructions... http://yorba.org/shotwell/install/
[12:47:26]<hefring>anarcat: I'll pass that on when ergonlogic is around.
[12:47:41]* jhedstrom has quit (Ping timeout: 240 seconds)
[12:47:54]<mig5>haha
[12:47:55]<mig5>wtf
[12:48:15]<anarcat>that was fast :P
[12:48:20]<mig5>drupal.org's git throwing 400s
[12:49:07]<anarcat>uh
[12:49:19]<anarcat>well, that doesn't seem to be the right url
[12:49:28]<anarcat>git://git.drupalcode.org/project/provision works
[12:49:59]<anarcat>done
[12:50:07]<mig5>i think drupal.org have weird restrictions
[12:50:08]<mig5>oh
[12:50:10]<mig5>git clone --branch 6.x-1.x http://git.drupal.org/project/provision.git
[12:50:13]<mig5>according to the project page
[12:50:33]<mig5>maybe git.drupal.org reacts differently to git.drupalcode.org? surely not
[12:50:59]<anarcat>meh
[12:51:02]<anarcat>whatever
[12:51:12]<anarcat>i used git:// and changed the cron to be at 55 :P
[12:51:18]<mig5>:)
[12:51:18]<anarcat>so 5 more minutes
[12:51:29]<anarcat>besides, i heard that git:// was faster on d.o
[12:51:39]<mig5>that's interesting
[12:53:31]<anarcat>i heard that on c.a.o so take that with a grain of salt :P
[12:54:41]<mig5>anarcat - 888 commits
[12:54:43]<mig5>woot
[12:54:55]<mig5>on provision
[12:56:03]<anarcat>yay :)
[12:56:08]<anarcat>our times are not sync, it seems
[12:56:39]<mig5>not quite
[12:56:43]<mig5>12:55pm just now on the server
[12:56:52]<anarcat>13 Apr 22:56:48 ntpdate[28003]: adjust time server 209.167.68.100 offset -0.010929 sec
[12:56:56]<mig5>can be a little delay on jenkins
[12:57:02]<anarcat>scratch that, *your* clock is not synced :P
[12:57:03]<mig5>i 'm not sure this server's running ntp, it probably should be
[12:57:06]<mig5>yeah
[12:57:45]<mig5>hmm
[12:57:47]<anarcat>it's not fetching the git repo again...
[12:58:05]<mig5>Started on 14/04/2011 12:55:51 PM
[12:58:05]<mig5>No workspace is available, so can't check for updates.
[12:58:05]<mig5>Done. Took 2 ms
[12:58:06]<mig5>No changes
[12:58:07]<anarcat>Started on 14/04/2011 12:55:51 PM
[12:58:07]<anarcat>No workspace is available, so can't check for updates.
[12:58:07]<anarcat>Done. Took 2 ms
[12:58:09]<anarcat>snap
[12:58:15]<anarcat>well, that's great ain't it
[12:58:22]<anarcat>i'll schedule a build
[12:58:23]<mig5>oh
[12:58:39]<anarcat>so we'll have a checkout at least
[12:58:48]<mig5>must be this stale workspace or something
[12:59:11]<mig5>looks more promising
[12:59:14]<anarcat>yeah
[12:59:21]<anarcat>it's doing *something* :P
[12:59:33]<mig5>clou:/var/lib/jenkins# ps aux | grep git
[12:59:33]<mig5>jenkins 2706 0.0 0.1 24544 1452 ? R 12:57 0:00 git merge-base ecc86f8d4be18c2296ceef8f520b8c0460f06ded b6c489b38c9459bacd868f465453de3595381cc2
[12:59:36]<mig5>root 2710 0.0 0.1 7784 844 pts/0 S+ 12:57 0:00 grep git
[12:59:37]<mig5>good
[12:59:40]<anarcat>wow, all branches
[12:59:53]<mig5>awesome
[12:59:54]<anarcat>maybe that's not what we want
[12:59:57]<mig5>hahaha
[12:59:59]<mig5>git --config
[13:00:11]<mig5>ok
[13:00:19]<anarcat>i'll stick it on the 6.x branch
[13:00:45]<anarcat>it's raining on jenkins
[13:00:58]<anarcat>that was quick :)
[13:01:03]<anarcat>i'll let you git --config
[13:01:07]<anarcat>and i should cook rice ffs
[13:01:07]<mig5>done
[13:01:42]<anarcat>Commencing build of Revision dae2f1097d13f243d3836030d7cd2182e5602cee (origin1/6.x-1.x)
[13:01:45]<mig5>sweet
[13:01:48]<mig5>and it's building a vps
[13:01:59]<anarcat>man that rocks
[13:02:01]<mig5>i'm still not sure how those two tasks will interact
[13:02:36]<anarcat>i think it just checks out to figure its things out with git then goes on as usual
[13:02:50]<anarcat>it parses commitlogs to get your email to spam you too :P
[13:03:04]<mig5>don't worry my phone is beeping away at me every time there's a failed build :)
[13:03:05]<anarcat>i love this shit
[13:03:06]<mig5>i use notifo.com
[13:03:15]<anarcat>yeah, i noticed/notifo'd
[13:03:20]<mig5>i use that for nagios too now
[13:03:36]<anarcat>notifo?
[13:03:37]<anarcat>why?
[13:03:44]<mig5>notifo app on my iPhone
[13:03:48]<mig5>equivalent of sms alerts
[13:03:54]<mig5>without having to run an sms gateway
[13:04:01]<mig5>Push notifications
[13:04:09]<mig5>if you're a sell out like me, with an iPhone, it's neat
[13:04:15]<anarcat>i see
[13:04:16]<mig5>there's a notifo app being built for android too
[13:04:48]<anarcat>dude, you're even testing d7 installs in there... wow
[13:04:56]<mig5>yep
[13:04:58]<mig5>and openatrium.
[13:05:11]<mig5>all built from makefiles
[13:05:13]<mig5>i wondered about removing openatrium as it takes longer than the others
[13:05:19]<mig5>but it's not a bad test of a distro
[13:05:34]<anarcat>haaaalleluia!
[13:06:14]<anarcat>so i guess we'll see the real value of this when i'll break the shit out of head :P
[13:06:26]<mig5>so if you're quick enough, or don't destroy the node at all, you can go in there and see it, i love how we expose the login url http://173-203-107-194.static.cloud-ips.com/user/reset/1/1302750229/ad5a...
[13:06:30]<mig5>haha
[13:06:42]<mig5>yes this is our anarcrap/migression spotter
[13:06:53]<anarcat>oops :)
[13:07:04]<anarcat>there's a short window there :)
[13:07:06]<mig5>i don't mind that
[13:07:11]<mig5>you don't get long :)
[13:07:24]<mig5>although
[13:07:40]<mig5>if the build fails before the end, it might leave the server running
[13:07:41]<mig5>not sure
[13:07:43]<mig5>:)
[13:07:54]<mig5>as long as we don't break aegir, i'm sure it will be fine
[13:08:00]<mig5>hehe
[13:08:41]<mig5>what i might do is set a firewall allowing port 80 access only from the jenkins server
[13:08:46]<mig5>in an earlier fabric call
[13:08:48]<mig5>that's safest
[13:09:00]<anarcat>yeah
[13:09:02]<mig5>i know *I* could then get inside if i needed too
[13:09:12]<mig5>if you ever do, we could stick your IP in there too
[13:09:15]<anarcat>i rescheduled the git poll for 10 * * * *
[13:09:20]<anarcat>and pushed a dummy cmomit again
[13:09:28]<anarcat>ok
[13:09:34]<mig5>clou:/var/lib/jenkins# curl icanhazip.com
[13:09:34]<mig5>2001:4b98:dc0:51:216:3eff:fe61:39e5
[13:09:37]<mig5>i love that service
[13:09:48]<mig5>clou:/var/lib/jenkins# curl ipv4.icanhazip.com
[13:09:48]<mig5>95.142.164.178
[13:09:50]<mig5>that's better
[13:09:57]<anarcat>hehe
[13:13:23]<anarcat>hehe that worked, it's building a new install
[13:13:38]<anarcat>
[13:13:39]<anarcat>Un changement dans la base de code a provoqué le lancement de ce job
[13:13:42]<anarcat>i'll stop now :)
[13:14:35]<mig5>awesome
[13:15:06]<anarcat>i tweaked the git settings so that jenkins distinguises clearly between the two repos (instead of saying "origin" and "origin1")
[13:16:04]<anarcat>Config user.name Value
[13:16:05]<anarcat> Config user.email Value
[13:16:10]<anarcat>you could have set that there too...
[13:16:57]<mig5>oh for .deb's ?
[13:17:05]<mig5>oh
[13:17:11]<mig5>i could set the IP there in fact!
[13:17:14]<anarcat>no, in the job git config
[13:17:18]<mig5>oh right
[13:17:23]<mig5>but it just occurred to me
[13:17:33]<mig5>i can do a config set of the IP since i am already using configParser
[13:17:38]<mig5>which means i could chain jobs
[13:18:02]<mig5>that can be a job for another day :)
[13:18:07]<anarcat>hehe
[13:19:12]<anarcat>as my next test commit to make sure the repo rename didn't break the build, i think i'll cleanup example.drushrc :P
[13:20:51]<anarcat>and i scheduled the git checks for "every 15 minutes" instead of every hour + changing the time all the time :P
[13:21:17]<mig5>little firewall in there now
[13:21:28]<mig5>cool
[13:21:44]<anarcat>should we test the upgrade path like crazy too?!
[13:22:22]<anarcat>i guess that would be a bit abusive
[13:22:33]<mig5>hehe
[13:22:35]<anarcat>it could make sense once we have chained commands
[13:22:40]<mig5>well, it's cheap
[13:22:41]<mig5>yeah
[13:23:12]<anarcat>oops
[13:23:19]<anarcat>i accidentally provoked a new build :P
[13:23:23]* rovo has joined #aegir
[13:25:04]<mig5>:)
[13:25:08]<mig5>it's good!
[13:25:25]<anarcat>rah now all the branches are dup'd
[13:25:31]<mig5>No change to record in branch provision/6.x-1.x
[13:25:34]<mig5>isn't that odd
[13:25:38]<mig5>should have seen a change?
[13:25:48]<mig5>hmm
[13:25:50]<mig5>dns fail on the vps
[13:26:06]<mig5>ooh
[13:26:09]<mig5>it'll be my firewall
[13:26:10]<mig5>stupid
[13:26:23]<anarcat>hehe
[13:26:51]<mig5>i just commented out the firewall for now
[13:26:53]<mig5>will fix later
[13:34:56]<anarcat>okay, have fun, i'm going to eat and stepping away
[13:36:10]* gusaus has quit (Ping timeout: 260 seconds)
[13:38:18]* FatGuyLaughing has joined #aegir
[13:40:16]* FatGuyLaughing has quit (Client Quit)
[13:49:36]* kvanderw has quit (Quit: Computer has gone to sleep)
[13:49:38]<mig5>sigh
[13:50:08]<rovo>mig5: whats going on
[13:50:18]* nlcluigi_ has quit (Quit: nlcluigi_)
[13:50:24]<mig5>rackspace cloud api being shit
[13:50:36]<rovo>hm yeah
[13:50:45]<rovo>i don't deal with that
[13:50:55]<rovo>but do use them for all my sites at work
[13:50:59]<mig5>that's more like it http://jenkins.greenbeedigital.com.au:8080/job/aegir%206.x-1.0-rc3%20to%...
[13:51:06]<rovo>but personally been using aws.amazon as you know
[13:51:33]* kvanderw has joined #aegir
[13:53:49]* josh_k has quit (Ping timeout: 246 seconds)
[14:22:14]* scientist has quit (Ping timeout: 276 seconds)
[14:22:23]<grape>is the koumbit-unstable repo still the best place to get the debs?
[14:23:16]<bgm>grape: for aegir, yes. but drush is now in unstable / squeeze-backports
[14:24:30]<grape>ok i was going to grab it from mercer's ppa
[14:24:57]<grape>im setting up a lucid system
[14:25:16]* lolmaus has joined #aegir
[14:25:20]<grape>bgm, thanks
[14:25:39]<bgm>ah ok, i don't know for ubuntu
[14:25:50]* rovo has left #aegir ()
[14:30:20]* EclipseGc has joined #aegir
[14:33:51]* simesy has quit (Quit: simesy)
[14:37:17]* scientist has joined #aegir
[14:41:06]* arianek_afk has quit (Quit: Smell ya later)
[14:41:28]* lolmaus has quit ()
[14:50:31]* drapergeek has quit (Ping timeout: 246 seconds)
[14:50:42]* drapergeek has joined #aegir
[14:51:10]* kvanderw is now known as kvanderw|afk
[14:54:03]* simesy has joined #aegir
[14:56:27]* trainer has quit (Quit: trainer)
[14:57:08]* sepehr has joined #aegir
[15:10:35]* sepehr has quit (Ping timeout: 240 seconds)
[15:15:33]* L-four has joined #aegir
[15:17:01]<L-four>Hi, does anyone know if there is a easy way to export sites from aegir as a standalone drupal website?
[15:20:56]* markwk has joined #aegir
[15:20:57]<bgm>L-four: if you create your platforms with drush make, re-create the platform on the other server, then untar the site (in sites/ directory), then import the database (which is in the aegir backup)
[15:25:06]* sepehr has joined #aegir
[15:26:38]<L-four>bgm, Thanks for the help. I should probably should explained a bit better that I am looking for a way that i can have a single click export of website (platform and site folder) as a single tar/zip. similar to what they have on the drupal gardens website.
[15:29:15]* scientist has quit (Ping timeout: 252 seconds)
[15:29:32]<bgm>when we have to do that (export a site to another host), we provide the makefile and the aegir backup. it's maybe not one click to deploy, but that depends on the other end
[15:30:16]<bgm>http://community.aegirproject.org/node/436 - is the only reference i found to that topic
[15:30:33]<bgm>might be worth posting there if you have any ideas or suggestions of how it should be done
[15:30:47]<bgm>(i'm not an aegir developer)
[15:35:23]* beautifulmind has joined #aegir
[15:35:56]<mig5>http://drupal.org/node/1047992
[15:40:41]* snlnz has quit (Ping timeout: 240 seconds)
[15:47:20]* markwk has quit (Read error: Connection timed out)
[15:49:00]* markwk has joined #aegir
[15:51:51]<L-four>bgm, Thanks a bunch i'll look have a look at that link
[15:56:46]* markwk has quit (Read error: Operation timed out)
[15:57:08]* markwk has joined #aegir
[16:01:30]* lolmaus has joined #aegir
[16:01:30]* beautifulmind has quit (Quit: Leaving.)
[16:02:05]* mikl has joined #aegir
[16:02:05]* mikl has quit (Changing host)
[16:02:05]* mikl has joined #aegir
[16:06:24]* markwk has left #aegir ()
[16:06:50]* ngnp has joined #aegir
[16:11:55]* patcon|myplanet has quit (Quit: patcon|myplanet)
[16:12:17]* mikl has quit (Quit: Connection reset by Peer Gynt)
[16:13:40]* NeoID has quit (Ping timeout: 260 seconds)
[16:16:37]* sepehr has quit (Ping timeout: 246 seconds)
[16:22:10]* NeoID has joined #aegir
[16:22:28]* troubado` has joined #aegir
[16:24:25]* lolmaus has quit ()
[16:32:11]* jonhattan has joined #aegir
[16:34:44]* realityloop has quit (Remote host closed the connection)
[16:37:42]* EclipseGc has quit (Quit: EclipseGc)
[16:47:59]* mrfelton has joined #aegir
[16:51:35]* penyaskito has joined #aegir
[17:13:20]* Pasqualle has joined #aegir
[17:19:41]* bertodsera has quit (Ping timeout: 276 seconds)
[17:20:33]* Chipie has joined #aegir
[17:26:33]* lolmaus has joined #aegir
[17:27:35]* L-four has quit (Remote host closed the connection)
[17:32:39]* bertodsera has joined #aegir
[17:39:55]* skwashd has quit (Ping timeout: 246 seconds)
[17:40:09]* mrfelton_ has joined #aegir
[17:43:11]* mrfelton has quit (Ping timeout: 240 seconds)
[17:43:11]* mrfelton_ is now known as mrfelton
[17:44:02]* eft has left #aegir ()
[17:50:59]* siliconmeadow has joined #aegir
[17:55:06]* skwashd has joined #aegir
[17:59:11]* V1ntage_work has joined #aegir
[18:02:16]* lolmaus has quit ()
[18:04:17]* NeoID has quit (Ping timeout: 240 seconds)
[18:11:29]* NeoID has joined #aegir
[18:16:20]* sk33lz has quit (Quit: Leaving.)
[18:18:00]* lolmaus has joined #aegir
[18:25:23]* sdrycroft has joined #aegir
[18:44:26]* Artusamak_afk is now known as Artusamak
[19:03:55]* Slydder has joined #aegir
[19:20:22]* HaloFX has quit (Ping timeout: 246 seconds)
[19:23:30]* adrinux has joined #aegir
[19:25:47]* Pasqualle has quit (Ping timeout: 276 seconds)
[19:29:14]<mig5>hefring: tell anarcat dig +short aegirproject.twitter.any.io TXT
[19:29:14]<hefring>mig5: I'll pass that on when anarcat is around.
[19:31:48]<adrinux>morning
[19:32:12]<adrinux>I get busy with a project and suddenly we are 5 rc's further on :/
[19:33:20]<mig5>haha
[19:33:23]<mig5>yeah
[19:33:35]<mig5>but i gather there were a couple quite forgettable ones there..
[19:33:39]<mig5>i was out of it too
[19:33:41]<mig5>rc7 is good
[19:33:52]<mig5>i think the next will be 1.0
[19:34:15]<adrinux>how close are we to full release? I've been meaning to work up a minor eldir theme change for ages
[19:34:57]<mig5>snappo
[19:35:03]<mig5>we are very close, probably within a week
[19:35:17]<mig5>starting to work on release notes
[19:35:33]<mig5>i will likely miss it, as i'm on a plane by sunday night
[19:35:46]<mig5>and won't be around much for the 5 weeks after that
[19:36:15]<mig5>what's your change?
[19:48:06]<adrinux>I want to change the background image
[19:48:14]<adrinux>that grey with white dots
[19:48:45]<adrinux>might find time for it today
[19:49:17]* Slydder has quit (Quit: Leaving.)
[19:49:42]* Slydder has joined #aegir
[19:50:49]<darthsteven>adrinux: We can still fix bugs post 1.0, just not going to change the API
[19:51:24]<adrinux>oh yeah, just thought it would be nice to get it in
[19:54:40]<darthsteven>adrinux: sure, there's still a ton of documentation that needs to get into Aegir at some point, so I can foresee a 1.1 release within a little while
[19:56:15]<adrinux>shush. I've been thinking about this for months. A deadline is good motivation :P
[19:58:17]<darthsteven>adrinux: you have one hour
[19:58:22]<darthsteven>;)
[19:58:23]<adrinux>lol
[19:59:00]<darthsteven>adrinux: what's the issue with the dots?
[19:59:32]<adrinux>personal dislike :)
[19:59:50]<adrinux>also I think it makes it harder to read the heading in white text
[20:00:02]<adrinux>I said it was *minor*
[20:00:46]<darthsteven>adrinux: indeed :)
[20:01:15]<darthsteven>hefring: help
[20:01:15]<hefring>Detailed information is available with "BOTNAME: help <feature>" where <feature> is one of: Aggregator, Botagotchi, Factoids, Karma, Logging, Reminders, Seen, Tell, Timezones.
[20:01:46]<darthsteven>hefring: help aggregator
[20:01:46]<hefring>Enables the reporting of aggregated feed items to configured channels.
[20:05:42]<mig5>yes cron hasn't run on that site in weeks
[20:05:47]<mig5>due to a broken production environment
[20:05:48]<mig5>fixing it now
[20:06:32]<hefring>community => Is this a bug? => http://community.aegirproject.org/discuss/bug
[20:06:32]<hefring>community => Started from scratch => http://community.aegirproject.org/discuss/started-scratch
[20:06:32]<hefring>community => 1.0-rc7 released! => http://community.aegirproject.org/discuss/10-rc7-released
[20:06:48]<hefring>community => SSL questions => http://community.aegirproject.org/discuss/ssl-questions
[20:06:48]<hefring>community => Platform not able verify on hostmaster rc7 => http://community.aegirproject.org/discuss/platform-not-able-verify-hostm...
[20:06:48]<hefring>community => Mute point => http://community.aegirproject.org/discuss/mute-point
[20:07:04]<hefring>community => what was that step => http://community.aegirproject.org/discuss/what-was-step
[20:07:04]<hefring>community => 1.0rc7 install on Gentoo segfaults => http://community.aegirproject.org/discuss/10rc7-install-gentoo-segfaults
[20:07:04]<hefring>community => Issues Enabling SSL => http://community.aegirproject.org/discuss/issues-enabling-ssl
[20:07:20]<hefring>community => Issues with apt-get install on Debain => http://community.aegirproject.org/discuss/issues-apt-get-install-debain
[20:10:02]<darthsteven>log?
[20:17:11]<mig5>logs?
[20:17:11]<hefring>yeah, ok, i admit, i'm logging this channel in http://hefring.mig5.net/bot/log/aegir - ask me about log bookmark too ;)
[20:25:17]<adrinux>mig5, where does eldir code live now? still http://drupal.org/project/eldir ?
[20:26:27]<adrinux>just confused by last commits 1 year ago :/
[20:31:26]<mig5>adrinux: it's inside hostmaster
[20:31:34]<omega8cc>adrinux: http://drupalcode.org/project/hostmaster.git/tree/HEAD:/themes/eldir
[20:31:56]<adrinux>thought i might be
[20:31:59]<adrinux>it
[20:32:17]<omega8cc>adrinux: note - I like the dots! :p
[20:32:40]<adrinux>you might like what I have in mind better :)
[20:33:02]<omega8cc>adrinux: I hope! :)
[20:52:26]* realityloop has joined #aegir
[21:01:06]* realityloop has quit (Remote host closed the connection)
[21:08:38]<hefring>Provision => 993944 rollback http configs in site installs failures => http://drupalcode.org/project/provision.git/commitdiff/ab78897d7ec881bb7...
[21:08:38]<hefring>Provision => rename get_certificate_ip() to assign_certificate_ip() before I go insane => http://drupalcode.org/project/provision.git/commitdiff/314f01ce183ca8b8d...
[21:08:38]<hefring>Provision => #1016890 by joestewart - drop setContext function in the cluster service to fix the... => http://drupalcode.org/project/provision.git/commitdiff/18d01fc575d51ca07...
[21:08:54]<hefring>Hostmaster => 1093436 MariaDB support: Use VARCHAR on the contact_type field => http://drupalcode.org/project/hostmaster.git/commitdiff/b932e1692468c25b...
[21:08:54]<hefring>Hostmaster => fix regression introduced in previous commit => http://drupalcode.org/project/hostmaster.git/commitdiff/1b3885d50773a7ed...
[21:08:54]<hefring>Hostmaster => *REALLY* fix regression introduced in previous commit => http://drupalcode.org/project/hostmaster.git/commitdiff/dae2f1097d13f243...
[21:09:09]<hefring>Provision => 1081536 rollback platform apache configs => http://drupalcode.org/project/provision.git/commitdiff/f7a071135d626bcef...
[21:09:09]<hefring>Provision => Revert "1081536 rollback platform apache configs" => http://drupalcode.org/project/provision.git/commitdiff/ccb4e985e0d33ec12...
[21:09:09]<hefring>Provision => try to rephrase things to provoke jenkins => http://drupalcode.org/project/provision.git/commitdiff/b6f1c6c9b63a2d2b9...
[21:09:09]<mig5>chouette
[21:09:26]<hefring>Provision => more docs in the sudoers file => http://drupalcode.org/project/provision.git/commitdiff/5cdc20e73a7e06d9a...
[21:09:26]<hefring>Provision => deprecate most of this file => http://drupalcode.org/project/provision.git/commitdiff/0428e052be139d797...
[21:16:01]* nlcluigi has joined #aegir
[21:18:34]* alfthecat has joined #aegir
[21:20:36]* Pasqualle has joined #aegir
[21:23:45]* Guest72069 is now known as SeanBannister
[21:27:46]* nlcluigi has quit (Read error: Connection reset by peer)
[21:28:01]* nlcluigi has joined #aegir
[21:35:41]* jonhattan has quit (Quit: llevaré una chaquete de guns'n'roses)
[21:37:30]* kvanderw|afk is now known as kvanderw
[21:40:19]* mig5 has quit (Ping timeout: 260 seconds)
[21:40:31]* adrinux has quit (Quit: Leaving.)
[21:42:53]* anarcat has quit (Ping timeout: 258 seconds)
[21:46:06]* lukus has joined #aegir
[21:46:57]* psynaptic is now known as psynaptic|away
[21:48:38]* beautifulmind has joined #aegir
[21:51:53]* beautifulmind has quit (Read error: Connection reset by peer)
[21:52:37]* beautifulmind has joined #aegir
[21:54:40]* anarcat has joined #aegir
[21:57:14]* JoeMurray has quit (Ping timeout: 260 seconds)
[21:58:24]* troubado` has quit (Read error: Connection reset by peer)
[21:58:46]* troubado` has joined #aegir
[22:05:03]<anarcat>gosh
[22:05:03]<hefring>anarcat: 2 hours 35 min ago <mig5> tell anarcat dig +short aegirproject.twitter.any.io TXT
[22:05:08]<anarcat>twitter through dns
[22:05:13]<anarcat>i'm not sure i wanted to know that :P
[22:09:04]* lukus has quit (Ping timeout: 246 seconds)
[22:10:28]* trainer has joined #aegir
[22:14:28]* anarcat quite happy to see jenkins working
[22:18:53]<anarcat>trying to clear the status on the builds
[22:20:08]<noecc>anarcat: What is jenkins executing here? [08:18:39]
[22:20:23]<anarcat>what's that?
[22:20:44]<noecc>What's being run for rc3 to 1.x upgrade?
[22:20:58]<anarcat>oh
[22:23:04]<anarcat>well, it installs rc3 and upgrades :P
[22:26:58]* redShadow[2V] has joined #aegir
[22:28:29]* snlnz has joined #aegir
[22:42:02]* lukus has joined #aegir
[22:46:41]* lukus has quit (Ping timeout: 252 seconds)
[22:57:48]* Slydder has quit (Ping timeout: 260 seconds)
[23:03:01]* EclipseGc has joined #aegir
[23:04:58]* lukus has joined #aegir
[23:06:10]* scientist has joined #aegir
[23:13:14]* NeoID_ has joined #aegir
[23:16:49]* NeoID has quit (Ping timeout: 264 seconds)
[23:17:07]* lukus has quit (Ping timeout: 252 seconds)
[23:19:53]* NeoID_ has quit (Quit: Forlater kanalen)
[23:20:13]* NeoID has joined #aegir
[23:23:28]* patcon|myplanet has joined #aegir
[23:23:52]* psynaptic|away is now known as psynaptic
[23:24:04]* sdrycroft has quit (Quit: Leaving.)
[23:24:28]* sdrycroft has joined #aegir
[23:31:00]* lukus has joined #aegir
[23:33:40]* lolmaus has quit ()
[23:37:31]* penyaskito has quit (Quit: Saliendo)
[23:40:50]* q0rban has joined #aegir
[23:41:24]* q0rban has quit (Remote host closed the connection)
[23:45:25]* q0rban has joined #aegir
[23:50:10]* obicke has joined #aegir
[23:50:18]* scientist has quit (Ping timeout: 260 seconds)
[23:51:18]<anarcat>what is *that* about: http://twitter.com/doraemonppc/statuses/58448597884207104
[23:51:26]<anarcat>"Hello @aegirproject I can't post in the discussion forums. It says that you don't have installed the spam-filter and its mandatory ....
[23:51:30]<anarcat>wtf?
[23:51:35]<anarcat>ergonlogic: ^?!
[23:52:25]<ergonlogic>anarcat: I'll look
[23:52:25]<hefring>ergonlogic: 11 hours 4 min ago <anarcat> tell ergonlogic that this is a nice way to present install instructions... http://yorba.org/shotwell/install/
[23:53:45]* NeoID has quit (Ping timeout: 248 seconds)
[23:54:15]<bgm>i can post without issues
[23:54:53]<bgm>although it skipped me the captcha, since i guess i passed it recently
[23:55:33]<anarcat>i never got the captcha
[23:55:38]<anarcat>ergonlogic: nevermind, i think it's pebkac
[23:55:53]<ergonlogic>anarcat: agreed
[23:57:16]<bgm>i did a secon post to another thread, and had the captcha this time, and still worked fine
[23:58:19]<anarcat>ok thanks folkds
[23:58:21]<anarcat>folks
[00:02:49]* kvanderw has quit (Ping timeout: 248 seconds)
[00:04:12]<hefring>community => Delete old pre-upgrade-* backups of drush provision and drush_make, delete old modules in drush-backups => http://community.aegirproject.org/discuss/delete-old-pre-upgrade-backups...
[00:04:39]* kvanderw has joined #aegir
[00:06:45]<darthsteven>hefring: Could that title be any longer?
[00:17:32]* NeoID has joined #aegir
[00:19:05]* lukus has quit (Ping timeout: 252 seconds)
[00:22:28]* lukus has joined #aegir
[00:23:48]* mattmcmanus has joined #aegir
[00:27:40]* q0rban is now known as q0rban|otp
[00:28:30]* AquaticDisorder has joined #aegir
[00:32:19]* simesy has quit (Quit: simesy)
[00:32:36]* ezra-g has joined #aegir
[00:33:07]* ezra-g has quit (Client Quit)
[00:33:48]* ezra-g has joined #aegir
[00:49:45]* scientist has joined #aegir
[00:53:52]* whatdoesitwant has joined #aegir
[00:55:31]* beautifulmind has quit (Quit: Leaving.)
[00:59:29]* totten has quit (Ping timeout: 240 seconds)
[01:00:32]* psynaptic is now known as psynaptic|food
[01:14:12]* totten has joined #aegir
[01:26:01]* smthomas has joined #aegir
[01:26:10]* jonhattan has joined #aegir
[01:31:26]* jhedstrom has joined #aegir
[01:36:14]* scientist has quit (Read error: Operation timed out)
[01:37:40]* Pasqualle has quit (Quit: Leaving)
[01:40:56]* Pasqualle has joined #aegir
[01:42:03]* q0rban|otp is now known as q0rban
[01:42:27]* V1ntage_work has quit (Read error: Connection reset by peer)
[01:49:17]* patcon|myplanet has quit (Quit: patcon|myplanet)
[01:53:12]* scientist has joined #aegir
[01:53:38]* ngnp has quit (Quit: ngnp)
[02:01:39]* patcon|myplanet has joined #aegir
[02:08:22]* q0rban has quit (Excess Flood)
[02:08:47]* q0rban has joined #aegir
[02:10:20]* q0rban is now known as q0rban|lunch
[02:16:52]* siliconmeadow has quit (Ping timeout: 246 seconds)
[02:17:25]* kepford has joined #aegir
[02:19:42]* j0nathan has joined #aegir
[02:23:28]* ezra-g is now known as ezra-g|afk
[02:25:23]* q0rban|lunch is now known as q0rban
[02:25:24]* redShadow[2V] has quit (Quit: Leaving)
[02:33:18]* Artusamak is now known as Artusamak_afk
[02:35:12]* penyaskito has joined #aegir
[02:38:02]* Slydder has joined #aegir
[02:40:04]* josh_k has joined #aegir
[02:40:42]* sdrycroft has quit (Quit: Leaving.)
[02:40:50]* Slydder has quit (Read error: Connection reset by peer)
[02:41:23]* Slydder has joined #aegir
[02:55:10]* sdrycroft has joined #aegir
[02:55:26]* sdrycroft has quit (Client Quit)
[03:00:41]* psynaptic|food is now known as psynaptic
[03:02:58]* ryanarmstrong has joined #aegir
[03:22:19]* penyaskito has quit (Ping timeout: 246 seconds)
[03:27:07]* scientist has quit (Ping timeout: 258 seconds)
[03:29:39]* jonathan has joined #aegir
[03:31:31]* j0nathan has quit (Ping timeout: 276 seconds)
[03:31:38]* drapergeek has left #aegir ()
[03:34:26]* kvanderw has quit (Ping timeout: 250 seconds)
[03:36:13]* kvanderw has joined #aegir
[03:37:13]* lolmaus has joined #aegir
[03:42:12]* sk33lz has joined #aegir
[03:42:46]* scientist has joined #aegir
[03:44:21]* gusaus has joined #aegir
[03:47:21]* chipotle has joined #aegir
[03:50:17]* ezra-g|afk is now known as ezra-g
[03:52:30]<chipotle>greetings. I am trying to upgrade aegir using the upgrade script i get the following error upgrade.sh: line 39: tempfile: command not found
[03:52:31]<chipotle>upgrade.sh: line 40: $TEMPFILE: ambiguous redirect
[03:57:38]<anarcat>https://drupal.org/project/evil
[03:57:47]<anarcat><sigh>
[03:57:50]<anarcat>chipotle: which platform?
[03:59:16]<chipotle>centos 5 using command su -s /bin/sh aegir -c "sh upgrade.sh"
[04:01:11]<anarcat>chipotle: tempfile was added there *just* for centos iirc
[04:01:27]<anarcat>ah wait, no
[04:01:34]<anarcat>chipotle: that was fixed: https://drupal.org/node/1123176
[04:01:44]<anarcat>it would be appreciated if you looked in the issue queue before asking :P
[04:02:32]<anarcat>unfortunately, this fix didn't make it in rc7, but will be in 1.0
[04:03:50]<chipotle>i ran a google search and nothing came up. sorry i am new to aergir, where is the issue queue? at drupal.org or the aegir community site?
[04:03:53]<chipotle>thanks for the help
[04:06:14]<anarcat>chipotle: it's in the topic
[04:06:33]* anarcat has changed the topic to Aegir hosting system 1.0-rc7 released! http://community.aegirproject.org/1.0-rc7 | Issue queue: http://is.gd/edalY | read this before asking: http://community.aegirproject.org/help | scrums on tuesday 2200UTC | 1.0 getting cooked for you here http://drupal.org/node/1126736 - smell the sweets, add your sugar!
[04:06:39]* anarcat has changed the topic to Aegir hosting system 1.0-rc7 released! http://community.aegirproject.org/1.0-rc7 | Issue queue: http://is.gd/edalY | read this before asking: http://community.aegirproject.org/help | scrums on tuesday 2200UTC | 1.0 getting cooked for you here http://drupal.org/node/1126736 - smell the sweets, add your sugar!
[04:16:18]* patcon|myplanet has quit (Ping timeout: 260 seconds)
[04:17:21]* patcon|myplanet has joined #aegir
[04:23:30]* patcon_ has joined #aegir
[04:24:21]* arianek_afk has joined #aegir
[04:24:42]* patcon|myplanet has quit (Ping timeout: 250 seconds)
[04:24:43]* patcon_ is now known as patcon|myplanet
[04:24:52]<chipotle>thanks anarcat. i hope my next question is not as obvious ;o) i have granted the user aegir all db admin permissions yet still am unable to generate db backups so tasks like clone, migrate, etc fail. I have reviewed all of the threads at drupal org i could find on the topic and nothing seems to answer my question. can you help?
[04:25:10]<anarcat>i am rather busy now, sorry
[04:25:16]* arianek_afk is now known as arianekWORK
[04:25:59]<chipotle>no problem
[04:38:31]* eft has joined #aegir
[04:42:37]<noecc>chipotle: which version of aegir?
[04:46:20]<chipotle>I believe it is 1.0-rc6
[04:48:29]<noecc>I've had a similar problem in alpha8, verifying the site fixed the unable to create backup problem.
[04:51:41]* scientist has quit (Ping timeout: 240 seconds)
[04:53:18]* mattmcmanus has quit (Ping timeout: 250 seconds)
[04:53:24]<chipotle>i will try that. thanks.
[04:58:37]* arianekWORK is now known as arianek_afk
[05:10:17]* NeoID_ has joined #aegir
[05:11:07]* penyaskito has joined #aegir
[05:12:41]* NeoID has quit (Ping timeout: 240 seconds)
[05:15:00]* scientist has joined #aegir
[05:19:44]* NeoID_ has quit (Quit: Forlater kanalen)
[05:20:13]* NeoID has joined #aegir
[05:20:35]* mrfelton has quit (Ping timeout: 258 seconds)
[05:25:56]* mrfelton has joined #aegir
[05:26:16]* psynaptic is now known as psynaptic|away
[05:27:51]* mrfelton_ has joined #aegir
[05:29:42]<chipotle>i still get this error all of the time Could not generate database backup from mysqldump. (error: ) user aegir has full db privileges.
[05:30:57]<noecc>chipotle: verify the credentials in drushrc
[05:31:02]* NeoID has quit (Quit: Forlater kanalen)
[05:31:07]* mrfelton has quit (Ping timeout: 246 seconds)
[05:31:07]* mrfelton_ is now known as mrfelton
[05:31:15]* Artusamak_afk is now known as Artusamak
[05:32:07]<chipotle>noecc: how do i do that?
[05:34:44]* NeoID has joined #aegir
[05:55:28]* scientist has quit (Ping timeout: 258 seconds)
[06:03:25]* chipotle has quit (Read error: Operation timed out)
[06:08:09]* scientist has joined #aegir
[06:08:33]* whatdoesitwant has quit (Quit: ChatZilla 0.9.86.1-rdmsoft [XULRunner 1.9.2/20100222071121])
[06:14:41]* Pasqualle has quit (Quit: Leaving)
[06:17:44]* chipotle has joined #aegir
[06:21:49]* jhedstrom has quit (Ping timeout: 276 seconds)
[06:22:52]* lolmaus has quit ()
[06:27:26]* jonhattan has quit (Quit: Ex-Chat)
[06:32:17]* sepehr has joined #aegir
[06:32:17]* sepehr has quit (Client Quit)
[06:36:57]* jhedstrom has joined #aegir
[06:38:43]* q0rban is now known as q0rban|afk
[06:41:26]* chipotle has quit (Remote host closed the connection)
[06:51:10]<bgm>how can I invoke a fail() in a provision verify hook?
[06:51:16]<bgm>(to trigger a rollback)
[06:51:57]<bgm>i see it's part of provisionChainedState, but i'm not sure how to call it, whether I need to extend it?
[06:53:33]* jhedstrom has quit (Read error: Connection reset by peer)
[06:54:13]<bgm>hum, although i'm not making sense.. cannot rollback a verify.
[06:54:17]* q0rban|afk is now known as q0rban
[06:54:49]<bgm>in other words, i have a module (civicrm) that needs to have its upgrade function called when a migrate is done
[06:55:49]* mattmcmanus has joined #aegir
[06:57:18]* SqyD has quit (Ping timeout: 260 seconds)
[06:58:41]* snlnz has quit (Ping timeout: 240 seconds)
[07:00:56]* snlnz has joined #aegir
[07:05:10]<darthsteven>bgm: you could implement a drush hook, for the migration task
[07:06:19]<darthsteven>bgm: drush_HOOK_post_provision_migrate
[07:06:31]<darthsteven>bgm: drush_db_post_provision_migrate() does this
[07:08:09]<bgm>darthsteven: thanks! i'll look at its code
[07:08:31]* chipotle has joined #aegir
[07:08:36]<darthsteven>bgm: It only has a single line of code though...
[07:08:47]<bgm>oh :)
[07:08:56]<bgm>i'm wondering how to trigger a rollback, if there is an error
[07:08:57]<darthsteven>bgm: and the context looks like it's still the old site not the new one?
[07:09:06]<bgm>civicrm is a bit infamous for failing upgrades
[07:09:43]<darthsteven>you just raise an error and provision will detect it and rollback
[07:09:51]<darthsteven>so, watchdog errors
[07:09:56]<darthsteven>or drush errors
[07:10:11]<bgm>ah ok, too easy ;-)
[07:10:24]<darthsteven>also if you return an error from the update hook, then provision will roll back
[07:10:44]<bgm>such as drush_log(dt('foo'), 'error') ?
[07:11:10]<chipotle>now i am at the drush command prompt and want to check db privileges for aegir to troubleshoot the this error; Could not generate database backup from mysqldump. (error: ) what command do I use? thanks
[07:11:56]<darthsteven>bgm: drush_error I think
[07:11:57]<bgm>chipotle: try using the mysql credentials used by the aegir user to run mysqldump manually (?)
[07:12:18]<darthsteven>bgm: but if you're writing an update hook, just return an error there as normal for Drupal
[07:12:43]<darthsteven>chipotle: drush sql-cli
[07:12:58]* AntiNSA has quit (Ping timeout: 246 seconds)
[07:13:01]<darthsteven>chipotle: or just 'sql-cli' if your on the drush command line
[07:13:19]<darthsteven>if you can connect to the DB, then the credentials are fine for drush
[07:14:02]<chipotle>do i need to be in the aegir site directory to run that?
[07:14:03]* jhedstrom has joined #aegir
[07:14:34]<darthsteven>chipotle: easier than that, if your site is www.example.com
[07:14:47]<darthsteven>chipotle: run drush @www.example.com sql-cli
[07:14:56]* AntiNSA has joined #aegir
[07:15:02]<darthsteven>(as the Aegir user)
[07:15:04]* mikl has joined #aegir
[07:15:22]<darthsteven>so most likely /var/aegir/drush/drush @www.example.com sql-cli
[07:16:18]* Slydder has quit (Quit: Leaving.)
[07:16:19]<chipotle>permission denied. i guess that's my problem but i have granted aegir user all db privileges.
[07:17:25]<darthsteven>hmm...
[07:17:46]<darthsteven>you're not on a remote server are you?
[07:18:06]* AntiNSA-AFK has joined #aegir
[07:18:29]<chipotle>no. virtualbox
[07:18:58]<darthsteven>ah, no I mean is your database server and the master aegir server the same machine?
[07:19:07]<chipotle>even trying to run the upgrade script all goes well accept the db backup error
[07:20:06]<chipotle>i guess so. i only have one server
[07:20:09]<anarcat>bgm: to set errors, you want drush_set_error()
[07:21:13]* AntiNSA has quit (Ping timeout: 248 seconds)
[07:21:49]<darthsteven>chipotle: Can you run mysqldump manually with the aegir credentials?
[07:22:25]<chipotle>i don't know the command for that. i can try it. what is the command?
[07:22:40]* alfthecat has left #aegir ()
[07:23:03]<darthsteven>Google mysqldump
[07:23:28]<chipotle>ok
[07:24:15]<darthsteven>Probably faster than me tapping all the details out here...
[07:24:25]* troubado` has quit (Read error: Connection reset by peer)
[07:25:51]* michaelfavia has joined #aegir
[07:26:16]<michaelfavia>are the two servers that are populated by defaul on my ubunut LTS server necessary?
[07:26:23]<michaelfavia>localhost and the FQDN
[07:28:05]* jonathan has quit (Ping timeout: 252 seconds)
[07:28:22]<darthsteven>As in why can't they be a single server?
[07:28:54]* troubado` has joined #aegir
[07:29:31]<michaelfavia>darthsteven, i have 1 that has the mysql server and one that has the apache.
[07:29:36]<michaelfavia>and i just wondered why.
[07:29:44]<michaelfavia>as they both refer tot the same physical machine.
[07:29:54]<michaelfavia>by different hostnames
[07:30:34]<michaelfavia>so i enabled the mysql on the fqdn one (boron.favish.com) and then disabled it on localhost
[07:30:50]<michaelfavia>now aegir is not loading.
[07:30:57]<michaelfavia>i assume that it was loaded on localhost.
[07:30:58]<darthsteven>Ah
[07:31:08]<michaelfavia>i dont know that to be true though.
[07:31:17]<michaelfavia>i dont have to ditch one just thought it was unnecessary
[07:32:34]<michaelfavia>i can reflash the box if i have to but is there an easy way to get aegir back up so i can reenable mysql on the host i deleted it from
[07:32:35]<michaelfavia>?
[07:33:08]<darthsteven>I think that having both on the same machine is just a special case
[07:33:24]<darthsteven>And so the installer doesn't try to care too much
[07:33:37]<michaelfavia>darthsteven, i suppose youre right.
[07:33:56]<darthsteven>Try popping your db creds in the settings.php for aegir
[07:35:08]<michaelfavia>darthsteven, thx will do.
[07:35:21]* michaelfavia needs to investigate the project backend a little.
[07:35:53]<darthsteven>Yeah, some kind of provision recovery tool would be cool
[07:37:33]<chipotle>with this command myslqdump aegir i get permission denied.
[07:38:02]<chipotle>mysqldump
[07:38:10]<darthsteven>chipotle: Right, well I'd look into that...
[07:38:20]* darthsteven waves good night
[07:38:58]<Zelfje>chipotle: mysqldump -u user -p database.* > yourbackupfile.sql
[07:39:57]* chipotle has quit (Read error: Operation timed out)
[07:40:15]<Zelfje>hope he had the chance to read ^^
[07:42:55]<Zelfje>Anyone have experiance with multiple solr instances on agir, i found : http://davehall.com.au/blog/dave/2010/06/26/multi-core-apache-solr-ubunt... would this be possible to create a solr config in aegir or is there a better method?
[07:44:14]* chipotle has joined #aegir
[07:46:38]* q0rban is now known as ericduran
[07:46:44]* ericduran is now known as q0rban
[07:47:56]* simesy has joined #aegir
[07:49:18]* arianek_afk is now known as arianek
[07:49:20]* q0rban has quit (Quit: Computer has gone to sleep.)
[07:50:19]* arianek is now known as arianek_afk
[07:54:52]* scientist has quit (Ping timeout: 250 seconds)
[08:01:04]* chipotle has quit (Read error: Operation timed out)
[08:03:01]* lukus2 has joined #aegir
[08:03:32]* lukus has quit (Disconnected by services)
[08:04:10]* lukus has joined #aegir
[08:04:14]* lukus has quit (Changing host)
[08:04:14]* lukus has joined #aegir
[08:12:48]* AquaticDisorder has quit (Remote host closed the connection)
[08:13:13]* josh_k has quit (Ping timeout: 264 seconds)
[08:15:37]* susie has quit (Ping timeout: 248 seconds)
[08:16:10]* susie has joined #aegir
[08:18:15]* secoif_ has joined #aegir
[08:19:17]* secoif has quit (Ping timeout: 276 seconds)
[08:19:18]* secoif_ is now known as secoif
[08:23:22]* penyaskito has quit (Quit: Saliendo)
[08:30:45]* smthomas has left #aegir ()
[08:40:07]* ezra-g has quit (Quit: COD: The Conference Organizing Distribution. http://usecod.com)
[08:45:02]<mvc>Zelfje: hmm, interesting idea
[08:46:30]<mvc>it would take a bit more work to do from aegir, we'd have to add a solr server class and then figure out a way to push conf files to it and restart it, like is currently done for http
[08:46:57]<mvc>should definitely be possible though
[08:47:37]* mvc has quit (Quit: Changing server)
[08:48:38]* mig5 has joined #aegir
[08:52:18]* Aurorus has joined #aegir
[08:52:33]* patcon_ has joined #aegir
[08:53:27]* mvc has joined #aegir
[08:53:46]* patcon|myplanet has quit (Ping timeout: 246 seconds)
[08:53:46]* patcon_ is now known as patcon|myplanet
[08:54:30]* realityloop has joined #aegir
[09:16:09]* jonathan has joined #aegir
[09:16:23]* arianek_afk is now known as arianek
[09:17:39]<anarcat>Zelfje: open a feature request! it's an idea that's been floating around forever and we need to merge the good tips in one place
[09:31:25]* ryanarmstrong has left #aegir ()
[09:32:13]* arianek is now known as arianekWORK
[09:40:39]* eft has left #aegir ()
[09:41:02]* arianekWORK is now known as arianek_afk
[09:46:09]* mrfelton has quit (Quit: mrfelton)
[09:53:33]* jonathan is now known as j0nathan
[09:53:48]* j0nathan has quit (Quit: ¡Hasta luego!)
[09:54:24]<mig5>anarcat: this might interest you btw http://drupal.org/node/1123302
[09:54:26]<mig5>tripped me up
[09:54:57]<mig5>also, drupalcon london session proposals have opened: perhaps we should get one in this time, or do you actually prefer the chicago method of 'lots of bogs'
[09:55:00]<mig5>er
[09:55:00]<mig5>bofs, haha
[09:55:18]<mig5>i do prefer lots of bogs, preferably evenly spaced apart
[09:55:50]<anarcat>lots of bugs!! ;)
[09:55:57]<anarcat>we should probably do a session of some sort
[09:56:01]<anarcat>what's the deadline?
[09:56:16]* realityloop has quit (Remote host closed the connection)
[09:56:40]<mig5>15th may http://london2011.drupal.org/conference/sessions
[09:58:06]* josh_k has joined #aegir
[09:59:41]<anarcat>gotcha
[09:59:51]<mig5>i wonder if i will miss the 1.0 aegir release :)
[09:59:58]<mig5>i might be in the air, or offline in any case