IRC logs for #aegir, 2011-02-22 (GMT)

‹ 2011-02-21
2011-02-23 ›
TimeNickMessage
[11:00:16]* talengix has quit (Ping timeout: 240 seconds)
[11:00:49]<omega8cc>or /var/spool/postfix/deferred, or something
[11:00:57]<omega8cc>or
[11:01:14]<omega8cc>just repair your postfix and ask for password again
[11:03:10]<EugenMayer>or elvis has left the building :)
[11:08:57]<omega8cc>skwashd: you managed to break the signup url again (I fixed it before) http://community.aegirproject.org/node/371
[11:09:16]<omega8cc>skwashd: there is a typo in the subdomain
[11:10:10]<omega8cc>the http://initimateaegir.eventbrite.com/ should be http://intimateaegir.eventbrite.com/
[11:10:19]<skwashd>omega8cc: thanks
[11:10:24]<skwashd>it was copied and pasted
[11:11:19]<skwashd>fixed
[11:11:23]<omega8cc>thanks
[11:12:51]* that0n3guy has joined #aegir
[11:13:15]<omega8cc>skwashd: hmm, it is fixed on the target page but not in c.a.o
[11:13:51]<skwashd>omega8cc: try it now
[11:14:45]<omega8cc>now fixed, maybe it was a cache in oa
[11:18:43]* snlnz has quit (Ping timeout: 255 seconds)
[11:18:57]<milk>i'm poking aegir again, but got a problem with a site created with it in that /user, /admin, etc. don't want to work (Not Found). was using the ubuntu script in the community site install subpage (which mig5 went WTF to), but i've reinstalled with the install.sh on a fresh ubuntu 10.4 server and getting the same problem. /platform.d files look fine (as much as i can tell anyway) and copying the .htaccess from the hostmaster folder does nowt.
[11:18:57]<milk>any thing i might be missing?
[11:21:04]* snlnz has joined #aegir
[11:23:39]* kvanderw has quit (Quit: kvanderw)
[11:24:33]* lukus has quit (Read error: Connection reset by peer)
[11:39:20]* berniecram___ has joined #aegir
[11:40:54]* dob_ has joined #aegir
[11:41:04]* berniecram has quit (Ping timeout: 240 seconds)
[11:41:04]* EclipseGc has joined #aegir
[11:41:04]* berniecram___ is now known as berniecram
[11:44:50]* sepehr1 has quit (Ping timeout: 246 seconds)
[11:47:06]* Chipie has quit (Ping timeout: 246 seconds)
[11:50:32]* samhassell has joined #aegir
[11:53:28]* SeanBannister has quit (Ping timeout: 240 seconds)
[11:55:03]* Chipie has joined #aegir
[11:56:03]* gusaus has quit (Read error: Connection reset by peer)
[11:56:13]* gusaus has joined #aegir
[11:56:44]* susie has joined #aegir
[11:59:04]* SeanBannister has joined #aegir
[11:59:50]* kvanderw has joined #aegir
[12:01:20]* ryanarmstrong has quit (Quit: Leaving.)
[12:07:03]* susie has quit (Ping timeout: 246 seconds)
[12:08:51]* susie has joined #aegir
[12:09:49]* samhassell has quit (Ping timeout: 250 seconds)
[12:09:55]* ezra-g has joined #aegir
[12:12:50]* psynaptic is now known as psynaptic|afk
[12:13:28]* berniecram has quit (Ping timeout: 240 seconds)
[12:14:12]* FatGuyLaughing has quit (Quit: FatGuyLaughing)
[12:14:13]* ezra-g has quit (Read error: Connection reset by peer)
[12:16:28]* ezra-g has joined #aegir
[12:16:51]* berniecram_ has joined #aegir
[12:19:21]* susie has quit (Ping timeout: 250 seconds)
[12:24:55]* EclipseGc has quit (Quit: EclipseGc)
[12:26:45]* susie has joined #aegir
[12:41:16]* ergonlogic has joined #aegir
[12:41:39]* ezra-g has quit (Quit: ezra-g)
[12:43:51]* mrchrisadams has quit (Quit: mrchrisadams)
[12:44:49]* ezra-g has joined #aegir
[12:45:05]<hefring>community => Thanks => http://community.aegirproject.org/node/378
[12:45:50]* mrchrisadams has joined #aegir
[12:51:00]* lukus has joined #aegir
[12:53:57]* josh_k has joined #aegir
[12:54:00]* EugenMayer1 has joined #aegir
[12:56:26]* EugenMayer has quit (Ping timeout: 276 seconds)
[13:05:02]<anarcat>omega8cc: Upgrade script - is this platform agnostic? (works also for Nginx)
[13:05:05]* omega8cc has quit (Remote host closed the connection)
[13:05:12]* omega8cc has joined #aegir
[13:05:14]<anarcat>omega8cc: i meant OS-agnostic...
[13:08:34]* ezra-g has quit (Quit: ezra-g)
[13:11:34]<omega8cc>anarcat: I believe it is also OS agnostic, and works both for Apache and Nginx based Aegir installs
[13:12:57]* mlncn has joined #aegir
[13:13:09]<omega8cc>anarcat: at least I don't see anything OS specific there
[13:19:29]<omega8cc>anarcat: hmm.. it looks it is using bash magic in this line http://git.aegirproject.org/?p=provision.git;a=blob;f=upgrade.sh.txt;h=2..., so not sure how it works under new sh on Ubuntu and Debian Squeeze, I hope the new default shell supports this? to be confirmed, anyway
[13:22:11]<anarcat>omega8cc: that should be okay...
[13:22:24]<omega8cc>I hope
[13:23:22]* milk has quit (Quit: baaaiiii)
[13:24:10]<anarcat>yeah, it works
[13:24:21]<anarcat>i test those things with dash
[13:25:29]<anarcat>http://pastebin.com/9r4Cufpf
[13:27:16]<omega8cc>ok, so it looks safe and OS agnostic
[13:28:28]<anarcat>we'll need to make sure the script refuses to run on debian-package-managed machines though
[13:31:18]* ezra-g has joined #aegir
[13:33:27]<omega8cc>anarcat: something like check for the result of "aptitude show provision" ?
[13:34:16]<anarcat>nah, just the location of files should be enuf
[13:34:23]<anarcat>boy the spelling
[13:39:58]<Met4physica>still looking for some ssl lovin
[13:40:19]<Met4physica>perhaps mcdonalds should sell mc-certificates
[13:45:28]* susie has quit (Ping timeout: 240 seconds)
[13:51:59]* omega8cc_ has joined #aegir
[13:52:44]* EclipseGc has joined #aegir
[13:53:39]* omega8cc_ has quit (Remote host closed the connection)
[13:53:47]* omega8cc_ has joined #aegir
[13:55:06]* that0n3guy has quit (Remote host closed the connection)
[13:55:33]* omega8cc has quit (Ping timeout: 250 seconds)
[13:55:33]* omega8cc_ is now known as omega8cc
[14:06:23]* lukus has quit (Ping timeout: 246 seconds)
[14:08:31]* susie has joined #aegir
[14:11:28]* mrchrisadams has quit (Quit: mrchrisadams)
[14:19:27]* EclipseGc has quit (Quit: EclipseGc)
[14:31:09]* arianek has joined #aegir
[14:42:35]* ezra-g has quit (Quit: ezra-g)
[14:45:27]* samhassell has joined #aegir
[14:50:30]* q0rban has joined #aegir
[14:51:09]* boztek has joined #aegir
[14:57:43]* Nautilus has joined #aegir
[15:04:38]* q0rban has quit (Quit: Computer has gone to sleep.)
[15:09:30]* gusaus has quit (Quit: gusaus)
[15:12:28]* gusaus has joined #aegir
[15:21:26]* EclipseGc has joined #aegir
[15:22:27]* EclipseGc has quit (Client Quit)
[15:28:39]* dipen has quit (Read error: Operation timed out)
[15:29:46]* Nautilus has left #aegir ("Client Exiting")
[15:31:54]* ezra-g has joined #aegir
[15:32:13]* ezra-g has quit (Client Quit)
[15:35:34]* skwashd has quit (Ping timeout: 272 seconds)
[15:45:28]* dipen has joined #aegir
[15:58:44]* merro has joined #aegir
[16:06:05]* dipen has quit (Ping timeout: 246 seconds)
[16:13:23]* samhassell has quit (Ping timeout: 276 seconds)
[16:15:34]* ergonlogic has quit (Ping timeout: 260 seconds)
[16:25:07]* dob_ has quit (Quit: dob_)
[16:26:30]* gusaus has quit (Quit: gusaus)
[16:32:38]* yuracagran has joined #aegir
[16:49:28]* Irishgringo has quit (Ping timeout: 240 seconds)
[16:52:10]* nick_santa has quit (Quit: nick_santa)
[16:57:35]* gusaus has joined #aegir
[16:59:34]* Slydder1 has joined #aegir
[17:22:13]* xk has joined #aegir
[17:35:49]<xk>http://drupal.pastebin.com/xsFAQJhN
[17:35:49]<xk>hmm
[17:44:22]* mikl has joined #aegir
[17:45:12]* mrchrisadams has joined #aegir
[17:47:45]* arianek has quit (Remote host closed the connection)
[17:48:29]* talengix has joined #aegir
[17:49:13]<xk>weird database issue... wonder how that happened.
[17:49:51]<xk>doesnt matter; fixed now. added the user with full perms, then added another user it complained about.
[17:52:20]* yuracagran has quit (Read error: Connection reset by peer)
[17:56:46]* Slydder1 has quit (Quit: Leaving.)
[18:06:15]* feldmand has joined #aegir
[18:11:06]* ngnp has joined #aegir
[18:11:33]* gusaus has quit (Quit: gusaus)
[18:13:09]* boztek has quit (Remote host closed the connection)
[18:15:32]* berniecram_ has quit (Quit: berniecram_)
[18:20:36]* mrchrisadams has quit (Quit: mrchrisadams)
[18:29:13]* boztek has joined #aegir
[18:31:02]<boztek>Argh. My new aegir box has all of a sudden stopped provisioning due to email send fails - "WD mail: Error sending e-mail" - not sure when it was last working. Would love some troubleshooting tips.
[18:33:01]* josh_k has quit (Quit: This computer has gone to sleep)
[18:33:37]<eft>hefring: tell EugenMayer that his recent revisions to http://community.aegirproject.org/node/71 might be better placed in a new page at same level in handbook - just a suggestion
[18:33:37]<hefring>eft: I'll pass that on when EugenMayer is around.
[18:38:43]<boztek>so fekin weird - postfix was uninstalled - wtf
[18:38:50]<boztek>who's been touching my server!
[18:38:53]<boztek>:)
[18:49:53]* mikl_ has joined #aegir
[18:53:17]* mikl has quit (Ping timeout: 276 seconds)
[18:56:52]* sepehr has joined #aegir
[19:01:38]* ngnp_ has joined #aegir
[19:01:55]* josh_k has joined #aegir
[19:01:55]* josh_k has quit (Changing host)
[19:01:55]* josh_k has joined #aegir
[19:01:58]* gusaus has joined #aegir
[19:01:59]* josh_k has quit (Remote host closed the connection)
[19:02:40]* eft has left #aegir ()
[19:05:13]* ngnp has quit (Ping timeout: 250 seconds)
[19:05:13]* ngnp_ is now known as ngnp
[19:06:18]<mig5>boztek: oh, about that.. :)
[19:07:56]* mikl_ has quit (Read error: Connection reset by peer)
[19:10:13]* boztek waves fist
[19:17:20]* recidive has quit (Quit: zzz)
[19:22:37]<EugenMayer1>morning
[19:22:49]* EugenMayer1 waves 3rd
[19:24:30]* kylemathews has quit (Ping timeout: 240 seconds)
[19:25:08]* talengix has quit (Ping timeout: 276 seconds)
[19:34:33]* omega8cc_ has joined #aegir
[19:37:52]* omega8cc has quit (Ping timeout: 240 seconds)
[19:37:53]* omega8cc_ is now known as omega8cc
[19:44:07]* Aciid has joined #aegir
[19:52:00]* siliconmeadow has joined #aegir
[19:54:39]* mikl has joined #aegir
[19:54:39]* mikl has quit (Changing host)
[19:54:39]* mikl has joined #aegir
[20:10:12]* gusaus has quit (Quit: gusaus)
[20:13:24]* redShadow[2V] has joined #aegir
[20:14:15]* jonhattan has joined #aegir
[20:15:50]* boztek has quit (Ping timeout: 276 seconds)
[20:20:16]* boztek has joined #aegir
[20:21:17]* xk has quit (Quit: BitchX: time to make the donuts.)
[20:23:12]* V1ntage_work has joined #aegir
[20:26:52]<sdboyer>anarcat mig5: problem that i'd forgotten about: http://pastebin.com/G3KL7d8P
[20:27:21]<sdboyer>anarcat mig5: if those release nodes are useless, and not even good for anything historically, then i can just delete them and we don't have to worry about it
[20:27:30]<sdboyer>but i don't want to do that without the go-ahead from y'all
[20:27:47]* dipen has joined #aegir
[20:28:00]<mig5>that's the legacy stuff from drupal.org cvs that was migrated into git
[20:28:08]<mig5>what's the implication, just release nodes?
[20:28:31]<sdboyer>yes, release nodes and any packages generated therefrom
[20:28:40]<mig5>hmm
[20:29:00]<sdboyer>basically i either have to delete them or not do the clone for you guys
[20:29:01]<mig5>well, http://drupal.org/node/196005/release
[20:29:03]<mig5>they did exist
[20:29:04]* Artusamak_afk is now known as Artusamak
[20:29:05]<mig5>but they're mega old
[20:29:08]<sdboyer>right
[20:29:15]<mig5>and i pity the fool who's still looking for those :)
[20:29:19]<sdboyer>indeed :)
[20:29:24]<mig5>so i don't know that i care, but i t would be good to get anarcat's opinion too
[20:29:26]<sdboyer>b/c as-is, if those aren't addressed, then we'll have a big hole in packaging
[20:29:27]<sdboyer>kk
[20:29:40]<sdboyer>as long as y'all get me a decision within the next 24 hours, it's cool
[20:29:43]<mig5>is there a workaround we could do to 'fix' it?
[20:29:58]<sdboyer>you could create tags in your current repo corresponding to all those
[20:30:01]<sdboyer>s/repo/repos/
[20:30:37]<sdboyer>whether they're real (as in, actually point to the place the release used to point to) or not i don't really care
[20:30:51]<sdboyer>other than that, not really any options
[20:31:58]<mig5>it's odd because i thought we had such tags
[20:32:01]<mig5>http://git.aegirproject.org/?p=provision.git;a=tags
[20:32:15]<mig5>ah they're named like the cvs 6-- stuff
[20:34:36]<sdboyer>ohh
[20:34:37]<sdboyer>heh
[20:34:51]<sdboyer>yeah, cloned repos miss the transform pass...
[20:34:52]<sdboyer>hmm
[20:35:23]<sdboyer>mig5: hang on a sec, i might be able to move the clone step earlier and solve this problem easily
[20:35:39]<sdboyer>that'd also get you keyword & translation stripping
[20:35:44]* dipen has quit (Ping timeout: 260 seconds)
[20:37:10]<sdboyer>mig5: yeah, i can take care of htis
[20:37:18]<sdboyer>forget my concerns :)
[20:38:53]<mig5>"no probs
[20:38:54]<mig5>you rock
[20:38:57]* sdboyer grins
[20:39:14]* TrevorT has quit (Ping timeout: 276 seconds)
[20:39:52]* kvanderw has quit (Ping timeout: 240 seconds)
[20:52:41]* instanceofjamie has joined #aegir
[20:52:57]* instanceofjamie has left #aegir ("Leaving")
[20:55:34]* jonhattan has quit (Remote host closed the connection)
[21:06:51]* TrevorT has joined #aegir
[21:10:53]* dipen has joined #aegir
[21:17:16]<feldmand>Installation profile creation literature is regrettably sparse - given its necessity to robust/reliable platform/site dev. What best practices have you , the Aegir user, found in developing install profiles? Organize the profile into 10+ different sections? Use a test bed and profile master? Start with a defalut profile (ie OpenAtrium, OpenEnterprise..)? Check api helper fxs? Query db to find values? Thanks.
[21:20:54]<mig5>use initiative :)
[21:21:09]<mig5>all of the above
[21:21:50]* sepehr1 has joined #aegir
[21:21:52]<mig5>don't wait to be spoon fed, that's my best advice. also, check out http://drupal.org/project/profiler, which promises to be a great boon those struggling to get install profiles over the line.
[21:23:23]<mig5>and means not having to depend on install_profile_api (whose functions i read in order to learn install profiles)
[21:24:03]* sepehr has quit (Ping timeout: 250 seconds)
[21:25:58]<feldmand>mig5: Thanks for the reference to profiler and advice. I was hoping that there was some better documentation available. I have installed and have been studying the OpenEnterprise distribution, but have been unable to figure out the make file and profile interact. By what mechanism does a profile invoke a profile makefile?
[21:26:52]<mig5>it's the other way around
[21:27:02]<mig5>drush make invokes a makefile that might fetch core + a profile
[21:27:15]<mig5>a proifile has its own makefile, anddrushMake automatically findsand recurses into that to fetch dependencies
[21:27:27]<mig5>sorry, spacebar slightly broken.
[21:28:03]<mig5>install profiles themselves aren't 'aware' of makefiles, but they can have them.it's drush_make's job to find and build based on makefiles
[21:28:56]<feldmand>mig5: I got that much. But we may need to clarify our terms. You have been using a stub makefile to invoke a profile. But where in the profile does it make reference to a profile makefile (the one that gets called recursively)?
[21:29:37]* omega8cc has quit (Remote host closed the connection)
[21:29:44]* omega8cc has joined #aegir
[21:29:44]<feldmand>mig5: so the profile and the profile makefile, should have the same base names to help drush make?
[21:29:51]* ngnp has quit (Remote host closed the connection)
[21:30:26]* lukus has joined #aegir
[21:30:27]<mig5>ideally, but technically it doesn't matter.
[21:30:33]<mig5>let me find you an example
[21:30:46]<feldmand>mig5: That would be great. Thanks.
[21:30:58]<mig5>stub makefile: https://github.com/mig5/builds/blob/master/mig5_net.build
[21:31:17]* ngnp has joined #aegir
[21:31:19]<mig5>the profile repo https://github.com/mig5/mig5_net
[21:31:26]<mig5>which has a .profile and a .make with the same base name
[21:31:36]<feldmand>mig5: I guess as long as the profile makefile is located within the profile/ directory then drush make should be able to recursively find it, right?
[21:31:58]<mig5>yep, drush_make will look for anything with a .make extension I believe.
[21:32:07]<mig5>out of convention it helps to name them the same (perhaps)
[21:32:21]<EugenMayer1>feldmand: i think before you get into profiles to deeply, you should start with features. The profile API is horrible for everything past initial deployment. Be sure to check the profile API extensions developed by quicksketch ( also used by aegir ). But it is more or less obsolete with features
[21:32:25]<mig5>i don't think the 'stub' file and the profile makefile can share the same name.
[21:33:03]<EugenMayer1>I think the best paradigm here is: Do as much as possible with features, fill the gaps with profiles and finally automate installation with the core profile API
[21:33:15]<feldmand>mig5: is that one of the reasons why you use the .build suffix for the stub makefile?
[21:33:17]<mig5>yep, but that's nowhere near the originalquestion.
[21:33:23]<mig5>feldmand: yes, partially
[21:33:35]<mig5>just my own convention.
[21:35:43]* omega8cc_ has joined #aegir
[21:36:02]<EugenMayer1>mig5: my answer was related to his Q @ (11:17:17) feldmand: I
[21:36:41]<EugenMayer1>the one you are talking now is kind of the raw "shipping and building" part with recursive make files (which are also important). But that is by far not the issue with profiles :)
[21:36:53]<mig5>agreed
[21:37:14]<mig5>but you need the profile as a sort of 'bootstrapper' in order to fire off the features. my own profiles are *very* bare for this reason
[21:37:16]<feldmand>EugenMayer1: Thanks for the input. However, I must admit that from what I have understood so far your input creates a bit of cognitive dissonance. My understanding - Features are developed for custom modules. For features to work with Drush make a profile is required. Profiles configure the contributed modules, themes, features, etc. helping to create a distribution. I will need to study your comments in greater detail.
[21:37:22]<mig5>i rarely even install the site with that profiler, i just drush vset to it later
[21:37:36]<mig5>s/profiler/profile/
[21:37:45]<EugenMayer1>jap
[21:37:49]* boztek has quit (Quit: Computer has gone to sleep.)
[21:37:51]<EugenMayer1>Anything else is suicide :)
[21:38:04]<mig5>it'd drupal: *everything* is suicide :)
[21:38:08]<mig5>it's* :)
[21:38:21]<EugenMayer1>feldmand: i just wanted to shake you a bit. Drupalianer praise the "profiles" called "distributions a lot
[21:38:31]<EugenMayer1>but to be honest, profiles in D6 are pure sh*t.
[21:38:53]<EugenMayer1>The way d.o thinks about them is a DB dump import in code-form..
[21:39:04]* omega8cc has quit (Ping timeout: 240 seconds)
[21:39:04]* omega8cc_ is now known as omega8cc
[21:39:11]* jonhattan has joined #aegir
[21:39:36]<mig5>lol, that's a Feature too, just with a configurable UI
[21:39:46]<mig5>anything with a database backend is shit.
[21:40:20]<EugenMayer1>so you really have to step back and look at features here, which has by far the better solutions for profiles. Take it that way :
[21:40:20]<EugenMayer1>- make files are kind of the "get codebase from there, that version, put it there"
[21:40:20]<EugenMayer1>- Features: install and maintain a set of modules, set some initial setting, control overrides
[21:40:33]<EugenMayer1>mig5: not at all :)
[21:40:52]<feldmand>EugenMayer1: Thanks for the shake up. Actually a month ago I was planning to develop in D7, but as I looked for contributed modules, I found that much had not been ported. I felt safer sticking with D6. That said, if D6 profiles are sh*t then I need to find another way. You recommend using features in D6 for not only custom modules, but also for contributed mondules?
[21:41:05]<EugenMayer1>The intersting part on features are the detection of "overrides" and the "revert" of those or the upgrade. but actually thats not yet working to good ;-)
[21:41:19]* realityloop has joined #aegir
[21:41:26]<EugenMayer1>use feature for _everything.
[21:41:47]<EugenMayer1>A Feature is a container, mostly not a module for itself ( the own code is small actually )
[21:41:54]* realityloop has quit (Client Quit)
[21:42:24]<EugenMayer1>Its a container for installing a set of modules, configure them a specific way and track changes later, being able to redeploy new settings using code ( which is great )
[21:42:58]<EugenMayer1>90% of my features just use drupal.org modules, maybe add some CCK fields to content types, add some views for grouping and mostly holding configuration
[21:43:05]* skwashd has joined #aegir
[21:43:19]<EugenMayer1>A feature is most like what you want with profile in a "micro" way :)
[21:44:18]<feldmand>EugenMayer1: So are you suggesting that in D6 the best workflow would involve just enough of a profile to allow drush make to function and do all initial settings for both custom and contributed code with features?
[21:44:28]<EugenMayer1>yes
[21:44:41]<EugenMayer1>Golden rule: avoid stuffin any code into the profile.module.
[21:45:04]<EugenMayer1>There are things not supported by features, you have to do them in your profile. But limit that to close to zero
[21:45:08]<feldmand>EugenMayer1: Great! We understood each other rather quickly. I must have gotten enough sleep.
[21:45:10]<EugenMayer1>be sure to look at strongarm
[21:45:17]<EugenMayer1>and record_features
[21:45:30]<EugenMayer1>That will make your day ;-)
[21:45:35]<EugenMayer1>gn8
[21:46:02]<feldmand>Would I be using strongarm and record_features in the Features world?
[21:46:10]<EugenMayer1>oh yes :)
[21:46:36]<EugenMayer1>mig5: to that --update rsyn think .. i would be suprised if there are differen implementations of rsync. On my debian lenny / ubuntu 10.10
[21:46:41]<feldmand>How is the documentation for Features? Better than for profiles?
[21:47:02]<EugenMayer1>rsync -a source dest always copies 100% of the files, no matter what
[21:47:24]<EugenMayer1>feldmand: hard to say. Profiles has no UI, features is a lot more UI
[21:48:01]<mig5>i have never noticed that, only the first time. the whole point of rsync is that it's incremental
[21:48:02]<EugenMayer1>so you need less docs. But i would rather say both are not to well documented. Compared to the docs of the aegir guys and girls in here, there is no docs for profiles / features :)
[21:48:25]<EugenMayer1>mig5: jap, but stuff like rsnapshot and derwish use --update
[21:48:29]<mig5>it only syncs what has changed, otherwise there'd be no use for it
[21:48:49]<mig5>and i have watched many a provision-verify --debug and seen it only syncchanges
[21:49:06]<mig5>anyway
[21:49:06]* skwashd has quit (Ping timeout: 252 seconds)
[21:49:18]<EugenMayer1>odd.
[21:49:19]<mig5>my main concern is if --update breaks the concept of the spoke model, documentation notwithstanding
[21:49:52]<mig5>i don't think i'd *ever* have used rsync if it did a *full* sync *every* time
[21:50:03]<mig5>i'd almost call it the most useless tool if that was the case
[21:50:11]<mig5>that's basically an scp.
[21:50:14]<EugenMayer1>well as you explained, it can if people modify the dest by hand. But that is against the spoke model by definition. i think the main difference is "enforce it" or not, or?
[21:50:20]<mig5>so i *must* be mistundestanding
[21:50:39]<EugenMayer1>mig5: just try it :)
[21:50:42]<mig5>dude
[21:50:48]<mig5>i've been running rsync for years :)
[21:50:51]<EugenMayer1>heh
[21:50:52]<feldmand>mig5: Thanks for the help.
[21:50:56]<EugenMayer1>i dont doubt that :)
[21:51:00]<mig5>i would not have touched it for more than 30 seconds if i found it did a full sync of *everything* every time.
[21:51:05]<EugenMayer1>i was really really confused seeing that either
[21:51:06]<mig5>we must be using different versions of rsync
[21:51:13]<feldmand>EugenMayer1: Thanks for the shake up and help.
[21:51:17]<EugenMayer1>yw
[21:52:05]<EugenMayer1>mig5: my test case : su -s /bin/bash aegir
[21:52:09]<EugenMayer1>cd ~/platforms
[21:52:29]<EugenMayer1>time rsync --stat someplaftoms someremote:dest
[21:52:50]<EugenMayer1>then do that one again. Look at the stats and the real transfered files - always 100%
[21:52:56]<mig5>and it really syncs every single file? replaces every file entirely even if it hasn't changed?
[21:53:05]<EugenMayer1>( actually i was doing this to benchmark compress..)
[21:53:34]<EugenMayer1>but then i was shoked by "WTF, all files? why does it take 40 seconds each time...there should be a diff of zero"
[21:53:48]<EugenMayer1>man rsync..."--update" is not defaul?!!?! ... tried ..worked.
[21:54:04]<EugenMayer1>so believe me iam the same way confused as you are mig5.
[21:54:30]<EugenMayer1>but i have tried it. aegir master is a debian lenny, remote debiena lenny or 10.04 / 10.10
[21:54:45]<EugenMayer1>and no, its not a frankendebin :D
[21:55:57]<EugenMayer1>Some my intention was not to find out about --update but rather see how compress benchmarks compares to plain ( and suprisinly --compress is 15% slower..)
[21:56:09]<EugenMayer1>eventhough the compress rate is 2/3
[21:56:33]<EugenMayer1>or, 0.3 rather :)
[21:57:44]<EugenMayer1>mig5: you maybe just try it with some platform and report back. Iam glad if you can teach me wrong, but i rather see you coming back shocked :)
[21:58:54]* timmillwood has joined #aegir
[21:58:57]* EugenMayer1 rsyncs himself into the office ..laters
[22:00:30]* sepehr1 has quit (Quit: Leaving.)
[22:00:47]* sepehr has joined #aegir
[22:01:34]* nomad411 has joined #aegir
[22:04:12]* EugenMayer1 has quit (Ping timeout: 276 seconds)
[22:09:55]* timmillwood has quit (Quit: timmillwood)
[22:11:59]* boztek has joined #aegir
[22:15:10]<omega8cc>mig5: EugenMayer1 is right, rsync will copy *everything* on every run if you don't use -u flag, see https://gist.github.com/838524
[22:17:09]* adrinux has joined #aegir
[22:23:12]* EugenMayer has joined #aegir
[22:23:29]* szczym has joined #aegir
[22:23:34]<EugenMayer>Hello. mig5 any updates? :)
[22:23:34]<hefring>EugenMayer: 3 hours 49 min ago <eft> tell EugenMayer that his recent revisions to http://community.aegirproject.org/node/71 might be better placed in a new page at same level in handbook - just a suggestion
[22:24:35]<omega8cc>EugenMayer: http://drupal.org/node/1068126#comment-4123870
[22:25:10]<EugenMayer>mig5: omega8cc: hehe mig, welcome to the new world order.
[22:25:24]<EugenMayer>Well iam also very very suprised that this not the default with rsync
[22:25:33]<EugenMayer>omega8cc: did you knew about that in general?
[22:25:34]<omega8cc>it isn't
[22:25:38]<omega8cc>yes
[22:25:53]<omega8cc>but I don't use remote stuff in Aegir yet
[22:26:09]<omega8cc>so I didn't realize it doesn't use -u
[22:26:46]<EugenMayer>well i guess iam currently really the only one testing that remote stuff in real production?
[22:26:54]<omega8cc>rsync by default just transfer files
[22:26:59]<mig5>i'm not convinced -u is a good idea per spoke model, but i'll ponder it a bit more
[22:27:05]<EugenMayer>that would actually explain why iam the only one complaining that much, e.g. over migrate :D
[22:27:05]<mig5>be interesting to see what anarcat says
[22:27:37]<EugenMayer>mig5: well verify takes ~4 minutes from root to root
[22:27:39]<mig5>reminds me of http://drupal.org/node/998484
[22:27:57]<EugenMayer>mig5: thats what iam working on
[22:29:03]<EugenMayer>mig5: if you have a second, http://drupal.org/node/1068114
[22:29:46]* jonhattan has quit (Quit: llevaré una chaquete de guns'n'roses)
[22:29:48]<EugenMayer>that is kind of very important for my work. When that API change is not going to be part of provision i have to rework stuff / write it differently now. So it would be great to know upfront if you consider it or not
[22:30:13]* skwashd has joined #aegir
[22:30:41]<EugenMayer>Basically the fr is aout passing that extra argument. The bigger part of the patch is adding missing docs about the nginx hooks and stuff to provision.api
[22:30:54]<mig5>what's normally in the $config attribute?
[22:31:04]<EugenMayer>that the chole config object
[22:31:04]<mig5>argument I mean
[22:31:07]<mig5>ok
[22:31:26]<omega8cc>mig5: we should use -u for all files except configuration (vhost + settings.php) and use force option for config files
[22:31:28]<EugenMayer>provisionConfig or more generalized ones
[22:31:44]<EugenMayer>omega8cc: thats what i have also though. Settings must be enforced
[22:31:57]<EugenMayer>vhost / settings / ssl certs and so fort
[22:32:07]<mig5>they all share the same function provision_sync_site() or something
[22:32:09]<mig5>so that will be tricky
[22:32:11]<mig5>re: -u
[22:32:16]<EugenMayer>omega8cc: but even then, if you change anything on the master, it will get rsynced because of the date omega8cc
[22:32:35]<mig5>yes, -a will preserve the timestamps. that's ok
[22:32:40]<omega8cc>and not just because someone could manually touch config on remote - also because servers can use different timezones
[22:32:46]<mig5>that *cannot* be a huge bandwidth cost :)
[22:32:55]<EugenMayer>so basically -u will enforce any changes. The only thing happens is the other way arround. somebody changes something on the remote and the master files will not get synced
[22:33:00]<EugenMayer>"again"
[22:33:27]<EugenMayer>my settings.php is 5mb! :)
[22:33:34]<mig5>gah
[22:33:34]<EugenMayer>i put my contrib module code in there. You dont?!
[22:33:49]<EugenMayer>;-)
[22:33:57]* timmillwood_ has joined #aegir
[22:33:57]<omega8cc>what?
[22:34:04]* timmillwood_ has quit (Excess Flood)
[22:34:09]<omega8cc>you are crazy
[22:34:18]<EugenMayer>Comeon, it was a joke!
[22:34:21]<mig5>:)
[22:34:24]<EugenMayer>Hell what are you thinking about me!
[22:34:31]<omega8cc>same
[22:34:33]<mig5>EugenMayer is lightening up, it's good
[22:34:36]<adrinux>lol
[22:34:39]* timmillwood_ has joined #aegir
[22:34:44]<EugenMayer>;-)
[22:35:16]<EugenMayer>mig5: http://community.aegirproject.org/node/71
[22:35:32]<EugenMayer>i have edited that one, added the docs about the hook ( was missing ).
[22:35:44]<omega8cc>so, we need to change our rsync logic and split it for files and config probably?
[22:35:55]<EugenMayer>i need to remove the $config part though, was just slipping through
[22:36:27]<EugenMayer>omega8cc: i think we need an interface
[22:36:30]* timmillwood_ is now known as timmillwood
[22:36:50]<EugenMayer>->forcePush() ->updatePush .. --compressedPush
[22:37:13]<EugenMayer>and use and strategy pattern. This will help a lot also with speeding up clone and migration stuf
[22:38:02]<EugenMayer>or decorators, i dont mind. But it would be very handy to decide what kind of push is done. When you start to work with remotes and a customer is doing inhouse-hosting you will remember that _)
[22:38:32]<omega8cc>btw, we should also stop doing things like tar and gzip all files just to move site between local platforms, we should use rysnc locally instead
[22:38:44]<EugenMayer>omega8cc: implemented it.
[22:38:51]<omega8cc>where
[22:39:00]<EugenMayer>its even much worse then that. We tar it, transfer it back to the aegir master
[22:39:05]<EugenMayer>then we untar it, and sync it back
[22:39:09]<EugenMayer>_that is cool_
[22:39:14]<EugenMayer>migrate-ng
[22:39:27]<EugenMayer>the issue here is the backup task we utilize
[22:39:43]<EugenMayer>that was what i told you with backup-slim and migrate-ng
[22:40:03]<omega8cc>where did you implement it?
[22:40:13]<EugenMayer>private repo right now.
[22:40:30]<omega8cc>are you in open source or not? share the stuff
[22:40:32]<EugenMayer>Yes
[22:40:40]<EugenMayer>It will be release
[22:40:49]<omega8cc>do it
[22:41:22]* snlnz has quit (Ping timeout: 255 seconds)
[22:42:13]<EugenMayer>i want to finish it first. backup-slim has no UI task yet. Cloning is not improved at all
[22:42:32]<EugenMayer>there is no clone-without-files task at all ( which is very important for quick stage trys )
[22:42:47]<EugenMayer>or setting up a customer on the laptop for dev.
[22:43:24]<EugenMayer>omega8cc: any feedback on vhost extra / site settings? I have them working in production now, works fine.
[22:43:44]<omega8cc>the work doesn't need to be finished before it can be shared, there is an alpha for this, share the stuff
[22:43:50]<EugenMayer>also that chain cert stuff is working, but i guess useless for you due you use nginx
[22:45:03]<omega8cc>yes, nginx doesn't use separate file for cert bundles
[22:45:30]<EugenMayer>useful.
[22:46:05]<omega8cc>I will try that anyway
[22:48:18]<EugenMayer>I also have added a lot more informations and output on errors to migrate
[22:48:28]<EugenMayer>what i miss in the current implemenation is "where".. which server
[22:48:29]* snlnz has joined #aegir
[22:48:52]<EugenMayer>"could not connect to database" .. hey cool. Where..from aegir master, the other remote or the source remote..?
[22:49:27]<EugenMayer>Remotes really introduce some piece of complexity to the model, but its greate we have them
[22:53:52]<mig5>a shame the people who wrote all that part don't work on aegir anymore
[22:53:54]<mig5>and wrote no docs
[22:54:21]<mig5>but those are drush_logs, should be easy to add % variables defining the hosts
[22:54:31]<mig5>*should* be :)
[22:55:29]<EugenMayer>sure its easy :)
[22:55:42]<EugenMayer>IF you know how to determine the current servers context
[22:55:46]<EugenMayer>and that is fecking not easy :)
[22:56:37]<EugenMayer>esp in the migration step / clone. Because you have 2... and hey .. what is d() right now ? :)
[22:56:58]<mig5>exactly
[22:57:01]<EugenMayer>and no, d()->service('http')->remote_host wont work
[22:57:21]<EugenMayer>and also not d($platform)->service(..)->remote_host
[22:57:47]* timmillwood has quit (Quit: timmillwood)
[22:57:50]<mig5>we have a lot of $this->server->remote_host
[22:57:51]<EugenMayer>d(d($platform)->config_data()['server'])
[22:57:58]<mig5>hanging off the server context
[22:58:02]<mig5>gah
[22:58:05]<mig5>:)
[22:58:11]<EugenMayer>funny or? Was funny for me
[22:58:15]<mig5>yes
[22:58:19]<EugenMayer>with my huge provision knowledege...
[22:58:26]<mig5>you know more than me how it works
[22:58:27]<mig5>true
[22:58:53]<EugenMayer>there are def. some bugs digged
[22:59:09]<EugenMayer>in the whole context part. It is not consistent
[22:59:19]<EugenMayer>What makes the code enormous hard to read
[22:59:51]<EugenMayer>d($platform)->server->remote_host is always "aegir" )
[23:00:05]<EugenMayer>($youraegirmasterhostname
[23:00:14]<mig5>strange
[23:00:16]<EugenMayer>no
[23:00:22]<EugenMayer>because it is the _local_ one :)
[23:00:29]* noecc has joined #aegir
[23:00:43]<EugenMayer>which will be synced to the remote later .. but that way platforms have those 2 contextes
[23:00:50]<EugenMayer>and yes it sounds wrong to me either
[23:01:09]<EugenMayer>Well you will see it, when i release the migrate-ng code
[23:01:25]<EugenMayer>i really way big PITA...esp without guidience
[23:05:52]<mig5>http://developmentseed.org/blog/2011/feb/22/open-atrium-and-managing-new...
[23:07:10]* volkan has joined #aegir
[23:08:01]* nomad411 has quit (Quit: Back to the unreal world...)
[23:12:00]* Artusamak is now known as Artusamak_afk
[23:13:53]<EugenMayer>mig5: to be expected
[23:13:59]<EugenMayer>since the news of drupalradar
[23:16:29]* boztek_ has joined #aegir
[23:19:09]* psynaptic|afk is now known as psynaptic
[23:19:10]* boztek has quit (Ping timeout: 272 seconds)
[23:22:42]* q0rban has joined #aegir
[23:22:58]* psynaptic has quit (Remote host closed the connection)
[23:24:17]<Vertice>and just like that. it's over
[23:24:20]<Vertice>god i hate NDA's
[23:27:26]<mig5>pff
[23:27:49]<mig5>well NDAs are meaningless when actions speak for themselves no? :)
[23:28:01]<mig5>it's a lot of dressed up hoohah :)
[23:28:12]<Vertice>yeah
[23:28:16]<mig5>'we're getting rid of this drupal crap' :)
[23:28:21]<mig5>i mean
[23:28:24]* psynaptic has joined #aegir
[23:28:26]<mig5>inroads, alignments, something something
[23:28:31]<mig5>:)
[23:28:37]<skwashd>Vertice: the big module ownership reassignment last week was a bit of a big give away
[23:29:02]<Vertice>yeah. but we couldnt re-assign open atrium and other
[23:29:02]<Vertice>s
[23:29:11]<Vertice>until the contracts got sorted out
[23:29:22]<Vertice>it really is the best of all possible outcomes
[23:29:50]<mig5>the best of what was inevitable
[23:29:51]<EugenMayer>Vertice: you are part of devseed or?
[23:29:52]<skwashd>i need to check some patches is for admin too :)
[23:29:59]<Vertice>yup
[23:30:21]<EugenMayer>We are following this pretty close because OA is somekind near to drupal-wiki.com
[23:30:25]<skwashd>EugenMayer: he is part of devseed 2.0
[23:30:51]<EugenMayer>and we really liked your drupal developments. To be more presized, those stuff is the only thing we use in Drupal. Very good, innovative and well written
[23:30:52]<mig5>i thought devseed 2.0 was phase2 tech :)
[23:30:53]<adrinux>the reboot ;)
[23:31:10]<Vertice>devseed 2.0 is a node.js company
[23:31:11]<EugenMayer>i hope those things dont stop with the move to phase2
[23:31:26]<mig5>i want djevseed 2.0
[23:31:34]<Vertice>phase 2 will actually maintain the stuff
[23:31:35]<EugenMayer>Vertice: are the devs of OA / MN moving to phase2 then?
[23:31:58]<Vertice>no. they are taking over maintainership
[23:32:01]<EugenMayer>damn.
[23:32:10]<Vertice>the devs of oa/mn are switching to node.js
[23:32:15]<Vertice>and aegir for that matter
[23:32:25]<Vertice>at least, i am
[23:32:31]<mig5>aegir is switching to node.js? :)
[23:32:37]<mig5>i joked about that today but it backfired, damn
[23:32:40]<skwashd>the yhanh visual style and consistency will be sadly missed
[23:32:44]<mig5>stupid youtube
[23:32:51]<EugenMayer>skwashd: signed.
[23:32:52]<Vertice>yeah/ have you guys seen tilemill
[23:32:57]<Vertice>sweet jesus, it is amazing
[23:33:21]<mig5>i'll be frank, i've no interest in mapping stuff
[23:33:25]<mig5>but it looks nice and clean
[23:33:43]<EugenMayer>Vertice: we will really miss you.
[23:33:55]<adrinux>how realistic is it to run Tilemill in a vm?
[23:33:57]<EugenMayer>Those things devseed create really took drupal forward.
[23:34:08]<skwashd>Vertice: i still have to explain to people why after 3 yrs devseed doesn't recognise kosovo's independence
[23:34:10]<EugenMayer>Much more then D7 or D8
[23:35:04]<EugenMayer>Vertice: what licence is nodejs?
[23:35:10]<Vertice>mit
[23:35:20]<Vertice>adrinux: easy
[23:35:37]<EugenMayer>Vertice: ah...thats were the storm comes from :)
[23:35:55]<Vertice>no. it comes from performance
[23:36:05]<Vertice>drupal (especially 7) isnt performant for the stuff we do
[23:36:12]<EugenMayer>Vertice: eventhough drupal is far not perfect
[23:36:15]<Vertice>we work with LARGE data
[23:36:24]<Vertice>large geo-tagged data
[23:36:35]<mig5>'night folks
[23:36:43]<Vertice>and trying to conform to drupal's node system and stick things in mysql became counter productive
[23:36:45]<EugenMayer>just switching to nodejs from sake of Perf. is kind of strange move for me. It is the "dev" the "toolhunter" way, no way buissiness decision
[23:36:47]<EugenMayer>mig5: nu8
[23:36:49]<skwashd>cya mig5
[23:37:02]<Vertice>devseed is a R&D team
[23:37:13]<EugenMayer>R&D?
[23:37:18]<Vertice>research and development
[23:37:25]<EugenMayer>ah. ok that fits.
[23:38:14]<skwashd>EugenMayer: at devseed the kids run the toy factory :)
[23:38:15]<EugenMayer>Vertice: is the main scale on nodejs as it is cloud by defintion?
[23:38:29]<Vertice>no
[23:38:31]<EugenMayer>every new visitor is adding himself to the cloud, the to client side JS.
[23:38:36]<Vertice>it's because it has asynchronous
[23:38:45]<Vertice>and has no blocking io
[23:39:00]* Artusamak_afk is now known as Artusamak
[23:39:02]<Vertice>it can handle orders of magnitude more requests than drupal
[23:39:16]<Vertice>and doesnt have to start up the whole process on every request
[23:39:33]<EugenMayer>yeah, thats an issue
[23:39:43]<Vertice>and doesnt need 10 million layers of caching
[23:39:46]<skwashd>EugenMayer: right now i'm fighting with drupal trying to do crazy joins due to limitations of CCK and users ... it takes 40mins+ to run an extract of 1.2mil records mapped to a subset of 2mil+ users
[23:39:47]<EugenMayer>over at Foswiki we created a Webserver kind of core
[23:39:59]<Vertice>also. it works very well with distributed key / data store like couch or mongo
[23:40:01]<EugenMayer>which then handled the request, was its "APC" cached and booted all the time.
[23:40:15]<Vertice>four kitchens did something similar
[23:40:31]<skwashd>Vertice: with what?
[23:40:37]<Vertice>the biggest issue with drupal is that it's written in php
[23:40:40]<Vertice>with drupal.
[23:40:47]<Vertice>ie: loads up and handles multiple requests
[23:40:54]<Vertice>but php isnt built for that kind of thing
[23:40:54]<EugenMayer>Well iam not sure were we are heading to right now. Iam pretty satisfied witht he custom multi-level caching i have implemented. Tht helps us scale right now
[23:40:58]<Vertice>memory leaks galore
[23:41:19]<skwashd>PHP - the VB of the web
[23:41:20]<EugenMayer>but we are arroudn 10k nodes (500k revisions) at max, so we are not at the millions yet skwashd
[23:41:30]<EugenMayer>skwashd: no, not that bad!
[23:41:34]<EugenMayer>lets say delphi :D
[23:41:53]<Vertice>it took us 64 hours to do our last import into a drupal 7 site
[23:41:56]<EugenMayer>and to be honest, its more drupals architectures fault, not PHP
[23:42:00]<skwashd>EugenMayer: if you look at it properly it is VB
[23:42:16]<Vertice>the model of bootstrapping the whole system on every request is broken
[23:42:24]<Vertice>and it forces you to do things like serialized strings in dbs
[23:42:28]<EugenMayer>when you addopt properly to OOP and the newer stuff like namespaces, you can see aht PHP can work and look quite prtty
[23:42:43]<Vertice>at it's core it still leaks memory like a sieve
[23:42:48]<EugenMayer>skwashd: iam working with php sind ~2.x
[23:42:51]<skwashd>EugenMayer: code doesn't have to look pretty to work
[23:43:10]<EugenMayer>and it developed a lot. It is drupal archticture at least 80%
[23:43:24]<EugenMayer>the other 20% is PHP compared to ruby or python
[23:44:01]<EugenMayer>if drupal would be better designed, that "bootstrap on every request" thing would have been handled much earlier and easier
[23:44:14]<skwashd>EugenMayer: i started with PHP3
[23:44:23]<EugenMayer>but we dont have those interfaces to wrap a service behind it. We have to kill the whole core code for it.
[23:44:30]<adrinux>drupal was designed?
[23:44:37]<EugenMayer>skwashd: dont take me wrong, i rather hate PHP then like it :)
[23:45:02]<skwashd>adrinux: druplicon is the most designed part of drupal
[23:45:15]<adrinux>lol
[23:45:16]<EugenMayer>skwashd: but "drupal people" tend to moan about PHP a lot, while its drupal architecture not even taking a small part into accoutn what PHP is capable of
[23:45:47]<EugenMayer>we are using PHP on the low-level part possible.
[23:45:58]<EugenMayer>Thanks to the "PHP4 compat" for years
[23:46:04]<EugenMayer>zero abstraction.
[23:46:08]<skwashd>EugenMayer: for me drupal is a tool which fits certain use cases
[23:46:19]<skwashd>I'm not a drupal is a hammer fanboi
[23:46:22]<EugenMayer>skwashd: for me drupal is a name ;-)
[23:46:48]<EugenMayer>Once i started to utilize Drupal because of the huge amount of contribs
[23:47:10]<EugenMayer>now i simply dont use any contribs except the main onse like cck / views and the stuff of devseed + aegir
[23:47:21]<Vertice>i never liked cck
[23:47:36]<EugenMayer>Its ok - not perfect
[23:47:55]<EugenMayer>Fits well with Views for that kind of abstraction. Could need a good redesign but iam fine with it
[23:48:21]<EugenMayer>compared to "taxonomy", or "comments", "files" in D6..
[23:48:28]<Vertice>it, along with most of drupal, has the main issue in that configuration stored in db tables should not be the primary way to do something
[23:48:38]<EugenMayer>or "private files" .. vvvvveeerry funny part.
[23:48:56]<EugenMayer>Vertice: true, true
[23:49:04]<Vertice>and then fields came along
[23:49:07]<Vertice>and took that even worse
[23:49:07]<EugenMayer>Features was a attempt...
[23:49:12]<Vertice>views is the only thing that does it right
[23:49:16]<Vertice>features was a bandaid
[23:49:20]<Vertice>it shouldnt have been necessary
[23:49:26]<EugenMayer>jap
[23:49:35]<EugenMayer>and its ill inside because its a bandaid
[23:49:56]<Vertice>and fields in d7
[23:50:00]<Vertice>*shivers*
[23:50:02]<EugenMayer>never used D7 yet.
[23:50:10]<EugenMayer>i like the streamwrappers
[23:50:17]<Vertice>just you wait 'enry'iggins , just you wait
[23:50:33]<EugenMayer>i like entities as a general thing, i like that fields can work on every entity
[23:50:54]<EugenMayer>All those things which are set in other OOP frameworks fro decades
[23:51:03]<Vertice>except the entire thing is built on cahing
[23:51:06]<skwashd>EugenMayer: no ... they on;y work on the entities the developer decides to make fieldable
[23:51:19]<Vertice>the only way to load the field content is from a cache
[23:51:24]<EugenMayer>skwashd: thats fine.
[23:51:28]<Vertice>and each extra field is 2 joins
[23:51:34]<skwashd>for example a file is an entity in D7 but it isn't fieldable
[23:51:51]<Vertice>but people will find out about this all soon enough
[23:51:58]<skwashd>Vertice: use mongodb - problem solved :)
[23:52:21]<EugenMayer>Vertice: are your designer rentable? :)
[23:52:25]<EugenMayer>*s
[23:52:34]<Vertice>send a mail to eric@developmentseed.org
[23:52:35]<Vertice>and ask
[23:52:43]<Vertice>we have a designer that may be available
[23:52:47]<Vertice>but not young
[23:52:52]<Vertice>young is our designer
[23:52:58]<Vertice>who ends up writing most of our code
[23:53:06]<skwashd>Vertice: "devseed rent boys"
[23:53:11]<EugenMayer>like searchlight
[23:53:48]<EugenMayer>which i like but still stuck to apachesolr-crappynismus
[23:54:13]<EugenMayer>"hey lets index nodes as guest, thats awesomness"
[23:54:30]<EugenMayer>"you cant leak undwanted informations that way".
[23:54:49]<EugenMayer>Yeah, def. because on private sites, you cant leak informations. Search is 100% zero empty.
[23:55:15]* q0rban has quit (Remote host closed the connection)
[23:55:41]* q0rban has joined #aegir
[23:57:08]<EugenMayer>Will miss you devseed guys :/
[23:57:40]<EugenMayer>What about you swapping places with Acquia and they switch over to node.js? :)
[23:58:51]* NeoID has quit (Quit: Forlater kanalen)
[23:59:26]<EugenMayer>anybody in here offers drupal commons on aegir? omega8cc?
[23:59:42]<EugenMayer>just currious if anybody who knows OA ever would consider this
[00:01:36]<skwashd>EugenMayer: OA is more internal, Commons is more public facing
[00:01:55]<EugenMayer>skwashd: i doubt that :)
[00:02:21]<skwashd>in terms of approach they are
[00:03:26]<EugenMayer>community.aegirproject.org?
[00:03:34]<EugenMayer>Just in case of examples are needed
[00:03:43]* talengix has joined #aegir
[00:04:13]<EugenMayer>skwashd: i undestand what you are trying to say, but in term of QA its just a configuration or in the largest cases a small extension
[00:05:13]<EugenMayer>skwashd: i have no slighly idea why Acquia did this. Its like a personal flank on dev-seed. They could have invested into QA and have a similar system right now, being one gazzilion better
[00:05:45]<EugenMayer>skwashd: well, but its to everybody else to think about this :)
[00:07:59]* j0nathan has quit (Quit: ¡ Hasta luego !)
[00:08:00]<skwashd>EugenMayer: think of drupal.org as debian, acquia as ubuntu ... and devseed as arch or LFS
[00:08:15]<EugenMayer>comeon :D
[00:08:16]<omega8cc>Vertice: pm?
[00:08:36]<EugenMayer>arch is so much similar to drupal
[00:09:01]<EugenMayer>i mean, devseed stuff is useable OOTB. Nice interface, nice UI, nice concept
[00:09:09]<EugenMayer>that is rather ubuntu then arch :)
[00:10:31]<Vertice>?
[00:10:59]<Vertice>not having to worry about acquia anymore is nice
[00:11:40]<EugenMayer>Vertice: well yes. The way they help the community is the way that we are not releasing a single module :)
[00:12:00]<omega8cc>Vertice: so you are leaving Aegir or rewriting it to node.js ? you are a father :)
[00:12:36]<Vertice>well. i'm still around in a advisory capacity , as time permits
[00:12:47]<Vertice>but the truth is i havent needed aegir myself for many years
[00:13:09]<Vertice>and one of the goals has always been that it would be able to live on without me
[00:13:24]<Vertice>so i will officially step down at some point in the future
[00:13:33]<Vertice>the project is in really good hands though
[00:14:04]<EugenMayer>Args
[00:14:20]<Vertice>truth is i havent really been around much in months
[00:14:37]<Vertice>but we will make a proper announcement when the time comes
[00:16:37]<omega8cc>Vertice: child grows up and builds his own home, but he is always a child, and you are always a father, so I hope you will never step down completely :)
[00:16:39]<EugenMayer> help documenting provision before you leave :)
[00:16:59]<Vertice>oh. i'll be around
[00:17:01]<Vertice>dont you worry
[00:17:01]<Vertice>=)
[00:17:13]<omega8cc>oh, thanks :)
[00:17:23]<Vertice>EugenMayer: that is one of the things on my todo list, but i've been so fscking busy
[00:17:32]* talengix has quit (Read error: Connection reset by peer)
[00:20:25]<EugenMayer>Vertice: trust me and trust your inner voice
[00:20:29]<EugenMayer>will never change! :)
[00:23:50]<EugenMayer>ssl_error_rx_record_too_long
[00:23:52]* boztek_ has quit (Ping timeout: 240 seconds)
[00:24:18]<EugenMayer>getting this on a new remote after deploying a SSL cert ( the other was working ) key / crt have the same modulus. Any ideas?
[00:27:34]* mattmcmanus has joined #aegir
[00:28:40]* boztek has joined #aegir
[00:32:30]<anarcat>so the cat is out
[00:33:28]<anarcat>sdboyer: what do those warnings mean?
[00:34:20]<anarcat>sdboyer: oh nevermind, i see you can take care of this
[00:34:35]<lukus>Vertice, is tilemill a GIS authoring platform?
[00:34:43]<Vertice>yes
[00:34:48]<Vertice>part of one
[00:35:23]<Vertice>http://www.govloop.com/profiles/blogs/tilemill-realizing-the
[00:35:38]<lukus>ah .. so it would be possible to run it locally? (or even desirable?)
[00:37:24]<lukus>looks really interesting .. I've been working with raphael.js to produce custom maps lately
[00:38:10]<lukus>partly because it's difficult to get aesthetically pleasing results using simple mapserver implementations
[00:40:04]<lukus>its a shame devseed won't be drupal anymore - need a new posterchild .. but the mapping stuff does look interesting and worthwhile .. 2GB requirement seems quite heavy tho
[00:41:31]* talengix has joined #aegir
[00:43:28]* mattmcmanus has quit (Ping timeout: 240 seconds)
[00:46:23]* NeoID has joined #aegir
[00:46:48]<talengix>would somebody be kind enough to help me with this? when I try to run the command php /var/aegir/drush/drush.php dl drupal-6.19 I get this error Could not open input file: /var/aegir/drush/drush.php
[00:47:17]* skwashd has quit (Ping timeout: 250 seconds)
[00:48:26]* mattmcmanus has joined #aegir
[00:54:28]<omega8cc>talengix: maybe su - aegir first and then type: drush dl drupal
[00:55:10]<omega8cc>lunchtime, bbl
[00:57:35]* Irishgringo has joined #aegir
[00:58:45]<talengix>it said success with drush dl drupal
[00:58:58]<talengix>bon apetite ;o)
[01:17:44]* talengix has quit (Remote host closed the connection)
[01:18:29]* NeoID has quit (Quit: Forlater kanalen)
[01:28:31]* talengix has joined #aegir
[01:33:05]<EugenMayer> anarcat: http://community.aegirproject.org/node/360 .. i dont think it is a good idea to release any stable out of 0.4 :/
[01:36:19]* EclipseGc has joined #aegir
[01:38:48]* scientist has joined #aegir
[01:40:39]<anarcat>EugenMayer: why is that?
[01:49:02]* ergonlogic has joined #aegir
[01:51:44]* j0nathan has joined #aegir
[01:51:47]* EugenMayer has quit (Quit: Leaving.)
[01:55:49]* totten has joined #aegir
[01:59:45]* kvanderw has joined #aegir
[02:02:04]* gezick has quit (Quit: Leaving.)
[02:04:39]* FatGuyLaughing has joined #aegir
[02:05:54]* JoshBenner has joined #aegir
[02:06:36]* Artusamak is now known as Artusamak_afk
[02:09:55]* Artusamak_afk is now known as Artusamak
[02:15:58]* dipen has quit (Ping timeout: 268 seconds)
[02:17:04]* talengix has quit (Quit: Leaving)
[02:18:19]* jonhattan has joined #aegir
[02:19:14]* ghankstef has joined #aegir
[02:23:28]* TrevorT has quit (Ping timeout: 272 seconds)
[02:27:29]* shrop has left #aegir ()
[02:27:51]* shrop has joined #aegir
[02:27:53]* TrevorT has joined #aegir
[02:42:13]* feldmand has quit (Ping timeout: 272 seconds)
[02:46:05]* recidive has joined #aegir
[02:51:28]* boztek has quit (Ping timeout: 246 seconds)
[02:51:33]* dipen has joined #aegir
[02:58:37]* eft has joined #aegir
[03:02:16]* NeoID has joined #aegir
[03:08:37]* psynaptic is now known as psynaptic|break
[03:09:03]* mikl has quit (Quit: Computer has gone to sleep.)
[03:11:04]* V1ntage_work has quit (Ping timeout: 240 seconds)
[03:11:23]* obrienmd has joined #aegir
[03:12:47]* V1ntage_work has joined #aegir
[03:16:37]* Slydder1 has joined #aegir
[03:21:04]* V1ntage_work has quit (Ping timeout: 240 seconds)
[03:22:52]* V1ntage_work has joined #aegir
[03:24:34]* milk__ has joined #aegir
[03:24:46]* V1ntage_work has quit (Client Quit)
[03:30:33]* dipen has quit (Ping timeout: 246 seconds)
[03:34:40]* FatGuyLaughing has quit (Ping timeout: 240 seconds)
[03:35:38]* jhedstrom has joined #aegir
[03:36:17]* milk__ is now known as milk
[03:36:54]* ngnp has quit (Quit: ngnp)
[03:40:44]* FatGuyLaughing has joined #aegir
[03:43:10]* psynaptic|break is now known as psynaptic
[03:43:35]* dipen has joined #aegir
[03:52:23]* NeoID has quit (Read error: Connection reset by peer)
[03:52:51]* NeoID has joined #aegir
[03:58:56]* kostajh has quit (Read error: Connection reset by peer)
[04:00:54]* Brandonian has joined #aegir
[04:02:22]* adrinux wonders if $id$ will finally be removed from aegir source files with the move back to d.org
[04:02:40]<adrinux>siliconmeadow's fault
[04:05:24]* redShadow[2V] has quit (Remote host closed the connection)
[04:06:33]* kvanderw has quit (Quit: kvanderw)
[04:06:44]* siliconmeadow blames it on the naughty ones in #drupaluk
[04:13:09]* volkan has quit (Quit: volkan)
[04:13:33]* ryanarmstrong has joined #aegir
[04:17:30]* ergonlogic has quit (Ping timeout: 240 seconds)
[04:25:02]<ryanarmstrong>Has anyone here had issues where the migrate command doesn't work, with it saying that "The external command could not be executed due to an application error." I've searched the web for others having this issue and they all said that upping the execution time and memory limit solved their problem. But my server is running a 256M limit and a 120 execution time. Is that not enough?
[04:25:43]<adrinux>might not be
[04:25:49]<adrinux>ra
[04:26:07]<adrinux>ryanarmstrong: have you run the command with —debug ?
[04:26:51]<ryanarmstrong>No I haven't. I'm just doing all of this within the Aegir backend
[04:26:59]<ryanarmstrong>I wasn't aware there was a debug option
[04:27:56]<ryanarmstrong>I'm assuming I'd have to run that from the cmd line?
[04:29:12]* NeoID has quit (Quit: Forlater kanalen)
[04:29:42]<adrinux>backend, yeah, just add the —debug flag then
[04:29:54]<adrinux>hopefully will give you more error info
[04:30:20]<ryanarmstrong>ok, I've never run any of the Aegir commands from the cmd line.
[04:30:43]* siliconmeadow has quit (Ping timeout: 250 seconds)
[04:30:58]* kvanderw has joined #aegir
[04:31:06]<ryanarmstrong>adrinux: is there a faq or readme that goes over those commands?
[04:31:25]<ryanarmstrong>whoops
[04:31:28]<ryanarmstrong>I think I found it
[04:31:31]* psynaptic is now known as psynaptic|food
[04:31:31]* obrienmd has quit (Read error: Connection reset by peer)
[04:35:03]* mikl has joined #aegir
[04:35:03]* mikl has quit (Changing host)
[04:35:03]* mikl has joined #aegir
[04:36:16]* ergonlogic has joined #aegir
[04:38:43]* obrienmd has joined #aegir
[04:41:35]* NeoID has joined #aegir
[04:41:51]* nicholasalipaz has joined #aegir
[04:42:54]<omega8cc>obrienmd: any chance you could post your ssl how-to on c.a.o?
[04:44:45]* nicholasalipaz has quit (Remote host closed the connection)
[04:45:46]* adrinux has quit (Quit: Leaving.)
[04:46:09]* omega8cc has quit (Remote host closed the connection)
[04:46:16]* nicholasalipaz has joined #aegir
[04:46:17]* omega8cc has joined #aegir
[04:46:45]<obrienmd>omega8cc: buried in work today, will try to get it up by the end of the weekend
[04:46:45]<hefring>obrienmd: 22 hours 32 min ago <omega8cc> tell obrienmd: could you post your SSL related how-to in the c.a.o handbook and link it in response to this comment? https://github.com/omega8cc/nginx-for-drupal/issues/8#comment_792144 Thanks in advance!
[04:47:25]<omega8cc>obrienmd: ok, thanks!
[04:47:44]* jackinloadup has joined #aegir
[04:48:07]<obrienmd>oemga8cc: although, to be fair, I don't know how different it will be from your first comment https://github.com/omega8cc/nginx-for-drupal/issues/8
[04:48:23]<obrienmd>s/oemga8cc/omega8cc/
[04:51:20]<omega8cc>obrienmd: it is always easier to understand when it is explained by some native speaker, plus, I know this stuff too well probably and then I use 'shortcuts' others find hard to follow etc, so I will really appreciate your how-to taken from your own experience and posted in the docs
[04:52:52]<obrienmd>omega8cc: will do!
[04:53:12]<omega8cc>obrienmd: thanks a lot!
[04:53:56]<obrienmd>omega8cc: by the way, as we only use custom platforms and really want to reduce complexity, we're thinking about transitioning back to barracuda only, and bringing all the nice global.inc (cache, etc) stuff over from octopus, but not deploying an real octopus intances... does this make sense?
[04:58:04]<omega8cc>obrienmd: I recommend to not use hostmaster in /var/aegir (unless you are migrating from old system using it), as it is responsible for managing main Nginx config and all hostmasters created by Octopus are symlinked/included there, so maybe another switch allowing you to skip installing default platforms at all, when installing Octopus hostmaster? I think I could introduce it easily.
[05:01:24]<omega8cc>obrienmd: also, the hostmaster created by Barracuda is by design left in the almost vanilla version (just some blocks hidden in the UI) to keep it compatible with any older/migrated installations, but it using the same global.inc
[05:02:52]<obrienmd>omega8cc: ah, let me think about this a bit
[05:03:05]<obrienmd>omega8cc: for simplicity's sake, "almost vanilla version" doesn
[05:03:11]<obrienmd>'t sound so bad :)
[05:04:44]<omega8cc>obrienmd: the idea behind Octopus is also to allow you move sites to the next Octopus "node" on the same system if you will break your hostmaster for any reason, and if you will move this to the Barracuda hostmaster, then you have no rollback option left
[05:05:32]* TrevorT has quit (Ping timeout: 268 seconds)
[05:07:09]<omega8cc>obrienmd: and it is also to add another layer of privileges separation, so you can safely restrict access per hostmaster instance
[05:09:01]<obrienmd>omega8cc: got it
[05:09:30]<obrienmd>omega8cc: I think given those things, that barracuda vanilla is probably best suited for my environment
[05:10:38]<obrienmd>omega8cc: all warnings appreciated and understood, though :)
[05:11:20]<omega8cc>obrienmd: then you can add the contrib modules you found useful in Octopus and it will use the same global.inc anyway
[05:13:04]* Irishgringo has quit (Read error: Connection reset by peer)
[05:14:47]<obrienmd>omega8cc: right, I certainly understand that, and octopus is an awesome system, it just seems like another layer of abstraction that we don't need given our processes... I think it's a good fit for 90% of situations though
[05:15:12]<obrienmd>omega8cc: what barracuda gives us is a solid nginx/aegir base to build from
[05:15:43]<omega8cc>obrienmd: Octopus is using custom feature to enable and tune some Aegir modules/settings and UI for new users to make the first experience easier, while Barracuda gives you Aegir as-is, with default factory settings
[05:16:30]<obrienmd>omega8cc: understood
[05:16:48]<omega8cc>so it is better when you want to adjust things for your needs
[05:17:03]* AquaticDisorder has joined #aegir
[05:17:04]<obrienmd>omega8cc: which we appreciate :)
[05:18:40]* EclipseGc has quit (Ping timeout: 240 seconds)
[05:18:49]<omega8cc>obrienmd: thanks :)
[05:21:41]* EclipseGc has joined #aegir
[05:25:36]* josh_k has joined #aegir
[05:25:36]* josh_k has quit (Changing host)
[05:25:36]* josh_k has joined #aegir
[05:26:16]* eft_ has joined #aegir
[05:29:28]* scientist has quit (Ping timeout: 240 seconds)
[05:32:49]* sepehr has quit (Ping timeout: 246 seconds)
[05:37:18]* sepehr has joined #aegir
[05:41:20]* milk has quit (Ping timeout: 276 seconds)
[05:42:03]* milk has joined #aegir
[05:46:47]* milk has quit (Read error: Connection reset by peer)
[05:56:41]* scientist has joined #aegir
[05:57:14]* NeoID has quit (Quit: Forlater kanalen)
[06:02:30]* mlncn has quit (Remote host closed the connection)
[06:11:07]* FatGuyLaughing has quit (Read error: Connection reset by peer)
[06:11:57]* gusaus has joined #aegir
[06:13:51]* FatGuyLaughing_ has joined #aegir
[06:15:01]* LloydPearsonIV has quit (Quit: is going somewhere he can swear and not get booted)
[06:20:35]* NeoID has joined #aegir
[06:25:02]* omega8cc_ has joined #aegir
[06:28:16]* omega8cc has quit (Ping timeout: 240 seconds)
[06:28:16]* omega8cc_ is now known as omega8cc
[06:29:40]* sepehr has quit (Ping timeout: 240 seconds)
[06:29:46]* mlncn has joined #aegir
[06:39:37]<szczym>folks, i know its off topic but i really cant stop my self from passing you the link: http://www.agileapproach.com/blog-entry/open-atrium-and-managing-news-fi...
[06:40:46]* TrevorT has joined #aegir
[06:43:58]<omega8cc>szczym: it is hot since morning
[06:44:07]<szczym>i hope hope hope oa and mn will be not bloated like op is now ... and they will not put the "be awesome" banners
[06:44:53]* noecc has left #aegir ("pax")
[06:45:03]<szczym>ohh, im so late, just looked on our mn
[06:45:06]<omega8cc>szczym: I already wrote to Eric that I will kill Phase2 if they will try to screw that up
[06:45:20]<omega8cc>lol
[06:45:27]<szczym>;) good move !
[06:45:27]<omega8cc>see http://twitter.com/#!/drupalfacts/status/40107646547791872
[06:45:37]* joestewart|afk is now known as joestewart
[06:48:01]<szczym>and i hope my personal pet (mn) will be not such depressing, like the damen tattler was
[06:49:16]<joestewart>szczym: I wrote the same thing in an email here earlier. There was more discussion earlier. mostly about DS move to node.js
[06:53:52]* szczym is investigating the node.js and propably will get sadlly drunk today, also becouse we faild 2nd time with upgrade of our custom oa from 1 year old code ...
[06:58:40]* omega8cc has quit (Ping timeout: 240 seconds)
[07:00:00]* omega8cc has joined #aegir
[07:02:54]<ryanarmstrong>joestewart: DS?
[07:03:09]<obrienmd>development seed
[07:03:18]<ryanarmstrong>Ah ok
[07:03:20]<ryanarmstrong>Thanks
[07:03:58]* catatonicrelapse has quit (Ping timeout: 240 seconds)
[07:09:44]* noecc has joined #aegir
[07:13:50]* scientist has quit (Ping timeout: 240 seconds)
[07:17:20]* j0nathan has quit (Remote host closed the connection)
[07:19:10]* j0nathan has joined #aegir
[07:19:46]* adrinux has joined #aegir
[07:27:10]<eft_>adrinux: hey
[07:27:14]* catatonicrelapse has joined #aegir
[07:27:19]<adrinux>hello
[07:27:19]<hefring>hey
[07:27:44]<adrinux>you. the one that clutters up my issue queue ;)
[07:29:44]<eft_>just giving your project some seo love, baby ;)
[07:31:55]<ryanarmstrong>lol
[07:34:14]<adrinux>glad you got it working easily
[07:36:43]<eft_>adrinux: suggest you rename http://drupal.org/node/817394 to "check for whether imagemagick imageAPI is enabled" and close http://drupal.org/node/1063594
[07:37:47]<adrinux>the latter one should close itself after two weeks
[07:38:11]<adrinux>it's marked 'fixed', what more do you want?
[07:38:15]<adrinux>:)
[07:38:19]* talengix has joined #aegir
[07:38:22]<adrinux>you're right about the first one though
[07:40:15]<adrinux>retitled them both
[07:41:49]* EclipseGc has quit (*.net *.split)
[07:41:49]* ergonlogic has quit (*.net *.split)
[07:41:49]* recidive has quit (*.net *.split)
[07:41:50]* dfrt has quit (*.net *.split)
[07:41:51]* Artusamak has quit (*.net *.split)
[07:41:51]* mig5 has quit (*.net *.split)
[07:45:31]* EclipseGc has joined #aegir
[07:53:23]<eft_>nice - they throw flags for adopters, even if the fault is not that of the maintainter :)
[07:57:51]<eft_>adrinux: here's an example http://oi53.tinypic.com/10xa3kk.jpg
[08:00:03]<eft_>the flipped image is behind the title text - it may be hard to see depending on your monitor brightness - still fiddling with the opacity
[08:00:11]* ergonlogic has joined #aegir
[08:00:12]* recidive has joined #aegir
[08:00:12]* dfrt has joined #aegir
[08:00:12]* Artusamak has joined #aegir
[08:00:12]* mig5 has joined #aegir
[08:00:16]* szczym has quit (Ping timeout: 240 seconds)
[08:00:18]* grugnog has quit (Excess Flood)
[08:00:26]<eft_>point is that all I had to do was install your module and add -flip as an im_raw preset
[08:00:37]* grugnog has joined #aegir
[08:01:45]<adrinux>oh yep, I see it at the bottom of my monitor but not at the top
[08:01:56]* jackinloadup has quit (Remote host closed the connection)
[08:04:06]<eft_>also, you should change the project name to im_adrinux_here_me_raw :)
[08:04:38]<eft_>FWIW s/here/hear
[08:05:00]<adrinux>heh
[08:05:18]* snlnz has quit (*.net *.split)
[08:05:18]* Aciid has quit (*.net *.split)
[08:05:18]* mariooo has quit (*.net *.split)
[08:05:18]* grugnog_office has quit (*.net *.split)
[08:05:19]* catatonicrelapse has quit (*.net *.split)
[08:05:19]* NeoID has quit (*.net *.split)
[08:05:19]* mweltin has quit (*.net *.split)
[08:05:20]* TrevorT has quit (*.net *.split)
[08:05:20]* eft has quit (*.net *.split)
[08:05:20]* scyrma has quit (*.net *.split)
[08:05:22]* basic` has quit (*.net *.split)
[08:05:22]* FatGuyLaughing_ has quit (*.net *.split)
[08:05:22]* univate has quit (*.net *.split)
[08:05:23]* Letharion has quit (*.net *.split)
[08:05:23]* Zelfje has quit (*.net *.split)
[08:05:23]* eft_ has quit (*.net *.split)
[08:05:23]* mvc has quit (*.net *.split)
[08:05:24]* CitizenKane has quit (*.net *.split)
[08:05:26]* mariooo has joined #aegir
[08:05:27]* Aciid has joined #aegir
[08:05:50]* grugnog_office has joined #aegir
[08:06:26]* catatonicrelapse has joined #aegir
[08:06:26]* TrevorT has joined #aegir
[08:06:26]* NeoID has joined #aegir
[08:06:26]* eft has joined #aegir
[08:06:26]* mweltin has joined #aegir
[08:06:26]* scyrma has joined #aegir
[08:06:26]* basic` has joined #aegir
[08:06:54]* eft_ has joined #aegir
[08:06:54]* mvc has joined #aegir
[08:06:54]* CitizenKane has joined #aegir
[08:07:43]* FatGuyLaughing_ has joined #aegir
[08:07:44]* univate has joined #aegir
[08:07:44]* Letharion has joined #aegir
[08:07:44]* Zelfje has joined #aegir
[08:07:58]* eft has quit (Ping timeout: 266 seconds)
[08:14:14]* EugenMayer has joined #aegir
[08:14:35]<EugenMayer>Hello
[08:14:35]<hefring>salut
[08:15:16]* EugenMayer has quit (Client Quit)
[08:21:37]* obrienmd has left #aegir ()
[08:22:17]* snlnz has joined #aegir
[08:23:25]* EugenMayer has joined #aegir
[08:24:49]* griz has joined #aegir
[08:24:58]<griz>hello
[08:24:58]<hefring>que tal
[08:25:38]<griz>hello?
[08:25:38]<hefring>hello
[08:26:05]<EugenMayer>Hi
[08:26:16]<griz>I have a problem... I've been trying to figure it out but I've never had anything quite like it before.
[08:26:37]<griz>It seems like something minor but it's seriously getting in my way.
[08:27:11]<griz>I've installed a couple of sites, no problems. Now with the most recent site images are behaving strangely.
[08:27:46]<griz>I tried installing a new platform ('drush dl drupal' as aegir user) and then a new site to test and its doing the same thing
[08:28:33]<griz>Images won't show up. If I hover over the img tag in firebug the image is displayed, and when I use the URL it works, but the images won't show up in the page.
[08:29:00]<griz>any ideas anybody>
[08:29:01]<griz>?
[08:29:20]* talengix has quit (Read error: Operation timed out)
[08:29:47]<griz>I don't think its permissions - I've: chmod -R aegir:www-data files
[08:29:56]<griz>i mean chown
[08:29:57]* talengix has joined #aegir
[08:30:15]<griz>and chmod 2775 -R files
[08:31:06]* scientist has joined #aegir
[08:31:06]<griz>images work on my other sites and I can't find any difference between them
[08:32:41]<griz>new files and folders that get uploaded through node/add are owned by www-data - is that normal?
[08:33:29]<griz>I've looked at the file structure page in the handbook and all seems well
[08:34:42]* talengix has quit (Ping timeout: 246 seconds)
[08:35:18]* adrinux has left #aegir ()
[08:38:24]<griz>also, I get these wierd classes appended to my images: "piiqgcarokgbupzobwou"
[08:38:25]<griz>anyone seen anything like that before?
[08:38:25]<griz>this is on a clean install
[08:39:41]* ergonlogic has quit (Ping timeout: 260 seconds)
[08:40:35]<griz>is there anyone here who worlks on aegir? if so can I suggest that the folder permissions in the handbook are given in the form of numbers as well?
[08:41:41]* FatGuyLaughing_ is now known as FatGuyLaughing
[08:42:20]<griz>I don't think its perm.s though, I've set them to aegir:www-data on the files and I'm still getting the same thing.
[08:42:49]<omega8cc>griz: it is for imagecache?
[08:43:09]<griz>yep (thankyou for the response)
[08:43:31]<omega8cc>griz: you are using apache or nginx?
[08:43:39]<griz>apache2
[08:50:24]<omega8cc>griz: it is for both imported and new sites? on all sites or just one platform?
[08:50:24]<griz>oh, and the only way I can get images to display is to set the upload path to the files directory - if I use a subdirectory they don't show.
[08:50:24]<griz>all recent sites - that is, It happened today so I set up a new platform to test... same thing.
[08:50:24]<omega8cc>any images are expected to live in the sites/domain/files
[08:50:24]<omega8cc>not sure what you mean by subdir
[08:50:25]<anarcat>sdboyer: you have another git rebel to help out - i hope we didn't open a too big can of worms here :) http://groups.drupal.org/node/128479#comment-419269
[08:50:25]<griz>in the field settings you can choose a subdir to upload files to. it works on other sites.
[08:50:25]<omega8cc>griz: you should use path to files you can find in the settings.php file, it is forced by Aegir anyway, so maybe you need to purge caches
[08:50:25]* joestewart|afk has joined #aegir
[08:50:26]* Aciid has quit (*.net *.split)
[08:50:26]* j0nathan has quit (*.net *.split)
[08:50:26]* nicholasalipaz has quit (*.net *.split)
[08:50:26]* Slydder1 has quit (*.net *.split)
[08:50:26]* q0rban has quit (*.net *.split)
[08:50:26]* SeanBannister has quit (*.net *.split)
[08:50:27]* darthsteven has quit (*.net *.split)
[08:50:27]* recidive has quit (*.net *.split)
[08:50:28]* dfrt has quit (*.net *.split)
[08:50:28]* Artusamak has quit (*.net *.split)
[08:50:28]* mig5 has quit (*.net *.split)
[08:50:29]* EugenMayer has quit (Ping timeout: 276 seconds)
[08:50:29]* joestewart has quit (Ping timeout: 276 seconds)
[08:50:29]* EugenMayer has joined #aegir
[08:50:30]<omega8cc>griz: if it works on other sites, then the problem is not related to Aegir, but to your sites settings, probably
[08:50:32]* Aciid has joined #aegir
[08:50:32]* mig5 has joined #aegir
[08:50:32]* Artusamak has joined #aegir
[08:50:32]* dfrt has joined #aegir
[08:50:32]* recidive has joined #aegir
[08:50:32]* j0nathan has joined #aegir
[08:50:32]* nicholasalipaz has joined #aegir
[08:50:32]* Slydder1 has joined #aegir
[08:50:32]* q0rban has joined #aegir
[08:50:32]* SeanBannister has joined #aegir
[08:50:32]* darthsteven has joined #aegir
[08:50:46]<EugenMayer>14minutes for migrating a normal platforms with a 80MB DB and local file copy. Man thats long :/
[08:51:35]<griz>Hmmm. I've used aegir to install sites and imagecache works... I now use aegir to install sites in the same eay, yet imagecache doesn't work.
[08:51:35]<EugenMayer>griz: i can confirm that aegir works with imagecache
[08:52:05]<griz>eugen: so can I! :) I don't know what's up.
[08:52:13]<griz>test.brokendesign.net
[08:53:00]<griz>if inspect the field-item in firebug, it's there
[08:53:16]<omega8cc>griz: you need to analyze the difference between settings and/or permissions on your sites, I doubt it is related to Aegir
[08:53:44]* dipen has quit (Ping timeout: 276 seconds)
[08:53:54]<omega8cc>with such domain it is expected to be broken ;)
[08:54:00]<griz>would it be permissions that would cause that?
[08:54:28]<griz>yeah, it's meant to be a name with cahones, but it doesn't really work if it is >actually< broken :(
[08:54:28]<EugenMayer>how would it. You wrote you can access the file calling the direct URL
[08:54:29]<EugenMayer>..
[08:54:29]<griz>lol
[08:54:32]* skwashd has joined #aegir
[08:54:43]<griz>eugen: that's what I thought
[08:55:21]<omega8cc>griz: I don't see any broken images there?
[08:56:06]<griz>the top post, where it says 'mung mung mung...' there is an image there, it's just not being displayed.
[08:56:12]<griz>firebug it.
[08:57:05]<omega8cc>no, there is an image
[08:57:21]<griz>?? excuse me while I open chrome...
[08:57:31]<omega8cc>displayed, at least I see that (one image)
[08:57:52]<griz>oooooo, now I'm angry.
[08:57:57]<griz>thanks for your time everynody.
[08:58:09]<griz>angry, yet relieved.
[08:58:56]<griz>wierd thing... the preview images don't show up when I upload a file. Never mind, obviously not something to do with aegir.
[08:59:06]<griz>PBKAC...
[09:02:08]* omega8cc_ has joined #aegir
[09:02:09]* psynaptic|food is now known as psynaptic
[09:02:37]<griz>well, thanks again omega and eugen.
[09:02:57]* catatonicrelapse has quit (Ping timeout: 268 seconds)
[09:03:24]* Brandonian has quit (Quit: Brandonian)
[09:03:40]<griz>I can't believe I spent ages checking perms etc, searching the issue queue.
[09:03:49]<griz>:O
[09:04:16]* omega8cc has quit (Ping timeout: 240 seconds)
[09:04:16]* omega8cc_ is now known as omega8cc
[09:06:55]<EugenMayer>omega8cc: on UMTS? :)
[09:08:04]* lukus has quit (Ping timeout: 246 seconds)
[09:09:51]<omega8cc>EugenMayer: no, but moving in the space and our two access points are connected via different VPNs
[09:10:14]<EugenMayer>ah
[09:11:24]* V1ntage has joined #aegir
[09:11:42]<omega8cc>so, nodejs.org is using Nginx, nice
[09:12:24]<EugenMayer>omega8cc: as i told you, most of the newer projects do. But it seems like its not enterprise ready at all
[09:12:43]<EugenMayer>(yet). Enterprise not in terms of scale or stability, rather of extensions
[09:12:45]<omega8cc>what is not ready
[09:13:14]<omega8cc>nginx?
[09:13:14]<EugenMayer>I would run into issues with kerberos or with shibboleth auth.
[09:13:39]<EugenMayer>i think shib would be easier, why kerberos with e.g. multidomain is pretty pain in the ass AFAIR
[09:14:03]* Slydder1 has quit (Quit: Leaving.)
[09:14:10]<EugenMayer>s/why/while
[09:14:14]<omega8cc>ah, I don't think it is designed to bring you a cold beer, and it is mature enough for what it is designed
[09:14:42]<EugenMayer>Cold beer and kerberos. Maybe you used to web-hosting
[09:15:01]<EugenMayer>But those SSO things are kind of state-of-the-art in enterprise env.
[09:15:13]<omega8cc>it is designed for high performance web hosting
[09:15:18]<omega8cc>I don't care
[09:15:23]<EugenMayer>and its good that nginx is specialized, i guess thats what makes it perform so well
[09:15:31]<omega8cc>of course
[09:16:03]* catatonicrelapse has joined #aegir
[09:16:53]* omega8cc has quit (Remote host closed the connection)
[09:17:01]* omega8cc has joined #aegir
[09:18:40]<EugenMayer>uh, rc1 does not chown files anymore?
[09:20:03]<EugenMayer>ah, its more my migrate-ng ... i see
[09:20:04]<omega8cc>it shouldn't, aegir is not a root
[09:21:16]<omega8cc>only 8 tickets left! see who will be there! Intimate Ægir http://intimateaegir.eventbrite.com
[09:21:19]<EugenMayer>it does before.
[09:22:29]<EugenMayer>(when its aegir:aegir, see docs)
[09:22:29]* lukus has joined #aegir
[09:22:29]<omega8cc>it was there but it never worked, how? it is not a root
[09:22:41]<EugenMayer>sure it worked
[09:22:46]<EugenMayer>aegir is in the www-data group
[09:22:58]<EugenMayer>so it can chown to aegir:www-data when a file is aegir:aegir
[09:23:26]<omega8cc>so it is not a chown, but chgrp
[09:24:01]<EugenMayer>well yes with the correct terminology i guess you are right
[09:24:17]<EugenMayer>i was more thinking about the command, not the effect, but yeah, chgrp
[09:24:54]* EugenMayer starts to move his last client to aegir
[09:25:06]* EugenMayer crosses fingers
[09:25:35]* kvanderw has quit (Ping timeout: 276 seconds)
[09:25:38]<EugenMayer>the site-import into aegir is still pain :)
[09:25:47]* kvanderw has joined #aegir
[09:26:01]<EugenMayer>create site, manually import db, cc -type all, updatedb without update hooks
[09:26:21]<EugenMayer>mysqldump system table, whipe db, replay original db
[09:26:23]<omega8cc>well, but what if there are files owned by www-data and with write access only for www-data?or even some uploaded on import? it is a mess
[09:26:40]<EugenMayer>replace system table, migrate platform to fix file paths - done
[09:27:12]<EugenMayer>omega8cc: in my case those files exist for the math extension in wYSIWYG ( latex )
[09:27:15]<omega8cc>we need it automated
[09:27:22]<EugenMayer>they use a advanvde version of drutex
[09:27:41]<EugenMayer>omega8cc: well people tend to not care about this :)
[09:28:01]* JoshBenner has quit (Read error: Connection reset by peer)
[09:28:21]<EugenMayer>as i have written in the 0.4 stable announcement, i dont think 0.4 is very close to a stable yet and it would be not a good sign marking it that way
[09:28:46]<EugenMayer>people will expect it to work - which will then flood the issue queue with stuff like "not working here, there"
[09:28:59]<omega8cc>I mean automated import - upload compatible site archive and run import task in the front-end, bah, done
[09:29:47]<omega8cc>EugenMayer: you are doing it again
[09:30:54]<EugenMayer>omega8cc: uploading using your client is kind of bad by desing
[09:31:18]<EugenMayer>ok fine, if its a 50MB page it "could work". anything else will be a day of waiting
[09:31:43]<omega8cc>I believe this kind of suggestions/decisions are reserved for core devs, you can express opinions, submit issues, but never try to say "you stupid dev, this code is not ready for the prime time!"
[09:31:56]<EugenMayer>Stop teaching me.
[09:32:05]<EugenMayer>I never said this - i suggested.
[09:32:20]<omega8cc>EugenMayer: really? don't like it?
[09:32:26]<omega8cc>EugenMayer: so don't do that
[09:32:28]<ryanarmstrong>lol
[09:32:39]<ryanarmstrong>I poked my head in at an interesting time
[09:38:28]* q0rban has quit (Quit: Computer has gone to sleep.)
[09:39:13]* catatonicrelapse has quit (Ping timeout: 240 seconds)
[09:40:17]* q0rban has joined #aegir
[09:43:51]<ryanarmstrong>Quick q, I migrated some sites from one platform to the next. If I navigate to the old platform, the sites folders are still there. Inside of them the themes, and for some, the files folder are still there. The sites folders are also present on the new platform, including the themes and files folders, and the sites are running the new platform. Is this expected behavior for the migrate command?
[09:45:42]<omega8cc>ryanarmstrong: yes, it is expected when you didn't set correct permissions on uploaded stuff, so Aegir will not be able to delete old files and will leave zombies there
[09:46:20]* psynaptic is now known as psynaptic|afk
[09:46:34]<ryanarmstrong>Damn zombies!!
[09:46:38]<ryanarmstrong>*grabs shotgun*
[09:47:57]<EugenMayer>ryanarmstrong: have fun. Dont forget to check the aegir master, remove them there before you verify the next time :)
[09:51:27]* V1ntage has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
[09:51:48]<ryanarmstrong>thanks for the help guys!
[09:54:33]<omega8cc>and girls
[09:54:37]<omega8cc>lol
[09:54:41]* FatGuyLaughing has quit (Quit: FatGuyLaughing)
[09:55:44]<omega8cc>time to get some sleep, night all
[09:56:09]* jonhattan has quit (Quit: Ex-Chat)
[09:56:16]* nick_santa has joined #aegir
[10:03:29]* omega8cc has quit (Quit: Cheers! It's Time for Offline Reality)
[10:07:45]* omega8cc has joined #aegir
[10:09:15]* Artusamak is now known as Artusamak_afk
[10:15:20]* realityloop has joined #aegir
[10:22:06]* ghankstef has quit (Quit: ghankstef)
[10:27:10]* mlncn has quit (Remote host closed the connection)
[10:30:57]* psynaptic|afk is now known as psynaptic
[10:32:43]* j0nathan has quit (Quit: ¡ Hasta luego !)
[10:36:33]<griz>hey, omega...
[10:37:05]<griz>when you had a look at the test site I put up what browser were you using?
[10:37:36]<griz>or eugen, for that matter
[10:48:46]* mlncn has joined #aegir