--- Log opened Thu Jun 18 00:00:00 2009 00:52 -!- the_rainbow [n=the_rain@41.209.167.147] has left #t2 [] 00:58 -!- the_rainbow [n=the_rain@41.209.167.147] has joined #t2 03:42 < the_rainbow> does somebody know how to copy symbolic link files when making a root-overlay? 04:49 -!- the_rainbow [n=the_rain@41.209.167.147] has left #t2 [] 05:24 -!- Dr_Willis [n=Dr_Willi@c-69-243-158-200.hsd1.in.comcast.net] has joined #t2 05:33 -!- Dr_Willis [n=Dr_Willi@c-69-243-158-200.hsd1.in.comcast.net] has left #t2 [] 09:11 -!- synchris_ is now known as synchris 09:46 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has joined #t2 09:47 < mpp> moinsen 09:58 < mpp> hey rxr 09:58 < mpp> i did a fresh svn co + build on target/generic/minimal 09:58 < mpp> ncurses still breaks with the error posted yesterday 10:01 < CIA-22> rene * r33374 /trunk/package/base/sdparm/sdparm.desc: * updated sdparm (1.03 -> 1.04) 10:02 < CIA-22> rene * r33375 /trunk/package/security/gnupg2/gnupg2.desc: * updated gnupg2 (2.0.11 -> 2.0.12) 10:04 < CIA-22> rene * r33376 /trunk/package/database/sqlite/sqlite.desc: * updated sqlite (3.6.14.2 -> 3.6.15) 10:39 < CIA-22> rene * r33377 /trunk/package/network/squid/squid.desc: * updated squid (3.0.STABLE15 -> 3.0.STABLE16) 11:39 -!- michael-i [n=michael-@141.41.40.153] has joined #t2 13:42 < michael-i> I've looked and looked but haven't found anything. Is there a way to override the download function to use a file local to the package definition dir? 13:42 -!- T-One [n=T-One@62.218.122.66] has joined #t2 13:42 < michael-i> I'm adding a package whose source consists of a single .c file 14:00 < mjungwirth> hello 14:16 < rxr> hi michael-i 14:16 < michael-i> hi, rxr 14:16 < michael-i> I think I got this one figured out...just overridding everything with mainfunction 14:22 < rxr> we have simillar packages already 14:22 < rxr> even the main function can be avoided 14:22 < rxr> makeopt= 14:22 < rxr> makeinstopt= 14:22 < rxr> hook_add inmake 5 "gcc hello.c -o hello" ... 14:23 < rxr> hook_add postmake 5 "install hello $root$bindir/" 14:23 < rxr> or so - --typos 14:27 < rxr> custmain is highly deprecated, with hooks targets can better hook in their own overrides / tweaks 14:28 < michael-i> nice :) yeah that is a bit cleaner 14:28 < rxr> maybe I should spent a week removall all use of custmain ... :-) 14:29 < rxr> if just we had not so many package utilizing it left, and my desk is so populated ... 14:35 < michael-i> i know the feeling...as I port this project over to Linux I see SOOO many places where it could be improved 14:44 < rxr> yeah - there are also million of more interesting things than cleaning custmain using packages ... :-) 15:34 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"] 15:34 -!- koan [n=koan@koan.xlshosting.net] has joined #t2 15:41 -!- the_rainbow [n=the_rain@41.209.167.148] has joined #t2 16:03 < michael-i> checkin' out....see you later 16:04 -!- michael-i [n=michael-@141.41.40.153] has quit [] 17:02 -!- the_rainbow [n=the_rain@41.209.167.148] has quit [Remote closed the connection] 17:27 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has quit [Remote closed the connection] 18:42 -!- the_rainbow [n=the_rain@41.209.167.148] has joined #t2 19:14 -!- the_rainbow [n=the_rain@41.209.167.148] has quit [Remote closed the connection] 19:15 -!- the_rainbow [n=the_rain@41.209.167.148] has joined #t2 19:28 -!- the_rainbow [n=the_rain@41.209.167.148] has quit [Remote closed the connection] 19:29 -!- the_rainbow [n=the_rain@41.209.167.148] has joined #t2 20:43 < the_rainbow> hello? 20:45 < rxr> hi 20:45 < rxr> the_rainbow: the bas4 "warning" is know, review and patch welcome 20:46 < rxr> though t2 still builds with bash4 I have it on my test server without a hick (despite this strange "warning/error") 20:50 < the_rainbow> it's okay I went back to bash 3 20:50 < the_rainbow> ah by the way rxr, I modified one of the T2 scripts 20:51 < the_rainbow> in order to make copying symbolic links possible when doing a root overlay 20:52 < the_rainbow> I modified the function: copy_and_parse_from_source () 20:52 < the_rainbow> in the file functions.in 20:53 < rxr> ah yes 20:54 < the_rainbow> wanna have a look? 20:54 < rxr> usually one wants to version control them - so we have this special #! unix shebang like syntax to create symlinks (and chmod, rm etc.) from normal files 20:54 < rxr> yes, please paste or better yet send to the mailing list 20:55 < rxr> e.g. 20:55 < rxr> target/archivista/rootfs/var/opt/mysql:#!ln -sf /home/data/archivista/mysql 20:55 < rxr> and others 20:55 < the_rainbow> ah 20:55 < rxr> but please post or mail your changes as well .-) 20:56 < the_rainbow> okay, it solved my problem, I'll send it to the mailing list 20:56 < rxr> thanks! 20:56 < rxr> surry I still had no time for the aufs update 20:56 < the_rainbow> I know there will be much people having similar troubles 20:56 < the_rainbow> it's okay 20:56 < the_rainbow> I updated unionfs 20:56 < rxr> nice 20:56 < the_rainbow> to the latest version 20:56 < rxr> works? 20:56 < the_rainbow> yep 20:56 < the_rainbow> fine 20:56 < the_rainbow> with 2.6.30 20:56 < rxr> maybe you can send a patch of that as well to the list 20:57 < the_rainbow> yes, I changed the desc file 20:57 < rxr> please send in, too :-) 20:57 < the_rainbow> I'll send it 20:58 < CIA-22> rene * r33378 /trunk/package/mail/mutt/mutt.desc: * updated mutt (1.5.19 -> 1.5.20) 20:58 < the_rainbow> now I'm having another trouble 20:58 < rxr> oh :-( 20:58 < the_rainbow> there's no unionfs for kernel 2.6.29.5 20:58 < rxr> and you need that kernel? 20:58 < rxr> btw. which country / time-zoon are you in ? 20:58 < the_rainbow> it's not the prpblem of t2 20:58 < the_rainbow> yes I'm obliged 20:59 < the_rainbow> I'm from Algeria, GMT+1 20:59 < rxr> ah 20:59 < rxr> then near 20:59 < the_rainbow> what about you? 20:59 < the_rainbow> germany? 20:59 < rxr> yep, .de 20:59 < the_rainbow> okay 20:59 < the_rainbow> grsecurity doesn't have a patch for 2.6.30 21:00 < the_rainbow> so I have to use kernel 2.6.29.5 21:00 < rxr> yeah, all this externel kernel sutff always is a pain ... 21:00 < the_rainbow> yep right 21:01 < the_rainbow> still worse, I'll consider using 2.6.29.4 21:01 < CIA-22> rene * r33379 /trunk/package/develop/bazaar/bazaar.desc: * updated bazaar (1.15 -> 1.16) 21:01 < the_rainbow> because there's no unionfs for 2.6.29.5 21:02 < the_rainbow> ah, by the way, the "install CD" target acts like a livecd, excpet that it can be installed through "stone", right? 21:03 < the_rainbow> except* 21:05 < rxr> not exactly 21:05 < rxr> the install has only a minimal system that is loaded into RAM on boot 21:05 < rxr> without unionfs or so 21:06 < rxr> very classic 21:06 < the_rainbow> because I noticed that the "livecd" target isn't installable 21:06 < rxr> the livecd is not loaded into RAM, but the original files overlayed with the write-layer by any unionfs a-like 21:06 < rxr> yeah - we really should add some tiny "rsync all files to partition" script 21:06 < rxr> the archivista target has an example (with some archivista-ism) 21:07 < rxr> target/archivista/rootfs/home/archivista/install.sh 21:07 < the_rainbow> isn't there any target that allows me to install the system on disk? 21:07 < rxr> as install image any target should allow installation 21:07 < rxr> we really should add a install script the live flavor 21:08 < the_rainbow> I thought there was already one 21:08 < the_rainbow> as I downloaded T2 a few months ago 21:08 < the_rainbow> I could install it through "stone" 21:08 < rxr> not the live, only the "classic" install 21:09 < the_rainbow> will "install CD" target allow me install the system through stone? 21:10 < rxr> yep 21:11 < rxr> but you can also copy the archivista install script for your live image and remvoe the few archivista specificas 21:11 < the_rainbow> cool 21:11 < the_rainbow> install.sh, is it sufficient? 21:11 < rxr> it requires xdialog and rsync 21:12 < rxr> you can give it a try, e.g. in some virtual machine 21:12 < rxr> (or spare box) 21:12 < the_rainbow> yep right 21:14 < the_rainbow> I think that the fact that the system loads only in RAM won't affect it too much 21:15 < the_rainbow> as I intend to make a system that is meant to be installed, and the live is just to try it out 21:15 < the_rainbow> do you think that the classical method will be enough? 21:16 < the_rainbow> bbl 22:25 -!- the_rainbow0 [n=the_rain@41.209.167.148] has joined #t2 22:36 -!- the_rainbow [n=the_rain@41.209.167.148] has quit [Read error: 113 (No route to host)] 22:41 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"] 22:41 -!- koan [n=koan@unaffiliated/koan] has joined #t2 22:49 -!- the_rainbow [n=the_rain@41.209.167.148] has joined #t2 22:56 -!- the_rainbow0 [n=the_rain@41.209.167.148] has quit [Read error: 113 (No route to host)] --- Log closed Fri Jun 19 00:00:00 2009