--- Log opened Fri Aug 22 00:00:04 2008 01:22 -!- Netsplit calvino.freenode.net <-> irc.freenode.net quits: Enqlave, axionix, TobiX, Dallur, dsoul 01:22 -!- Netsplit over, joins: Enqlave, Dallur, axionix, dsoul, TobiX 04:51 -!- Cyda [n=cyrusm@c-66-229-118-244.hsd1.fl.comcast.net] has quit [Read error: 113 (No route to host)] 08:40 < rxr> moin 08:54 < CIA-5> rene * r30157 /branches/7.0/package/base/acl/install.patch: 08:54 < CIA-5> * fixed the acl package to not use cp, but install to install 08:54 < CIA-5> binaries, otherwise mangles shared libraries result in nearly 08:54 < CIA-5> every program on an libacl enabled system getting broken during 08:54 < CIA-5> a libacl rebuild 09:05 < rxr> oh - damn, that was already fixed in trunk 09:38 -!- Enqlave [i=stealth@sourcemage/guru/stealth] has quit [Read error: 145 (Connection timed out)] 09:45 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2 09:47 < CIA-5> rene * r30158 /trunk/package/network/mtr/ (mtr.conf mtr.desc): * updated mtr (0.72 -> 0.74) 09:48 < CIA-5> rene * r30159 /trunk/package/filesystem/libburn/libburn.desc: * updated libburn (0.4.4 -> 0.5.2.pl00) 09:48 < CIA-5> rene * r30160 /trunk/package/base/mktemp/mktemp.desc: * updated mktemp (1.5 -> 1.6) 09:48 < CIA-5> rene * r30161 /trunk/package/graphic/gimp/gimp.desc: * updated gimp (2.4.6 -> 2.4.7) 09:52 < CIA-5> rene * r30162 /trunk/package/filesystem/uif2iso/ (uif2iso.conf uif2iso.desc): * updated uif2iso (0.1.2 -> 0.1.6) 09:56 < CIA-5> rene * r30163 /branches/7.0/package/base/acl/ (install.patch install.patch): 09:56 < CIA-5> * replaced the r30157 acl/install.patch with the fix from trunk 09:56 < CIA-5> (same fix, just to have a bit-identical comment, etc.) 10:11 -!- CuE1 [n=root@g230131161.adsl.alicedsl.de] has left #t2 [] 10:11 < CIA-5> rene * r30164 /trunk/package/network/ccrtp/ccrtp.desc: * updated ccrtp (1.5.0 -> 1.6.2) 12:13 < CIA-5> rene * r30165 /trunk/package/network/madwifi/madwifi.desc: 12:13 < CIA-5> Adam Stirk : 12:13 < CIA-5> * updated madwifi (r3375 -> r3837) 12:13 < CIA-5> rene * r30166 /trunk/package/network/hostapd/hostapd.conf: 12:13 < CIA-5> Adam Stirk : 12:13 < CIA-5> * added madwifi support to hostapd 12:14 < CIA-5> rene * r30167 /trunk/package/develop/php/php.desc: 12:14 < CIA-5> Adam Stirk : 12:14 < CIA-5> * new php/phar file download 12:14 < CIA-5> rene * r30168 /trunk/package/network/openntpd/ (openntpd.init postsysfiles.in): 12:14 < CIA-5> Adam Stirk : 12:14 < CIA-5> * added openntpd user/group and init script 12:22 < CIA-5> rene * r30169 /trunk/package/network/linux-igd/ (5 files): 12:22 < CIA-5> Adam Stirk : 12:22 < CIA-5> * added linux-igd (2008-07-20) - Linux Internet Gateway Device 13:05 -!- Enqlave [i=stealth@sourcemage/guru/stealth] has joined #t2 13:43 < CIA-5> rene * r30170 /trunk/package/lua/lua/lua.desc: * updated lua (5.1.4-rc1 -> 5.1.4) 14:01 < CIA-5> rene * r30171 /trunk/target/generic/pkgsel/50-minimal-desktop.in: 14:01 < CIA-5> Martin Papadopoulos : 14:01 < CIA-5> * added python to minimal-desktop for libgsf 14:14 -!- mpp [n=user@i53877B63.versanet.de] has joined #t2 15:31 -!- Cyda [n=cyrusm@c-66-229-118-244.hsd1.fl.comcast.net] has joined #t2 15:32 < rxr> hi mpp & Cyda 15:32 < Cyda> hey rxr 15:33 < mpp> hey rxr 15:33 < mpp> hey cyda 15:34 < mpp> rxr could you pass me a build-target / config for your initramfs / xen you talked about ? 15:36 < rxr> I do not have a xen thing, just an other initrmfs target 15:36 < mpp> could you pass me that one 15:36 < rxr> I could create some example for the initrd, yes 15:36 < mpp> cause my images are far to big 15:36 < mpp> i used generic target 15:36 < mpp> with bootstrap selection 15:36 < rxr> that's probably too big to fit into the initrd 15:37 < mpp> but after unpacking the kernel hang cause it exhausted over 1 gig 15:37 < mpp> yep 15:37 < mpp> i am trying to build a rescue for pxe 15:37 < mpp> thats all 15:37 < rxr> have you tried using rescue as base? that's probably way tinier than the bootstrap selection 15:38 < mpp> yes but the rescue only suggests livecd 15:38 < mpp> should i manually change it to initramfs ? 15:38 < mpp> i mean in the target ? 15:38 < mpp> btw i now have dedicated server for building t2 15:38 < mpp> 24/7 15:38 < mpp> i really needed that one cause it's quiet cpu heavy 15:39 < rxr> yes, the manual change could work 15:39 < mpp> what did you choose for your initramfs ? 15:39 < mpp> or did you strip down a built target ? 15:40 < rxr> I have some custom target for a company 15:40 < mpp> ah - 15:40 < rxr> but I'll try the rescue in some minutes 15:40 < mpp> cool so do i 15:40 < rxr> we can also change the rescue target to offer a selection, whether to build a livecd or an initrd 15:40 < mpp> if it works 15:40 < mpp> thats the point i don't know 15:40 < rxr> (I just wire up livecd for the rescue target a year ago so it works out of the box and folks do not accidently build an install image .-) 15:41 < mpp> makes sence :-) 15:41 < rxr> you can also just remove the hardwiring from the rescue target and manually select initrd and give it a spin 15:41 < rxr> might already just work .-) 15:41 < rxr> the config fixing is in target/rscue/config.in 15:41 < mpp> i'll do so right know 15:41 < mpp> now 15:41 < mpp> yep .. found that some days ago 15:42 < mpp> but didn't have time cause i'm on project for several months now with a lot of travelling 15:42 < mpp> but now i'll stick around until tuesday 15:42 < rxr> :-) 15:44 < mpp> btw. i'll post the stuff related to the chrooting later on today ;-) 15:44 < mpp> i am grateful for t2 15:45 < mpp> as an intermediate skilled linuxer i relay on real pros 15:45 < mpp> big thanx !!! 15:45 < rxr> yw 15:54 < mpp> hey rxr . what script is called on a build process to create the initramfs file itself ? 15:55 < mpp> i could modify the rootfs before it is packed 15:55 < mpp> that way i could strip down a built target. 15:56 < mpp> if i recall correctly squashfs is also involved .. which i have no idea how it's used 15:56 < mpp> in the initram processing / building 16:05 < rxr> the output scripts are in target/share/*/ 16:05 < rxr> squashfs is only used for livecd 16:15 < mpp> does it make sense to modify the embedded target for a pxe initramfs ? 16:15 < mpp> it's fairly small 16:17 < rxr> embedded includes some cross only stuff and is notoriosly harder to debug/track 16:17 < rxr> I would use rescue and go from there 16:17 < rxr> I hope I can commit the config switch for it in some seconds 16:19 < mpp> for 8.0-trunk or 7.0 ? 16:21 < rxr> trunk :-) 16:21 < rxr> aka "some day 8.0" 16:21 < rxr> so - done 16:21 < rxr> CIA info pending 16:21 < rxr> t2/trunk:HEAD results in a 73MB initrd for me 16:22 < mpp> i am running atm the modified rescue initramfs build 16:22 < rxr> if you stip out all the X fluff and co it will be significantly smaller :-) 16:22 < CIA-5> rene * r30172 /trunk/target/rescue/ (build.sh config.in): * allow choosing whether to build the rescue target as livecd or initrd 16:22 < mpp> i thought so 16:22 < mpp> cudos ! 16:23 < rxr> btw, mpp & Cyda, new IRC log rendering: http://www.t2-project.org/irclogs/t2-2008-08-22.html 16:25 < mpp> nice :-) 16:26 < mpp> how is it done ? 16:26 < rxr> it's rendered by a CGIs cript, now 16:26 < mpp> it's really cute ? 16:26 < mpp> ! 16:26 < rxr> in the past it was just a dumb file dump 16:26 < rxr> which looked ugly and had no site menu integration 16:26 < rxr> plain black on white text 16:26 < mpp> yea this is more eye candy 16:27 < mpp> web2.0 compatible .... 16:27 < mpp> or so 16:27 < rxr> now the people that randomly hit the logs thru google have a chance to explore the t2 website (without manually editing the URL field ) 16:27 < mpp> whooo. 16:27 < mpp> you 're into pr stuff 16:28 < mpp> which is obviously a wise choice for your company ! 16:28 < rxr> have been for some time now, already ... 16:28 < mpp> let google do the work 16:28 < rxr> heh, lol, and t2 in general to be considered between Gentoo, Debian, and the other pre-built stuff 16:29 < mpp> actually it's on top 16:29 < mpp> not in between :-) 16:29 < rxr> yeah - that's what we try to tell the world for some time now .-) 16:30 < mpp> the prophet in his own land... yadayadayda LOL 16:30 < mpp> Gee U aR U 16:30 < mpp> ;-) 16:31 < mpp> we should stick with N.I.K.E. 16:31 < mpp> just do it ! 16:31 < rxr> :-) 16:32 < mpp> so hows the weather in berlin 16:32 < mpp> ? 16:32 < mpp> http://de.youtube.com/watch?v=UxPSz54l3ps 16:32 < mpp> berlinerisch für anfänger 16:32 < mpp> this is fun 16:33 < mpp> LOL :-)) 16:33 < mpp> ike jet jetse schrippen koofen 16:33 < mpp> mamfmamf 16:35 < rxr> :-) 16:35 < mpp> cheerz mate 16:35 < rxr> gut das ich kein Berliner binn :-) 16:35 < rxr> ick meine ich natuerlich :-) 16:35 < rxr> anyway :-) 16:35 < mpp> ne war 16:35 < rxr> there was something I wanted to say - ah yes 16:35 < mpp> ei verbimsch noch mal 16:35 < rxr> PR: So go spread the word! 16:35 < mpp> go on 16:36 < mpp> i'll do so 16:36 < rxr> good :-) 16:37 < mpp> t2 is supreme ---- i obey --- exterminate exterminate .... 16:37 < mpp> okay cutting the crap now 16:37 < mpp> back to topic 16:38 < rxr> yeah - best blog and fill forums with it, and best with direct back again 17:40 < CIA-5> rene * r30173 /trunk/package/develop/git/git.desc: * updated git (1.5.6.5 -> 1.6.0) 17:48 < mpp> what exactly is nxterm ? 17:49 < mpp> in contrary to xterm.. 17:50 < rxr> I guess you talk about just the TERM variable ? 17:50 < mpp> yep 17:50 < rxr> I think the xterm authors introduced this new name when they made incompatible changes to their ANSI escape character support ... 17:50 < mpp> since in the mc the pos1 and other keys dont't work 17:50 < rxr> (some 5 years or more ago) 17:50 < mpp> ah 17:51 < rxr> maybe soner termcap / ncurses files missing? 17:51 < mpp> so that could be the reason for some keys not functioning inside mc 17:51 < mpp> i built it on trunk and 7.0 17:51 < rxr> I ususally use rxvt-unicode exlusively, and on all my boxes with lots of windows I seldomely have problems ... 17:51 < mpp> both show the same behaviour 17:51 < mpp> whats that rxvt-unicode 17:51 < mpp> ? 17:52 < rxr> a package 17:52 < mpp> i guess i would just want to know how i get these keys working (again) 17:53 < rxr> http://t2-project.org/packages/rxvt-unicode.html 17:53 < mpp> aha 17:53 < mpp> so other distros use other emulation - ? 17:53 < rxr> nope :-) 17:54 < rxr> they all ship the same xterm, rxvt, rxvt-unicode, konsole, gnome-terminal 17:54 < mpp> come on throw me a fricking bone will ya ? 17:54 < rxr> btw. all of them implement terminal emulation differently 17:54 < rxr> so different behaviour in corner cases, feature completness ... 17:54 < mpp> thats what i am looking for 17:54 < rxr> but even xterm should just work 17:54 < mpp> what do i need to modifiy on a running t2-host ? 17:55 < rxr> guess maybe either you selected a very strange or unexisting locale or soe termcap file is missing 17:55 < rxr> no idea what's broken in your case 17:55 < rxr> what did you install? 17:55 < mpp> ahm... 17:56 < mpp> t2 7.0 generic minmal build 17:56 < mpp> i meant generic minimal basic desktop 17:57 < rxr> self built ? 17:57 < mpp> yes and no 17:57 < mpp> i tried the rc2 and a self built 17:58 < rxr> and on both xterm did not work fully ? 17:58 < mpp> yes 17:58 < mpp> strange thing 17:59 < rxr> in my trunk (from some weeks) desktop xterm home and end keys work fine 17:59 < mpp> what target did you user for building ? 17:59 < rxr> desktop 18:00 < mpp> okay i should try that one instead 18:00 < rxr> but I just remeber we recently had broken keyboard layouts due to x.org transition from xkbdata to xkeyboard-config 18:00 < rxr> maybe 7.0 is affected and your x server just does not handle all key mappings correctly .. 18:00 < mpp> but it would be cool if it worked on the console of a minimal build 18:00 < rxr> testing that part of 7.0 is on my todo 18:00 < mpp> no i don't mean the x server 18:00 < mpp> i am using mc on the console 18:00 < rxr> the 7.0.0 release is emminent, x86* are built, ppc*, likewie, sparc64 pending ... 18:00 < mpp> all the time - sorry - 18:00 < rxr> have to test install them soonish ... 18:01 < rxr> and last minute fix what I find on-the-way 18:01 < mpp> allright 18:01 < mpp> wait a minute 18:02 < mpp> i just ran the learn keys inside the mc 18:02 < mpp> so it works finally 18:02 < mpp> puuh 18:02 -!- mpp [n=user@i53877B63.versanet.de] has left #t2 [] 18:03 -!- mpp [n=user@i53877B63.versanet.de] has joined #t2 18:03 < mpp> oops - 18:03 < mpp> got kicked out 18:03 < mpp> can you repeat rxr ? 18:04 < rxr> with part ? 18:04 < rxr> which part even? 18:04 < mpp> the last 2 posts 18:04 < mpp> my irc client skrewed up 18:04 < mpp> so i missed the part after the learning keys 18:06 < rxr> 18:00 < rxr> testing that part of 7.0 is on my todo 18:06 < rxr> 18:00 < rxr> the 7.0.0 release is emminent, x86* are built, ppc*, likewie, sparc64 pending ... 18:06 < rxr> 18:00 < rxr> have to test install them soonish ... 18:06 < rxr> 18:01 < rxr> and last minute fix what I find on-the-way 18:06 < mpp> okay 18:19 < mpp> the mailing list had also this posting about libx11 18:19 < mpp> pkgprefix: package libx11 is not present 18:19 < mpp> was there a fix for that message - maybe it can be ignored ? 18:20 < mpp> i get this message during a build 18:20 < mpp> i am running on console only 18:21 < rxr> yeah - wanted to fix this already 18:21 < rxr> wanted to finish some other work today, though 18:21 < rxr> I'll take a look into a fix, now 18:21 < rxr> you can ignore it for now 18:21 < rxr> (should be harmless, beside it should be fixed .-) 18:22 < mpp> okay then - you should skip this for now . 18:24 < mpp> an issue i have found with trunk - which is really annoying - is allthoug i configured the eth0 interface with dchp , the interface is not configured with dhcp after reboot 18:24 < mpp> kinda funny 18:24 < mpp> the network service is being started arcording to the bootscreen . 18:25 < mpp> so i created an rc.local for now which does the dhcpclient for eth0 18:25 < mpp> prio is kinda low though 18:27 < rxr> I think dhcpd version 4.0.0 was just recently fixed 18:27 < rxr> does it work if you re-emerge the dhcp packaga ? 18:27 < rxr> unfortunatly they broke / change quite a lot internally in this dhcp-4.x release ... :-(((( 18:29 < mpp> ill give it a try 18:29 < mpp> my 2 hosts are building at the moment generic + rescue 18:30 < mpp> so ill report later on that 18:30 < mpp> back in the days of rocklinux there was a target called router .. i think 18:30 < mpp> is that correct ? 18:30 < mpp> i could use that one in t2 too :-) 18:31 < mpp> for turning a old / thin equipped pc into a router/proxy/vpn etc. 18:31 < rxr> yeah - the router thing never was really maintained 18:31 < mpp> if not i'll create a target for that purpose 18:31 < rxr> we removed it after some years of equally lingering in rock and t2 ... 18:32 < mpp> i think i'll make a new one 18:32 < rxr> rescue should be quite close, and then mika pesu started this wrt2, wireless router t2, intended for AP like devices 18:32 < mpp> yep 18:33 < mpp> for those kind of forks / mods i wished sometimes t2 was on monotone to easily branch / fork the sourcetree 18:33 < mpp> and commit back 18:33 < rxr> or hg (which I personally prefer these days) 18:33 < rxr> recently tried to start a read-only hg mirror 18:34 < rxr> though our SVN tree complexity does not make the converter too happy 18:34 < mpp> hg ? 18:34 < rxr> mercurial 18:34 < mpp> ah --- 18:34 < mpp> i browsed throug the package list of t2 which has it on 18:34 < rxr> http://www.t2-project.org/packages/mercurial.html 18:34 < mpp> yep but thing about monotone is that it only needs a db sync 18:34 < mpp> all other operations are done offline 18:35 < mpp> ideal for traveling with no connectivity 18:35 < mpp> thats all really 18:35 < rxr> yeah 18:35 < mpp> i could build a monotone repo from the synced svn on my machine 18:35 < rxr> thing is back in time when I converted "ROCK Linux" from nothing (or better rsyncable snapshots) to SVN there was not much else beside, CVS (yuck), BitKeeper, some commercial ones (perforce at al.) and the upcomming SVN 18:36 < mpp> but i'll have to figure out a way to commit the changes as comment on monotone 18:36 < mpp> is doable i think 18:36 < rxr> we where like the first big project to adapt SVN around version 0.17 or so ... :-) 18:36 < mpp> i remember 18:36 < mpp> since i met you guys on a cebit many years ago first time 18:36 < rxr> so we where happy to have somthing as advanced and useful as SVN at all :-) 18:37 < rxr> :-) 18:37 < mpp> cliff was there and off course you 18:37 < mpp> sure svn is good . period ! 18:38 < mpp> monotone is very compact on ther other hand and easy to user for multiple forks and branches even of different contributers 18:38 < mpp> i'll check on it asap 18:38 < rxr> what I usually do with any of our projects, make a local hg repo, work^work^work, when it's ready commit glued hg changesets ... 18:39 < mpp> sounds slick 18:39 < mpp> i don't quite know mercurial . 18:40 < mpp> monotone has an advantage that is has one single db file 18:40 < mpp> all checkouts / commits can then be extracted / injected 18:43 < mpp> monotne also signs the transactions which is like signing packets and codes 18:43 < mpp> anyways 18:43 < mpp> svn brought us here - so it's cool . 18:49 < rxr> :-) 18:52 < mpp> another thing i am interested in t2 is the distcc 18:52 < mpp> does it work out of the box to build with several machines a t2 target ? 18:55 < rxr> in the past yes, since multi cores boxes I do not use it too often 18:56 < mpp> i think using comodity pc's for a t2 compile cluster 18:56 < mpp> thats way cool 19:00 < rxr> stray warning of libx11 fixed 19:00 < rxr> svn up and have fun 19:00 < CIA-5> rene * r30174 /trunk/package/xorg/xorg-server/parse-config: * silenced r30153 pkgprefix warning if libx11 was not yet built 19:01 < mpp> thanx 19:30 < mpp> on the chrooting error on a non t2-host 19:31 < mpp> chroot . 19:31 < mpp> mount /dev 19:31 < mpp> mount: none has wrong device number or fs type devfs not supported 19:31 < mpp> mount /proc 19:32 < mpp> since the /dev is not mounted succesfully the stone cannot start - as expected. 19:41 < mpp> looks like my ubuntu host does not support devfs anymore 19:53 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 104 (Connection reset by peer)] 19:56 < mpp> for the bind package it would obviously be meaningfull to have 2 configuration files in the package - so bind can work as a caching only nameserver . 19:56 < mpp> db.cache an db.127.0.0 19:59 -!- koan [n=koan@unaffiliated/koan] has quit [Read error: 110 (Connection timed out)] 20:29 < rxr> devfs is long gone everywhere will take a look 20:29 < rxr> patch for bind welcome 20:29 < mpp> cool 20:29 < mpp> patch for bind is easy 20:30 < mpp> where do i post it ? 20:30 < rxr> t2@t2-project.org or so 20:30 < mpp> k 20:30 < rxr> official mailing list 21:42 -!- mp2 [n=user@i53875610.versanet.de] has joined #t2 21:59 -!- mpp [n=user@i53877B63.versanet.de] has quit [Read error: 110 (Connection timed out)] 22:04 -!- mp2 [n=user@i53875610.versanet.de] has quit [Read error: 110 (Connection timed out)] 22:33 < Ragnar|away> rxr: heeelp! 22:34 < Ragnar|away> how do I get the initrd to find a md device? --- Log closed Sat Aug 23 00:00:10 2008