[02:25] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) got netsplit.
[02:25] SMP (n=stefanp@ got netsplit.
[02:25] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) returned to #rocklinux.
[02:25] SMP (n=stefanp@ returned to #rocklinux.
[04:19] kasc_ (n=kasc@dslb-084-060-103-251.pools.arcor-ip.net) joined #rocklinux.
[04:21] stf^rocklinux (n=user@M1025P026.adsl.highway.telekom.at) left irc: Read error: 110 (Connection timed out)
[04:32] kasc (n=kasc@dslb-084-060-105-040.pools.arcor-ip.net) left irc: Read error: 110 (Connection timed out)
[04:32] Nick change: kasc_ -> kasc
[06:17] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) got netsplit.
[06:17] SMP (n=stefanp@ got netsplit.
[06:17] mnemoc (n=amery@kilo105.server4you.de) got netsplit.
[06:17] daja77 (n=daja77@ got netsplit.
[06:17] ravelo (n=userm@ got netsplit.
[06:18] ravelo (n=userm@ returned to #rocklinux.
[06:18] daja77 (n=daja77@ returned to #rocklinux.
[06:18] mnemoc (n=amery@kilo105.server4you.de) returned to #rocklinux.
[06:18] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) returned to #rocklinux.
[06:18] SMP (n=stefanp@ returned to #rocklinux.
[09:25] stf^rocklinux (n=user@M1187P004.adsl.highway.telekom.at) joined #rocklinux.
[09:49] <stf^rocklinux> hi
[09:50] <stf^rocklinux> th: btw. linux-libc-headers have been replaced with linux26-headers in trunk
[09:54] ija_ (n=ija@i59F70F24.versanet.de) joined #rocklinux.
[09:54] <stf^rocklinux> th: I'll also start a crystal and bootdisk build from trunk today to see if there are any problems
[10:01] <th> stf^rocklinux: well coreutils and boehm-gc failed for me in crystal
[10:01] <th> i noticed we changed to glibc26 and gcc42
[10:02] <stf^rocklinux> yes, I've already done a bootdisk and crystal build with those
[10:02] <th> and your coreutils worked?
[10:02] <stf^rocklinux> do you have the exact error message for failed packages?
[10:02] <th> some stupid detail at the end when installing help/man stuff
[10:03] <th> i can provide this later today
[10:03] <stf^rocklinux> oh, maybe I've had this error, too, but only in a cross-build afaik
[10:03] <stf^rocklinux> it was something about the touch.1 manpage
[10:06] <th> exactly - sounds familiar
[10:07] <th> crossbuild and crossnative options are off
[10:07] <th> Updating man page touch.1
[10:07] <th> help2man: can't get `--help' info from touch.td/touch
[10:07] <th> make[1]: *** [touch.1] Error 1
[10:07] <th> INSERT = '--specs=specs.cross -Os -pipe -march=i486'
[10:07] <th> hmm??
[10:07] <th> why is this cross?
[10:08] <th> i'm talking of stage1
[10:08] <th> yesterday i got this with gcc42 - now i have the same with gcc40
[10:11] <stf^rocklinux> afaik stages 0 and 1 always use a pseudo-cross compiler
[10:12] ija (n=ija@i59F74293.versanet.de) left irc: Read error: 110 (Connection timed out)
[10:12] <th> stf^rocklinux: so coreutils failed in stage1 for you as well?
[10:12] <stf^rocklinux> the problem in my cross build was that help2man tried to execute the cross-built touch, don't know why it doesn't work in native builds
[10:13] <stf^rocklinux> th: no, worked for me afair
[10:13] <th> hmmm
[10:14] <stf^rocklinux> well, I can tell you more in a few hours, when 1-coreutils has built
[10:15] <th> well i found the reason
[10:15] <th> ./src/touch: /lib/libc.so.6: version `GLIBC_2.4' not found (required by ./src/touch)
[10:15] <th> it's because glibc26 is now the default
[10:15] <stf^rocklinux> I only know that executing help2man is not required for man pages to be installed
[10:15] <th> so a glibc23 as build host is not sufficient
[10:16] <th> so this is a chicken egg problem
[10:16] <stf^rocklinux> I simply disabled the help2man run and still got all man pages including touch.1
[10:16] <th> how would i get a build host?
[10:16] <th> can we do this by patching coreutils?
[10:17] <stf^rocklinux> yes, I already had a simple patch for that, will rewrite it today and send it to SM
[10:17] <th> thanks
[10:17] <th> so this explains why only my testbuild fails.
[10:17] <th> because i've still glibc23 on the buildhost
[10:18] <stf^rocklinux> yes, that seems plausible, I'm using 2.5 for some months already
[10:22] <th> yes another userland incompatibily step. 
[10:23] <th> but we propably have to take it
[10:23] <th> damn
[10:23] <th> and i can still run windows95 binaries on windows vista...
[10:31] <stf^rocklinux> th: the patch is 2007062810305120346
[11:03] <th> thanks a lot
[11:03] <th> i think i can apply this without the full build
[11:03] <th> it cant make the full build worse
[11:03] <th> stf^rocklinux: agreed?
[11:04] <th> well testing does not hurt
[11:12] <daja77> hi th :)
[11:23] <th> hey hi daja77 
[11:23] <daja77> thx for applying the git hack :)
[11:23] <th> == 06/28/07 02:49:46 =[1]=> Aborted building package coreutils.
[11:24] <th> stf^rocklinux: that did not work out so smoothly - are you sure a .patch.cross is applied in stage1?
[11:24] <th> daja77: no problem - it had positive votecheck.
[11:24] <daja77> th: that package even knocked clifford down :)
[11:24] <th> how that?
[11:25] <daja77> you need a complex toolchain including tetex and stuff to build the documentation
[11:25] <daja77> and one of the tools is always complaining that it isn't executed on fedora
[11:26] <th> haha
[11:26] <th> talking of coreutils?
[11:26] <daja77> that's why i built the docs on fedora which i have in qemu for other reasons anyway 
[11:27] <daja77> i have no build running atm so i don't have that coreutils problem
[11:27] <th> "that package" -> talking of coreutils?
[11:27] <daja77> git
[11:27] <daja77> hm no
[11:27] <daja77> xmlto that was
[11:27] <th> stf^rocklinux: Apply patch /mnt/rock/trunk/package/base/coreutils/dont-run-help2man.patch.cross ...
[11:27] <th> hmm it is applied
[11:28] <th> /bin/sh: -c: line 8: syntax error: unexpected end of file
[11:28] <th> haha
[11:28] <daja77> O_o
[11:28] <th> ++      fi                          \
[11:28] <th> i guess the "\" might be the problem
[11:34] <stf^rocklinux> th: there's a \ missing here:
[11:34] <stf^rocklinux> ++          if false ; then
[11:34] <th> the fi \ should stay?
[11:35] <stf^rocklinux> yes
[11:35] <th> retrying...
[11:35] <th> == 11:44:00 =[1]=> Building base/coreutils [6.9 0].
[11:37] <th> == 06/28/07 11:45:33 =[1]=> Aborted building package coreutils.
[11:38] <th> /bin/sh: -c: line 20: syntax error near unexpected token `;;'
[11:38] <th> /bin/sh: -c: line 20: `    && rm -rf touch.td ;;                                        \'
[11:38] <th> gahh
[11:38] <stf^rocklinux> ups.
[11:38] <th> +       && rm -rf $t ;;                 \
[11:38] <th> ++      fi                          \
[11:38] <th> fi ;; ??
[11:38] <stf^rocklinux> yep
[11:39] <stf^rocklinux> sry for the inconvenience...
[11:39] <stf^rocklinux> and ;; has to be removed from the && rm -rf $t line, of course
[11:45] <th>             && rm -rf $t                                        \
[11:45] <th> +           fi ;;                                               \
[11:46] <th> == 11:54:36 =[1]=> Building base/coreutils [6.9 0].
[11:46] <th> !> patching file man/Makefile.in
[11:46] <th> !> Hunk #2 FAILED at 871.   
[11:46] <th> stupid me
[11:48] <stf^rocklinux> ; missing?
[11:48] <th> nope
[11:48] <th> -           && rm -rf $t ;;                                     \
[11:48] <th> +           && rm -rf $t                                        \
[11:48] <th> +           fi ;;                                               \
[11:48] <th> the - line was missing
[11:48] <th> it's a patch after all
[11:49] <stf^rocklinux> ok, but  it must read && rm -rf $t ; \
[11:49] <th> oh ok
[11:51] <th> i'll rediff the 2007062810305120346 once it works ;)
[11:51] <stf^rocklinux> good idea :)
[11:52] <th> == 12:00:25 =[1]=> Building base/coreutils [6.9 0].
[11:54] <th> == 06/28/07 12:02:22 =[1]=> Finished building package coreutils.
[11:57] <th> rediffed in 2007062811574221443
[11:58] <th> and applied
[12:00] <th> now let's give trunk another chance
[12:00] <th> might be easier with `mkdir` available.
[12:01] <th> Downloading download/mirror/I/ImageMagick-6.3.4-1.tar.bz2 ...
[12:01] <th> this makes trouble in crystal, syn?
[12:07] <stf^rocklinux> th: btw. I agree with applying the coreutils patch
[12:07] <th> 11:55:25 < th> and applied
[12:07] <th> done
[12:08] <stf^rocklinux> th: new versions of ImageMagick are released very often, with older versions being deleted from the official download site. It's hard to get a version on the ROCK mirrors before it disappears :/
[12:09] <daja77> indeed
[12:10] <th> perhaps we can find that missing file somewhere and inject it directly to our mirrors?
[12:10] <th> who has the file?
[12:10] <daja77> which?
[12:10] <th> 11:58:11 < th> Downloading download/mirror/I/ImageMagick-6.3.4-1.tar.bz2 ...
[12:10] <th> ImageMagick-6.3.4-1.tar.bz2
[12:11] <stf^rocklinux> better download the newest version and upload it to the mirrors if you can
[12:11] <th> thus updating a CORE package
[12:11] <stf^rocklinux> but I have the file if you need it
[12:11] <th> which needs testbuild and testinstall ;)
[12:12] <daja77> ImageMagick-6.3.3-5.tar.bz2 is the latest i git
[12:12] <th> ImageMagick-6.3.3-5.tar.bz2 is on our mirror
[12:12] <daja77> nah crap just update that damn imagemagick
[12:12] <th> stf^rocklinux: could you make 6.3.4-1 wgettable?
[12:12] <daja77> it is the same thing anyway just 10 cvs commits later
[12:13] <stf^rocklinux> th: not without some effort
[12:13] <stf^rocklinux> but updating imagemagick is pretty harmless, I suppose
[12:13] <th> can you make it available somehow else?
[12:13] <th> hmm
[12:13] <th> ok can you just throw the update in sm then?
[12:14] <th> i just dont like having a version referenced in trunk which is not on the mirrors
[12:14] <stf^rocklinux> I'll send an update patch
[12:14] <th> great
[12:17] <stf^rocklinux> th: it's in 2007062812172218720
[12:18] <th> 1st thing i do: manually put the file in the mirror
[12:38] stf^rocklinux (n=user@M1187P004.adsl.highway.telekom.at) left irc: Read error: 110 (Connection timed out)
[12:48] stf^rocklinux (n=user@M1523P031.adsl.highway.telekom.at) joined #rocklinux.
[13:06] rocklogbot (n=rocklogb@ left irc: Remote closed the connection
[13:06] rocklogbot (n=rocklogb@pallas.crash-override.net) joined #rocklinux.
[14:54] markuman (n=supermar@dslb-084-063-174-097.pools.arcor-ip.net) joined #rocklinux.
[16:19] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) left irc: Remote closed the connection
[16:19] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) joined #rocklinux.
[16:40] <th>  [5] base/ncompress                      [5] public/pcmcia-cs
[16:40] <th> Found shared files with other packages:
[16:40] <th> usr/bin/uncompress: gzip ncompress
[16:41] <th> and pcmcia-cs is some real issue
[16:41] <th> /usr/src/linux/include/linux/types.h:158: error: conflicting types for 'blkcnt_t'
[16:41] <th> /usr/include/sys/types.h:235: error: previous declaration of 'blkcnt_t' was here
[17:26] kasc (n=kasc@dslb-084-060-103-251.pools.arcor-ip.net) got netsplit.
[17:27] kasc (n=kasc@dslb-084-060-103-251.pools.arcor-ip.net) returned to #rocklinux.
[18:00] mnemoc (n=amery@kilo105.server4you.de) got netsplit.
[18:00] daja77 (n=daja77@ got netsplit.
[18:00] ravelo (n=userm@ got netsplit.
[18:00] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) got netsplit.
[18:00] SMP (n=stefanp@ got netsplit.
[18:02] ubijtsa2 (n=ubijtsa@karlsson.force9.co.uk) returned to #rocklinux.
[18:02] ravelo (n=userm@ returned to #rocklinux.
[18:02] daja77 (n=daja77@ returned to #rocklinux.
[18:02] mnemoc (n=amery@kilo105.server4you.de) returned to #rocklinux.
[18:02] SMP (n=stefanp@ returned to #rocklinux.
[18:43] Nick change: ija_ -> ija
[19:19] <th> re
[20:24] markuman (n=supermar@dslb-084-063-174-097.pools.arcor-ip.net) left irc: Read error: 110 (Connection timed out)
[20:25] markuman (n=supermar@dslb-084-063-177-009.pools.arcor-ip.net) joined #rocklinux.
[20:27] markuman (n=supermar@dslb-084-063-177-009.pools.arcor-ip.net) left irc: Client Quit
[20:44] <th> small router target:
[20:44] <th> 146 builds total, 143 completed fine, 3 with errors.
[20:44] <th>  [5] base/ncompress                      [5] public/pcmcia-cs
[20:44] <th>  [5] base/nfs-utils                      
[20:45] <th> checking for libwrap... configure: error: *** libwrap missing
[20:45] <th> that's for nfs
[20:45] <th> so might be a target issue
[20:46] <th> so where is libwrap?
[20:47] <th> ahh simple tcpwrapper
[20:53] <th> ok - adding tcp_wrappers to pkgsel/common.set (which includes nfs-utils
[20:55] stf^rocklinux (n=user@M1523P031.adsl.highway.telekom.at) left irc: Read error: 110 (Connection timed out)
[20:57] stf^rocklinux (n=user@M1081P014.adsl.highway.telekom.at) joined #rocklinux.
[20:59] <th> so gzip provides uncompress (which conflicts with package ncompress)
[21:00] <th> but gzip does not provide "compress"?
[21:34] <stf^rocklinux> hm, gzip 1.3.5 did not have uncompress
[21:35] <stf^rocklinux> th: otoh, ncompress is not core, so this is not an urgent problem
[21:35] <th> true
[21:35] <th> but still one of our targets
[21:35] <th> (router
[21:37] <th> its not urgent - i'm just curious what's up with gzip
[21:37] <stf^rocklinux> I thought the router target is only hosted, but not maintained by ROCK Linux
[21:38] <th> yea but gzip is core
[21:38] <th> and it's strange when it provides uncompress but no compress, no?
[21:38] <stf^rocklinux> it seems gzip has added a program for uncompressing compress'd files, so that you don't need to have ncompress
[21:38] <stf^rocklinux> I guess noone wants to use compress these days...
[21:39] <th> hmmm sounds reasonable
[21:39] <th> perhaps i should remove ncompress from router pkgsel
[21:40] <stf^rocklinux> the desc file of ncompress  says it's fast, but even contemporary routers may be fast enough for gzip, so it should be ok to remove it
[21:49] <th> stf^rocklinux: ok - you have an idea for:
[21:49] <th> In file included from ide_info.c:44:
[21:49] <th> /usr/src/linux/include/linux/types.h:158: error: conflicting types for 'blkcnt_t'
[21:49] <th> /usr/include/sys/types.h:235: error: previous declaration of 'blkcnt_t' was here
[21:49] <th> 5-pcmcia-cs.err 
[21:50] <daja77> ah nice that's why we had those cleaned kernel headers
[21:51] <th> why did we drop them?
[21:51] <th> have gone too old?
[21:51] <daja77> yep
[21:56] <th> so the real bug is that pcmcia-cs includes one of these headers right?
[21:56] <th> so it should not include both
[22:01] <daja77> i think so
[22:10] <th> http://pcmcia-cs.sourceforge.net/
[22:10] <rocklogbot> Linux PCMCIA Information: http://pcmcia-cs.sourceforge.net/
[22:10] <th> The Linux pcmcia-cs package is officially deprecated. It can only be used with 2.4 and older kernels. Current information on PCMCIA support for recent 2.6 kernels is available here.
[22:10] <th> so just to be dropped
[22:10] <th> and perhaps replaced by PCMCIAutils
[22:11] <th> http://kernel.org/pub/linux/utils/kernel/pcmcia/pcmciautils-014.tar.gz
[23:30] <SMP> th: nobody needs compress anymore. it's enough when gzip can uncompress the stuff that was obsolete 10 years ago
[23:31] <th> step by step i notice that the router target might not be even good enough as a scratch
[23:31] <th> but the compress thing was in the common.set
[23:31] <th> but that was ok until gzip learned to uncompress
[23:32] <SMP> I'd say the best place to start would be rescue 
[23:34] <th> perhaps generic
[00:00] --- Fri Jun 29 2007