**** BEGIN LOGGING AT Wed Aug 17 02:59:57 2011
Aug 17 03:19:30 <CIA-89> jow * r28024 /trunk/target/linux/brcm47xx/base-files/etc/init.d/netconfig: [brcm47xx] fix the default network configuration for the ASUS WL-330gE (#8751)
Aug 17 03:20:32 <CIA-89> jow * r28025 /branches/backfire/target/linux/brcm-2.4/base-files/etc/init.d/netconfig: [backfire] merge r28024
Aug 17 04:19:17 <philipp64|laptop> jow_laptop: ping
Aug 17 06:48:10 <_trine> the MD5 checksum for boxbackup is wrong
Aug 17 06:48:51 <_trine> cshore, ping
Aug 17 06:54:03 <_trine> cshore, I corrected the MD5 checksum and now I get this :-
Aug 17 06:54:06 <_trine> http://dpaste.com/596261/
Aug 17 06:55:51 <_trine> boxbackup-0.11.1.tgz seems to be corrupt
Aug 17 06:56:24 <_trine> gzip: stdin: not in gzip format
Aug 17 06:56:24 <_trine> tar: Child returned status 1
Aug 17 06:56:24 <_trine> tar: Error is not recoverable: exiting now
Aug 17 07:17:14 <KanjiMonster> _trine, cshore: the source url is also wrong; it should be http://www.boxbackup.org/trac/export/2978/box/packages/
Aug 17 07:58:05 <loswillios> jow_laptop: what I forgot yesterday, pulseaudio requires uclibc-0.9.32 so it's not available in backfire anyway.
Aug 17 08:51:10 <CIA-89> florian * r28026 /trunk/package/kernel/modules/crypto.mk:
Aug 17 08:51:10 <CIA-89> [package] kernel: add kmod-crypto-user
Aug 17 08:51:10 <CIA-89> Add kernel crypto user-space interface (af_alg). Includes interface for
Aug 17 08:51:10 <CIA-89> hash algorithms (algif_hash) and skcipher algorithms (algif_skcipher).
Aug 17 08:51:10 <CIA-89> Signed-off-by: Lars Hjersted <lars@hjersted.com>
Aug 17 09:17:28 <swalker> [florian]: done
Aug 17 09:18:17 <[florian]> swalker: thanks, now you should be able to update patches status on patchwork
Aug 17 09:18:37 <swalker> thepeople: is rriggio another maintainer that does not have access to the package he's supposedly maintaining?
Aug 17 09:28:51 <swalker> [florian]: the patchwork validation url still needs fixing, iirc someone mentioned the issue before
Aug 17 09:34:42 <CIA-89> swalker * r28027 /packages/utils/dbus/patches/: [packages] dbus: remove empty patches directory
Aug 17 09:36:10 <[florian]> swalker: you are correct
Aug 17 10:15:49 <CIA-89> florian * r28028 /packages/libs/elfutils/Makefile: [package] elfutils: add back libelf, package is marked BROKEN
Aug 17 10:15:51 <CIA-89> florian * r28029 /packages/net/siproxd/ (4 files in 3 dirs):
Aug 17 10:15:51 <CIA-89> [package] siproxd: update to 0.8.1
Aug 17 10:15:51 <CIA-89> Upgrade siproxd, fix init script, link statically with libosip2 (size)
Aug 17 10:15:51 <CIA-89> and add description.
Aug 17 10:15:51 <CIA-89> Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
Aug 17 10:15:52 <CIA-89> florian * r28030 /packages/net/fwknop/ (. Makefile):
Aug 17 10:15:52 <CIA-89> [package] add fwknopd
Aug 17 10:15:52 <CIA-89> Signed-off-by; Jonathan Bennett <jbscience87@gmail.com>
Aug 17 10:15:53 <CIA-89> florian * r28031 /packages/multimedia/minidlna/ (3 files in 2 dirs):
Aug 17 10:15:53 <CIA-89> [package] minidlna: update to 1.0.21
Aug 17 10:15:54 <CIA-89> 010-genconfig-checks.patch adjusts two configure checks to use ICONV_PREFIX and INTL_PREFIX. These changes were already made previously elsewhere in the build process and looks like it was just overlooked here.
Aug 17 10:15:54 <CIA-89> 040-configuration-tweak.patch was to rebase.
Aug 17 10:15:55 <CIA-89> Minidlna's changes were primarily bug fixes and compatibility with newer ffmpeg.
Aug 17 10:15:55 <CIA-89> Signed-off-by: Ian Leonard <antonlacon@gmail.com>
Aug 17 10:26:04 <EqUaTe> anyone else here using oidentd ?
Aug 17 10:27:16 <stintel> o/
Aug 17 10:27:22 <EqUaTe> with trunk?
Aug 17 10:27:38 <stintel> ah hmm didn't see the channel, not sure I am using it on openwrt
Aug 17 10:27:49 <EqUaTe> heh
Aug 17 10:28:14 <stintel> ah yes, seems I am using it on my rspro (ar71xx), running trunk
Aug 17 10:28:31 <EqUaTe> which rev?
Aug 17 10:28:38 <EqUaTe> as i'm finding that it won't forward requests on anymore
Aug 17 10:28:48 <EqUaTe> or do lookups in the local masq file either
Aug 17 10:30:25 <EqUaTe> just building strace to have a look at things and see what i might be missing
Aug 17 10:30:35 <stintel> r27930
Aug 17 10:30:42 <EqUaTe> this forum thread seemed relevant, but doesn't appear to have fixed it for me..
Aug 17 10:30:43 <EqUaTe> https://forum.openwrt.org/viewtopic.php?id=30293
Aug 17 10:30:48 <stintel> but package revision no idea
Aug 17 10:30:59 <stintel> oidentd - 2.0.8-1
Aug 17 10:31:00 <EqUaTe> i'm on 27976
Aug 17 10:31:05 <EqUaTe> yeah, that's the same.
Aug 17 10:31:12 <EqUaTe> forwarding requests works for you?
Aug 17 10:31:18 <stintel> I'll have to test
Aug 17 10:31:46 <stintel>  2574 root       796 S    /usr/sbin/oidentd -m -f 113
Aug 17 10:31:53 <stintel> that's how it appears in ps
Aug 17 10:31:59 <stintel> same for you ?
Aug 17 10:32:00 <EqUaTe> yup, same for me
Aug 17 10:32:13 <EqUaTe> though i've been running it by hand with -i -d as well
Aug 17 10:32:17 <EqUaTe> so i can see the debug messages
Aug 17 10:32:20 <stintel> heh lol I have to open it in my firewall
Aug 17 10:32:23 <stintel> geez
Aug 17 10:32:24 <EqUaTe> lol
Aug 17 10:32:42 <stintel> I guess I am still using oidentd from the day I was ircing from home
Aug 17 10:33:06 <EqUaTe> i still do irc from home.. and run a few other things where having ident is useful
Aug 17 10:33:12 <EqUaTe> eggdrops for example :P
Aug 17 10:33:55 <stintel> yeah
Aug 17 10:34:11 <stintel> but I'm ircing and running eggdrops from colocation now :)
Aug 17 10:34:40 <stijn> fail :)
Aug 17 10:35:01 <EqUaTe> hm
Aug 17 10:35:04 <stintel> food now, bbl
Aug 17 10:35:07 <EqUaTe> kk
Aug 17 12:58:43 <_trine> is there anyone who can explain why the maximum TX power available in current trunk is only 17dBm on the wndr3700
Aug 17 12:59:56 <_trine> it appears the maximum could be as much as 31dBm
Aug 17 13:29:32 <WillieNL> _trine: hmm not upstream modified driver defaults ?
Aug 17 13:38:01 <jow_laptop> _trine: 31dbm is wishful thinking, hw limit is 17
Aug 17 13:38:26 <jow_laptop> _trine: some older versions showed 31dbm but that was the regulatory limit, not the physical/driver limit
Aug 17 13:41:26 <builder_> build #72 of at91 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/72
Aug 17 13:52:35 <Hauke> why doesn't CIA-89 report new commits?
Aug 17 13:53:19 <Hauke> I just got a little confused with the time ;-)
Aug 17 13:56:19 <builder_> build #70 of ubicom32 is complete: Failure [failed compile_3]  Build details are at http://buildbot.openwrt.org:8010/builders/ubicom32/builds/70
Aug 17 13:56:47 <loswillios> Hauke: sometimes he has hickups
Aug 17 14:39:59 <CIA-89> florian * r28032 /packages/libs/elfutils/Makefile: [package] elfutils: rename libelf to libelf1
Aug 17 14:40:02 <CIA-89> florian * r28033 /trunk/package/perf/Makefile: [package] perf: depends on libelf1
Aug 17 14:41:31 <fullstop> ls
Aug 17 14:41:37 <fullstop> ugh, sorry, wrong window
Aug 17 14:43:18 <KanjiMonster> fullstop: popular mistake :P
Aug 17 14:44:21 <builder_> build #47 of mpc52xx is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/47
Aug 17 14:49:04 <fullstop> KanjiMonster: at least it was not a password.  ;-)
Aug 17 15:06:46 <Chocky> hunter2
Aug 17 15:08:10 <KanjiMonster> Chocky: why did you write so many asterisks?
Aug 17 15:08:19 <Chocky> IRC does that automatically
Aug 17 15:08:25 <Chocky> didn't you know?
Aug 17 15:16:35 <loswillios> how can I depend on either package a or b?
Aug 17 15:19:29 <loswillios> @PACKAGE_a||PACKAGE_b works
Aug 17 15:21:21 <jow_laptop> loswillios: check how its propagated down to the .ipk control file, might trip up opkg
Aug 17 15:22:03 <loswillios> ok
Aug 17 15:24:25 <KanjiMonster> loswillios: the problem is that this only works as a "depends"; i.e. one of these packages needs to be selected first; but not as a "selects" (like the normal +<package> dependencies)
Aug 17 15:24:55 <loswillios> yeah I saw it, but that's ok
Aug 17 15:25:32 <KanjiMonster> loswillios: also the "proper", not very well working way is letting both possible packages provide a meta package/feature with "PROVIDES:=<bla>" and then use that as the dependency
Aug 17 15:26:27 <loswillios> interesting, need to look into that
Aug 17 15:27:12 <KanjiMonster> note there is no concept of conflicts or recommendations or default selections for meta-dependencies
Aug 17 16:14:11 <karlp> whee, the package I just got into owrt trunk is now being used by facebook.  (though not on owrt presumable) http://mosquitto.org/2011/08/facebook-using-mqtt/
Aug 17 17:47:53 <dangole> i'm building a simple file upload/download app for luci
Aug 17 17:48:20 <dangole> i'm still novice to lua and try to use part of fon's filebrowser app to realise the thing
Aug 17 17:48:22 <dangole> see http://trac.fonosfera.org/fon-ng/browser/trunk/luci/applications/luci-fonbrowser/luasrc/controller/fon_browser.lua
Aug 17 17:48:24 <dangole> however
Aug 17 17:48:49 <dangole> i got stuck because http.splice(nixio.open(file)) does stream file to the browser
Aug 17 17:48:50 <dangole> but
Aug 17 17:49:30 <dangole> no matter in which order stuff is written in lua, the headers somehow endup after the content instead of before :(
Aug 17 17:50:50 <dangole> my first guess was that the order of execution is resolved using a dependency tree, so i wrapped the instructions in if-clauses to make them dependend and therefore have prepare_content and headers be done before splice
Aug 17 17:50:58 <dangole> but that didn't help either...
Aug 17 17:52:01 <dangole> when using http.write instead, it seems to work, but i don't want to read-buffer-write the whole file
Aug 17 17:52:19 <dangole> would you do that using ltn12?
Aug 18 01:03:01 <builder_> build #72 of s3c24xx is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/72
**** ENDING LOGGING AT Thu Aug 18 02:59:56 2011