**** BEGIN LOGGING AT Sun Dec 12 02:59:58 2010 Dec 12 03:10:14 vchange -a y ? Dec 12 03:10:42 vgchange -a y I mean Dec 12 03:10:54 I think that's what you're looking for Dec 12 03:11:30 in a boot script Dec 12 03:11:58 or maybe it's lvchange -a y Dec 12 03:12:26 try it and see Dec 12 03:15:36 yea its lvchange -ay Dec 12 03:15:48 can i just add it to an init script? Dec 12 03:15:55 jarvis: yes Dec 12 03:15:57 kk Dec 12 03:16:25 cheer Dec 12 03:16:29 *cheers Dec 12 03:16:46 that's essentially all the desktop distros do, it's just in their early boot scripts Dec 12 03:17:40 (i.e. before getting it init or upstart or whatever) Dec 12 03:17:44 *to Dec 12 03:17:57 aye, i just wasn't sure about the boot system in openwrt Dec 12 03:18:10 i've only got around to looking at lvm today Dec 12 03:18:22 jarvis: if you need it before init.d then you need to do a preinit thing Dec 12 03:18:32 but for most uses init.d is fine Dec 12 03:18:40 aye init should be fine Dec 12 03:18:51 cheers :) Dec 12 03:21:58 build #38 of ixp4xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/38 Dec 12 03:25:53 ah i think i found out why .. not i've changed to a jffs2 install, instead of squashfs , the initscript is there .. before when it was squashfs it wasnt Dec 12 03:27:11 jarvis: did you enable it with /etc/init.d/lvm enable ? Dec 12 03:27:23 or build it into the image? Dec 12 03:27:39 gg Dec 12 03:27:51 if you don't do one of those (i.e. install it later but don't enable it) then it won't start Dec 12 03:28:12 ah Dec 12 03:28:15 that'd be why Dec 12 03:28:22 i assumed installing it would automatically enable it Dec 12 03:29:04 cheers for that cshore_ Dec 12 03:31:41 * Jarvis should really write a howto for this Dec 12 03:32:10 please do ! Dec 12 03:32:19 build #38 of kirkwood is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/kirkwood/builds/38 Dec 12 03:32:23 i knew you were gonna say that :P Dec 12 03:33:14 :) I hit everyone up to contribute to the wiki Dec 12 03:39:58 build #37 of mpc52xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/mpc52xx/builds/37 Dec 12 04:09:29 ok, interesting, looking at the init rc layout, i can see a slight issue Dec 12 04:10:00 with the ordering with regards to USB/LVM and fstab Dec 12 04:11:11 S15lvm2 , S20fstab , S39usb Dec 12 04:15:31 ideally , if we want to have roots on a usb drive , shouldn't usb be before all of them? Dec 12 04:15:46 ie S13 USB S15 lvm2 S20fstab Dec 12 04:15:59 ugh its 4am.... Dec 12 04:21:13 if you want rootfs on lvm you will need to hack preinit Dec 12 04:22:27 and S39usb isn't the modules, it some other thing for usb Dec 12 04:22:48 oh? Dec 12 04:22:50 ah Dec 12 04:23:19 isn't usbfs deprecated in later kernels? Dec 12 04:23:38 probably...it might be left over from 2.4 Dec 12 04:23:43 ah k Dec 12 04:26:26 right , thats for that cshore_ i'm off for some sleep before i fall asleep on my keyboard Dec 12 04:40:29 build #44 of ubicom32 is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/ubicom32/builds/44 Dec 12 04:57:46 cshore * r24510 /trunk/tools/sstrip/src/sstrip.c: [tools] sstrip: Was missing include of Dec 12 05:00:31 cshore * r24511 /trunk/tools/sstrip/src/sstrip.c: Dec 12 05:00:31 Revert "[tools] sstrip: Was missing include of " Dec 12 05:00:31 This reverts commit bcb48716d097c16834f222c267c73e847f4a8dd0. Dec 12 05:03:19 cshore * r24512 /trunk/tools/sstrip/Makefile: [tools] sstrip: Correct include path to really get endian.h Dec 12 05:22:33 build #37 of au1000 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/37 Dec 12 05:33:38 build #37 of iop32x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/iop32x/builds/37 Dec 12 05:51:25 cshore * r24513 /trunk/tools/Makefile: [tools] Makefile: Put lzma before squashfs so compile of squashfs4 succeeds (depends on lzma) Dec 12 06:14:33 build #50 of brcm63xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/50 Dec 12 06:24:39 build #37 of rdc is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/37 Dec 12 06:28:57 build #42 of ps3 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/42 Dec 12 06:43:04 build #44 of pxcab is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/pxcab/builds/44 Dec 12 09:04:13 * russell-- ran into a couple blocking build failures libxkbfile (can't find libXau) and another one X related Dec 12 09:05:36 libXfixes is the other one Dec 12 09:29:02 was missing libXau.la, building it manually seems to fix it Dec 12 10:25:11 cat /aux/src/openwrt/build_dir/linux-brcm47xx/vmlinux | /aux/src/openwrt/staging_dir/host/bin/lzma e -si -so -eos -lc1 -lp2 -pb2 > /aux/src/openwrt/build_dir/linux-brcm47xx/vmlinux.lzma Dec 12 10:25:16 /aux/src/openwrt/staging_dir/host/bin/lzma: invalid option -- 's' Dec 12 10:28:05 russell--: -h ?! :) Dec 12 10:33:49 wait, what? Dec 12 10:37:44 russell@frieda[/src/openwrt] $ /aux/src/openwrt/staging_dir/host/bin/lzma -V Dec 12 10:37:47 xz (XZ Utils) 5.0.0 Dec 12 10:37:56 russell--: make tools/lzma/{clean,install} Dec 12 10:38:04 i fixed that one yesterday Dec 12 10:38:43 or maybe tools/lzma-old/{clean,install} Dec 12 10:42:13 i think the first one did it, i'm rebuilding now Dec 12 10:53:52 fixt, danke schön Dec 12 10:59:33 build #46 of s3c24xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/s3c24xx/builds/46 Dec 12 11:59:16 build #37 of octeon is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/octeon/builds/37 Dec 12 13:15:06 is there an easy way to xor a file with a value? Dec 12 13:28:49 <[florian]> cshore_: ping Dec 12 13:31:16 <[florian]> cshore_: r24512 breaks sstrip compilation on most systems, what's the reason for your change, does it fix something for you? Dec 12 14:32:20 build #49 of atheros is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/atheros/builds/49 Dec 12 14:35:01 build #47 of brcm47xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx/builds/47 Dec 12 14:52:23 build #39 of uml is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/uml/builds/39 Dec 12 14:52:45 agb * r24514 /packages/libs/libsamplerate/Makefile: [packages] libs/libsamplerate: use autoreconf PKG_FIXUP Dec 12 14:52:50 agb * r24515 /packages/utils/triggerhappy/Makefile: [packages] utils/triggerhappy: fix path to linux/input.h Dec 12 14:54:06 florian * r24516 /trunk/tools/sstrip/Makefile: Dec 12 14:54:06 Revert "[tools] sstrip: Correct include path to really get endian.h" (#8447) Dec 12 14:54:06 This reverts commit ac0f793ae985db73caf99916336569c211a759f8. Dec 12 16:53:44 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/uscan.shtml Dec 12 17:31:35 build #48 of ppc40x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc40x/builds/48 Dec 12 17:38:43 jow * r24517 /packages/utils/collectd/patches/ (2 files): [packages] collectd: fix bogus -ltdl dependency Dec 12 18:00:58 jow * r24518 /packages/sound/lame/patches/ (. 001-automake-compat.patch): [packages] lame: fix autoreconf Dec 12 18:49:08 jow * r24519 /packages/libs/libcroco/patches/ (. 001-automake-compat.patch): [packages] libcroco: fix autoreconf Dec 12 18:55:01 jow * r24520 /packages/libs/libinklevel/patches/001-configure.patch: [packages] libinklevel: fix autoreconf Dec 12 18:56:38 * Jarvis thinks hostapd has a reauth bug :( Dec 12 19:13:03 jow * r24521 /packages/Xorg/lib/libast/ (3 files in 2 dirs): [packages] libast: fix autoreconf Dec 12 19:38:21 jow * r24522 /packages/libs/libmikmod/patches/ (002-missing-macros.patch 003-automake-compat.patch): [packages] libmikmod: fix autoreconf Dec 12 19:52:20 jow * r24523 /packages/libs/libmpcdec/patches/001-automake-compat.patch: [packages] libmpcdec: fix autoreconf Dec 12 20:06:49 hi Dec 12 20:07:21 does openwrt include runlevels? did i just didnt find them or are there no runlevels? Dec 12 20:07:30 there are none Dec 12 20:08:03 actually they exist but are not used at all (just boot and halt) Dec 12 20:08:11 i think runlevels as a concept is broken by design Dec 12 20:08:23 hmm is there a way to determine if a init script is stopped by the user or while the system powers off or reboots? Dec 12 20:08:42 not yet, but it's easy to add Dec 12 20:08:51 nbd: how? Dec 12 20:09:12 nbd: systemd is about to solve that Dec 12 20:09:26 systemd ?! Dec 12 20:09:49 whats systemd? Dec 12 20:10:21 tripolar: /etc/inittab calls /etc/init.d/rcS K stop Dec 12 20:10:27 make it call rcS K shutdown instead Dec 12 20:10:35 and add a shutdown() wrapper that calls stop to rc.common Dec 12 20:10:49 that way init scripts can override the shutdown() implementation if they want special handling for that Dec 12 20:11:30 http://www.freedesktop.org/wiki/Software/systemd Dec 12 20:12:43 nbd but all initscipt would need a shutdown routine? Dec 12 20:13:10 no, read again what i wrote Dec 12 20:13:11 tripolar: 21:09 < nbd> and add a shutdown() wrapper that calls stop to rc.common Dec 12 20:13:14 :) Dec 12 20:13:29 :) Dec 12 20:19:03 systemd will be as cool as pulseaudio Dec 12 20:19:43 xMff: you mean the seamless integration in openwrt ;-) Dec 12 20:20:46 yes, and the beautiful look of all that bloat... like a big shitpile shining in the moonlight Dec 12 20:20:46 xMff: :D Dec 12 20:22:23 well, maybe systemd is small if we remove dbus support and replace it with ubus Dec 12 20:22:26 ;) Dec 12 20:25:22 nbd: and the rest with something else we wrote Dec 12 20:25:30 :) Dec 12 20:26:00 nbd i tried it there is allready a shutdown routine in rc.common i just had to replace the return 0 with a stop could u make/commit this in trunk? Dec 12 20:26:55 ah, didn't know that we already had a shutdown Dec 12 20:27:04 maybe we shouldn't even make it default to using stop() then Dec 12 20:27:17 i'm thinking that most services actually don't need stopping on shutdown Dec 12 20:27:26 and those that do can specify it explicitly Dec 12 20:27:46 yeah but every prog that doesnt need special handling could stay as it is and use the stop routine Dec 12 20:29:07 depends on what you define as special handling Dec 12 20:29:15 i need it for sshd Dec 12 20:29:17 i consider the need for a graceful stop on shutdown special handling Dec 12 20:29:17 openssh Dec 12 20:33:14 nbd: but every prog that has a STOP= line in its init script would need a shutdown routine Dec 12 20:34:54 hm Dec 12 20:35:05 oh, right. we removed STOP= for most packages Dec 12 20:35:08 ok, add the wrapper then Dec 12 20:36:50 nbd i send the patch to the ml Dec 12 20:37:00 ok Dec 12 20:38:22 the biggest patch ever - two lines ;) Dec 12 20:38:28 sent it Dec 12 20:45:01 tripolar: signed-off-by missing *whine* ;) Dec 12 20:45:21 :) Dec 12 20:46:09 I don't have any commit rights, so feel free to ignore me ;) Dec 12 20:47:33 KanjiMonster what for is the signed-off-by needed? Dec 12 20:47:37 i never used it Dec 12 20:48:39 tripolar: afaik its for keeping a record where some code came from; it gets part of the commit message (usually) Dec 12 20:50:36 okay Dec 12 20:52:34 nbd did u look at it? Dec 12 21:11:19 tripolar: did u send this to the mailing list ? Dec 12 21:11:50 OutBackDingo yes Dec 12 21:12:31 OutBackDingo: why? Dec 12 21:13:38 tripolar: just curious as to what the modification was, i see its a single like Dec 12 21:14:03 loswillios: whos working on the systemd ? Dec 12 21:14:33 tripolar: u have a need for shutdown on an AP ? Dec 12 21:14:41 yes Dec 12 21:14:55 openssh for example need differnt treading for stop and shutdown Dec 12 21:15:00 need=needs Dec 12 21:15:03 tripolar: got ya Dec 12 21:16:01 OutBackDingo: and as there are no runlevels on openwrt it cant be determined in stop if stop is called from user or on shutdown/reboot Dec 12 21:32:55 nbd did u look at it? Dec 12 21:41:49 agb * r24524 /branches/backfire/include/unpack.mk: Dec 12 21:41:49 [backfire] backport r22517 Dec 12 21:41:49 [include] support unpacking of .tar.xz archives, no prereq on xzcat for now Dec 12 21:46:23 build #43 of rb532 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/rb532/builds/43 Dec 12 21:58:18 hi i got an device here where i am not sure if there is alinux installed, anyone an idea how to find out more, if it has to be relesed open source ? Dec 12 22:02:59 phaidros: try to nmap it Dec 12 22:03:10 nmap -O ipofdevice Dec 12 22:06:58 OutBackDingo: you mean integration into openwrt? nobody afaik Dec 12 22:07:22 loswillios: as ok Dec 12 22:07:26 ahhh Dec 12 22:08:49 fedora would be pissed if openwrt'd be the first distro which shipped systemd :p Dec 12 22:09:10 loswillios: im looking at the feasibility of it, and screw federa :) Dec 12 22:10:07 reminds me a bit of launchd Dec 12 22:12:17 I've seen someone on the systemd-ml integrating it into his "own embedded" distro Dec 12 22:18:51 loswillios: its and interesting concept, though for what it does, not sure its worth it on an embedded system Dec 12 22:29:30 jow * r24525 /packages/net/znc/files/znc.init: [packages] znc: fix script error in r24218 Dec 12 22:58:00 blogic * r24526 /trunk/ (66 files in 14 dirs): [lantiq] adds new lantiq kernel. once the codebase is fully tested and know to be working on all the devices previously supported by ifxmips, we will drop ifxmips support. Dec 12 23:17:58 jow * r24527 /packages/libs/liboil/patches/ (001-disable-docs-tests-examples.patch 001-no_docs.patch): [packages] liboil: disable docs, tests and examples - solves -lintl linking error spotted by buildbot Dec 12 23:23:05 build #43 of ramips is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ramips/builds/43 Dec 12 23:33:09 jow * r24528 /packages/libs/flac/patches/002-no-utility.patch: [packages] flac: disable command line utilities, they're not packaged anyway and introduce libiconv dependencies Dec 12 23:41:52 urgh....can't get the darn thing to compile the toolchain....it's not doing uClibc before gcc/final Dec 12 23:43:16 anyone else having such troubles with a clean checkout? Dec 12 23:46:39 I've been fighting this since yesterday....the problem is I don't have a clue what could be wrong, as I don't see anything in the commit logs to explain it. Dec 12 23:47:56 gcc/final seems to ignoring the dependency on uClibc for some reason Dec 12 23:51:39 did you update your host system recently? Dec 12 23:51:53 other weird errors have been popped up lately Dec 12 23:52:06 yes, it's debian Dec 12 23:52:22 was there a libtool update recently? :) Dec 12 23:52:28 that broke mpfr for example Dec 12 23:53:16 xMff: I didn't notice it, but you don't happen to know of a place to check the update history in debian, do you? Dec 12 23:53:39 no but you could check the modtimes of stuff in /var/lib/apt/cache Dec 12 23:57:48 jow * r24529 /packages/libs/xmlrpc-c/patches/ (3 files): [packages] xmlrpc-c: fix autoreconf Dec 12 23:58:50 perl-base, gcc and a ton of libraries that are things that could be relevant I guess Dec 12 23:59:23 also can you supply a log? Dec 12 23:59:24 oh and base-files Dec 12 23:59:35 yep Dec 13 00:00:02 generating log now Dec 13 00:00:36 I have BUILD_LOG logs, but I haven't done a V=99 in a while Dec 13 00:01:32 You should not have jested about upating ubuntu to introduce breakage....I think you're being punished :P Dec 13 00:04:42 lovely, got a configure here which produces coredumps while it runs Dec 13 00:04:59 its then unable to remove them... Dec 13 00:05:40 oh nvm Dec 13 00:05:58 they just have a directory called core and autofail tries to clean it up Dec 13 00:06:18 hmmm pastebin doesn't like the build.log Dec 13 00:06:43 jow * r24530 /packages/multimedia/peercast/Makefile: [packages] peercast: a new twist, they hid their libtool in acinclude.m4, kill it as well Dec 13 00:07:14 xMff: it's been a wonder znc compiled until now, -> https://dev.openwrt.org/browser/packages/net/znc/Makefile#L48 - there's no \ at the end of the file (note: adding it breaks configuration) Dec 13 00:07:24 file->line Dec 13 00:07:40 heh Dec 13 00:07:54 got a log? Dec 13 00:08:02 compiling c++ makes my cpu melt Dec 13 00:08:25 xMff: I think I'm going to have to trim my build log...pastebin is complaining it's too big Dec 13 00:08:36 xMff: it can't find libcrypto.0.9.8.so or something when testing whether the g++ compiler works Dec 13 00:08:55 PKG_BUILD_DEPENDS:=libgcrypt Dec 13 00:08:59 or opensl ? Dec 13 00:09:01 *openssl Dec 13 00:09:12 0.9.8 sounds like openssl Dec 13 00:10:14 cshore_: no webspace for uploading? Dec 13 00:10:37 cshore_: you can get a userdir on luci.subsignal.org if that helps Dec 13 00:11:57 xMff: I've not had my web server up for a few months....this build is actually working on setting up a router as my web server with a USB drive for archive storage. My ISP has webspace but you have to get them to enable it and I haven't done that yet. Dec 13 00:12:44 xMff: I suppose I could use GoogleDocs for that matter Dec 13 00:16:14 xMff: adding -lcrypto to the libs was enough (openssl is already a dependency) Dec 13 00:17:26 now let's see if it actually compiles... Dec 13 00:17:50 cshore_: okay, whatever works Dec 13 00:18:10 xMff: https://docs.google.com/uc?id=0B07LcVZMfvdkYjNkZGQ0ZjctYmNjNS00ODNiLTgzZjMtMDY3OTk0NzA5MTgx&export=download&authkey=CPGs65IE&hl=en - it's just because I already have the account and everything Dec 13 00:18:53 sorry https://docs.google.com/leaf?id=0B07LcVZMfvdkYjNkZGQ0ZjctYmNjNS00ODNiLTgzZjMtMDY3OTk0NzA5MTgx&sort=name&layout=list&num=50 Dec 13 00:19:46 I really want to get my web server up and running so I can have real urls Dec 13 00:20:24 Sorry, the page (or document) you have requested is not available. Dec 13 00:21:01 the first link worked Dec 13 00:21:15 oh, strange Dec 13 00:21:42 as long as you got it, is all I care about Dec 13 00:27:38 make[3]: Leaving directory `/home/daniel/Build/openwrt/fionavar/celidon-20101211/openwrt/toolchain/gcc/initial' Dec 13 00:27:38 make[3]: Entering directory `/home/daniel/Build/openwrt/fionavar/celidon-20101211/openwrt/toolchain/gcc/final' Dec 13 00:27:38 is the problem...it should be doing uClibc in there AIUI Dec 13 00:28:42 nbd: ^ Dec 13 00:29:47 are you sure you don't have any issues in your .config? Dec 13 00:29:58 nbd: like? Dec 13 00:30:22 it is an old config I think Dec 13 00:30:25 CONFIG_LIBC containing crap Dec 13 00:30:37 never use old configs with new systems, always run at least make oldconfig Dec 13 00:30:48 oh, that I did Dec 13 00:30:50 hm Dec 13 00:31:01 so what's CONFIG_LIBC set to? Dec 13 00:31:19 "uClibc" Dec 13 00:31:21 hm Dec 13 00:32:06 can't find any reason for this issue Dec 13 00:32:13 try a fresh config Dec 13 00:32:43 ok, I couldn't see anything in the Makefiles that would explain it, unless the dependencies are being ignored Dec 13 00:33:45 of course I haven't figured out the conversion of x/x/compile to a real make target yet Dec 13 00:37:15 jow * r24531 /packages/net/peerguardian/ (3 files in 2 dirs): [packages] peerguardian: libnetfilter-queue-libipq has been dropped upstream, switch back to libipq Dec 13 00:41:06 that's all done by include/subdir.mk Dec 13 00:48:53 build #47 of at91 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/at91/builds/47 Dec 13 00:49:17 jow * r24532 /packages/libs/radiusclient-ng/patches/001-cross_compile.patch: [packages] radiusclient-ng: fix autoreconf Dec 13 00:55:15 ok, new .config worked Dec 13 00:55:58 xMff: that fixed the pulled in dependency to libstdcpp, now I have to test that it works Dec 13 01:20:54 jow * r24533 /packages/net/yafc/ (Makefile patches/): [packages] yafc: force full autoreconf, fix cross compilation after reconf Dec 13 01:35:52 yo yo yo, brcm47xx does not boot a wrt350n!!! why oh why? Dec 13 01:45:12 jow * r24534 /trunk/include/autotools.mk: [include] autotools.mk: support autoreconf at different paths, introduce PKG_AUTOMAKE_PATHS which defaults to '.' Dec 13 01:48:20 jow * r24535 /trunk/package/ead/src/tinysrp/ (acinclude.m4 aclocal.m4): [package] ead: fixup some AC_DEFINE() invocations that will make recent automak bail Dec 13 01:49:05 xMff: you're up late... Dec 13 01:49:30 yeah Dec 13 01:50:16 what's wrong with brcm47xx? Dec 13 01:50:28 who broke it? Dec 13 01:50:46 idk, I guess Hauke is your man Dec 13 01:51:03 personally I'm solely focused on libtool/automake issues right now Dec 13 01:52:09 grrr, he seems to be offline Dec 13 01:54:00 yeah, he's in CET Dec 13 01:55:57 build #41 of ar7 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/41 Dec 13 02:02:40 jow * r24536 /packages/libs/libmpd/Makefile: [packages] libmpd: fix path to libiconv and libintl Dec 13 02:10:12 i'll bbiab and hopefully someone will know then what happened Dec 13 02:22:59 jow * r24537 /packages/libs/libvncserver/patches/ (001-automake-compat.patch 100-noexamples.patch): [packages] libvncserver: fix autoreconf Dec 13 02:50:29 jow * r24538 /packages/net/peerguardian/Makefile: [packages] peerguardian: I accidentally marked it as broken, remove this dep **** ENDING LOGGING AT Mon Dec 13 02:59:58 2010