**** BEGIN LOGGING AT Sat Feb 24 02:59:58 2007
Feb 24 03:01:05 <CIA-6> 03nail * r5593 10optware/trunk/platforms/packages-ts101.mk: ts101-packages: Remove a bunch of broken packages
Feb 24 03:10:02 <CIA-6> 03nail * r5594 10optware/trunk/platforms/packages-ts101.mk: ts101-packages: remove uclibc-opt
Feb 24 03:11:00 <CIA-6> 03nail * r5595 10optware/trunk/platforms/packages-ts101.mk: ts101-packages: remove buildroot
Feb 24 03:13:36 <CIA-6> 03nail * r5596 10optware/trunk/platforms/packages-ts101.mk: ts101-packages: remove lhda
Feb 24 03:18:49 <CoreDump|home> mwester: ;)
Feb 24 03:18:52 <CoreDump|home> NAiL: ACK
Feb 24 03:19:01 <CoreDump|home> n8 all
Feb 24 03:19:10 <CIA-6> 03bzhou * r5597 10optware/trunk/platforms/packages-ts101.mk: promoted bsdmainutils, swi-prolog, libnetfilter*, moblock, man for ts101
Feb 24 03:42:57 <CIA-6> 03bzhou * r5598 10optware/trunk/make/snownews.mk: snownews: stage, link & depend on libiconv if standalone libiconv
Feb 24 03:53:45 <rwhitby> ts101 feed is uploading now
Feb 24 03:58:50 <CIA-6> 03bzhou * r5599 10optware/trunk/platforms/packages-ts101.mk: promoted snownews for ts101 with libiconv fix
Feb 24 04:02:05 <CIA-6> 03bzhou * r5600 10optware/trunk/platforms/packages-ts101.mk: promoted snownews for ts101 with libiconv fix
Feb 24 04:03:05 <CIA-6> 03bzhou * r5601 10optware/trunk/platforms/packages-ts101.mk: 
Feb 24 04:03:05 <CIA-6> promoted for ts101:
Feb 24 04:03:05 <CIA-6> ed, flip, id3lib, irssi, lsof, man python25, py-simpy, vim lha;
Feb 24 04:03:05 <CIA-6> all builds fine on nudi
Feb 24 04:10:09 <CIA-6> 03bzhou * r5602 10optware/trunk/ (make/coreutils.mk platforms/packages-ts101.mk): coreutils: stage, link & depend on libiconv if standalone libiconv; promoted for ts101
Feb 24 04:20:40 <CIA-6> 03bzhou * r5603 10optware/trunk/platforms/packages-ts101.mk: ruby: promoted for ts101
Feb 24 05:02:56 <CIA-6> 03bzhou * r5604 10optware/trunk/ (make/git-core.mk platforms/packages-ts101.mk): git-core: stage, link & depend on libiconv if standalone libiconv; promoted for ts101
Feb 24 05:25:52 <CIA-6> 03bzhou * r5605 10optware/trunk/platforms/packages-ts101.mk: promoted for ts101: flac hexcurse netcat nmap ntop php* rc upslug2
Feb 24 05:46:16 <CIA-6> 03bzhou * r5606 10optware/trunk/platforms/packages-ts101.mk: promoted for ts101: transcode vlc zip
Feb 24 10:25:11 <CIA-6> 03blaster8 * r727 10kernel/trunk/patches/2.6.18/ (5 files): Update to 2.6.18.8, 2.6.18 no longer maintained
Feb 24 10:26:24 <CIA-6> 03blaster8 * r728 10kernel/trunk/README: Update README with Bugzilla link (2.6.21-rc1 build failure)
Feb 24 10:30:04 <CIA-6> 03blaster8 * r729 10kernel/trunk/patches/2.6.19/ (5 files): Cleanup 2.6.19 directory
Feb 24 11:11:48 <CIA-6> 03blaster8 * r730 10kernel/trunk/ (7 files in 2 dirs): Backport Redboot parser changes from 2.6.21-rc1 to 2.6.20, compile but NOT runtime tested
Feb 24 11:40:08 <CIA-6> 03blaster8 * r731 10kernel/trunk/ (5 files in 2 dirs): Backport kexec from 2.6.21-rc1 to 2.6.20
Feb 24 13:10:08 <CIA-6> 03blaster8 07ts101_support * r732 10kernel/ (23 files in 4 dirs): Add ts101_support branch
Feb 24 13:12:35 <CIA-6> 03blaster8 07ts101_support * r733 10kernel/: Update svn_ignore property for ts101_support branch
Feb 24 13:14:15 <CIA-6> 03blaster8 * r734 10kernel/trunk/ (Makefile ts101/): Remove ts101 support in trunk, fully migrated to ts101_support branch
Feb 24 13:18:33 <CIA-6> 03blaster8 * r735 10kernel/trunk/Makefile: Revert trunk Makefile
Feb 24 14:00:50 <CIA-6> 03blaster8 07ts101_support * r736 10kernel/Makefile: Swap to powerpc ARCH, build/copy uImage
Feb 24 15:06:54 <CIA-6> 03nail * r5607 10optware/trunk/sources/ts101-kernel-modules/ (. kernel.defconfig): ts101-kernel-modules: Add defconfig
Feb 24 15:10:44 <CIA-6> 03nail * r5608 10optware/trunk/make/ts101-kernel-modules.mk: ts101-kernel-modules: Add ts101-kernel-modules package. It doesn't actually build any specific modules quite yet, that will have to be enabled as people want modules
Feb 24 15:11:49 <CIA-6> 03nail * r5609 10optware/trunk/platforms/packages-ts101.mk: packages-ts101: Add ts101-kernel modules to the build
Feb 24 15:24:43 <CIA-6> 03nail 07ts101_support * r737 10kernel/Makefile: Makefile: Fixup vmlinuz/uImage stuff
Feb 24 15:49:10 <CIA-6> 03blaster8 07ts101_support * r738 10kernel/ (. .cvsignore): Update svn:ignore for uImage fixup
Feb 24 15:59:00 <CIA-6> 03blaster8 07ts101_support * r739 10kernel/Makefile: powerpc => ppc (I'll stop interfering now)
Feb 24 16:01:43 <CIA-6> 03nail 07ts101_support * r740 10kernel/Makefile: Fix toolchain breakage caused by interference ;)
Feb 24 17:10:41 <eno> NAiL: libgcrypt is the key to unbreak more packages for ts101
Feb 24 17:12:44 <eno> compared to ds101g, mpi/.libs/libmpi.a exported  __gcry_mpih_* instead of _gcry_mpih_*
Feb 24 17:12:57 <eno> any idea why?
Feb 24 17:29:06 <CoreDump|home> hi
Feb 24 17:29:26 <jason0_> How bad would it be to use a usb stick for swap & /var/log on an nslu2?
Feb 24 17:30:08 <NAiL> eno: Not the faintest
Feb 24 17:44:15 <eno> ah, there's a sources/libgcrypt/symbol-underscore.patch applied for ds101g
Feb 24 17:44:34 <eno> yep, that does it
Feb 24 17:44:44 <NAiL> ah, thanks
Feb 24 17:46:12 <NAiL> Will you fix it or should I?
Feb 24 17:47:51 <CIA-6> 03bzhou * r5610 10optware/trunk/make/libgcrypt.mk: libgcrypt: apply symbol-underscore.patch if powerpc
Feb 24 18:35:57 <CIA-6> 03bzhou * r5611 10optware/trunk/platforms/packages-ts101.mk: promoted for ts101: libgcrypt, gnutls, cherokee, lighttpd, loudmouth, ikselmel, asterisk14, wput
Feb 24 18:49:18 <CIA-6> 03bzhou * r5612 10optware/trunk/platforms/packages-ts101.mk: promoted for ts101: dcraw, mpd, nzbget pcre, postfix
Feb 24 19:01:09 <hillct> is there a bitbake recipie that can be run directly to build a slugosbe image?
Feb 24 19:01:24 <hillct> rather than building using the mastermakefile target?
Feb 24 19:01:37 * mwester wonders why someone would wish to do such a thing
Feb 24 19:02:19 <mwester> You'd have to read the MasterMakefile; I expect it does eventually resolve to a bitbake target, after doing much setup of the environment and other files.
Feb 24 19:03:11 <hillct> yeah
Feb 24 19:03:43 <hillct> I ws looking but there's not a clearcut recipie that can be called separately as far as I can tell
Feb 24 19:03:54 <hillct> the master makefile will work fine
Feb 24 19:04:05 <hillct> I just wanted to cut a few hours off the build
Feb 24 19:04:14 <hillct> by using the existing toolchain
Feb 24 19:04:23 <hillct> such is life
Feb 24 19:04:47 <mwester> Ah - now *that* I understand!  I too have wished for such a split -- toolchain separate from target
Feb 24 19:05:35 <hillct> well, I did document the release-specific toolchain build stuff last week
Feb 24 19:06:08 <hillct> the one remaining issue is, the makefile target setup-slugosbe doesn't include creation of topdir.conf
Feb 24 19:06:25 <mwester> I've spent my life in environments where once you have a stable, tested toolchain, it does not change; OE has taken a great deal of "getting used to" for me.  The idea that a minor patch by someone not related to my project can, without my knowlege, result in my toolchain building anew is distressing.
Feb 24 19:07:11 <hillct> it is an interesting approach to build management
Feb 24 19:07:23 <mwester> or lack thereof
Feb 24 19:07:34 <hillct> I can see where such an approach would be valuable in an enterprise enviroment
Feb 24 19:07:51 <hillct> but for OSS, it does seem to be an odd choice
Feb 24 19:10:16 <mwester> It has a certain "you *WILL* run the most recent everything" feel to it, which may be good for early development stages, and for prototyping, but it lacks a great deal when one is in the final development/release process.
Feb 24 19:10:33 <mwester> I'm very curious to see how this is handled with OpenMoko.
Feb 24 19:20:36 <hillct> mwester: who I need ot assign an OE ticket to, in order to get my new package recipe added to the slugOS feeds?
Feb 24 19:22:38 <hillct> mwester: http://bugs.openembedded.org/show_bug.cgi?id=1894 http://bugs.openembedded.org/show_bug.cgi?id=1895
Feb 24 19:48:10 <CoreDump|home> !oebug 1895
Feb 24 19:48:12 <cdbot2> * * Bug 1895, Status: NEW, Created: 2007-02-20 15:44
Feb 24 19:48:13 <cdbot2> * * hillct(AT)pinnacledigital.com: New Package: logrotate - binary implementation
Feb 24 19:48:13 <cdbot2> * * http://bugs.openembedded.org/show_bug.cgi?id=1895
Feb 24 19:49:32 <CoreDump|home> hillct: If no one beats me to it, I'll add it to OE. Dunno how to go about the slug feeds tho
Feb 24 19:50:50 <hillct> CoreDump|home: thanks
Feb 24 19:51:03 <Sinclair73de> when make slugosle-image i got ERROR: You must set MACHINE to nslu2le or nslu2be, any idea?
Feb 24 19:52:09 <mwester> hillct: NAiL handles the SlugOS feeds
Feb 24 19:52:51 <mwester> Sinclair73de: just "make slugosle" -- that will set the correct config file to set MACHINE to nslu2le for you.
Feb 24 19:53:44 * mwester didn't know we could do "oebug" on this channel, too
Feb 24 19:54:31 <Sinclair73de> mwester, same error
Feb 24 19:55:41 <mwester> Odd.  There was a short time where that happened, but if you have done a "make update" recently, it should be ok.
Feb 24 19:57:46 <mwester> Have you cleaned out your build directory, and done a complete rebuild since the OpenSlug->SlugOS name change happened?
Feb 24 19:59:09 <hillct> mwester, CoreDump|home I don't know if you were on when last I asked about this packaging error, but I can't seem to find the appropriate enviroment variable to set to get the proper include path in to place here: http://www.pastebin.ca/368351
Feb 24 20:02:14 <Sinclair73de> yes, i build slugosle-image on 2007-02-11, i try make update again...
Feb 24 21:00:57 <CoreDump|home> hillct: I've pushed your ntpclient and logrotate recipes
Feb 24 21:01:10 <hillct> thanks
Feb 24 21:01:22 <CoreDump|home> dunno about the PAM thingy tho
Feb 24 21:01:23 <CoreDump|home> np
Feb 24 21:01:50 <hillct> how exactly does that work? Is there a review process or do they now show up for anyone who does an openembedded update?
Feb 24 21:02:33 <hillct> the pam thing is really odd
Feb 24 21:03:08 <hillct> I gather the error message is reporting the wrong thing, or the variable is being overwritten somewhere in the build process
Feb 24 21:05:20 <CoreDump|home> hillct: they will show up for everyone after they run 'mtn pull' to update their local db
Feb 24 21:05:38 <hillct> cool
Feb 24 21:06:14 <hillct> I tested them but here's hoping I didn't screw them up too badly :)
Feb 24 21:08:00 <CoreDump|home> =)
Feb 24 21:08:38 <CoreDump|home> if they don't work good enough, we both will feel the wrath of "koen" hehe
Feb 24 21:14:30 <rwhitby> CoreDump|home: packages/meta/slugos-packages.bb is what automatically goes into the feeds.
Feb 24 21:14:43 <rwhitby> so if you add them to there they will show up.
Feb 24 21:15:22 <CoreDump|home> rwhitby: that's good to know! However, I'd rather not mess w/ the slug feeds just now heh
Feb 24 21:16:12 <rwhitby> it's ok - it adds to the unstable feed.  people using the binary images are pointed to the stable feed, which NAiL updates manually after the package is known good.
Feb 24 21:16:28 <rwhitby> so please do add it immediately to slugos-packages, so we can get some testing by users.
Feb 24 21:16:37 <CoreDump|home> ok then
Feb 24 21:25:09 <CoreDump|home> hillct: rwhitby pushed
Feb 24 21:31:51 * CoreDump|home continues working on his IRC log parser
Feb 24 21:41:32 <[g2]> CoreDump|home what kind of parser are you doing ?
Feb 24 21:42:31 <CoreDump|home> [g2]: Just adding some formatting to text/plain logfiles -> http://hentges.net/tmp/logs/irc/livelogs/%23nslu2-linux.livelog
Feb 24 21:43:34 <hillct> ooh. pretty!
Feb 24 21:43:58 <[g2]> CoreDump|home nice. what is the parser written in ?
Feb 24 21:44:12 <CoreDump|home> bash hehe
Feb 24 21:44:20 <CoreDump|home> well, sed mostly
Feb 24 21:44:21 <hillct> idea. It would be neat to link each nick to the homapage and/or email address recorded by nickserv
Feb 24 21:44:37 <[g2]> how are you collecting the logs ?
Feb 24 21:44:46 <CoreDump|home> [g2]: cdbot2 ;)
Feb 24 21:45:07 <[g2]> CoreDump|home that's a cmdline bot ?
Feb 24 21:45:14 <CoreDump|home> hillct: interesting idea but that would be kinda overkill =)
Feb 24 21:45:34 <CoreDump|home> ?? help > [g2] 
Feb 24 21:45:36 <cdbot2> CoreDump|home: User "\[g2\]" is not in this channel
Feb 24 21:45:42 <CoreDump|home> hehe
Feb 24 21:45:48 <[g2]> ~[g2]
Feb 24 21:45:50 <CoreDump|home> well, its an eggdrop
Feb 24 21:45:54 <jbot> rumour has it, [g2] is really cool, and an an all around nice guy!  Also he is a real funny guy, the master of NSLU2 Buildroot and a known Slug Sacrificer.  And editor of SlugNews.
Feb 24 21:47:10 <[g2]> ?? help > CoreDump|home 
Feb 24 21:47:44 <CoreDump|home> [g2]: he doesn't like your brackets :\
Feb 24 21:47:49 <[g2]> nod
Feb 24 21:48:01 <CoreDump|home> but only redirection should be affected
Feb 24 21:50:03 <[g2]> so cdbot's in OE ?
Feb 24 21:51:12 <CoreDump|home> it's a normal eggdrop, nothing fancy
Feb 24 21:51:21 * [g2] comes up pretty short on google for cdbot2
Feb 24 21:51:39 <CoreDump|home> google for eggdrop
Feb 24 21:51:53 * [g2] is familiar with eggdrop
Feb 24 21:53:08 <CoreDump|home> ?? help-database
Feb 24 21:53:08 <cdbot2> help-database[1]: Valid commands are: !learn !del !forget !lock !unlock !learnlist !locklist !lhelp 
Feb 24 21:53:10 <cdbot2> help-database[2]: Only -o users may use !lock !unlock !locklist 
Feb 24 21:53:10 <cdbot2> help-database[3]: ?? <TERM> prints the definition of <TERM> into the channel 
Feb 24 21:53:12 <cdbot2> help-database[4]: ?? <TERM> > <NICK> sends the definition of <TERM> to <NICK> via /msg 
Feb 24 21:53:14 <cdbot2> help-database[5]: ?? <TERM>:<INDEX> prints only entry # <INDEX> of <TERM>. Works with redirection as well. 
Feb 24 21:53:16 <cdbot2> help-database[6]: ?? <TERM WITH SPACE> is handled internally as <TERM_WITH_SPACE>, so to delete the term 'altboot manual' you'd have to !forget altboot_manual 
Feb 24 21:53:19 <cdbot2> help-database[7]: !insert <TERM> inserts another definition to a known term. 
Feb 24 21:53:20 <cdbot2> help-database[8]: !insert <TERM>:<INDEX> inserts a definition for <TERM> after <INDEX> 
Feb 24 21:54:56 <CoreDump|home> !learn cdbot-bugs Searchterm redirection is broken for nicknames containing filtered chars (brackets)
Feb 24 21:54:58 <cdbot2> Added "cdbot-bugs" to database
Feb 24 21:55:08 <hillct> this is not cinfidence inspiring. Building slugosbe-image, it appears that I've not updated prior to the build or something, yet I have... http://www.pastebin.ca/370829
Feb 24 21:56:11 <hillct> is there some special kind of update I need to do besides the usual, to get this to do the final image construction step?
Feb 24 21:57:07 <CoreDump|home> hillct: bitbake slugos-image w/ propper MACHINE and DISTRO works w/o problems here
Feb 24 21:57:58 <hillct> hmm
Feb 24 21:58:06 <hillct> I was using the master makefile
Feb 24 21:58:15 <hillct> make slugosbe-image
Feb 24 21:58:26 <hillct> guess I can try bb
Feb 24 22:03:04 <CoreDump|home> ah, I thought you were already using bb
Feb 24 22:23:34 <rwhitby> let's get one thing straight.  the master makefile simply sets up some environment variables and config files, then calls bitbake.
Feb 24 22:24:08 <CoreDump|home> yeah, my wording was sub-optimal, sorry
Feb 24 22:24:18 <rwhitby> so if bitbake fails when running the makefile, it's usually cause bitbake will fail exactly the same way if called manually.
Feb 24 22:25:05 <CoreDump|home> rwhitby: I built an LE image from scratch yesterday w/o probs =)
Feb 24 22:25:18 <rwhitby> hillct: OE is a quickly moving target - it often breaks.  if you're not comfortable with that, use the source release for 3.10 instead.
Feb 24 22:58:23 <CIA-6> 03nail 07ts101_support * r741 10kernel/patches/original/ (52 files): Add original patchset
Feb 24 23:02:05 <CIA-6> 03nail 07ts101_support * r742 10kernel/patches/original/ (102 files): rename all to remove silly prefix
Feb 24 23:10:14 <CIA-6> 03nail 07ts101_support * r743 10kernel/patches/original/ (8 files): Remove fluffy cruft
Feb 24 23:12:02 * mwester approves; silly prefixes and fluffy cruft :-D have no place in software!  :-p
Feb 24 23:12:32 <NAiL> hehe
Feb 24 23:12:34 <hillct> what? No dirty jokes in comments?
Feb 24 23:12:53 <hillct> I guess Google Code Search put an end to that sort of stuff
Feb 24 23:13:02 <NAiL> 33 < CIA-6> blaster8 ts101_support * r739 kernel/Makefile: powerpc => ppc (I'll stop interfering now)
Feb 24 23:13:10 <NAiL> 35 < CIA-6> nail ts101_support * r740 kernel/Makefile: Fix toolchain breakage caused by interference ;)
Feb 24 23:13:37 <NAiL> commit messages are fun :-P
Feb 24 23:14:48 <hillct> NAiL: shouldn't the timezones package be included in released images?
Feb 24 23:14:51 <CIA-6> 03nail 07ts101_support * r744 10kernel/patches/original/ (3 files): Remove usb-hackery from a very warm place
Feb 24 23:15:02 <NAiL> hillct: no. They can be installed.
Feb 24 23:15:15 <hillct> NAiL: and have one of the turnup questions be what timezone are you in
Feb 24 23:15:29 <hillct> just seems like a standard thing all systems should have
Feb 24 23:17:47 <CIA-6> 03nail 07ts101_support * r745 10kernel/patches/original/ (drivers_scsi_sd.c drivers_scsi_sg.c): Remove scsi-hacks. *sigh*
Feb 24 23:20:45 <CIA-6> 03nail 07ts101_support * r746 10kernel/patches/original/include_linux_syscalls.h: ooooh, silly syscall!
Feb 24 23:21:57 <CoreDump|home> hmm
Feb 24 23:28:05 <CIA-6> 03bzhou * r5613 10optware/trunk/platforms/ (packages-ddwrt.mk packages-oleg.mk packages-uclibc.mk): 
Feb 24 23:28:05 <CIA-6> moved gambit-c, htop & par2cmdline from uclibc_broken to ddwrt & oleg broken,
Feb 24 23:28:05 <CIA-6> since these work on ts101
Feb 24 23:29:54 <CIA-6> 03bzhou * r5614 10optware/trunk/platforms/packages-ts101.mk: 
Feb 24 23:29:54 <CIA-6> promoted tons of packages for ts101, all of which just need clean rebuild,
Feb 24 23:29:54 <CIA-6> also removed duplicate uclibc_broken
Feb 24 23:51:45 <CoreDump|home> hillct: reload http://hentges.net/tmp/logs/irc/livelogs/%23nslu2-linux.livelog and hover the mouse over your name
Feb 24 23:52:52 <hillct> CoreDump|home: ok. it's a link. Where's it suposed to go?
Feb 24 23:53:19 <CoreDump|home> hover the mouse over the link, a text should appear
Feb 24 23:53:29 <NAiL> heh
Feb 24 23:53:32 <NAiL> nice
Feb 24 23:53:36 <CoreDump|home> ?? hillct 
Feb 24 23:53:38 <cdbot2> hillct[1]: Is the guy who added logrotate and ntpclient 
Feb 24 23:53:55 <hillct> oh. cool
Feb 24 23:53:59 <CoreDump|home> =)
Feb 24 23:54:12 <CoreDump|home> !forget hillct 
Feb 24 23:54:13 <cdbot2> "hillct" deleted from database
Feb 24 23:54:20 <CoreDump|home> !learn hillct Some Text Here
Feb 24 23:54:21 <cdbot2> Added "hillct" to database
Feb 24 23:55:01 <hillct> !forget hillct
Feb 24 23:55:02 <cdbot2> "hillct" deleted from database
Feb 24 23:55:03 <NAiL> !learn NAiL Slug sacrificer
Feb 24 23:55:06 <cdbot2> Added "nail" to database
Feb 24 23:55:17 <CoreDump|home> hehe
Feb 24 23:55:32 <CoreDump|home> NAiL: how many slugs did you sac? 
Feb 24 23:55:34 <hillct> and it handles the case of no information properly
Feb 24 23:55:36 <hillct> cool
Feb 24 23:55:37 <NAiL> nah, just one
Feb 24 23:56:06 <NAiL> I've mangled one pretty bad, but it still lives
Feb 24 23:56:13 <chacko> Hello
Feb 24 23:56:17 <CoreDump|home> NAiL: =D
Feb 24 23:56:20 * hillct avoids sharing the sac with my slugs
Feb 24 23:56:21 <CoreDump|home> chacko: hi
Feb 24 23:56:34 <mwester> greetings chacko
Feb 24 23:56:37 <hillct> I love my slugs
Feb 24 23:56:41 <hillct> but not THAT way
Feb 24 23:57:00 <CoreDump|home> hillct: hehe
Feb 24 23:57:12 <CoreDump|home> as if loving "slugs" wasn't bad enough lol
Feb 24 23:57:23 <hillct> heh
Feb 24 23:57:40 <chacko> For my os/be, I see 9168 as the number from du in /initrd.  How much more can I put there safely?
Feb 24 23:58:31 * CoreDump|home dunnos
Feb 24 23:59:00 <rwhitby> don't go past 256 free
Feb 24 23:59:12 <rwhitby> dont go past 512 free if you want to be safe
Feb 24 23:59:49 <rwhitby> (jffs2 likes to have two blocks free for garbage collection, otherwise you might find jffs2 hangs)
Feb 25 00:00:53 <chacko> df show 5528/1000.  So I can use 700 of that?  !!
Feb 25 00:02:42 <CIA-6> 03bzhou * r5615 10optware/trunk/platforms/packages-ts101.mk: demoted openobex for ts101
Feb 25 00:03:35 <chacko> Why du shows 9168, but df shows 5528 used for initrd?
Feb 25 00:05:13 <mwester> Good question.
Feb 25 00:08:05 <chacko> mwester, was that a test for me ? :) or is it a puzzle to you too?
Feb 25 00:08:25 <mwester> Perhaps a busybox bug? It's handling /dev devices oddly?     (It's a puzzle...)
Feb 25 00:08:51 <mwester> my slugos/be 4.3 is showing something similar for the initrd.
Feb 25 00:09:37 <NAiL> jffs2 compression? ;)
Feb 25 00:10:27 <chacko> mwester, thanks.  How big is the flash really?  6M seems an odd number...
Feb 25 00:10:44 <mwester> 2M used for the kernel + bootloader
Feb 25 00:10:53 <mwester> makes total of 8MB
Feb 25 00:10:58 <chacko> Ah!!
Feb 25 00:11:16 <mwester> The df command should be considered the definitive answer on usage of the flash.
Feb 25 00:12:02 <mwester> Specifically, the available blocks, as rwhitby said, is the important measure -- needs to be 256 minimum, recommened 512.  I learned this the hard way :)
Feb 25 00:12:30 <NAiL> if you end up with less, you can't free up space 
Feb 25 00:12:41 <NAiL> and you'll have to reflash
Feb 25 00:14:38 <chacko> I would like to add a little bit of my own per-server specific config stuff to initrd;
Feb 25 00:14:57 <chacko> Where should I put this so that it will stay over upgrades?
Feb 25 00:15:05 <CIA-6> 03bzhou * r5616 10optware/trunk/ (make/kissdx.mk platforms/packages-ts101.mk): kissdx: use standalone libiconv for ts101
Feb 25 00:16:35 <chacko> can I just put this under /initrd/mydir ?
Feb 25 00:17:18 <rwhitby> chacko: what's your upgrade strategy if you do that?
Feb 25 00:18:59 <chacko> I dunno, i prefer not to have to re-add /initrd/mydir after upgrade, possible?
Feb 25 00:25:18 <chacko> rwhitby, any suggestion?
Feb 25 00:25:32 <rwhitby> nope
Feb 25 00:25:55 <NAiL> hmm
Feb 25 00:26:06 <NAiL> if they're tiny, why not edit "preserve"
Feb 25 00:26:07 <NAiL> ?
Feb 25 00:27:41 <CIA-6> 03bzhou * r5617 10optware/trunk/ (make/mediatomb.mk platforms/packages-ts101.mk): mediatomb: use standalone libiconv for ts101
Feb 25 00:28:06 <chacko> NAiL, 'preserve' ??  Yes my stuff will be very very tiny, the first is a server favicon.ico one file/ 1k
Feb 25 01:04:44 <chacko> Is the our newsgroup mail servers down?  I am not seeing any msgs for almost two days...
Feb 25 01:10:12 <CIA-6> 03nail 07ts101_support * r747 10kernel/patches/2.6.13/ (24 files): Add patchset for working 2.6.13 kernel
Feb 25 01:13:02 <chacko> Is there any downside to putting symbolic links for files like /etc/hostname, /etc/network/interface to their counterparts in /initrd ?
Feb 25 01:14:58 <chacko> NAiL, what is 'preserve' ??  Yes my stuff will be very very tiny, the first is a server favicon.ico one file/ 1k
Feb 25 01:19:41 <NAiL> there's a mechanism to let reflash preserve a few small files
Feb 25 01:19:45 <NAiL> look in /etc/ somewhere
Feb 25 01:19:54 <NAiL> Can't remember the details
Feb 25 01:22:20 <chacko> NAiL, thanks
Feb 25 01:24:07 <chacko> For my os/be, the installed kernel is 2.6.16-r6.4.  I see there is 2.6.16-r6.6.  Any reason to upgrade?
Feb 25 01:41:00 <CIA-6> 03nail 07ts101_support * r748 10kernel/patches/2.6.14/ (24 files): Add working patchset for 2.6.14
Feb 25 01:53:07 <CIA-6> 03bzhou * r5618 10optware/trunk/make/openobex.mk: openobex: no bin/openobex-config
Feb 25 01:56:17 <CIA-6> 03bzhou * r5619 10optware/trunk/platforms/packages-ts101.mk: promoted for ts101: openobex & bluez-utils
Feb 25 01:57:29 <NAiL> eno: great work
Feb 25 02:01:08 <CIA-6> 03bzhou * r5620 10optware/trunk/ (make/obexftp.mk platforms/packages-ts101.mk): obexftp: use libiconv if standalone; promoted for ts101
Feb 25 02:01:37 <eno> NAiL: i'm considering getting a ts101 myself
Feb 25 02:02:54 <NAiL> ah
Feb 25 02:02:58 <NAiL> it's a very nice device ;)
Feb 25 02:05:03 <eno> yes indeed
Feb 25 02:06:00 <NAiL> With the right hacks though, I think it'll make an impressive device
Feb 25 02:06:12 <NAiL> and since it's fanless, it's quiet too :-D
Feb 25 02:06:17 <HopsNBarley> also check out the iomega storcenter - single and dual drive.  2x a slug in disk performance and memory.
Feb 25 02:06:52 <NAiL> doesn't the storcenter have a fan?
Feb 25 02:07:11 <HopsNBarley> yeah, but we have a controller program (-;
Feb 25 02:07:14 <NAiL> hehe
Feb 25 02:07:23 <mwester> chacko: re: symbolic links: the only concern I know about is that /initrd is not always mounted -- such as when you reflash the device.  But then, you can address that by just doing "turnup ram" and rebooting anyway.
Feb 25 02:07:24 <HopsNBarley> the fan on the single unit is noisy.
Feb 25 02:07:43 <NAiL> the ts101's metal case leads away the heat extremely well
Feb 25 02:09:00 <HopsNBarley> the 201 fan is very quiet.  i've had it on my desk for a week and just realized it was cycling.
Feb 25 02:10:06 * NAiL tries booting 2.6.15
Feb 25 02:11:44 <NAiL> haha!
Feb 25 02:31:57 <theblue> Hi all.
Feb 25 02:34:03 <joshin> Howdy!  :)
Feb 25 02:34:36 <theblue> Hi.
Feb 25 02:38:29 <drif> hi
Feb 25 02:44:12 <theblue> drif: Do you use Debian NSLU2?
Feb 25 02:44:56 <theblue> I found the new key for the RC1 installer.
Feb 25 02:59:26 <drif> theblue: good, what was it?
Feb 25 02:59:39 <theblue> <dpkg> For the new apt key, download it with "gpg --recv-keys 6070d3a1", then verify in whatever way you normally do that it is signed by the Debian ftpmasters.  you can then use "apt-key" to install it.
Feb 25 02:59:48 <theblue> That's all I found out.
**** ENDING LOGGING AT Sun Feb 25 02:59:58 2007