**** BEGIN LOGGING AT Sun Sep 03 02:59:57 2006
Sep 03 03:33:59 <NAiL> ipkg install kernel-module-nfsd ?
Sep 03 03:34:43 <RastafaI> already installed (nfs-utils depend on it)
Sep 03 03:34:58 <NAiL> nfs-utils *used* to depend on it
Sep 03 03:35:26 <RastafaI> ok will try then
Sep 03 03:39:01 <RastafaI> hmmm
Sep 03 03:39:31 <RastafaI> root@JahStorage:~# ipkg install kernel-module-nfsd
Sep 03 03:39:31 <RastafaI> Package kernel-module-nfsd (2.6.16-r6.5) installed in root is up to date.
Sep 03 03:39:31 <RastafaI> Nothing to be done
Sep 03 03:39:31 <RastafaI> root@JahStorage:~# /etc/init.d/nfsserver start
Sep 03 03:39:31 <RastafaI> FATAL: Module nfsd not found.
Sep 03 03:39:36 <RastafaI> :(
Sep 03 03:40:33 <RastafaI> (btw, kernel-module-nfsd was already there with fresh install, ipkg update & upgrade, ipkg install nfs-utils)
Sep 03 03:41:52 <RastafaI> and the module is here: 
Sep 03 03:41:52 <RastafaI> root@JahStorage:~# cd /lib/modules/2.6.16/kernel/fs/nfsd/
Sep 03 03:41:52 <RastafaI> root@JahStorage:/lib/modules/2.6.16/kernel/fs/nfsd# ls
Sep 03 03:41:52 <RastafaI> nfsd.ko
Sep 03 03:46:40 <NAiL> odd
Sep 03 03:46:49 <NAiL> I'll look at it tomorrow
Sep 03 03:46:57 <NAiL> It's 6am here
Sep 03 03:48:34 <NAiL> btw, try depmod -a
Sep 03 03:54:03 <RastafaI> worked
Sep 03 03:54:13 <RastafaI> thanks !!!
Sep 03 03:54:21 <NAiL> np
Sep 03 04:50:18 <NCommander> Cant someone help me get scsi-idle working on the slug?
Sep 03 04:50:45 <NCommander> err, wrong channel. nm
Sep 03 05:02:35 <CIA-5> 03bzhou * r4005 10optware/trunk/make/nginx.mk: nginx: changed dir layout, did some runtime test
Sep 03 05:26:00 <eno-away> nginx is cool stuff
Sep 03 09:24:54 <RastafaI> i've a slight problem
Sep 03 09:24:56 <RastafaI> configure: error: no acceptable cc found in $PATH
Sep 03 09:25:06 <RastafaI> eventhough i did ipkg install gcc
Sep 03 09:25:14 <RastafaI> also tried with g++ same result
Sep 03 09:25:30 <RastafaI> i have no gcc nor g++ command available
Sep 03 09:25:43 <RastafaI> so i cannot set CC variable...
Sep 03 09:31:34 <RastafaI> got it
Sep 03 09:31:44 <RastafaI> package gcc-symlinks did it
Sep 03 09:32:52 <RastafaI> checking for gcc... gcc
Sep 03 09:32:52 <RastafaI> checking whether the C compiler (gcc  ) works... no
Sep 03 09:32:52 <RastafaI> configure: error: installation or configuration problem: C compiler cannot create executables.
Sep 03 09:33:00 <RastafaI> during a ./configure
Sep 03 10:19:14 <RastafaI> ok got it... had to ipkg install libc6-dev
Sep 03 10:21:18 <RastafaI> configure: error: ld is required to build the flex scanner!
Sep 03 10:21:20 <RastafaI> anyone ?
Sep 03 10:29:59 <drif> RastafaI: jah?
Sep 03 10:30:37 <drif> or just coincidence
Sep 03 10:33:13 <RastafaI> nope
Sep 03 10:33:16 <RastafaI> no coincidence
Sep 03 10:33:53 <drif> ok, so got you slug running?
Sep 03 10:33:57 <drif> +r
Sep 03 10:42:38 <RastafaI> yup
Sep 03 10:42:47 <RastafaI> openslug is ok now
Sep 03 10:42:59 <RastafaI> thanks to Nail
Sep 03 10:43:13 <RastafaI> now i'm trying to compile erlang....
Sep 03 10:43:22 <RastafaI> but confiugre stops asking for ld
Sep 03 10:43:30 <RastafaI> wich i don't have
Sep 03 10:43:40 <RastafaI> seems it's not included in busybox
Sep 03 10:58:41 <RastafaI> ipkg install binutils-symlinks....haha i=once again
Sep 03 11:33:12 <NAiL> RastafaI: ld is included in binutils 
Sep 03 11:33:26 <NAiL> if you want to compile stuff, installing slugos-native is a good idea
Sep 03 11:44:54 <gda_> hi
Sep 03 11:45:28 <gda_> is there a special reason why ipkg is not in the ipk feed for ds101g?
Sep 03 12:00:03 <CIA-5> 03gda * r4006 10optware/trunk/Makefile: ipkg: new cross package, current version segfaults on ds101g
Sep 03 12:05:36 <rwhitby> gda_: just to confirm - that new ipkg package is specifically only enabled for certain targets, right?
Sep 03 12:06:00 <rwhitby> (we had an incident last year where ipkg was put in an nslu2 feed, which broke a *lot* of people's slugs)
Sep 03 13:03:30 <gda_> rwithby: I made only a change for the ds101g, for all other plattform is should be the same as before
Sep 03 13:06:43 <gda_> rwithby: so if this version before my change was not enabled for all platforms, it should only be used for the ds101g
Sep 03 13:07:19 <gda_> but if this version was enabled for all platforms before, I see no reason to not enable my version for all
Sep 03 13:14:14 <gda_> rwithby: I see now it is in no feed, but we need for the ds101g urgend, no ipk updates with changed config files possible
Sep 03 13:18:04 <rwhitby> gda_: if someone was to move that from cross_packages_ready_for_testing to cross_packages, then all hell would break loose..
Sep 03 13:18:18 <rwhitby> Please add it to ds101g specific packages only
Sep 03 13:42:07 <gda_> k
Sep 03 13:44:04 <CIA-5> 03gda * r4007 10optware/trunk/Makefile: ipkg: moved to ds101g specific packages
Sep 03 14:15:16 <CIA-5> 03gda * r4008 10optware/trunk/Makefile: appweb: is not broken on the ds101g
Sep 03 16:03:43 <CIA-5> 03gda * r4009 10optware/trunk/ (4 files in 3 dirs): libgcrypt: updated to 1.2.3 and patched for ds101g
Sep 03 16:17:13 <CIA-5> 03gda * r4010 10optware/trunk/Makefile: cherokee, opencdk, gnutls not anymore broken for ds101g
Sep 03 16:20:25 <CIA-5> 03gda * r4011 10optware/trunk/Makefile: rsync works for ds101g
Sep 03 16:47:48 <artworkz> 1,00moin
Sep 03 16:48:12 <artworkz> 1,00ich suche ne liste der funktionierenden usb-hubs am unslung
Sep 03 16:48:17 <artworkz> 1,00gibts sowas?
Sep 03 16:51:22 <gda_> artworkz: die meisten der Leute hier sprechen kein deutsch und wenn du mal auf das Topic achtest, dann merkst du, dass du hier sowieso falsch bist
Sep 03 16:55:04 <artworkz> 1,00oh sorry
Sep 03 17:01:04 <gda_> hi eno
Sep 03 17:01:21 <eno> hi gda_ 
Sep 03 17:01:32 <gda_> eno: got libgcrypt working on the ds101g :)
Sep 03 17:02:08 <eno> excellent, i saw you upgraded ipk for ds101g specific
Sep 03 17:02:24 <eno> does that fix the segfault problem?
Sep 03 17:02:42 <gda_> yes
Sep 03 17:03:07 <gda_> I fixed it a week ago, but I didn't notice that it didn't get build
Sep 03 17:03:46 <eno> yeah, always need to be careful about ipkg
Sep 03 17:03:52 <gda_> k
Sep 03 17:04:01 <gda_> what was the problem lasttime
Sep 03 17:04:03 <gda_> ?
Sep 03 17:05:20 <gda_> btw. the upgrade didn't fix the segfault, I had to patch
Sep 03 17:06:37 <eno> ipkg was added in nslu2 feed which really shouldn't
Sep 03 17:07:21 <eno> on nslu2 we should only use /usr/bin/ipkg
Sep 03 17:21:15 <CIA-5> 03gda * r4012 10optware/trunk/Makefile: lsof works for ds101g
Sep 03 17:25:29 <gda_> does somebody still know why ntp is a broken package for the ds101g?
Sep 03 17:26:30 <gda_> maybe ldconfig is marked broken because the ds101g has already ldconfig?
Sep 03 17:27:49 <gda_> and I don't understand why xemacs is general in the package list
Sep 03 17:28:35 <gda_> and to use squid in this small memory boxes is strange
Sep 03 17:30:10 <NAiL> not really
Sep 03 17:30:19 <NAiL> squid doesn't have to have a lot of cache in memory
Sep 03 17:30:25 <NAiL> it can cache a lot on disk
Sep 03 17:30:39 <NAiL> of course, there are smaller alternatives, but squid is extremely flexible
Sep 03 17:30:48 <gda_> so you use it only as proxy and not as cache
Sep 03 17:31:06 <NAiL> cache on disk
Sep 03 17:31:19 <gda_> squid has the advantage of working with dansguardian
Sep 03 17:31:20 <NAiL> or transparent proxy, filtering proxy etc.
Sep 03 17:32:26 <gda_> for some friend I have setup routers with dansguardian, they have teenager sons that shall not see porn
Sep 03 17:33:24 <gda_> but to patch squid for cross seems to be a too  big job for this sunday evening
Sep 03 17:33:27 <NAiL> there's always a way through :-P
Sep 03 17:33:47 <gda_> not for them, currently
Sep 03 17:34:34 <gda_> they even have to ask me for getting a zip file
Sep 03 17:34:43 <NAiL> https?
Sep 03 17:34:55 <gda_> ?
Sep 03 17:35:07 <gda_> dansguardian is a content filter
Sep 03 17:35:22 <NAiL> when you use https through a proxy, you issue a CONNECT, allowing you to bypass a lot of proxies
Sep 03 17:35:54 <NAiL> easy tunneling through a proxy ;)
Sep 03 17:35:55 <gda_> have not seen much porn sides that use https recently
Sep 03 17:36:05 <NAiL> they don't need to use https
Sep 03 17:36:15 <NAiL> CONNECT can connect to other ports
Sep 03 17:36:22 <NAiL> eg. 80 
Sep 03 17:36:55 <gda_> I will not tell them
Sep 03 17:37:02 <NAiL> anyway, unless one has a certain level of persistence, it's irrelevant
Sep 03 17:37:26 <gda_> ?
Sep 03 17:37:58 <NAiL> there's a bunch of ways to bypass proxies, but the discussion is irrelevant
Sep 03 17:38:19 * NAiL just remembered all the ways he used to bypass the proxy at Siemens
Sep 03 17:56:33 <gda_> wanted to get the broken ipk list for the ds101g shorter, but there it seems there are no interesting candidates anymore
Sep 03 19:23:42 <blaster8> ping? NAiL
Sep 03 20:43:19 <NAiL> blaster8_: pong
Sep 03 20:43:26 <NAiL> been away helping my aunt move
Sep 03 20:44:21 <blaster8> allo
Sep 03 20:44:29 <blaster8> you have some unhappy customers
Sep 03 20:44:49 <blaster8> SlugOS feed seems broken
Sep 03 20:45:00 <blaster8> it's in the mailing list
Sep 03 20:55:01 <NAiL> ouch
Sep 03 20:55:57 <blaster8> good luck
Sep 03 20:56:08 <blaster8> C library breakage, I think...
Sep 03 20:56:13 <blaster8> not sure how, mind
Sep 03 20:58:02 <NAiL> ipkg doesn't handle upgrading libc
Sep 03 20:59:16 <blaster8> is that the problem?
Sep 03 21:01:08 <NAiL> yes
Sep 03 21:01:34 <blaster8> sheesh
Sep 03 21:05:36 <NAiL> doing the quick fix first :P
Sep 03 21:06:11 <NAiL> just deleting the offending ipks from the feeds...
Sep 03 21:06:49 <blaster8> we should have a chat when I get back next weekend
Sep 03 21:07:39 <NAiL> yeah
Sep 03 21:07:44 <blaster8> I'm willing to put some work into this project over the next month, but need to have an idea of what's possible
Sep 03 21:08:30 <blaster8> I'd like to get the ixp4xx kernel building without the access library
Sep 03 21:08:47 <CIA-5> 03nail 07slugos-3.10-beta * r352 10slugos/openembedded/packages/glibc/glibc_2.3.5+cvs20050627.bb: glibc: Un-bump PR so to avoid glibc breaking upgrades
Sep 03 21:09:09 <NAiL> hmm... so it's ready for the open driver?
Sep 03 21:09:42 <blaster8> not necessarily
Sep 03 21:09:55 <blaster8> but we need to test and fix the open driver, otherwise it'll never get done
Sep 03 21:10:20 <blaster8> so we need to maintain two kernel versions temporarily, and I'm willing to take on the totally open kernel during September at least
Sep 03 21:10:24 <NAiL> I've got a wishlist for SlugOS 4.x... glibc 2.4, 2.6.18, EABI, open ixp-driver, gcc 4.x 
Sep 03 21:10:40 <blaster8> that will jettison Ucslugc
Sep 03 21:10:49 <blaster8> but that is probably fine (controversially)
Sep 03 21:11:09 <NAiL> well, there's not many users. IIRC, eFfeM is working on uCslugC
Sep 03 21:11:17 <blaster8> Ucslugc is probably stuck where it is for the forseeable future
Sep 03 21:11:29 <CIA-5> 03bzhou * r4013 10optware/trunk/make/py-pyro.mk: new py-pyro package
Sep 03 21:11:38 <NAiL> well, it can upgrade kernel/ixp-driver and gcc. 
Sep 03 21:11:43 <blaster8> the thing is that Ucslugc userland doesn't need to change much
Sep 03 21:11:57 <NAiL> nope
Sep 03 21:11:58 <blaster8> it's emphasis is small, not up-to-date
Sep 03 21:12:15 <blaster8> and it's quite mature as of SlugOS 3.10 - I can continue to work on that codebase
Sep 03 21:12:23 <NAiL> do you have svn access?
Sep 03 21:12:39 <blaster8> no, but I haven't done much yet :)
Sep 03 21:12:46 <blaster8> trac access would be good though
Sep 03 21:13:01 <NAiL> If you want to work on the kernel svn repo, you should get access ;-)
Sep 03 21:13:13 <NAiL> with svn access comes trac. Uses the same cert.
Sep 03 21:13:15 <blaster8> I can set up a branch
Sep 03 21:13:48 <NAiL> imho, it's better if you use the official tree so that we can collaborate
Sep 03 21:14:06 <NAiL> "we" as in all of us 
Sep 03 21:14:27 <blaster8> OK, set up a branch in the official SVN so I can break things without annoying you :)
Sep 03 21:14:51 <blaster8> I'm going to have a dedicated build server soon
Sep 03 21:15:21 <blaster8> I'm setting up an OpenVZ server on my network so I can run lots of different linuxes totally seperately on one machine
Sep 03 21:15:26 <blaster8> handy for this sort of thing
Sep 03 21:15:35 <NAiL> just make a separate directory in the kernel repo, and you won't "annoy" us :-P
Sep 03 21:16:04 <blaster8> the lcdproc stuff made its way into OE bugtracker, btw
Sep 03 21:16:08 <NAiL> I saw
Sep 03 21:16:14 <blaster8> it's pretty complete, you jest need to check it builds
Sep 03 21:16:25 <blaster8> minor issue ;)
Sep 03 21:16:31 <NAiL> hehe
Sep 03 21:16:38 <NAiL> I've accepted the bug
Sep 03 21:16:47 <NAiL> so I'll try to work on it :)
Sep 03 21:16:50 <blaster8> cool, that can be sorted out at leisure then
Sep 03 21:16:52 <gda_> caplink811_log: for what is this rc.clamav good, adding group and user happens already in postinst
Sep 03 21:17:47 <gda_> caplink811_log: it is no problem that the clamd gets not started there, one daemon less and my amavis-new uses clamscan, but
Sep 03 21:18:06 <gda_> I would expect to see freshclam started there
Sep 03 21:19:06 <CIA-5> 03bzhou * r4014 10optware/trunk/ (Makefile make/py-pyro.mk): added py-pyro
Sep 03 21:25:31 <CIA-5> 03gda * r4015 10optware/trunk/make/amavisd-new.mk: amavisd-new: start before postfix
Sep 03 21:26:53 <gda_> fetchmail starts before postfix, that is too early, special whishes for the right SXXfetchmail ?
Sep 03 21:27:05 <gda_> S70 maybe?
Sep 03 21:36:27 <caplink811_log> gda_, if there are any issues, just change anything, so it fit your needs...
Sep 03 21:46:21 <gda_> caplink811_log: before I throw everything away, I wanted only to understand what was the intension of this
Sep 03 21:46:34 <NAiL> blaster8: btw, I sent a mail to the lcdproc devel list to see if anyone could help 
Sep 03 21:47:03 <blaster8> allo
Sep 03 21:47:06 <blaster8> ok
Sep 03 21:47:22 <NAiL> blaster8: I *think* it's a 5-minute job to separate the drivers, if you know your way around the autotools crap
Sep 03 21:47:27 <blaster8> they should separate the hd44780 driver, which would solve our problem
Sep 03 21:47:31 <blaster8> exactly
Sep 03 21:53:04 <NAiL> http://lists.omnipotent.net/pipermail/lcdproc/2006-September/011064.html
Sep 03 21:59:29 <NAiL> btw, I updated slugos firmware bug #11, and added a note about the ixp4xx driver you want to try. Put down your nick too.
Sep 03 21:59:51 <NAiL> http://trac.nslu2-linux.org/slugos/ticket/11
Sep 03 22:02:08 <blaster8> The big issue is not with the driver but with how to get the microcode
Sep 03 22:02:26 <blaster8> we can try and grep it from redboot automatically on the first boot
Sep 03 22:02:43 <NAiL> yeah
Sep 03 22:02:44 <blaster8> or we can install it in the package, which means we still have the IPL licence issues
Sep 03 22:03:00 <NAiL> Still, I'd prefer the firmware being newer than the version present in redboot...
Sep 03 22:03:10 <blaster8> good point, actually
Sep 03 22:03:26 <blaster8> hmm
Sep 03 22:03:43 <blaster8> probably best to keep in the install image for now
Sep 03 22:04:02 <blaster8> not sure how debian will do it, but that is their problem :)
Sep 03 22:04:05 <NAiL> From my POV, I'd like APEX to get this driver, pull firmware from a separate partition (which can then be used after boot too)
Sep 03 22:04:23 <blaster8> that is best, definitely
Sep 03 22:04:44 <NAiL> Actually... that's not the worst idea
Sep 03 22:04:52 <blaster8> ?
Sep 03 22:05:31 <NAiL> if we can have a separate firmware partition... Then users can upgrade to slugos, which includes the firmware, then use the debian installer image afterwards without the firmware attached
Sep 03 22:06:04 <blaster8> all we have to do is co-ordinate slugos, apex and debian :)
Sep 03 22:06:19 <NAiL> of course, then installing debian would require installing slugos first 
Sep 03 22:06:37 <blaster8> well, better than having to muck around with USB Ethernet
Sep 03 22:06:42 <NAiL> definitely
Sep 03 22:07:02 <blaster8> lots of possibilities
Sep 03 22:07:27 <blaster8> e.g. use upslug2 to write the microcode separately from the debian installer
Sep 03 22:07:36 <blaster8> so they are not distributed together
Sep 03 22:07:38 <NAiL> ah, good point
Sep 03 22:07:58 <blaster8> I'll try and get some contact with the guy who runs Cyrius.com
Sep 03 22:08:13 <NAiL> Cyrius.com?
Sep 03 22:08:14 <blaster8> we're both at the same uni so it shouldn't be too hard
Sep 03 22:08:21 <NAiL> oh
Sep 03 22:08:27 <blaster8> http://www.cyrius.com/debian/nslu2/
Sep 03 22:08:31 <NAiL> that's dwery IIRC
Sep 03 22:08:35 <NAiL> (on irc)
Sep 03 22:08:38 <blaster8> ok
Sep 03 22:08:40 <NAiL> core developer ;)
Sep 03 22:08:45 <blaster8> ooh, big cheese
Sep 03 22:09:26 <NAiL> hmm
Sep 03 22:09:42 <NAiL> or maybe I'm wrong ;)
Sep 03 22:10:24 <blaster8> he is officially the link between nslu2-linux and debian
Sep 03 22:10:33 <blaster8> so he is definitely the guy to chat to
Sep 03 22:10:58 <NAiL> yeah
Sep 03 22:11:04 <blaster8> right, I need to sleep
Sep 03 22:11:07 <blaster8> very early start
Sep 03 22:11:17 <blaster8> I'll be back Saturday
Sep 03 22:11:51 <NAiL> ok, see you then ;)
Sep 03 22:11:55 <blaster8> cya
Sep 03 22:13:56 * NAiL notes that Michlmayr is tbm
Sep 03 23:00:30 <CIA-5> 03nail 07slugos-3.10-beta * r353 10slugos/openembedded/packages/meta/slugos-native.bb: slugos-native: Remove libc6-dev dep, so the build doesn't stop
Sep 04 00:04:26 <CIA-5> 03bzhou * r4016 10optware/trunk/make/amavisd-new.mk: amavisd-new: S79amavisd -> S60amavisd
Sep 04 00:10:38 <placid_> bzhou: is there are any reason why the same package doesn't support 64 bit file handling?
Sep 04 00:10:48 <placid_> sorry, samba package
Sep 04 00:11:03 <NAiL> hmm.. it doesn?T
Sep 04 00:11:08 <placid_> nope
Sep 04 00:11:21 <placid_> copy bombs after 2GB
Sep 04 00:11:22 <NAiL> optware or slugos?
Sep 04 00:11:31 <placid_> unslung
Sep 04 00:11:40 <NAiL> optware then ;)
Sep 04 00:11:43 <placid_> yep
Sep 04 00:12:20 <NAiL> not my field, unfortunately
Sep 04 00:12:25 <placid_> I suspect it is because the make file sets: samba_cv_HAVE_OFF64_T=no
Sep 04 00:12:42 <placid_> I'm recompiling the package now to see what messages it emits
Sep 04 00:13:00 <NAiL> I remember using a few days getting samba to accept large files on openslug. 
Sep 04 00:13:20 <placid_> yeah it is doable because I did the same for ushare
Sep 04 01:06:40 <CIA-5> 03nail 07slugos-3.10-beta * r354 10slugos/openembedded/packages/litestream/ (. litestream_1.3RC3.bb): litestream: Add package. litestream is a shoutcast-compatible streamer
Sep 04 01:07:23 <CIA-5> 03nail 07slugos-3.10-beta * r355 10slugos/openembedded/packages/meta/slugos-packages.bb: slugos-packages: Add litestream to feed
Sep 04 01:33:54 <CIA-5> 03nail 07slugos-3.10-beta * r356 10slugos/openembedded/packages/meta/slugos-native.bb: slugos-native: Bump PR to force rebuild
Sep 04 01:37:20 <CIA-5> 03nail 07slugos-3.10-beta * r357 10slugos/openembedded/packages/meta/slugos-packages.bb: slugos-packages: Add mysql to feed
**** ENDING LOGGING AT Mon Sep 04 02:59:56 2006