**** BEGIN LOGGING AT Tue May 02 09:59:56 2006
May 02 19:51:59 <phatmonkey> my opendebianslug suddenly decided to stop booting
May 02 19:52:11 <phatmonkey> running turnup under openslug:
May 02 19:52:38 <phatmonkey> turnup: /tmp/rootfs.1096: partition does not seem to be a valid root partition
May 02 19:52:44 <phatmonkey>   /tmp/rootfs.1096/.recovery exists: fix the partition then remove it
May 02 19:54:23 <phatmonkey> what to do?
May 02 19:59:24 <evildevil_> phatmonkey, check the filesystem and if it is okay, mount it somewhere and remove the .rmrecover file in the filesystem's root
May 02 19:59:56 <phatmonkey> filesystem is fine, thanks
May 02 20:00:05 <phatmonkey> I'm sure there was something I needed to remove, I just didn't know what it was
May 02 20:00:31 <evildevil_> s/.rmrecover/.recovery/
May 02 20:00:31 <jbot> evildevil_ meant: phatmonkey, check the filesystem and if it is okay, mount it somewhere and remove the .recovery file in the filesystem's root
May 02 20:01:03 <evildevil_> "  /tmp/rootfs.1096/.recovery exists: fix the partition then remove it" hmmm.. you posted it already
May 02 20:01:28 <phatmonkey> I was looking for /tmp/rootfs.1096/.recovery on the filesystem :)
May 02 20:08:12 <cbrake_> anyone else getting this: padBytes error: data (1065372) is longer than total_len (1048576)
May 02 20:08:22 <cbrake_> when building openslug/debianlslug?
May 02 20:08:34 <cbrake_> http://pastebin.com/694921
May 02 20:27:15 <phatmonkey>  /join #ubuntu
May 02 20:27:18 <phatmonkey> whoops
May 03 02:27:01 <placid_> if I port glibc to unlsung do I install it in /opt/lib since there is already the native linksys libs in /lib?
May 03 02:50:41 <rwhitby> placid_: why would you be porting glibc to unslung?  the slug already has a C library.
May 03 02:50:55 <rwhitby> The idea of Unslung is that you don't change the kernel or standard libraries.
May 03 02:51:12 <rwhitby> It's sounding like you should be using OpenSlug or DebianSlug ....
May 03 03:01:32 <placid_> it is missing some headers
May 03 03:01:53 <placid_> but you may be right 8)
May 03 03:13:50 <placid_> rhwhitby: so is the conclusion that if a package is reliant on something that is missing from core libraries then it probably isn't a candidate for unslung?
May 03 03:14:04 <rwhitby> it depends.
May 03 03:14:36 <rwhitby> we do actually add some additional libraries and stuff to the firmware imaage.
May 03 03:14:42 <rwhitby> (from glibc)
May 03 03:15:14 <rwhitby> but we definitely do not replace any libraries that are already on the device, in case the linksys binaries for which we do not have source code fall over.
May 03 03:15:25 <rwhitby> (yes, that did happen when we tried it)
May 03 03:15:31 <placid_> ha
May 03 03:17:13 <rwhitby> e.g. we put librt on the device from the libc6-unslung package
May 03 03:17:24 <placid_> ok
May 03 03:17:39 <rwhitby> (which is just the ipk we emit from glibc-2.2.5)
May 03 03:18:25 <placid_> so at a minimum if I were interested in something not currently included it needs to be based on glibc-2.2.5
May 03 03:18:40 <rwhitby> yep
May 03 03:19:00 <rwhitby> in particular, it has to be something built by glibc_2.2.5.bb in OE
May 03 03:19:12 <rwhitby> which we then add to the libc6-unslung package
May 03 03:19:29 <rwhitby> (or we create a new package if it's of specific interest rather than general interest)
May 03 03:19:36 <placid_> ok
May 03 03:19:41 <placid_> will investigate further
May 03 03:19:47 <rwhitby> or a -dev package if it's for building rather than runtime
May 03 03:19:57 <placid_> yeah that's what I'm after
May 03 03:20:08 <rwhitby> building?
May 03 03:20:11 <placid_> yep
May 03 03:20:20 <placid_> dependency on glibc-dev
May 03 03:20:38 <placid_> ifaddrs.h
May 03 03:20:56 <rwhitby> is that just a build-time dependency, or a run-time dependency?
May 03 03:21:16 <rwhitby> (and can it be cross-compiled instead?)
May 03 03:21:17 <placid_> I believe it is only build but won't know until I can get it built
May 03 03:21:23 <placid_> I only cross compile
May 03 03:22:19 <rwhitby> so you're looking in tmp/staging/armeb-linux/include ?
May 03 03:22:41 <rwhitby> oh, this is an optware package, right?
May 03 03:22:46 <placid_> yep
May 03 03:23:30 <rwhitby> so /home/slug/optware/nslu2/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/armv5b\
May 03 03:23:30 <rwhitby> -softfloat-linux/include/
May 03 03:26:36 <placid_> is .../slug/unslung/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/armv5b-softfloat-linux/include on my setup
May 03 03:26:50 <placid_> and the header file isn't present
May 03 03:41:42 <placid_> rwhitby, found the answer.. getifaddrs wasn't added to glibc until 2.3 so it won't be available under unslung
May 03 05:10:27 <CIA-4> 03placid * 10unslung/ (3 files in 3 dirs): nfs-utils: update to 1.0.7
May 03 05:41:09 <placid_> rwhitby, I've updated a package and modified to Makefile to move it into testing since it is a version upgrade. how do i get it built?
May 03 05:43:21 <rwhitby> placid_: if it's an existing package, then do no move it into testing, or else it will be deleted from the feed
May 03 05:43:29 <placid_> ah
May 03 05:45:01 <CIA-4> 03placid * 10unslung/Makefile: nfs-utils: restore make to cross packages
May 03 05:45:59 <rwhitby> also please note any changes to top-level Makefile separately in the checkin comment.
May 03 05:46:26 <placid_> ok
May 03 06:09:57 <placid_> I reverted my mod to the Makefile so new version of nfs-utils should build in next run
May 03 07:21:34 <CIA-4> 03oleon * 10unslung/make/gift-ares.mk: gift-ares: revert from CVS to stable version
May 03 07:42:08 <CIA-4> 03placid * 10unslung/make/bzip2.mk: bzip2: upgrade to 1.0.3 (security fix plus others)
May 03 09:46:18 <CIA-4> 03placid * 10unslung/make/logrotate.mk: logrotate: upgrade to 3.7.1
**** ENDING LOGGING AT Wed May 03 09:59:56 2006