Bug#481350: mysql-server-5.0: On arm, 5.0.32-7etch5 segfaults frequently, where 5.0.32-7etch1 is ok.

2008-05-15 Thread Brian Brunswick
Package: mysql-server-5.0 Version: 5.0.32-7etch1 Severity: grave Justification: renders package unusable I'm running a mythtv setup using diskless machines as the front and back ends, and a DNS-323 with a chrooted debian etch installation as the file and mysql server. I just tried an update on

Bug#416664: libnss-ldap: If nsswitch.conf already configured with ldap, installation breaks and leaves system broken

2007-03-29 Thread Brian Brunswick
Package: libnss-ldap Version: 251-7.5 Severity: normal Setting up a new system, I copied its config over from an old system, before having installed the necessary packages. So nsswitch.conf had entries 'ldap files' for passwd etc. This was completely harmless in the absence of libnss-ldap

Bug#408325: sfsauthd: fatal: Should not be reached - server fails on arm nslu2

2007-02-16 Thread Brian Brunswick
I would say if the server binary has ever worked on any arm machine, then keep it. I did manage to start investigating this. I recompiled the package with some extra trace (very slowly, using qemu) and got some additional information. Unfortunately, I haven't had time to continue and follow

Bug#410597: user-mode-linux: Please enable all the crypto modules

2007-02-11 Thread Brian Brunswick
Package: user-mode-linux Version: 2.6.18-1um-2 Severity: normal I wanted to mount a luks partition from inside uml, but found that the sha256 kernel module wasn't built in the debian package. Rebuilding the package with this enabled went fine, so please consider it for the standard config. --

Bug#410154: user-mode-linux: crashes almost immediately with Kernel mode signal 7

2007-02-07 Thread Brian Brunswick
Package: user-mode-linux Version: 2.6.18-1um-2 Severity: grave Justification: renders package unusable It crashes out almost immediately. The crash can be slightly varied by changing the memory size, or it will even get almost all the way through a simple boot up, but always dies: Only thing

Bug#408325: sfsauthd: fatal: Should not be reached - server fails on arm nslu2

2007-02-05 Thread Brian Brunswick
On 03/02/07, Jaakko Niemi [EMAIL PROTECTED] wrote: On Wed, 24 Jan 2007, Brian Brunswick wrote: Package: sfs-server Version: 1:0.8-0+pre20060720.1-1 Severity: grave Justification: renders package unusable While sfs-client seems to work ok onto another server, I get sfsauthd crashing out

Bug#409872: On slow arch apt just wastes cpu while waiting for dpkg to run

2007-02-05 Thread Brian Brunswick
Package: apt Version: 0.6.46.4 Severity: normal I'm running apt from an nfs-mounted chroot of the real disk inside a qemu-system-arm instance. This is quite slow, but better than the real machine (nslu2) which has infeasibly little memory (32MB) [When I was young...], at least for modern big

Bug#408325: sfsauthd: fatal: Should not be reached - server fails on arm nslu2

2007-01-24 Thread Brian Brunswick
Package: sfs-server Version: 1:0.8-0+pre20060720.1-1 Severity: grave Justification: renders package unusable While sfs-client seems to work ok onto another server, I get sfsauthd crashing out and exiting leaving the above message in the syslog, when I do: sfskey register while trying to set up

Bug#403012: Using UUID= for a root partition in /etc/fstab fails to boot

2007-01-14 Thread Brian Brunswick
On 05/01/07, Martin Michlmayr [EMAIL PROTECTED] wrote: Then I put /dev/disk/by-uuid/7750fe81-888e-4bf4-9c87-94d54bcb3291 into /etc/fstab. /dev/disk/by-uuid/7750fe81-888e-4bf4-9c87-94d54bcb3291 / ext3defaults,errors=remount-ro 0 Ah, thats probably the difference. I used UUID=xxx /

Bug#403426: linux-image-2.6.18-3-ixp4xx: kernel corrupts LUKS partition header on arm NSLU2

2006-12-16 Thread Brian Brunswick
Package: linux-image-2.6.18-3-ixp4xx Version: 2.6.18-8 Severity: critical Justification: causes serious data loss This is on an NSLU2, I wanted to use it to access some disks that I had used previously on another system that had encrypted partitons. However, when I tried cryptsetup luksOpen, I

Bug#403012: Using UUID= for a root partition in /etc/fstab fails to boot

2006-12-13 Thread Brian Brunswick
Package: nslu2-utils Version: 0.10+r71-5 Severity: normal I wanted to make booting more reliable for multiple devices plugged in, and use UUIDs to find the root device, as I do on my server system. But it appears not to boot after regenerating the initrd with dpkg-reconfigure kernel with the root

Bug#384045: util-linux: Please package a version with ionice from schedutils merge

2006-08-21 Thread Brian Brunswick
Package: util-linux Version: 2.12p-4sarge1 Severity: important The ionice functionality has been in the kernel since 2.6.13, it would be nice to have the user space tools to match. ionice is part of schedutils 1.5, which the schedutils maintainer has said is being merged with util-linux. Please

Bug#384066: java-package: Please generate a debian specific fontconfig.properties to enable unicode fonts

2006-08-21 Thread Brian Brunswick
Package: java-package Version: 0.24 Severity: important Most international fonts don't work inside java applets on debian because it falls back to a generic minimal fontconfig.properties. There are much more complete ones provided for other distributions, but none for debian. I took the

Bug#381929: e2fsprogs: clean fsck should say when a check will next be forced

2006-08-07 Thread Brian Brunswick
Package: e2fsprogs Version: 1.37-2sarge1 Severity: wishlist It would be very useful to for-instance laptop users if fsck said what the force check date/mount count is when it skips a clean filesystem. -- System Information: Debian Release: 3.1 Architecture: i386 (i686) Kernel: Linux

Bug#358649: initramfs-tools: Allow passing nfs root mount options on kernel cmdline

2006-03-23 Thread Brian Brunswick
Package: initramfs-tools Version: 0.56 Severity: normal Tags: patch Here is a suggested patch. Might also be worth documenting that the klibc nfsmount takes udp instead of proto=udp arguments. --- /usr/share/initramfs-tools/init-dist2006-03-23 13:06:34.0 + +++

Bug#353484: Removal of /var/run/apache2 causes it to fill disk with logged errors

2006-02-18 Thread Brian Brunswick
Package: apache2 Severity: normal Please don't ship subdirectories of /var/run in packages at all. Created them in the init.d script. -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (50, 'unstable') Architecture: i386 (i686) Kernel: Linux 2.6.15-1-686-smp Locale:

Bug#352960: postinst doesn't use full path to apt-show-versions

2006-02-15 Thread Brian Brunswick
Package: apt-show-versions Version: 0.08 Severity: important We had an old version of apt-show-versions in /usr/local/bin that doesn't support the -i option. So the postinst script bombed out. It should at least use the full path to its version. -- System Information: Debian Release: 3.1

Bug#353015: dpkg: start-stop-daemon --name should warn about too long names

2006-02-15 Thread Brian Brunswick
Package: dpkg Version: 1.10.28 Severity: wishlist The kernel has a maximum length for process names that appears to be 15 (linux 2.6) The --name argument should produce a warning or error when a name longer than this is used, since it will never match. This is particularly bad since --start

Bug#318247: cron: Please (optionally?) log end of job as well as start

2005-07-14 Thread Brian Brunswick
Package: cron Version: 3.0pl1-86 Severity: wishlist It would be handy to have syslog entries for the end of jobs as well as the start, to sort out long-running crontab entries. -- System Information: Debian Release: 3.1 Architecture: i386 (i686) Kernel: Linux 2.4.27-2-386 Locale: LANG=C,

Bug#307977: udev unmounts devpts and does not remount it

2005-05-06 Thread Brian Brunswick
Package: udev Version: 0.056-2 Severity: critical Justification: breaks unrelated software During startup, udev runs at a priority 4 after the initial run of mountvirtfs at pri 2, and it unmounts /dev/pts while it fribs the ramdisk onto /dev, but doesn't remount /dev/pts afterwards. This means

Bug#292986: initrd-netboot-tools: Fails to include hostname binary in initrd, results in failure to set hostname from dhcp

2005-01-31 Thread Brian Brunswick
On Mon, 31 Jan 2005 12:08:54 -0800, Vagrant Cascadian [EMAIL PROTECTED] wrote: Package: initrd-netboot-tools Version: N/A; reported 2005-01-31 Severity: important /etc/mkinitrd/scripts/netboot doesn't include /bin/hostname in initrd_exe list. This results in failure to set the

Bug#292986: initrd-netboot-tools: Fails to include hostname binary in initrd, results in failure to set hostname from dhcp

2005-01-31 Thread Brian Brunswick
Package: initrd-netboot-tools Version: N/A; reported 2005-01-31 Severity: important /etc/mkinitrd/scripts/netboot doesn't include /bin/hostname in initrd_exe list. This results in failure to set the hostname from dhcp when booting with nfs root. It is fixed by simply adding it to the list. My