Package: hugs
Version: 98.200503.08-1
While having fun exploring the language, I discovered something
unexpected: Hugs interprets (*2+1) as as (*(2+1)), in contravention of
http://www.haskell.org/onlinereport/exps.html#sect3.5
# Syntactic precedence rules apply to sections as follows. (op e) is
#
Package: subversion
Version: 1.1.4-2
Severity: minor
While attempting to build a bleeding-edge subversion package with the
1.2.0 tarball and python2.4 (I don't have python2.3-dev installed), I
discovered that the autogenerated control file contained an unexpected
blank line:
> Package: python2.4-
Package: bittorrent
Version: 3.4.2-3
Severity: wishlist
The trackerless support is very interesting.
A proper experimental package would be a Very Nice Thing.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: hat-ghc6
Version: 2.02-12
The hat-ghc6 in unstable is uninstallable:
The following packages have unmet dependencies:
hat-ghc6: Depends: ghc6 (< 6.2.2+) but 6.4-3 is to be installed
E: Broken packages
I presume it just needs a rebuild...
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
Package: ntp-simple
Version: 1:4.2.0a+stable-8
Severity: minor
[[ SOLVED: still submitted in case it helps anyone else.
The problem was a combination of a 127.127.1.0 refclock and
the loopback interface not being configured due to not having
uncommented the line in /etc/network/interfaces
Package: gnome-session
Version: 2.10.0-4
Since upgrading to 2.10.0-4, I've had a long timeout on the default
gnome login. It hangs after with the splash screen up, after displaying
the icon and text for "metacity window manager", but I don't know if
that means that metacity is the problem, or if
I created a brand new user, tried logging in: no delay.
I copied all the newly created files to the hanging account's .gnome2
directory (bub didn't delete any other files already there): delay.
I deleted the session manager entry from .gnome2/session: delay
I deleted .gnome2/session entirely: NO
> Could you send us that .gnome2/session file?
Er, sorry... I forgot to save a copy while blowing it away.
I got the idea when I saw that the new test user's .gnome2 directory
didn't *have* a session file, so obviously it wasn't essential.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a s
Package: evince
Version: 0.3.0-2
To reproduce:
- Download and unzip
ftp://60.248.88.208/RaidCards/Documents/Manual_Spec/Manual_02182005.zip
(This is the areaca.com.tw ftp site; I have no idea why they prefer to use
a numeric IP address.)
- Try to execute "evince Manual_02182005.pdf"
- Observe
Package: qmail-src
Version: 1.03-38
Recent versions of dovecot are *very* fast at pulling mail messages out of
the new/ directory of a maildir. (Maybe they're using dnotify/inotify?)
Anyway, qmail-local.c contains the code:
> if (link(fntmptph,fnnewtph) == -1) goto fail;
> if ((fd = open(fnnew
Package: chase
Version: 0.5.2-1
The great C++ ABI transition has taken us to libgc1c2.
chase is actually unaffected; it only uses C calls, so the existing
binary works fine with the new library.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EM
Package: libautotrace3
Version: 0.31.1-8
No biggie, it's just not installable now that the big GCC 4.0 migration
has begin. Looking at the dependency list, you're probably waiting
until all the other libraries get upgraded, but I figured I'd create a
bug for it anyway.
Thanks!
--
To UNSUBSCRI
Package: aptitude
Version: 0.2.15.9-3
Of late, aptitude has been really sluggish. Inquiring, I tried using
strace on it, and found that one cursor-down resulted in over a
thousand stat64() calls to /var/lib/apt/lists/*_Packages.
For example, 135 stat64() calls to each of the 6 files:
/var/lib/a
Package: lsh-utils
Version: 2.0.1-4
Assuming it's not a bug that libgmp3 and libgmp3c2 conflict, lsh
needs to link to the latter to work in unstable.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: libcln3
Version: 1.1.9-1
Assuming it's not a bug that libgmp3 and libgmp3c2 conflict, libcln3
needs to link to the latter to work in unstable.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: qalc
Version: 0.7.2-4
Assuming it's not a bug that libgmp3 and libgmp3c2 conflict, qalc
needs to link to the latter to work in unstable.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: abiword
Version: 2.2.7-3
Thanks to the GCC 4 transition, a bunch pf packages that link to
libaspell15c2 conflict with abiword -> libenchant1 -> libaspell15
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: xterm
Version: 6.8.2.dfsg.1-8
The various man pages (xterm, uxterm, lxterm, koi8xterm) are installed as
".1x.gz", but the postinst script sets up links to ".1.gz", which dangle.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTE
Package: mbr
Version: 1.1.5-2
The only problem is the "harness" directory, which uses vm86
to operate a test harness for simulating the operation of the
MBR. AMD64 in long mode doesn't support VM86 mode, so the vm86()
syscamm isn't even available and even compiling the harness driver
fails.
Howe
Package: smpeg-xmms
Version: 0.3.5-5
Severity: important
The great library ABI transition. It needs to be bumped to libsmpeg0c2.
Thanks.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: lyx-qt
Version: 1.3.4-2
Severity: important
The great library ABI transition. It needs to be bumped to libaiksaurus-1.2-0.
Thanks.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: xmms-modplug
Version: 2.05-8
Severity: important
The great library ABI transition. It needs to be bumped to libmodplug0c2.
Thanks.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: easytag
Version: 1.99.7-1
Severity: important
The great library ABI transition. It needs to be bumped to libid3-3.8.3c2.
Thanks.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: mysql-server-4.1
Version: 4.1.13a-1
Trying to install mysql-server_4.1.13a-1 is quite paradoxical, as it
manages, via mysql-server-4.1, to conflict with itself.
I think the right fix is to add a version dependency to the conflict in
mysql-server-4.1, so I'm submitting the bug here. But
> Hello anonymous user ;-)
Sorry; I'm scrupulous about keeping my main e-mail address off of publicly
trollable archives, and I can either use an obvious alias or make up a
plausible one. I'm never sure which is more polite.
> This is more or less intended. mysql-server (4.0.x) is now obsolete.
> No, only libsmpeg0 should exist. My problem is that I don't have an
> operational Debian machine with my private key on it. I'll see if I can
> rectify that today, and revert the NMU.
Thanks!
You might want to send a heads-up to the maintainers of packages that
have been polluted with the pac
Package: libdjvulibre1
Version: 3.5.14-6
Severity: serious
Any number of packages that link to /usr/lib/libdjvulibre.so.14
(I hit evince 0.3.0-2 first) break horribly when libdjvulibre.so.15 is
installed ahead.
I'm not sure how this should be handled, but neither the changelog
nor the version nu
Package: iproute
Version: 20061002-3
I recently discovered (by locking myself out of a remote server :-()
that the semantics of "ip rule flush" were recently changed to include
deleting the default rule pointing at the "main" routing table.
This is perhaps more of a documentation problem than any
Package: adjtimex
Version: 1.21.1-1
Severity: wishlist
Tags: patch
Just for fun, I came up with the following, which may be useful to
someone else. The "1 /" is in there to discard fractional digits.
You can substitute backquotes `...` for $(...) for portability to
older shells.
DRIFT=/var/lib/
> Does this crash still happen with libgnomeui 2.18?
Yes. You get an infinite loop of blank windows popping up and:
gnome_segv:
/home/dajobe/dev/debian/cairo/libcairo-1.4.4/src/cairo-image-surface.c:199:
_cairo_format_from_pixman_format: Assertion `NOT_REACHED' failed.
Error: Cairo 1.4.4 does no
> * Use 8bit encoding for the mails, so that they are greppable again
>(Closes: 211714)
Thank you very much! Greatly appreciated!
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: apache2.2-common
Version: 2.2.3-1
This is AMD64, if it matters.
I got some annoying errors trying to remove apache2-common:
Removing apache2-common ...
Stopping apache 2.0 web server...grep: /etc/apache2/conf.d/[^.#]*: No such file
or directory
apache2: Syntax error on line 116 of /etc/
Package: backuppc
Version: 2.1.2-5
combined with
Package: apache2
Version: 2.2.3-1
Starting web server (apache2)...Syntax error on line 9 of
/etc/backuppc/apache.conf:
Invalid command 'AuthGroupFile', perhaps misspelled or defined by a module not
included in the server configuration
failed!
T
I needed to add
LoadModule auth_basic_module /usr/lib/apache2/modules/mod_auth_basic.so
LoadModule authz_groupfile_module
/usr/lib/apache2/modules/mod_authz_groupfile.so
LoadModule authn_file_module /usr/lib/apache2/modules/mod_authn_file.so
LoadModule authz_user_module /usr/lib/apache2/modules/m
Oh, I'm sorry it wasn't obvious.
The problem is that
1) apache2-common's prerm runs /etc/init.d/apache2 stop, which
2) runs /usr/sbin/apache2 -t, which
3) parses the config files, and does a full apache setup, which
4) tries to load all the modules, but
5) apache2-mpm-prefork has already been rem
Package: cupsys
Version: 1.2.2-1
Severity: important
After an upgrade from 1.2.2-1 to 1.2.2-2, printing stopped working if
the destination was a printer class, and error_log filled up with
E [25/Aug/2006:17:20:48 +] Class "ps" hasn't any printers!
E [25/Aug/2006:17:20:48 +] Class "pcl" has
I figured after all the complaining, some praise was in order...
> * 62_classes_crash: fix incorrect code. old code returned an error
> even if user use the correct class configuration. (closes: #380663,
> #384654)
It seems to work fine, thanks!
--
To UNSUBSCRIBE, email to [EMAIL PROTEC
Package: ntpdate
Version: 4.2.2+dfsg-2.1
The relevant lines from my ntp.conf file:
peer 192.35.100.1 key 101 minpoll 4 iburst
server 127.127.29.0 minpoll 1 maxpoll 4 iburst
Result in ntpdate trying uselessly to sync up to 127.127.29.0
while failing to query the perfectly good server at 192.35.100
Package: logrotate
Version: 3.7.1-3
.. for syslogd's pid. It's checking /var/log/syslog.pid, when
/etc/init.d/sysklogd puts it in /var/run/syslog.pid.
This also causes /etc/cron.daily/logrotate to exit with status 1,
which was helpful in this case, but might not be in general.
--
To UNSUBSCRI
Package: mbr
Version: 1.1.8-1
Severity: wishlist
I've been avoiding installing 1.1.8-1 because it wants to pull in libc6-i386,
which is several megabytes of crap I don't need.
So I created the following patch to build a 64-bit install-mbr binary.
It also fixes a few compile warnings. (The = to :
Package: imgsizer
Version: 2.7-1
Depends: imagemagick | file, imagemagick | libjpeg-progs, python (>= 2.3),
python (<< 2.4)
But the default python in unstable is now 2.4.3-11.
Sorry to only ever complain; it just seemed worth a comment.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a sub
Package: dovecot-common
Version: 1.0.rc10-1
Severity: brown-paper-bag ;-)
There's an close brace in the "protocol lda" whose opening mate is
commented out.
This causes doevecot not to start.
The fix is obvious.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Tr
Package: mcelog
Version: 100-1
Architecture: i386
Severity: important
I think this is what caused #738927 to be noticed, but this is a separate issue.
# strace -efile,read,ioctl,exit_group mcelog
execve("/usr/sbin/mcelog", ["mcelog"], [/* 15 vars */]) = 0
access("/etc/ld.so.nohwcap", F_O
The fix was committed in version 0.2.3; version 2.3 is a long long
way in the future.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: nis
Version: 3.17-32
I'm not sure what happened recently; I am running sid (unstable) and a
reasonably recent kernel (3.1.0). I just upgraded to perl 5.14.
The machine is an NIS client, and for some reason ypbind is not starting.
This causes all sorts of problems with userid->name mappin
This may be a non-bug. My machine crashed (kernel panic), and on reboot it has
gone away.
If it's random hardware flakiness, sorry about the noise.
I'll try to remember to close it in a week if it hasn't reoccurred.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a
Package: bind9
Version: 1:9.8.1.dfsg-1
Severity: serious
Tags: security upstream
As you have probably heard, someone has found a way to remotely crash a bind9
server:
http://isc.sans.edu/diary.html?storyid=12049
https://www.isc.org/software/bind/advisories/cve-2011-4313
A stopgap patch (9.8.1-p1
> Thanks, I fixed this now in my bzr tree and it will be part of the
> next upload.
Er, assuming that's revision 2269:
http://anonscm.debian.org/loggerhead/apt/debian-sid/revision/2269/debian/apt.cron.daily
You don't need the final "cut -c1-5". It is true that cksum returns an
unsigned 32-bit nu
> Thanks for checking this. I like this, its more compact than the
> current code. However the reason why I did not use this, was that the
> following fragment:
>
> $ while true; do res=$(od -N2 -d /dev/urandom | cut -s -d' ' -f2); echo
> $res; if [ -z "$res" ]; then break; fi ; done
> 61581
> 42
> I confirm this. The solution is to upgrade your liboctave1 package to
> version 3.6.3-2.
Thanks for the tip; I would have upgraded that at the same time, but
managed to overlook it since it's a long way away in an alphabetical
list.
That does indeed fix it.
--
To UNSUBSCRIBE, email to debian
Thank you! I'm looking forward to trying it once it filters through the FTP
servers.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: orage
Version: 4.8.3-2
Architecture: i386
Despite the claims in the changelog:
> orage (4.8.3-3) experimental; urgency=low
>
> * Rebuild against xfce4-panel 4.10.
> * debian/control:
> - update standards version to 3.9.3.
> - update debhelper build-dep to 9.
>
> -- Yves-Ale
E: Sub-process /usr/bin/dpkg returned an error code (1)
A package failed to install. Trying to recover:
Setting up postgresql-client-9.1 (9.1.4-2) ...
update-alternatives: error: alternative pg_basebackup.1.gz can't be slave of
psql.1.gz: it is a slave of postmaster.1.gz
dpkg: error processing po
Package: cups
Version: 1.5.0-16
Severity: grave
[521]$ echo "Hello, world" | lpr
lpr: Unsupported document-format "text/plain".
The split off of a separate cups-filters package omits the "texttops"
command which is called for in /etc/cups/mime.convs.
# dpkg-deb -c cups-filters_1.0~b1-3_i386.deb
>> # grep texttops /etc/cups/mine.convs
>> application/x-cshell application/postscript 33 texttops
> This is a local configuration file, not shipped by cups. So this does
> not break the package for everybody, but nevertheless is a rather
> important upgrade issue. But I don't want to argue
Package: buildcross
Version: 0.0.12
This is due to line 1095 of /usr/lib/buildcross/functions, which
defines the em_cross function:
case ${1} in
# (other cases omitted)
gcc-3.[3-4]|gcc-4.[0-6])
GCCVER=${1#gcc-}
myecho INFO "I: Building GCC ${GCCVER}
> is the "su - nobody" necessary or do you see this as regular user,
> too?
Me, I see this as a regular user.
> The whole thing seems to depend on the specific
> /var/cache/locate/locatedb (please make backup copy), I cannot reproduce
> it here. - Does locate 4.4.2-5 work with this locate-db?
I
>s> Re-installing 4.5.11-1 produces a segfault.
> OK I'm glad you guys can reproduce it! I'll leave it in your hands.
Er... I'm just a second schlub who reported the same symptoms.
Even though the maintainer wrote to you, and not me, I figured
I'd volunteer the requested information.
But same sym
>> You're still not off the hook. Does 4.5.11-1 word for you?
> That's the version I reported.
Oops, my cut-and-paste error. I meant 4.4.2-5.
> P.S., try strace on it. (Which is also broken for me due to
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=702309 so I can't.)
Already done in my
Control: tags 704193 patch
> Does
> http://git.savannah.gnu.org/cgit/findutils.git/commit/?id=4b7c8a448651fe96b72fd1e48fe0003778efe85a
> fix the issue for you? For your convenience I have generated test
> packages with this patch included and have uploaded these to
> http://people.debian.org/~amet
Package: kernel-package
Version: 12.036+nmu3
Tags: patch
As of commit 70730bca1331fc50c3caacaea00439de1325bd6e
("kernel: Replace timeconst.pl with a bc script")
Linux kernel building depends on bc(1). Without it,
building fails with:
CC kernel/exit.o
CC kernel/itimer.o
BC ke
Package: postgresql-9.2
Version: 9.2.1-1
Severity: minor
The above kernel complaint is generated when postgres starts up.
Presumably, it would be nice to shut it up.
(Kernel is 3.6.7, i686, FWIW.)
The PID does not correspond to any running process, but is in the range:
PID TTY STAT TIME COM
Package: apt
Version: 0.9.7.6
Severity: minor
/etc/cron.daily/apt does:
random_sleep()
{
RandomSleep=1800
eval $(apt-config shell RandomSleep APT::Periodic::RandomSleep)
if [ $RandomSleep -eq 0 ]; then
return
fi
if [ -z "$RANDOM" ] ; then
# A fix for shells that
I don't know Ruby AT ALL, but I did a bit of googling and this appears
to make unhide.rb work with 1.9:
--- unhide.rb.orig 2012-12-06 23:53:57.0 -0500
+++ unhide.rb 2012-12-06 23:52:51.0 -0500
@@ -29,7 +29,11 @@
# Support for libc functions not covered by the standard Ruby
Control: found 704193 4.5.11-1
Architecture: i386
Here's an strace. The segfault happens just AFTER locatedb is closed.
It also happns on a successful lookup. I'm running i386 sid userland
on an x86_64 kernel.
execve("/usr/bin/locate", ["locate", ""], [/* 24 vars */]) = 0
brk(0)
Package: debhelper
Version: 9.20130518
Severity: serious
Since this went in, packages are appearing which depend depend (via
${misc:Depends}) on sysv-rc (>= 2.88dsf-24), which makes them difficult
to install by people who don't have sysv-rc installed.
For example, I am unable to upgrade past open
Package: debhelper
Version: 9.20130518
Severity: serious
(I already sent this with a different title, but submit@ appended it to
the closed bug 708720, which isn't helpful, as this is a different bug;
the connection is just that the fix for that introduced this.)
Since this version, packages are
Package: cpp-4.8
Version: 4.8.0-7
Tags: patch
X-Debbugs-CC: ad_deb...@joonet.de
The lilo (1:23.2-4) build system does some moderately evil things with
the preprocessor. src/Makefile does:
gcc -E -C -traditional -DLILO_ASM -o common.s common.h
Then as86 is run on first.s, which includes common.s
Package: rtorrent
Version: 0.8.6-1
Architecture: i386
Toggling peer snubbing off with * seems to leave that peer permanently
choked, even when it wakes up and starts feeding me data.
The only way I can find to bring the connection back to life is to kick
off the peer and let it reconnect.
libto
Package: libc6
Version: 2.13-6
My poor confused brain thinks "the C library" is libc-2.13.so which is NOT
in /lib (it's in /lib/i386-linux-gnu/ and /lib/i386-linux-gnu/i686/cmov/)
so the error message is not terribly useful.
The check can't be found in /var/lib/dpkg/info/libc6.preinst (from -5),
Package: xpdf
Version: 3.03-3
I thought this was Bug#640967, but I guess it's something else.
Consider the PDF file at
http://www.state-machine.com/arm/Building_bare-metal_ARM_with_GNU.pdf
When I open it, I get page after blank page containong only
images and background shading. Ignoring text in
Package: iw
Version: 3.1-1
Severity: minor
It appears to *work*, but the message is annoying.
Here's an strace around the error message:
set_thread_area({entry_number:-1 -> 6, base_addr:0xb75816c0, limit:1048575,
seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0,
us
Package: deskbar-applet
Version: 2.32.0-1+b2
python_2.7.2-7 has been uploaded to unstable, which breaks deskbar-applet.
An update would be appreciated.
Thank you! I'm sorry to only ever complain in bug reports.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subj
Package: prelink
Version: 0.0.20090925-1
prelink.bin: ../../src/conflict.c:763: prelink_build_conflicts: Assertion `j <
ndeps' failed.
This assert fail is triggered by the following files:
/usr/bin/aptitude-curses from aptitude_0.6.4-1_i386
/usr/bin/inkscape from inkscape_0.48.1-2.1+b1_i386
/us
Package: prelink
Version: 0.0.20090925-1
Severity: serious
On i386, consider the following series of commands:
# dpkg -i psmisc_22.14-1_i386.deb
(snipped)
# /usr/sbin/prelink -y --md5 /usr/bin/pstree
bb296a950a089ec8a837a1c97b47eeb9 /usr/bin/pstree
# /usr/sbin/prelink -mR -T /usr/bin/pstree
#
Thanks for this!
Of course, since this report, the 2010 CODATA values came out, and while
the ones I reported have been updated, there are still some discrepancies.
Units DB2010 CODATA value
planckmass 2.17644e-8 kg 2.17651(13)
http://physi
Package: apt
Version: 0.9.9.3
Architecture: i386
Severity: important
When proxying through a squid3 cache, an attempt to "apt-get update"
fails with all errors.
The old apt package, which worked, sent the following request
to the proxy:
GET http://http.us.debian.org/debian/dists/unstable/InRelea
Thanks for the rapid feedback! Unless you specifically want me to test
the patch (seems unnecessary, given how easy it is to reproduce), I'll
just wait for 0.9.9.4 to show up.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact
Package: pdl
Version: 1:2.4.11-4
Architecture: i386
I'm trying to recompile pdl with perl 5.18, and it's failing in the
make test phase with:
> cc -g -O2 -fstack-protector --param=ssp-buffer-size=4 -Wformat
> -Werror=format-security -Wall -ffunction-sections -Wl,-z,relro
> -Wl,--as-needed -o bl
Just to send a "me, too!", I also recently bumped udev to the
systemd version and also had a non-booting machine.
Now, most of my machines had a sufficiently well-populated static /dev
that I could boot and actually didn't notice the lack of udevd.
On the non-booting machine, it was getting somew
Control: tags 722604 + patch
udev 204-3 will work, with a hand-rolled kernel and no initramfs,
if you do the following:
1. Enable CONFIG_DEVTMPFS=y. DEVTMPFS_MOUNT is not required.
2. Apply the following patch to /etc/init.d/udev.
There are five parts to this patch:
1. Add the quotes to `mount
> Or you just update to 204-4 which has been uploaded a few hours ago.
Well, yes, thank you, but it wasn't even in the BTS when I started work
on the bug, and it still isn't on ftp.debian.org as of a few seconds
before I send this message.
ftp> ls udev_*
200 PORT command successful. Consider usin
I've been installing 204-4 on various machines, and I wanted to thank you
for both the nice clear informative warning *and* the override provision
so that I can do the upgrade before rebooting into the new kernel.
Much appreciated!
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debi
> Thank you very much, your positive feedback is appreciated as well.
>
> When you do this in your spare time a nice word now and then makes a
> huge difference.
Although I tried not to be a complete asshole about it, I was pretty
pissed off at the beginning, and in particular I saw some friction
Package: libc6
Version: 2.17-6
Architecture: i386
Severity: important
This problem cropped up while compiling the kernel; it breaks
arch/x86/tools/relocs.c. I've tracked it down and reduced
it to a small test case.
With the given arguments, regcomp() corrupts the byte past the end of
its supplie
Package: libc6-i686
Version: 2.17-6
Architecture: i386
Severity: important
With dpkg 1.16.10, the output for "dpkg-query -l libc6-i686" is
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Statu
> It does break stuff. The nvidia driver for example -- it loads
> /usr/lib/i386-linux-gnu/libnvidia-tls.so.319.23 instead of
> /usr/lib/i386-linux-gnu/tls/libnvidia-tls.so.319.23, and the non-tls library
> writes to %gs and segfaults when starting X, making the system very much
> unusable.
Ah, I
Pehaps a better solution would be to update the symlink resolution code
at the top of sh.vim (line 30 et seq.) to know about dash.
That seems like it would have a better chance of being accepted upstream.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "u
Package: vim-runtime
Version: 2:7.3.923-2
I ran into this on line 289 of /var/lib/dpkg/info/libc6:i386.preinst:
if [ "${DPKG_MAINTSCRIPT_ARCH:-$(dpkg --print-architecture)}" = "armhf" ];
then
All lines after this have the syntax messed up, apparently due to losing
track of the double-quote s
If you need to get a kernel compiling in the meantime, the
following small kernel patch works around the buffer overrun.
diff --git a/arch/x86/tools/relocs.c b/arch/x86/tools/relocs.c
index f7bab68..9978f8b 100644
--- a/arch/x86/tools/relocs.c
+++ b/arch/x86/tools/relocs.c
@@ -99,7 +99,7 @@ static
> I am personally not able to to reproduce the issue with your test
> program on an up to date i386 sid system. Could you please provide us
> the command you are using to compile the test code and the locale you
> are using?
Very interesting! I'm also using an up-to-date sid system.
No locale is
Okay, I did a bit of comparison. The machine the binary was *compiled*
on seemed to matter, not the one it was *run* on. A bit of extra
printing revealed the obvious culprit:
[539]$ ./a.out
sizeof regex_t = 28
array[1] confirmed all-zero
J'accuse! buf[0] = 0x18
Bug! array[1] has been overwritt
Thanks, I'll test it as soon as it reaches the FTP site.
Sorry I haven't been following all the activity this last month...
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: munin-doc
Version: 1.999.4603-1
dpkg -i /var/cache/apt/archives/munin-doc_1.999.4603-1_all.deb
(Reading database ... 287067 files and directories currently installed.)
Unpacking munin-doc (from .../munin-doc_1.999.4603-1_all.deb) ...
dpkg: error processing /var/cache/apt/archives/munin-do
Package: keytouch-editor
Version: 1:3.2.0~beta-3
Architecture: i386
When I click "new" and press a remote control key mapped to
513 # KEY_NUMERIC_1
keytouch-editor segfaults. Other keys I've tried, not in the
KEY_NUMERIC_x range, work as expected.
An strace log beginning at the read of the inpu
> In case you're not following the upstream ticket:
I'm not; I don't even know where their bug tracker is.
> Would you be able to build libtorrent [1] and rtorrent [2] from source,
> in their latest revision from github, and see if the issue is still
> there?
Done. Exact git commits are
libtorr
Package: console-tools
Version: 0.2.3dbs-70
Tags: patch
The bug is in debian/patches/260_kernel26fix.patch, in this hunk:
diff -ruN console-tools-0.2.3-old/kbdtools/dumpkeys.c
console-tools-0.2.3/kbdtools/dumpkeys.c
--- console-tools-0.2.3-old/kbdtools/getkeycodes.c 2004-06-26
11:53:20.000
Package: ghc-doc
Version: 7.0.4-8
ghc-doc_7.0.4-7 has an uncompressed size of 30.0 M. -8 is 96.6 M.
Supposedly the only difference is removing something:
> ghc (7.0.4-8) unstable; urgency=low
>
> * Remove ghci support on powerpc (Closes: #631073)
>
> -- Joachim Breitner Sun, 30 Oct 2011 17:
Some advice on the spamassassin mailing lists suggests that there's a
bug in NetAddr::IP 4.049 (Debian package libnetaddr-ip-perl).
The recommended fix is to use 4.048 or 4.055+.
(Bug #601601 has similar symptoms, but I think it's different.)
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ.
Package: libnetaddr-ip-perl
Version: 4.056+dfsg-0
Severity: serious
CPAN bug #71925 is preventing spamassassin from functioning
correctly. The bug was fixed in 4.053. An additonal buglet
(affecting only Perl 5.6.1 and older) was fixed in 4.054, which
doesn't affect Debian, but the author recomme
1 - 100 of 229 matches
Mail list logo
