Neil McGovern wrote:
> Package name: libphp-clam
In spite of inconsistent current practice, the proposed PHP policy is in
favor of naming such packages "php-clam".
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: ssh
Version: 1:3.8.1p1-8.sarge.4
Severity: minor
Tags: patch
When using ssh-copy-id with -i it prints out a random number to the
terminal, e.g.,
$ ssh-copy-id -i ~/.ssh/id_dsa.pub [EMAIL PROTECTED]
25
Now try logging into the machine, with "ssh '[EMAIL PROTECTED]'", and check in:
.ssh
Michael Deegan wrote:
> licq fails to start when its PID file (~/.licq/licq.pid) exists, and
> a process with that PID exists, even if that process isn't licq. I
> believe a simple check that the process has the right name (and
> owner!!) is in order.
I don't think it's possible to do that portabl
Package: kernel-image-2.6.8-1-k7
Version: 2.6.8-10
Severity: normal
I have a motherboard with an nForce2 chipset (ASUS A7N8X Deluxe). Using
the stock Debian kernel, I cannot enable DMA on the /dev/hda hard disk.
# hdparm -d 1 /dev/hda
gives
HDIO_SET_DMA failed: Operation not permitted
As I ha
Emmanuel Charpentier wrote:
> The current PostgreSQL ODBC driver still behaves as a V2 ODBC driver,
> which translates in various annoying limitations (no long integers,
> no schema use, no view creation, etc ...).
If you have a plan to magically transform it into a V3 driver, let me
know how to
maximilian attems wrote:
> the modular-ide patch got reworked for the 2.6.10 kernel-image from
> unstable, could you check if that one works for you?
Unfortunately, that new kernel did not help.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMA
Package: dpatch
Version: 2.0.15
Severity: minor
dpatch-edit-patch may produce patches that have locale-dependent variants
of "\ No newline at end of file" in them. This is not a functional issue
but it looks weird anyway. Perhaps the locale should be overridden while
diff is being run.
-- Syste
I don't think this removal was such a good idea. Certainly, cleaning up
the archive is a valid goal, but breaking dozens of packages along the
way is not. The submitter of this bug did not offer the affected
packages any upgrade path. Some of the affected packages have reached
maintenance st
Package: abakus
Version: 0.90-4
Severity: serious
Hi
The package needs to be rebuild against the new kdelibs4-dev so that the
dependency against kdelibs4c2 changes to kdelibs4c2a.
(This is because of the libstdc++ allocation transition).
Greetings
Peter
-- System Information:
Debian Release: tes
1/debian/changelog
+++ xbsql-0.11/debian/changelog
@@ -1,3 +1,11 @@
+xbsql (0.11-4.2) unstable; urgency=low
+
+ * NMU
+ * Removed automake1.6 build dependency, patching Makefile.in directly
+(closes: #335148)
+
+ -- Peter Eisentraut <[EMAIL PROTECTED]> Tue, 29 Nov 2005 21:46:59 +0100
+
xbsql (
Package: xoscope
Version: 1.12-1
Severity: serious
This package installs a file /usr/bin/bitscope, which is also installed
by package bitscope. This is not allowed by the release policy, and you
will have to rename the file.
-- System Information:
Debian Release: testing/unstable
APT prefers t
Is there anything holding up the fixing of this bug? I'm sure that you
have figured out that the patch for this bug is as below. I am faced
with a number of packages that don't clean up the PID file (because they
don't have one and I have to make it), so the integration of this patch
would be app
I suggest that the following be applied. This removes the devscripts
dependency altogether, as it's not needed.
diff -Nru /tmp/BKSAoZWDbV/cdbs-0.4.32/scripts/cdbs-edit-patch
/tmp/15PktvE0NM/cdbs-0.4.33/scripts/cdbs-edit-patch
--- /tmp/BKSAoZWDbV/cdbs-0.4.32/scripts/cdbs-edit-patch 2005-04-20
14
Package: libhsqldb-java
Version: 1.8.0.2-1
Severity: normal
$ hsqldb-querytool
Exception in thread "main" java.lang.NoClassDefFoundError:
org/hsqldb/util/QueryTool
$ jar tf /usr/share/java/hsqldb.jar | grep Query
org/hsqldb/SubQuery.class
-- System Information:
Debian Release: testing/unstable
Could the maintainers comment on this patch? As long as type-handling
exists, it should have a man page, I think.
Btw., I noticed that if you apply this patch, you should also fix the
description to no longer refer to README.Debian.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subjec
p; $arch ne "all") {
tag "invalid-arch-string-in-source-relation", "$arch [$field: $part_d_orig]"
diff -Nru /tmp/sgYfq3sg05/lintian-1.23.13/debian/changelog /tmp/WtvDGKGrHp/lintian-1.23.13pe/debian/changelog
--- /tmp/sgYfq3sg05/lintian-1.23.13/debian/changelog 200
tags 185176 + moreinfo
stop
The bug you filed about the Debian samba package was about a package
version that is now more than two years old and belongs to a
pre-release state of the old stable distribution. Please try to
reproduce your problem in a current Debian release (stable, testing, or
Any Debian init script is affected by this bug. If you kill the process
in a way that the PID file is not removed, start-stop-daemon will
complain. Improving the PID file handling in samba won't change that
materially because kill -9 won't be affected by that. The question is
perhaps whether
There is no bug here. There is an ISO standard and associated national
standards that determine how strings should be sorted. glibc is just
following that standard. And of course coreutils has nothing to do
with that; it just defers to the user's locale setting. I suggest that
this bug can
I verified that this is fixed in 5.2.1.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: coreutils
Version: 5.2.1-2.1
Severity: important
The tech-ctte had ruled some time ago that md5sum from stdin should not
output a "-" [0]. It seems to be back to the old format:
$ md5sum http://lists.debian.org/debian-ctte/2004/06/msg00032.html
-- System Information:
Debian Release: te
tags 332778 + confirmed upstream
forwarded 332778 http://bugs.kde.org/show_bug.cgi?id=74607
stop
Michael Setzer wrote:
> But I'm unable to add ed2k:// links from konqueror directly to my
> remote running core by clicking on them:
It turns out this is a bug in konqueror if the link is in a frame.
Am Donnerstag, 1. Dezember 2005 17:01 schrieb Stephen Frost:
> Probably best to use Build-Depends-Indep in this case since building
> the docs adds alot of dependencies that aren't otherwise necessary.
But running binary-arch will build the docs anyway (it's all part of "make");
so this will
tags 285244 + patch
stop
Here's a patch for this:
diff -ur ../tmp/cdbs-0.4.32/1/rules/dpatch.mk.in ./1/rules/dpatch.mk.in
--- ../tmp/cdbs-0.4.32/1/rules/dpatch.mk.in 2005-05-18 11:50:40.0
+0200
+++ ./1/rules/dpatch.mk.in 2005-12-08 09:48:44.0 +0100
@@ -74,6 +74,7 @@
dea
Here is another case:
http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=egroupware;dist=stable;arch=source
Under normal outstanding bugs you will find #329597, which is already
archived, so it can never be really resolved.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubs
Package: devscripts
Version: 2.9.9
Severity: normal
According to http://www.debian.org/Bugs/server-control, it is valid to
write
found bugnr
without a version. But the bts command does not accept that:
$ bts found 329597 , found 329597 1.0.0.007-2.dfsg-2sarge1
bts found: found #329597 in what
Adam D. Barratt wrote:
> Not 100% intuitive, but
>
> found 329597
> found 329597 1.0.0.007-2.dfsg-2sarge1
>
> should achieve what you want.
This did not do at all what I wanted. It removed all information about
where the bug was *fixed*, but did not alter any information about
where the bug
tags 311464 + patch
stop
Here's a patch for this.
diff -ur ../tmp/cdbs-0.4.32/1/rules/debhelper.mk.in ./1/rules/debhelper.mk.in
--- ../tmp/cdbs-0.4.32/1/rules/debhelper.mk.in 2005-02-03 10:14:53.0 +0100
+++ ./1/rules/debhelper.mk.in 2005-12-08 11:28:29.0 +0100
@@ -213,7 +213,7 @@
Junichi Uekawa wrote:
> If it's harmless, I'd leave it. There's no reason to
> override efforts of translators.
The translations are for users that have configured their system to see
such translations. They are not for being put in source packages where
other developers are supposed to be able
rlewczuk wrote:
> I've sent a sponsoring request for xbsql some time ago (as I'm not a
> DD) but no one responded. Packages stripped of dpatch/autmake stuff
> (as it was redundant) and fixed are available as in a message below.
I can sponsor this for you, but please do the following things first:
Package: openoffice.org
Version: 2.0.0-4
Severity: minor
Attached are some small menu item spelling corrections.
diff -ur debian.orig/openoffice.org-base.menu debian/openoffice.org-base.menu
--- debian.orig/openoffice.org-base.menu2005-12-08 23:36:59.0
+0100
+++ debian/openoffice.
Package: postgresql-server-dev-8.0
Version: 8.0.4-4
Severity: important
When you compile the postgresql package with a non-standard compiler,
then packages using the pgxs build system will also attempt to use that
compiler. I don't think it's reasonable for maintainers of those
packages to keep t
The integer timestamp option is definitely "better". The only reason
it's not the default is that not all platforms have working 64-bit
integer types. I don't see any reason to change the current
compilation option; moving a data directory from a different operating
system isn't supported any
Steve Langasek wrote:
> I don't get it. Why should *any* Debian package be depending on pgxs
> to tell it what compiler to use? Why shouldn't postgresql-pljava not
> be fixed to call gcc directly, if it can be built with it?
Because that is how the build system of those packages works. You just
Paul Dwerryhouse wrote:
> After being logged in for a while, I find that all my ICQ contacts
> who were previously listed as online are suddenly listed as offline,
> even though they definitely are still online. If I force licq to go
> offline, and then online, they reappear as online again (until
reopen 342870
stop
> Also closing the bugs on digikam and licq, which have already been
> requeued for binNMUs with the new make
The licq bug is not because of make, it's just that the make problem prevented
me from uploading a fixed version. The binNMU isn't going to help. You can
stop it an
This has become an issue again. I'm building a module for a back version of
PostgreSQL. (I know we're not supposed to use 8.0 any more, but the same is
true for 7.4.) These fake "client" header files contain symbols relevant to
server version 8.1 but not present in earlier versions, so packag
Package: devscripts
Version: 2.9.10
Severity: important
There are apparently some dependencies that are not recorded by the package
system:
$ bts tags 343460 + confirmed , forwarded 343460
'http://sourceforge.net/tracker/index.php?func=detail&aid=1050400&group_id=254&atid=100254'
DB_File needs
Martin Pitt wrote:
> Please remove the build dependency postgresql-server-dev-8.0 from
> postgresql-pljava and build the package against -8.1 instead. This
> will allow us to remove 8.0 from the archive soon.
For the record: pljava 1.1.0 does not build with 8.1, and the new
version 1.2.0, which i
Martin Pitt wrote:
> Where does it record that information? Both the buildds and i386
> package (which is built in my sid pbuilder) only record 'CC=cc', no
> specific version.
/usr/lib/postgresql/8.0/lib/pgxs/src/Makefile.global
For example, the sparc version contains:
CC = gcc-3.3
CPP = gcc-3.3
Package: cdbs
Version: 0.4.32
Severity: normal
When running cdbs-edit-patch without a debian/patches directory, it
fails:
touch: cannot touch `debian/patches/disable_no_undefined.patch': No such file
or directory
cdbs-edit-patch should take care of creating that directory, I think.
-- System I
You wrote:
> On Nov 02, Marco d'Itri <[EMAIL PROTECTED]> wrote:
> > ide-disk should be loaded by /lib/hotplug/ide.agent. Please enable
> > events logging in /etc/udev/hotplug.rules, get an events log on
> > reboot (/dev/hotplug.log) and try to understand why ide.agent is
> > not being run or is not
Christian Perrier wrote:
> In fact, it turns out that the swat package embarks files which are
> also in samba-doc, namely the following directories:
> This sounds like a packaging error which we should be able to correct
> quite quickly and easily. This is probably the reason why Peter
> tagged t
Package: linda
Version: 0.3.16
Severity: normal
Building a new version of the licq package I get a linda warning like this:
W: licq-plugin-console; The .la file /usr/lib/licq/licq_console.la contains a
libdir which is different to its path.
The .la file shown above contains a line libdir=''. Th
Package: bugs.debian.org
Severity: normal
This is concerning bug #329597.
The user submitted the bug with the following headers:
Package: egroupware
Version: egroupware/1.0.0.007-2.dfsg-2sarge1
The version line was recorded literally, which throws off the found/fixed
calculation. I've
Aníbal Monsalve Salazar wrote:
> checking for moc... /bin/moc
> grep: /bin/moc: No such file or directory
Um, where the heck is /bin/moc coming from anyway? There is no package
that installs it, as far as I can see.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe"
Marco d'Itri wrote:
> ide-disk should be loaded by /lib/hotplug/ide.agent. Please enable
> events logging in /etc/udev/hotplug.rules, get an events log on
> reboot (/dev/hotplug.log) and try to understand why ide.agent is not
> being run or is not able to load ide-disk.
Here is the hotplug.log fil
Marco d'Itri wrote:
> Please report the output of:
>
> head /proc/ide/*/media
$ head /proc/ide/*/media
==> /proc/ide/hda/media <==
disk
==> /proc/ide/hdc/media <==
cdrom
==> /proc/ide/hdd/media <==
cdrom
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble?
Package: gphoto2
Severity: wishlist
Tags: patch
Please change your build dependency from libcdk-dev to libcdk5-dev. We
are trying to remove the old libcdk4 once there are no depending
packages left (and this would appear to be the last one). Thanks.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTE
Marco d'Itri wrote:
> Please add this at the top of /lib/udev/ide.agent, reboot and send me
> the log file:
>
> exec > /dev/ide-agent.log 2>&1
> echo "$DEVPATH"
$ cat /dev/ide-agent.log
/devices/pci:00/:00:09.0/ide1/1.1
That's all.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a s
This issue seems to be fixed in 2.6.14-1-686, so feel free to close this
bug when 2.6.12 is obsoleted.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Marco d'Itri wrote:
> Please add this at the top of /lib/udev/ide.agent, reboot and send me
> the log file:
>
> exec > /dev/ide-agent.log 2>&1
> echo "$DEVPATH"
[with >>]
$ cat /dev/ide-agent.log
/devices/pci:00/:00:09.0/ide0/0.0
Cannot read /proc/ide/hda/media!
/devices/pci:00/:00
Package: mldonkey-server
Version: 2.5.28.1-1
Severity: normal
Tags: patch
The init script has two slight flaws when the stop action is called:
1. It complains about a missing PID file if the server is not running.
Add the option --quiet to start-stop-daemon to avoid that.
2. It exits with sta
Package: ftp.debian.org
Severity: normal
Please remove the package licq-plugin-jonsgtk. This package is no
longer maintained upstream and does not compile with recent licq
versions. Alternative GUI plugins for licq are readily available.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a
Jens Müller wrote:
> History files seem to have moved from dir>/history/.history to /history/.Licq.history,
> and the update script does not move them. After moving one manually
> (I saw that it the .Licq.history one was created when typing
> something ...), I could see the history. So this is an
forwarded 310670
http://sourceforge.net/tracker/index.php?func=detail&aid=1291089&group_id=254&atid=100254
tags 310670 + unreproducible
stop
Christian Ludl wrote:
> "echo 'status offline' > ~/.licq/licq_fifo" sets status to '*dnd'
> i.e. dnd+invisible others seem to work correctly ('status online
Package: sudo
Version: 1.6.8p12-1
Severity: normal
# Provides: sudu
probably should be
# Provides: sudo
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable'), (1, 'experimental')
Architecture: i386 (i686
Am Dienstag, 10. Januar 2006 11:44 schrieb Ingo Weisemoeller:
> Since I upgraded to licq 1.3.2, german umlauts in nicknames are not
> displayed correctly in the contact list. For each umlaut two rectangles
> are displayed instead.
The internal format of nicknames has changed to UTF-8. Just enter
Package: libprelude
Severity: serious
Justification: no longer builds from source
The build dependency on libgnutls-dev needs to be (>= 1.0.17). Apparently,
libgnutls11-dev (currently at version 1.0.16) also provides libgnutls-dev,
which makes the build of this package fail if that older version
Package: libgnutls-dev
Version: 1.2.9-2
Severity: normal
This package needs a dependency on liblzo-dev because linking against
libgnutls-extra via libtool pulls in liblzo (see the .la file for details).
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (9
Uwe Steinmann wrote:
> I might be better to encapsulate the php specific configuration in
> /etc/egroupware/apache.conf into .
What about when PHP 5 is used instead (not currently possible, but
soon)?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Conta
Philipp Matthias Hahn wrote:
> In response to #332744, killproc() in /lib/lsb/init-functions from
> lsb-base was changed to --name instead of --exec, because on upgrades
> daemons need to be killed by name and not by inode, since they change
> on upgrade. Therefor killproc() has to be called with $
reopen 332744
stop
I don't think just changing from --exec to --name is correct, because
LSB 3.0 says killproc takes a "pathname", not just a basename as is
apparently required now. Instead you should call start-stop-daemon
with --name $(basename $1). That would address the concern of the
or
Package: libpgjava
Version: 8.0-312-1
Severity: wishlist
As long as you are randomly renaming the package, shouldn't you use the
upstream package name? That would be "postgresql-jdbc" so the package
should be called libpostgresql-jdbc-java. That would also have the nice
side effect that people l
tags 222116 + moreinfo
stop
There is no longer any Debian suite that has Samba linked against gnutls7.
Please check whether you check reproduce the bug in a current system.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: libmx4j-java
Version: 2.1.1-2
Severity: wishlist
C-JDBC apparently needs the remote and tools support in MX4J, so I'm
making this known here so I can track the progress. If there are
reasons why this support is currently not included in the package, it
might be worth adding them here.
Package: wnpp
Severity: wishlist
* Package name: tribe
Version : 0.3
Upstream Author : ObjectWeb consortium
* URL : http://tribe.objectweb.org/
* License : LGPL
Description : Java-based group communication library
Tribe is a Java-based group communication
Rolf Leggewie wrote:
> I lost quite a few hours trying to install egroupware on my Debian
> testing system. Turns out the problem was that version 7.4 of
> postgresql does not enable tcpip_socket by default in
> postgresql.conf. That was necessary before I was able to use
> egroupware. Before tha
Package: udev
Version: 0.071-1
Severity: normal
After upgrading from the previous udev and hotplug to only udev, one of
my hard disks was no longer mounted at boot time. No device nodes for
/dev/hda* were created. I had to manually run modprobe ide_disk to fix
that. I suppose I could put that i
moved README.Debian
+
+ -- Peter Eisentraut <[EMAIL PROTECTED]> Wed, 2 Nov 2005 11:28:42 +0100
+
type-handling (0.2.16) unstable; urgency=low
* Added ugly hacks to make type-handling works for amd64 and hurd.
diff -Nru /tmp/pc5d7sIXZl/type-handling-0.2.16/debian/README.Debian /tmp/8Qw3
Package: lintian
Version: 1.23.13
Severity: normal
The test invalid-arch-string-in-source-relation complains about the following
architectures:
s390x
sh3
sh3eb
sh4
sh4eb
These are valid architectures according to /usr/share/dpkg/cputable. lintian
should probably refer to this file at runtime to
Package: apt
Version: 0.6.42.1
Severity: normal
Tags: l10n
In the German (de_DE) locale, apt asks:
Möchten Sie fortfahren [J/n]?
(Do you want to continue [Y/n]?)
but also:
Diese Pakete ohne Überprüfung installieren [y/N]?
(Install these packages without verification [y/N]?)
Note that i
Robert Millan wrote:
> I'm not the maintainer (anymore), but are you aware this utility is
> deprecated in favour of dpkg aliases now? (see #291939)
I was not aware of that, but judging from the discussion, it will still
be hundreds of years before type-handling disappears entirely.
--
To UNS
Package: kdesdk-scripts
Version: 4:3.4.2-2.1
Severity: normal
Tags: patch
kdemangen.pl parses the output of --help-all to detect the options of the
program
under scrutiny. However, it fails to parse options that only have a
single-letter
short form, e.g., "-f ". The attached patch fixes that,
Package: kdesdk-scripts
Version: 4:3.4.2-2.1
Severity: normal
Tags: patch
kdemangen.pl parses the description out of the debian/control file, but
does so incorrectly. If the depends/recommends/suggests line is continued
on a second line, that is included in the description. (Try package
kdissert
If there is no progress on this, may it would be better to compile with
gcc-3.4 right now. I've tried that locally and it manages to build
successfully.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: wnpp
Severity: wishlist
Owner: Peter Eisentraut <[EMAIL PROTECTED]>
* Package name: ggz
Version : 0.0.12
Upstream Author : Josef Spillner <[EMAIL PROTECTED]>
* URL : http://www.ggzgamingzone.org/
* License : GPL
Description : libr
Package: cdbs
Version: 0.4.32
Severity: wishlist
The autotools.mk class could automatically add --disable-dependency-tracking
to the configure options because that option is unnecessary for one-time
package builds, and the dependency tracking just wastes resources and makes
the build logs harder t
tags 212196 - wontfix
stop
With ant being in main now, I take the liberty to reactivate this bug.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Yuri Akudovich wrote:
> I'm always use invisible mode. And I put some users into Visible
> List. After upgrade to 1.3.2-2 some of this users see status as
> Offline. They are using Licq 1.3.2-2, Miranda and ICQ 2003а.
I suggest that you contact the upstream developers about this directly.
They c
owner 278810 !
stop
As there are too many people talking and no one acting, I'm going to get
this moving now, based on the preliminiary packages that have been
posted earlier.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Roland Stigge wrote:
> building the package postgresql-pljava in a clean sid build
> environment (with pbuilder) on i386 results in:
This is actually a symptom of a bug in the dependencies of the
postgresql packages, which is currently under discussion. It should be
cleared up in the next coupl
One year later... can we still expect this to happen? The packaged
docbook2man-spec.pl is really unusable. Please update it.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Svilen Stanoev wrote:
> Licq & plugins are installed using apt-get, with no errors.
> When trying to start licq with RMS plugin (-p rms) it crashes
> with the following error message:
Please check with dpkg -l licq-\* that your plugins are in sync with the
main version.
--
To UNSUBSCRIBE, emai
Package: lsb-base
Version: 3.0-11
Severity: normal
/lib/lsb/init-functions does not declare the $i used as a loop variable
at several places as local, which interacts in strange ways with init
scripts that also happen to use $i.
-- System Information:
Debian Release: testing/unstable
APT prefer
Package: selinux-policy-default
Version: 1:1.26-5
Severity: normal
Simply installing the package and purging it again leaves files lying
around, namely
$ ll /usr/share/selinux/policy/current/
assert.te flask/macros/ README
VERSION
attrib.t
Package: selinux-policy-default
Version: 1:1.26-5
Severity: normal
The installation of this package adds a line to /etc/fstab (which itself
might be frowned upon) but does not undo this change when purging the
package.
Furthermore, the package installation does not check whether the line is
alrea
Package: selinux-policy-default
Version: 1:1.26-5
Severity: normal
Prompts in the postinst script should be handled using debconf. This
applies in particular to the note about initrd. I guess the other
prompts about which policies to use come from the selinux internal
makefiles, but even those s
Package: postgis
Version: 1.0.0-1
Severity: wishlist
The PostgreSQL packagers have begun to establish a convention for the
names of PostgreSQL plugins. See the thread starting here:
http://lists.alioth.debian.org/pipermail/pkg-postgresql-public/2005-June/83.html
In light of that, please ren
Package: postgis
Version: 1.0.0-1
Severity: normal
Dependencies on the PostgreSQL server should now be
Depends: ... postgresql-7.4
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: libecpg4
Severity: grave
The package does not contain any files except some documentation.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: libecpg-compat1
Version: 8.0.3-5
Severity: grave
The package doesn't contain any files except some documentation.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: libecpg-dev
Severity: grave
The package contains shared library symlinks to (for instance) libecpg.so.5.0
but does not depend on the package which would provide that library.
Given that that package should be called libecpg5, it seems to be missing.
--
To UNSUBSCRIBE, email to [EMAIL P
Package: libpq-dev
Version: 8.0.3-5
Severity: normal
Some of the header files included in this package are not for libpq development
and should be moved to postgresql-server-dev, namely:
usr/include/postgresql/8.0/c.h libdevel/libpq-dev
usr/include/postgresql/8.0/cata
Martin Pitt wrote:
> These files have not been in the first libpq-dev packages, but I
> added them while doing the PostgreSQL client transition for main.
> Many client packages need e. g. pg_type.h. All of these files are
> required by at least one client package.
Clients that require pg_type.h ar
Am Dienstag, 24. Mai 2005 12:17 schrieb Matthias Wieser:
> * fix of buildfailure when a variable for option --build is not set
What does that mean?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Matthias Wieser wrote:
> The build-failure will only show up if thee variable
> DEB_HOST_GNU_TYPE is not set, like the case on my server. The fix
> makes a check to figuer out the state of the war and not set in the
> confflags the --build option.
OK, the rules file is actually supposed to set the
Package: cdbs
Version: 0.4.30
Severity: minor
Tags: sid
dpatch.mk currently contains this:
deapply-dpatches:
$(MAKE) -f debian/rules reverse-config
dpatch deapply-all
# after reverting patches, we don't need update-config because:
# - the tree is already in a cons
tags 306888 + confirmed
stop
This bug is waiting for the update of libksba (bug #306887).
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Am Freitag, 18. März 2005 20:30 schrieb Lars Wirzenius:
> Attached is a preliminary patch for docbook-dsssl to make it so that
> fontconfig can be built with both jade and openjade. Before today I did
> not know any DSSSL so this may be wildly incorrect. The resulting PDF
> for fontconfig (doc/font
1 - 100 of 1781 matches
Mail list logo
