Bug#532818: Fixed upstream

[Matt Zimmerman]

tags 532818 fixed-upstream
thanks

According to
https://bugs.edge.launchpad.net/ubuntu/+source/zsync/+bug/661308/comments/7
this bug is fixed upstream in zsync 0.6.2.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#423458: Status of dnscap ITP?

[Matt Zimmerman]

Robert, are you still working on this?  I'd like to see this in Debian and
am happy to help get it packaged if you want help.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#688808: don't drop util-linux from the blacklist

[Matt Zimmerman]

On Mon, Dec 17, 2012 at 09:23:25AM +0100, Tomas Pospisek wrote:
> (I'm Cc'ing Matt, checkrestart's original author here since he might
>  provide valuable insight)

Unfortunately, I cannot remember for certain why I added the check for
util-linux, nor can I find any record of the original repo (probably CVS)
for debian-goodies.

I think it was probably because it wasn't able to correctly guess how to
restart these processes (early versions assumed that the init script in the
package would do so).

In any case I am sure it is safe to remove if it generates useful output
now.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#674201: /usr/bin/qemu-i386: fork() doesn't work

[Matt Zimmerman]

I'm seeing the same problem.  There is a patch linked from
https://bugs.launchpad.net/qemu/+bug/739785/comments/32 which is reported to
fix the problem for some people.  The patch is at
http://patchwork.ozlabs.org/patch/45206/

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#679323: manpages-dev: clearenv(3) implies that it's a security tool

[Matt Zimmerman]

Package: manpages-dev
Severity: normal

I was researching ways to securely purge environment variables, e.g.
where they're used to pass credentials.  clearenv(3) says in the NOTES
section:

   Used  in  security-conscious  applications.   If  it is unavailable the
   assignment

   environ = NULL;

   will probably do.

The implementation in glibc just frees the memory and sets the environ
pointer to NULL.  Neither this, nor environ = NULL will actually
overwrite the memory used for the environment, so it's misleading to say
that it's used in security-conscious applications.  The environment
needs to be zeroed out in order to be securely cleared.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#587525: Status of npm package?

[Matt Zimmerman]

Has a package been created yet?  I'd like to see this in Debian and can help
with it if needed.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#587525: Status of npm package?

[Matt Zimmerman]

On Sun, Mar 20, 2011 at 02:23:21PM +0100, Jérémy Lal wrote:
> On 20/03/2011 12:17, Matt Zimmerman wrote:
> > Has a package been created yet?  I'd like to see this in Debian and can help
> > with it if needed.
> 
> It's waiting at :
> http://ftp-master.debian.org/new/npm_0.2.16-1.html
> 
> 0.2 branch for now.
> We'll probably upload 0.3 branch to experimental when 0.2 is accepted.

Great!  Thanks for making this happen.

-- 
 - mdz



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#24326: sysklogd: diff for NMU version 1.5-6.1, DELAYED/15

[Matt Zimmerman]

On Tue, May 10, 2011 at 10:46:18AM +0200, Martin Pitt wrote:
> Hello all,
> 
> sorry for the late answer, I didn't have much time for Debian
> recently.

Unfortunate timing, as the patch has now landed in Debian. :-/

> 
> Stefano Zacchiroli [2011-05-06 14:16 +0200]:
> > Thanks a lot for the review, Guillem. I've looked into it and I see no
> > blockers for letting the DELAYED NMU go ahead. Nonetheless, if Martin or
> > Matt want to update the patch to take into account some of your
> > comments, I'll be happy to sponsor another upload with the updated
> > patch.
> 
> TBH it's pretty obsolete. rsyslog is the standard logging daemon now,
> and since then the kernel also has been fixed to not require root
> privileges for reading (not opening) from /proc/kmsg any more (which
> was the reason for the dd process). So rsyslog never had the nasty dd
> hack, since at that point we rather got the kernel fixed.

The kernel has been fixed in Ubuntu, in Debian, or in upstream?

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#308832: Ubuntu patch

[Matt Zimmerman]

Just attaching the patch here in the BTS for reference (it was only linked
to before).

-- 
 - mdz
diff -u dhcp3-3.0.1/debian/dhcp3-server.postrm dhcp3-3.0.1/debian/dhcp3-server.postrm
--- dhcp3-3.0.1/debian/dhcp3-server.postrm
+++ dhcp3-3.0.1/debian/dhcp3-server.postrm
@@ -20,6 +20,12 @@
 	# Remove init.d configuration file
 	rm -f /etc/default/dhcp3-server
 
+# Remove pid directory
+rm -rf /var/run/dhcp3-server
+
+# Remove user
+deluser --system --quiet dhcpd
+
 	update-rc.d dhcp3-server remove >/dev/null
 	;;
 
diff -u dhcp3-3.0.1/debian/dhcp3-server.init.d dhcp3-3.0.1/debian/dhcp3-server.init.d
--- dhcp3-3.0.1/debian/dhcp3-server.init.d
+++ dhcp3-3.0.1/debian/dhcp3-server.init.d
@@ -16,13 +16,22 @@
 # should listen on.)
 . /etc/default/dhcp3-server
 
-DHCPDPID=/var/run/dhcpd.pid
+DHCPDPID=/var/run/dhcp3-server/dhcpd.pid
 
 case "$1" in
 	start)
+# allow dhcp server to write lease and pid file
+mkdir -p /var/run/dhcp3-server
+chown dhcpd:dhcpd /var/run/dhcp3-server
+[ -e /var/lib/dhcp3/dhcpd.leases ] || touch /var/lib/dhcp3/dhcpd.leases
+chown dhcpd:dhcpd /var/lib/dhcp3 /var/lib/dhcp3/dhcpd.leases
+if [ -e /var/lib/dhcp3/dhcpd.leases~ ]; then
+chown dhcpd:dhcpd /var/lib/dhcp3/dhcpd.leases~
+fi
+
 		echo -n "Starting DHCP server: "
 		start-stop-daemon --start --quiet --pidfile $DHCPDPID \
-			--exec /usr/sbin/dhcpd3 -- -q $INTERFACES
+			--exec /usr/sbin/dhcpd3 -- -q $INTERFACES -pf $DHCPDPID
 		sleep 2
 
 		if [ -f "$DHCPDPID" ] && ps h `cat "$DHCPDPID"` >/dev/null; then
@@ -35,6 +44,10 @@
 	stop)
 		echo -n "Stopping DHCP server: dhcpd3"
 		start-stop-daemon --stop --quiet --pidfile $DHCPDPID
+		# remove PID file if it is still present
+if [ -e $DHCPDPID ] && ! ps h $(< $DHCPDPID) > /dev/null; then
+rm -f $DHCPDPID
+fi
 		echo "."
 		;;
 	restart | force-reload)
diff -u dhcp3-3.0.1/debian/dhcp3-server.postinst dhcp3-3.0.1/debian/dhcp3-server.postinst
--- dhcp3-3.0.1/debian/dhcp3-server.postinst
+++ dhcp3-3.0.1/debian/dhcp3-server.postinst
@@ -9,7 +9,8 @@
 
 case "$1" in
 	configure)
-		# continue below
+# create system dhcpd user and group
+adduser --system --quiet --no-create-home --home /nonexistent --group dhcpd
 		;;
 
 	abort-upgrade|abort-remove|abort-deconfigure)
diff -u dhcp3-3.0.1/debian/changelog dhcp3-3.0.1/debian/changelog
--- dhcp3-3.0.1/debian/changelog
+++ dhcp3-3.0.1/debian/changelog
@@ -1,3 +1,32 @@
+dhcp3 (3.0.1-2ubuntu3) breezy; urgency=low
+
+  Derooted the DHCP server:
+  * Added debian/patches/droppriv.patch:
+- Provide function drop_privileges() in common/droppriv.c.
+- Interface: includes/droppriv.h.
+  * Added debian/patches/deroot-server.patch:
+- server/dhcpd.c: Immediately chown to dhcpd:dhcpd and only keep
+  CAP_NET_RAW and CAP_NET_BIND_SERVICE for the initialization phase.
+- Drop these remaining capabilities after opening the sockets.
+  * debian/dhcp3-server.init.d:
+- Create /var/run/dhcp3-server with proper permissions.
+- Set proper permissions of /var/lib/dhcp3 and
+  /var/lib/dhcp3/dhcpd.leases[~].
+- Change PID file path to /var/run/dhcp3-server/dhcpd.pid.
+  * debian/dhcp3-server.postinst: Create system user dhcpd.
+  * debian/dhcp3-server.postrm:
+- Remove /var/run/dhcp3-server on purge.
+- Remove system user dhcpd on purge.
+  * debian/control: Added build-dependency libcap-dev.
+
+  Bug fixes:
+  * debian/rules: Remove client/scripts/debian in "clean" since it is copied
+there on build.
+  * debian/dhcp3-server.init.d: Remove PID file after stop if it still exists
+(dhcpd does not remove it on its own).
+
+ -- Martin Pitt   Thu, 12 May 2005 16:19:48 +0200
+
 dhcp3 (3.0.1-2ubuntu2) breezy; urgency=low
 
   * Remove 1-second sleep from dhclient-script.  In my testing it isn't
diff -u dhcp3-3.0.1/debian/rules dhcp3-3.0.1/debian/rules
--- dhcp3-3.0.1/debian/rules
+++ dhcp3-3.0.1/debian/rules
@@ -42,7 +42,7 @@
 
 clean: unpatch
 	dh_testdir
-	rm -f build-stamp install-stamp
+	rm -f build-stamp install-stamp client/scripts/debian
 
 	# Add here commands to clean up after the build process.
 	-$(MAKE) distclean
diff -u dhcp3-3.0.1/debian/control dhcp3-3.0.1/debian/control
--- dhcp3-3.0.1/debian/control
+++ dhcp3-3.0.1/debian/control
@@ -3,7 +3,7 @@
 Priority: optional
 Maintainer: Eloy A. Paris 
 Uploaders: Matt Zimmerman , Andrew Pollock 
-Build-Depends: debhelper (>= 4.1.16), dpkg-dev (>= 1.7.0), groff
+Build-Depends: debhelper (>= 4.1.16), dpkg-dev (>= 1.7.0), groff, libcap-dev
 Standards-Version: 2.4.0.0
 
 Package: dhcp3-server
only in patch2:
unchanged:
--- dhcp3-3.0.1.orig/debian/patches/deroot-server.patch
+++ dhcp3-3.0.1/debian/patches/d

Bug#35325: Ubuntu patch

[Matt Zimmerman]

Just attaching the original patch here in the BTS for reference, rather than
relying on the old link (which may go away soon).

-- 
 - mdz
diff -u sysklogd-1.4.1/syslogd.c sysklogd-1.4.1/syslogd.c
--- sysklogd-1.4.1/syslogd.c
+++ sysklogd-1.4.1/syslogd.c
@@ -46,6 +46,10 @@
  * extensive changes by Ralph Campbell
  * more extensive changes by Eric Allman (again)
  *
+ * Wed Nov 24 2004 14:02:48 CET 2004: Martin Pitt
+ *  Added option "-u " to drop privileges to given user after
+ *  initialisation.
+ *
  * Steve Lord:	Fix UNIX domain socket code, added linux kernel logging
  *		change defines to
  *		SYSLOG_INET	- listen on a UDP socket
@@ -500,6 +504,9 @@
 #include 
 #endif
 
+#include 
+#include 
+
 #ifndef UTMP_FILE
 #ifdef UTMP_FILENAME
 #define UTMP_FILE UTMP_FILENAME
@@ -821,6 +828,11 @@
 	extern char *optarg;
 	int maxfds;
 
+/* user and group id to drop to */
+uid_t uid = 0;
+gid_t gid = 0;
+const char* username = NULL;
+
 #ifndef TESTING
 	chdir ("/");
 #endif
@@ -829,7 +841,7 @@
 		funix[i]  = -1;
 	}
 
-	while ((ch = getopt(argc, argv, "a:dhf:l:m:np:rs:v")) != EOF)
+	while ((ch = getopt(argc, argv, "a:dhf:l:m:np:rs:vu:")) != EOF)	
 		switch((char)ch) {
 		case 'a':
 			if (nfunix < MAXFUNIX)
@@ -877,6 +889,21 @@
 		case 'v':
 			printf("syslogd %s.%s\n", VERSION, PATCHLEVEL);
 			exit (0);
+		case 'u':
+if (optarg) {
+username = strdup (optarg);
+struct passwd *pw = getpwnam (username);
+if (!pw) {
+fprintf (stderr, "User %s does not exist, aborting.\n", username);
+exit (1);
+}
+uid = pw->pw_uid;
+gid = pw->pw_gid;
+} else {
+fputs ("Internal error: -u optarg == NULL!\n", stderr);
+exit (1);
+}
+			break;
 		case '?':
 		default:
 			usage();
@@ -1021,6 +1048,19 @@
 		kill (ppid, SIGTERM);
 #endif
 
+/*
+ * Drop privileges if -u was specified
+ */
+	if (username) {
+if (initgroups (username, gid) || 
+setgid (gid) || setuid (uid)) {
+perror ("Could not drop to specified user privileges");
+exit (1);
+}
+free (username);
+username = NULL;
+}
+
 	/* Main loop begins here. */
 	for (;;) {
 		int nfds;
@@ -1175,7 +1215,7 @@
 int usage()
 {
 	fprintf(stderr, "usage: syslogd [-drvh] [-l hostlist] [-m markinterval] [-n] [-p path]\n" \
-		" [-s domainlist] [-f conffile]\n");
+		" [-s domainlist] [-f conffile] [-u user]\n");
 	exit(1);
 }
 
diff -u sysklogd-1.4.1/debian/changelog sysklogd-1.4.1/debian/changelog
--- sysklogd-1.4.1/debian/changelog
+++ sysklogd-1.4.1/debian/changelog
@@ -1,3 +1,40 @@
+sysklogd (1.4.1-16.1) unstable; urgency=low
+
+  * Make syslogd and klogd run as normal user
+  * syslogd.c: support dropping root privileges to normal user
+- added command line option "-u user"
+- drop to specified user after opening the log files (so that the log
+  files can still be owned by root)
+- added changelog entry
+  * sysklogd.8: document -u option
+  * debian/postinst, debian/postrm: create/remove user 'syslog'
+  * debian/rc: 
+- use option "-u syslog" by default
+- reloading using the SIGHUP signal does not work with root-owned log
+  files any more; completely restart the daemon instead
+  * debian/control: depend on adduser (for the rc script modifications)
+  * klogd.c:
+- add option "-P file" to specify alternative kernel messages file
+  (instead of /proc/kmsg)
+- added macro PIDFILE_DIR which is preferred over _PATH_VARRUN
+- added changelog entry
+  * Makefile: added macro KLOGD_DEFINES for allowing to specify PIDFILE_DIR
+(and other macros) without changing the Makefile
+  * debian/rules: compile klogd to use pidfile in /var/run/klogd
+  * postinst.klogd: 
+- sleep a second after stopping the old klogd to avoid "klogd is already
+  running" failures on restart
+- add/remove system user and group "klog"
+  * postrm.klogd: remove system user and group "klog" on purge
+  * debian/rc.klogd:
+- create klogd pid file in /var/run/klogd/ (klog-writeable directory)
+- create a klog-owned fifo /var/run/klogd/kmsg
+- start/stop a dd process (running as root) which pipes /proc/kmsg to
+  /var/run/klogd/kmsg; this allows to run klogd entirely as user 'klogd'
+- replaced restarting by calls to stop and start
+
+ -- Martin Pitt   Wed, 24 Nov 2004 14:02:48 +0100
+
 sysklogd (1.4.1-16) unstable; urgency=medium
 
   * applied patch by cph
diff -u sysklogd-1.4.1/debian/control sysklogd-1.4.1/debian/control
--- sysklogd-1.4.1

Bug#532818: zsync redirect bug (and patch)

[Matt Zimmerman]

zsync 0.6.1 now attempts to handle redirects intelligently, but at least
some of the time, it fails.  Similar failures are described in Debian bug
532818 and Ubuntu bug 661308.

I have isolated the root cause of these failures.  Full details and a patch
are available here:

http://launchpad.net/bugs/661308

and a copy of the patch (lightly tested) is attached.

-- 
 - mdz
Only in zsync-0.6.1+bug661308: .deps
Only in zsync-0.6.1+bug661308: Makefile
Only in zsync-0.6.1+bug661308: base64.o
Only in zsync-0.6.1+bug661308: build-stamp
Only in zsync-0.6.1+bug661308: client.o
Only in zsync-0.6.1+bug661308: config.h
Only in zsync-0.6.1+bug661308: config.log
Only in zsync-0.6.1+bug661308: config.status
Only in zsync-0.6.1+bug661308/doc: Makefile
diff -ru zsync-0.6.1/http.c zsync-0.6.1+bug661308/http.c
--- zsync-0.6.1/http.c	2010-10-17 11:37:07.0 +0100
+++ zsync-0.6.1+bug661308/http.c	2010-10-17 11:35:54.804452314 +0100
@@ -731,6 +731,7 @@
 rf->sd = connect_to(rf->chost, rf->cport);
 rf->server_close = 0;
 rf->rangessent = rf->rangesdone;
+rf->buf_start = rf->buf_end = 0;
 }
 
 /* range_fetch_getmore
Only in zsync-0.6.1+bug661308: http.o
Only in zsync-0.6.1+bug661308/librcksum: .deps
Only in zsync-0.6.1+bug661308/librcksum: Makefile
Only in zsync-0.6.1+bug661308/librcksum: hash.o
Only in zsync-0.6.1+bug661308/librcksum: librcksum.a
Only in zsync-0.6.1+bug661308/librcksum: md4.o
Only in zsync-0.6.1+bug661308/librcksum: range.o
Only in zsync-0.6.1+bug661308/librcksum: rsum.o
Only in zsync-0.6.1+bug661308/librcksum: state.o
Only in zsync-0.6.1+bug661308/libzsync: .deps
Only in zsync-0.6.1+bug661308/libzsync: Makefile
Only in zsync-0.6.1+bug661308/libzsync: libzsync.a
Only in zsync-0.6.1+bug661308/libzsync: sha1.o
Only in zsync-0.6.1+bug661308/libzsync: sha1test
Only in zsync-0.6.1+bug661308/libzsync: sha1test.o
Only in zsync-0.6.1+bug661308/libzsync: zmap.o
Only in zsync-0.6.1+bug661308/libzsync: zsync.o
Only in zsync-0.6.1+bug661308: make.o
Only in zsync-0.6.1+bug661308: makegz.o
Only in zsync-0.6.1+bug661308: progress.o
Only in zsync-0.6.1+bug661308: stamp-h1
Only in zsync-0.6.1+bug661308: url.o
Only in zsync-0.6.1+bug661308/zlib: .deps
Only in zsync-0.6.1+bug661308/zlib: Makefile
Only in zsync-0.6.1+bug661308/zlib: adler32.o
Only in zsync-0.6.1+bug661308/zlib: compress.o
Only in zsync-0.6.1+bug661308/zlib: crc32.o
Only in zsync-0.6.1+bug661308/zlib: deflate.o
Only in zsync-0.6.1+bug661308/zlib: inflate.o
Only in zsync-0.6.1+bug661308/zlib: inftrees.o
Only in zsync-0.6.1+bug661308/zlib: libdeflate.a
Only in zsync-0.6.1+bug661308/zlib: libinflate.a
Only in zsync-0.6.1+bug661308/zlib: trees.o
Only in zsync-0.6.1+bug661308/zlib: zutil.o
Only in zsync-0.6.1+bug661308: zsync
Only in zsync-0.6.1+bug661308: zsyncmake


signature.asc
Description: Digital signature

Bug#405915: init script problems

[Matt Zimmerman]

On Fri, Nov 09, 2007 at 01:54:36PM +0100, Philipp Kern wrote:
> On Fri, Nov 09, 2007 at 01:43:18PM +0100, arno wrote:
> > I installed sobby, and tried to use part of your patch to make sobby run as 
> > a 
> > daemon. I found a few problems with your init script:
> 
> The init script is not acceptable as-is, which is why it is not included
> (yet).

Perhaps you could give the contributor some feedback regarding the script,
so that they can resolve your concerns with it?

This is a server package, and it really ought to have an init script to
start the service.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#405915: init script problems

[Matt Zimmerman]

On Sat, Dec 08, 2007 at 12:25:06AM +0100, Philipp Kern wrote:
> On Fri, Dec 07, 2007 at 04:42:54PM +0000, Matt Zimmerman wrote:
> > On Fri, Nov 09, 2007 at 01:54:36PM +0100, Philipp Kern wrote:
> > > On Fri, Nov 09, 2007 at 01:43:18PM +0100, arno wrote:
> > > > I installed sobby, and tried to use part of your patch to make sobby 
> > > > run as a 
> > > > daemon. I found a few problems with your init script:
> > > The init script is not acceptable as-is, which is why it is not included
> > > (yet).
> > Perhaps you could give the contributor some feedback regarding the script,
> > so that they can resolve your concerns with it?
> 
> It does some things the wrong way, in a way that caused data loss on
> Canonical's servers when they deployed it.  I will rework and add an
> improved script ASAP (maybe late December).
> 
> The point why I did not see it as that critical is that most extended
> features that users desired (like autosave directory, script dir and
> stuff) are not sensible to use when sobby is run as a service.  (I.e. it
> should not be run as root, but I don't know if nobody's an option
> neither.)
> 
> So it works perfectly well when started as a user, deploying it as a
> service just needs some more thoughts and maybe improvements in the
> software.

The natural solution would be to have the sobby package create a sobby user,
and specify default global locations for the autosave directory, etc.  This
is how most other packages handle similar situations.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#448002: libvformat1: Package name does not match soname

[Matt Zimmerman]

On Thu, Nov 15, 2007 at 03:57:40PM +0100, Cyril Brulebois wrote:
> severity 448002 normal
> thanks
> 
> Matt Zimmerman <[EMAIL PROTECTED]> (25/10/2007):
> > Severity: serious
> > Justification: Policy 8.1
> 
> Err, no.
> 
> “serious
>   is a severe violation of Debian policy (roughly, it violates a must
>   or required directive), or, in the package maintainer's opinion,
>   makes the package unsuitable for release.”
> 
> > The soname of the library in this package is libvformat.so.0, so the
> > package name should be libvformat0, not libvformat1.
> 
> You said it, that's a “should”; the Policy requirement is about
> bumping/renaming when needed. Downgrading.

"The run-time shared library needs to be placed in a package whose name
changes whenever the shared object version changes."

If you do not pay attention to this issue, then when the soname of this
library changes, you will break upgrades for all of its users.  So
regardless of how severe you think this issue is with regard to the
requirements of Debian policy, please do fix it.

-- 
 - mdz

Bug#448002: libvformat1: Package name does not match soname

[Matt Zimmerman]

On Thu, Nov 15, 2007 at 07:41:26PM +0100, Cyril Brulebois wrote:
> Matt Zimmerman <[EMAIL PROTECTED]> (15/11/2007):
> > "The run-time shared library needs to be placed in a package whose
> > name changes whenever the shared object version changes."
> > 
> > If you do not pay attention to this issue, then when the soname of
> > this library changes, you will break upgrades for all of its users.
> 
> And? The soname version isn't present in the package name (rather soname
> version+1). I don't see what the policy violation is. As you quoted, the
> name as to change when the shared object version changes. And such a
> change didn't seem to have happened. What's the actual breakage?

This isn't a policy violation if the package was misnamed this way in the
first place, but it is awfully confusing and will be even more so if the
soname version is ever incremented to 1.  I'm surprised this made it into
the archive in its current form.

Are you maintaining this package now?  The maintainer is listed as Mathias
Palm <[EMAIL PROTECTED]> in the control file.

> > So regardless of how severe you think this issue is with regard to the
> > requirements of Debian policy, please do fix it.
> 
> The only issue I can see here is a cosmetic one, and gratuitously
> renaming packages sounds a bad idea to me.

A whopping one other package depends on libvformat1, so I don't think there
would be any significant disruption from this.

As you rightly point out, policy doesn't enforce a precise naming scheme,
and you could probably name the package frobnicator3.14159 if you wanted,
but it's common (and sensible) practice to name it after the library inside.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#448002: libvformat1: Package name does not match soname

[Matt Zimmerman]

On Thu, Nov 15, 2007 at 08:54:06PM +0100, Cyril Brulebois wrote:
> Matt Zimmerman <[EMAIL PROTECTED]> (15/11/2007):
> > As you rightly point out, policy doesn't enforce a precise naming
> > scheme, and you could probably name the package frobnicator3.14159 if
> > you wanted, but it's common (and sensible) practice to name it after
> > the library inside.
> 
> I know about best practices, but that doesn't make this bug RC, thus the
> severity downgrade. The goal is to let people walking through RC bugs
> concentrate on RC issues.

You'll note that I'm not arguing with you about the severity change, as I
agree with your reading of policy.  I do think this is more than a cosmetic
issue, though.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#451945: Accepted: libsvg 0.1.4-3ubuntu1 (source)

[Matt Zimmerman]

On Wed, Nov 21, 2007 at 09:19:45PM +, Matt Zimmerman wrote:
> On Wed, Nov 21, 2007 at 01:40:44AM -, Sarah Hobbs wrote:
> > Changes: 
> >  libsvg (0.1.4-3ubuntu1) hardy; urgency=low
> >  .
> >* Merge from debian unstable, remaining changes:
> >  - Add conflicts/replaces for libsvg on libsvg1.  Libsvg is from the
> >unofficial beryl packages, but it is easier to do this than close 
> > all of
> >the bug reports that will be filed on this issue.
> 
> This seems like something which should be submitted to Debian, in hope of
> avoiding future merges.  It's a trivial patch, and presumably applies
> equally well to Debian (given that it avoids bad interaction with an
> unofficial package).

I see someone has already suggested this to the maintainer without success:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=451945

He even acknowledges that Ubuntu has worked around this problem, but doesn't
seem to want to apply the same workaround in Debian.

Rene, can you explain why?  The problem is obviously caused by someone
else's broken package, but it seems harmless to add this workaround for the
benefit of users.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#95555: xserver-xfree86: [s3virge,ati/r128] SIGALRM spinlock between server generations when using Xinerama

[Matt Zimmerman]

On Tue, Jun 19, 2007 at 08:23:46PM +0200, Brice Goglin wrote:
> About 5 years ago, you replied to a bug in the Debian BTS regarding a
> SIGALRM spinlock between server generations when using Xinerama on a ATI
> r128 and a S3 Virge boards. Did any of you guys reproduce this problem
> recently? With Xorg/Etch? With latest xserver-xorg-core and drivers? If
> not, I will close this bug in the next weeks.

I've never used Xinerama, though I helped Joe analyze the problem when it
was happening.  I'm not in a position to say whether it is still
reproducible or not.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#448002: libvformat1: Package name does not match soname

[Matt Zimmerman]

Package: libvformat1
Version: 1.13-4
Severity: serious
Justification: Policy 8.1

The soname of the library in this package is libvformat.so.0, so the package
name should be libvformat0, not libvformat1.

mizar:[/tmp] dpkg -c libvformat1_1.13-4_i386.deb 
drwxr-xr-x root/root 0 2006-07-02 10:10 ./
drwxr-xr-x root/root 0 2006-07-02 10:10 ./usr/
drwxr-xr-x root/root 0 2006-07-02 10:10 ./usr/lib/
-rw-r--r-- root/root 20484 2006-07-02 10:10 ./usr/lib/libvformat.so.0.0.0
drwxr-xr-x root/root 0 2006-07-02 10:10 ./usr/share/
drwxr-xr-x root/root 0 2006-07-02 10:10 ./usr/share/doc/
drwxr-xr-x root/root 0 2006-07-02 10:10 ./usr/share/doc/libvformat1/
-rw-r--r-- root/root   357 2006-07-02 10:10 
./usr/share/doc/libvformat1/copyright
-rw-r--r-- root/root   349 2006-07-02 10:10 
./usr/share/doc/libvformat1/changelog.gz
-rw-r--r-- root/root   588 2006-07-02 10:10 
./usr/share/doc/libvformat1/changelog.Debian.gz
lrwxrwxrwx root/root 0 2006-07-02 10:10 ./usr/lib/libvformat.so.0 -> 
libvformat.so.0.0.0

--
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#311188: Apt repository interoperability (was: Bug#311188: Debian edu messed up my Ubuntu system.)

[Matt Zimmerman]

On Sun, Apr 20, 2008 at 03:09:38PM +0200, Herman Robak wrote:
>   Apt is an awesome package manager framework.  It has a lot of power!
> But it is a powertool with few safety features aimed at Joe Average.
> I don't think we want to advertise loudly the lack of such safety
> features.  But unless we do, I think it is the duty of Debian and its
> derivatives to improve the safety nets.
> 
>   Before anyone suggests more onerous warning dialogs telling the users
> that they are on their own (more washing of hands) I would like to
> propose "upstream" as a metadata item for apt.  debian-multimedia.org
> would have a debian.org apt source as it's upstream, for example.
> Basically, apt sources could declare binary interoperability with
> other apt sources.  Any thoughts?

Because quite a few different distributions use APT and .deb repositories,
and their packages aren't necessarily interchangeable, it would be useful if
APT knew which distribution you were running and checked this against the
Release file in the repository.  This would make it possible to display a
warning or otherwise behave intelligently.

This idea has been around for a while.  There is real work associated with
doing this, however, and it hasn't yet been a big enough priority for anyone
with the necessary time and skills.

-- 
 - mdz



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#509529: e2fsprogs: e2fsck prints spurious error on low-memory system

[Matt Zimmerman]

Package: e2fsprogs
Version: 1.41.3-1
Severity: minor

On an armel system with 32M of RAM and no swap, attempting e2fsck on a certain
filesystem (details below) resulted in:

m...@stuff:~$ sudo e2fsck /dev/sdc1
e2fsck 1.41.3 (12-Oct-2008)
e2fsck: Group descriptors look bad... trying backup blocks...
Media was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
Error allocating block bitmap (1): Memory allocation failed
e2fsck: aborted

Note the "Group descriptors look bad... trying backup blocks..." error.
It's failing later due to an allocation failure, so it wouldn't succeed in
any case.  This seemed worth reporting as the earlier error hints that there
may be an earlier allocation failure which is not recognized as such.

Enabling swap and then re-running the e2fsck command avoided this error, and
it proceeds to Pass 1 directly.

I didn't let it run to completion, and instead ran the fsck on a different
system (which found no errors in the filesystem).

The filesystem in question is:

Filesystem volume name:   Media
Last mounted on:  
Filesystem UUID:  5c0d3c28-427e-4ddd-a3dd-6a9aea10fe6f
Filesystem magic number:  0xEF53
Filesystem revision #:1 (dynamic)
Filesystem features:  has_journal ext_attr resize_inode dir_index filetype 
sparse_super large_file
Filesystem flags: signed_directory_hash 
Default mount options:(none)
Filesystem state: clean
Errors behavior:  Continue
Filesystem OS type:   Linux
Inode count:  178880
Block count:  183143637
Reserved block count: 1831436
Free blocks:  4795691
Free inodes:  156167
First block:  0
Block size:   4096
Fragment size:4096
Reserved GDT blocks:  980
Blocks per group: 32768
Fragments per group:  32768
Inodes per group: 32
Inode blocks per group:   1
Filesystem created:   Sun Apr 20 13:08:15 2008
Last mount time:  Mon Dec 22 17:02:31 2008
Last write time:  Mon Dec 22 17:02:41 2008
Mount count:  362
Maximum mount count:  34
Last checked: Mon Jun 16 14:16:43 2008
Check interval:   15552000 (6 months)
Next check after: Sat Dec 13 13:16:43 2008
Reserved blocks uid:  0 (user root)
Reserved blocks gid:  0 (group root)
First inode:  11
Inode size:   128
Journal inode:8
Default directory hash:   tea
Directory Hash Seed:  032f60b3-e9c9-41db-9484-49e3e3b104c3
Journal backup:   inode blocks


-- System Information:
Debian Release: 5.0
Architecture: armel (armv5tel)

Kernel: Linux 2.6.26-1-ixp4xx
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /bin/bash

Versions of packages e2fsprogs depends on:
ii  e2fslibs  1.41.3-1   ext2 filesystem libraries
ii  libblkid1 1.41.3-1   block device id library
ii  libc6 2.7-16 GNU C Library: Shared libraries
ii  libcomerr21.41.3-1   common error description library
ii  libgcc1   1:4.3.2-1  GCC support library
ii  libss21.41.3-1   command-line interface parsing lib
ii  libuuid1  1.41.3-1   universally unique id library

e2fsprogs recommends no packages.

Versions of packages e2fsprogs suggests:
pn  e2fsck-static  (no description available)
pn  gpart  (no description available)
pn  parted (no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#679323: clearenv(3): implies that it's a security tool

[Matt Zimmerman]

Thanks for following up.  My recommendation is to say something like:

This function DOES NOT securely erase the contents of the environment.
Security-conscious applications which need to do this should use 
instead.

On Thu, Feb 18, 2016 at 06:28:19PM +0100, Stéphane Aulery wrote:
> severity 679323 minor
> stop
> -
> 
> Hello Matt,
> 
> I dig your bug reports about the clearenv() function.
> 
> Does the sentence below would do, please?
> 
> 
>Used by security-conscious application, with the reservation
>that the memory is not zeroed by the glibc implementation
>before release.
> 
> 
> Regards,
> 
> -- 
> Stéphane Aulery

-- 
 - mdz

Bug#679323: clearenv(3): implies that it's a security tool

[Matt Zimmerman]

On Fri, Feb 19, 2016 at 12:59:05PM +0100, Michael Kerrisk (man-pages) wrote:
> On 18 February 2016 at 21:34, Matt Zimmerman  wrote:
> > Thanks for following up.  My recommendation is to say something like:
> >
> > This function DOES NOT securely erase the contents of the environment.
> > Security-conscious applications which need to do this should use 
> > instead.
> 
> So, I think this report is a little confused, but mainly because of
> the poor description in the man page.
> 
> The security-conscious applications in this context are those that
> want to precisely control the environment passed to an exec()ed
> program. clearenv() cannot, indeed must not, try to erase the buffers
> containing the environment definitions. (See putenv(3) to understand
> why.) I've adjusted the man page in away that I hope explains things
> better:
> 
>The  clearenv()  function  may  be  useful  in security-conscious
>applications that want to precisely control the environment  that
>is  passed  to  programs executed using exec(3).  The application
>would do this by first clearing the environment and  then  adding
>select environment variables.
> 
>Note that the main effect of clearenv() is to adjust the value of
>the pointer environ(7); this function does not erase the contents
>of the buffers containing the environment definitions.

Yes, that's much clearer, thank you!

-- 
 - mdz

Bug#228141: Have dh_clean remove temporary CVS files

[Matt Zimmerman]

It's no longer of interest to me, so feel free to close it. Thanks for
checking in.

On Jan 31, 2018 11:42, "Niels Thykier"  wrote:

> On Sun, 31 Dec 2017 22:53:00 + Niels Thykier 
> wrote:
> > Control: tags -1 moreinfo
> >
> > On Fri, 16 Jan 2004 15:26:55 -0500 Joey Hess  wrote:
> > > Matt Zimmerman wrote:
> > > > Yes, it does...seems like exactly the same reasons that an editor
> makes
> > > > backup files, and dh_clean deletes those.  But, if you don't like it,
> > > > whatever...
> > >
> > > I'm not unchangably opposed to it; I just don't know what's right.
> > >
> > > --
> > > see shy jo
> >
> > Hi Matt,
> >
> > It seems that 14 years have past on this bug without any reaction from
> > neither you nor Joey.
> >
> > I am pinging you to hear about your opinion on this bug.  If you still
> > think it is relevant and you think it is the proper thing to do, then I
> > am happy to apply the patch.
> >
> >   Disclaimer: I have zero-knowledge of this part of CVS, so I would be
> > trusting your judgement here.
> >
> > Alternatively, if you are no longer interested in this (e.g. because you
> > no longer use CVS), then I think I will take the liberty of closing the
> > bug as "wont do".
> >
> > Thanks,
> > ~Niels
> >
> >
> >
>
> Hi Matt,
>
> A friendly ping on this.
>
> If I have not heard from you (or anyone else with interest in this bug)
> by 2018-02-19, then I will consider this bug obsolete and close it
> without changing the behaviour of dh_clean.
>
> Thanks,
> ~Niels
>

Bug#348630: Package maxima depends on libgmp3 but it is not installable...

[Matt Zimmerman]

On Tue, Jan 17, 2006 at 08:27:11PM -0800, Rob Frohne wrote:
> Package: maxima
> Version: 5.9.1-7
> 
> Hi,
> 
> I'm trying to install maxima on kubuntu Breezy using apt-get, and even 
> though I have libgmp3-dev installed, it won't let me install maxima 
> because it depends on libgmp3.

Unless this is also the case in Debian, please report this bug to the Ubuntu
bug tracker instead.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#351298: moodle: Please consider adding php5 dependency

[Matt Zimmerman]

On Tue, Feb 07, 2006 at 02:01:56PM +, Gavin McCullagh wrote:
> On Fri, Feb 03, 2006 at 05:36:59PM -0500, Rudy Godoy wrote:
> 
> > Please consider adding php5 dependency since it's already in testing
> > and according upstream's website[0] it's supported since version 1.4.
> > I've tested and using it without problems. 
> 
> 
> I see the Ubuntu patch moves exmplicitly to php5.
> 
> -Depends: ${misc:Depends}, php4 (>= 4.1.0) | php4-cgi (>= 4.1.0), php4-pgsql 
> (>= 4.1.0) | php4-mysql (>= 4.1.0), php4-gd, apache | httpd, php4-cli, 
> wwwconfig-common (>= 0.0.7), mimetex
> +Depends: ${misc:Depends}, php5 | php5-cgi, php5-pgsql | php5-mysql, php5-gd, 
> apache2 | httpd, php5-cli, wwwconfig-common (>= 0.0.7), mimetex 
> 
> Would it not be better to depend on php4 or php5 rather than force people
> to upgrade to PHP5?

Possibly, but it is quite tricky to express this correctly using the
available dependency relationships.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#351298: moodle: Please consider adding php5 dependency

[Matt Zimmerman]

On Tue, Feb 07, 2006 at 10:27:30AM -0500, Rudy Godoy wrote:
> On 07/02/2006 at 10:21 Matt Zimmerman wrote...
> 
>  
> 
> > > Would it not be better to depend on php4 or php5 rather than force people
> > > to upgrade to PHP5?
> > 
> > Possibly, but it is quite tricky to express this correctly using the
> > available dependency relationships.
> > 
> 
> This works for me.
> Depends: ${misc:Depends}, php4 (>= 4.1.0) | php5 | php4-cgi (>=
> 4.1.0) | php5-cgi, php4-pgsql (>= 4.1.0) | php4-mysql (>= 4.1.0) | php5-mysql 
> |
> php5-pgsql, php4-gd | php5-gd, apache | httpd, php4-cli |  php5-cli,
> wwwconfig-common (>= 0.0.7), mimetex

It "works" in that on a system with no PHP installed, it will generally give
you php4 and the appropriate modules.  However, it is incorrect in that it
does not express the actual dependencies of the package, and so there are
many cases where it does not work.

For example, a system with php4-cgi and php5-mysql installed would satisfy
these dependencies, but would presumably not actually be able to run moodle.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#351298: moodle: Please consider adding php5 dependency

[Matt Zimmerman]

On Tue, Feb 07, 2006 at 05:08:12PM +0100, Isaac Clerencia wrote:
> On Tuesday, 7 February 2006 16:30, Matt Zimmerman wrote:
> > On Tue, Feb 07, 2006 at 10:27:30AM -0500, Rudy Godoy wrote:
> > > On 07/02/2006 at 10:21 Matt Zimmerman wrote...
> > > 
> > >  
> > > 
> > > > > Would it not be better to depend on php4 or php5 rather than force 
> > > > > people
> > > > > to upgrade to PHP5?
> > > > 
> > > > Possibly, but it is quite tricky to express this correctly using the
> > > > available dependency relationships.
> > > > 
> > > 
> > > This works for me.
> > > Depends: ${misc:Depends}, php4 (>= 4.1.0) | php5 | php4-cgi (>=
> > > 4.1.0) | php5-cgi, php4-pgsql (>= 4.1.0) | php4-mysql (>= 4.1.0) | 
> > > php5-mysql |
> > > php5-pgsql, php4-gd | php5-gd, apache | httpd, php4-cli |  php5-cli,
> > > wwwconfig-common (>= 0.0.7), mimetex
> I've moved to something like this and left to the user the responsibility to 
> get
> a sensible set of packages. If I get bug reports which are caused by these
> "loose" Depends: I'll move to the Depends: to php5 too.

If that doesn't work out, another possible solution would be to provide
-php4 and -php5 flavours (or metapackages for the dependencies).

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#339584: hpoj: ptal-mlcd reports "Access denied to parallel port!"

[Matt Zimmerman]

On Fri, Feb 10, 2006 at 10:20:36PM +1100, Mark Purcell wrote:
> Martin, Julian,
> 
> Has ubuntu been able to make any headway on this.
> 
> I am about at the point with debian/hpoj to revert hpoj to run as root, 
> allowing access to the parallel port.

We no longer support hpoj, and use hplip in its place.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#271032: Duplicate

[Matt Zimmerman]

reassign 271032 sbuild
severity 271032 normal
merge 271032 310863
thanks

Looks like this is actually sbuild bug #310863, which is fortunate because I
didn't see how apt could cause this behavior.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#354110: Firefox crashes with "bus error" on "File open"

[Matt Zimmerman]

On Thu, Feb 23, 2006 at 02:17:05PM +0100, Jarosław Tabor wrote:
> Package: firefox
> Version: 1.5.dfsg+1.5.0.1-2
> 
> Firefox reinstalled (with --purge before) crashes in same way. Firefox
> started as root works OK.
> Setting FIREFOX_DSP="none" fixes the problem (solution found on Ubuntu
> forum).
> Distribution unstable.

See https://launchpad.net/malone/bugs/29760

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#354838: Emails from apt-listchanges are not always RFC 2045 conform

[Matt Zimmerman]

reassign 354838 python2.3
thanks

On Wed, Mar 01, 2006 at 02:37:24PM +0100, Martin Schuster wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: minor
> 
> If the listed changes contain non-7bit-clean characters (like e.g. in
> a person's name), the mails that apt-listchanges generates are not
> RFC2045-clean, and some mailservers (e.g. courier) are not happy
> about this.
> 
> Example: The changes in nagios contain a line
> "Changed, thanks to Dagfinn Ilmari Manns[strange char]ker"
> which results in courier complaining
> "This message contains improperly-formatted binary content, or
> attachment."

I cannot confirm, but apt-listchanges simply uses Python's routines for
building the email.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#360895: ImportError: libstdc++.so.6: cannot handle TLS data

[Matt Zimmerman]

reassign 360895 python2.4
thanks

On Wed, Apr 05, 2006 at 12:32:28PM +0200, Bill Allombert wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: grave
> 
> apt-listchanges just fails with error message
> 
>   Traceback (most recent call last):
> File "/usr/bin/apt-listchanges", line 30, in ?
>   import apt_pkg
>   ImportError: libstdc++.so.6: cannot handle TLS data
> 
> which render the package useless.

...which cannot possibly be the fault of apt-listchanges.  It may not be
python's, either, but that's certainly a closer place to start.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#346068: apt-get barfs on locally rebuilt package

[Matt Zimmerman]

On Thu, Jan 05, 2006 at 12:46:53PM +0100, Gabor Gombas wrote:
> - apt-get does not honour the "hold" status

It does.

> - apt-get does not notice that the installed version _does_ satisfy the
>   dependencies (it depends on libdbus-1-2 and not on libdbus-1-1), and if

apt is good at some things, but imagination is not among them.  It must be
seeing a real dependency on libdbus-1-1 somewhere.  It's possible that the
error message is misleading.

What does apt-get -f install want to do?  tar up /etc/apt /var/lib/apt and
/var/lib/dpkg so that we can try to reproduce this.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#324497: ttf-dejavu: No more DejaVuSansMono-Roman.ttf: breaks librrd2

[Matt Zimmerman]

On Mon, Jan 09, 2006 at 05:52:08PM +0100, Laurent Fousse wrote:
> * Erwan David [2006-01-09]:
> > Since last uograde of ttf-dejavu package, all rrd dependant programs
> > give the error : 
> > failed to load /usr/share/fonts/truetype/ttf-dejavu/DejaVuSansMono-Roman.ttf
> > 
> > Thus this breaks all rrd programs.
> 
> Upstream changed font names and rrdtool sets the default font filename
> at configure time in debian/rules. I suggest to add a versioned
> dependancy on ttf-dejavu in librrd2 and change the filename in rrdtool
> accordingly.

Sounds reasonable to me; an NMU would be more than welcome.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#354925: apt-cacher: cached files of same name but from different path_maps break apt-cacher

[Matt Zimmerman]

On Thu, Mar 02, 2006 at 11:15:05AM +0100, Eduard Bloch wrote:
> #include 
> * Tony Lewis [Thu, Mar 02 2006, 03:17:18PM]:
> > Package: apt-cacher
> > Version: 1.5.1
> > Severity: normal
> > 
> > *** Please type your report
> > 
> > I am caching (amongst others) debian testing and ubuntu dapper 
> > repositories.  When a package 
> > exists in both, and the package has exactly the same filename, the second 
> > repository that tries 
> > to use it does not download it (because it is already cached), but it 
> > mismatches the MD5 sum, 
> > and won't install.

This is only one of many reasons why mixing repositories from different
distributions is not a good idea.  You wouldn't mix Fedora and SuSE RPMs
even though the packaging format is the same; it's equally problematic with
Debian and Ubuntu.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#354925: apt-cacher: cached files of same name but from different path_maps break apt-cacher

[Matt Zimmerman]

On Fri, Mar 03, 2006 at 08:19:45AM +1100, Tony and Robyn Lewis wrote:
> In case you're saying that I shouldn't be mixing debs from different 
> distros installed on the one box, you're right, and I'm not.  I'm 
> running a mixed Debian and Ubuntu network, and would love apt-cacher 
> goodness on both of these.

Yes, that's what I was saying.  I didn't realize apt-cacher could be used in
this configuration, in which case, I agree that its assumption is invalid.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#356501: python2.4-minimal: This package should not exist

[Matt Zimmerman]

On Sun, Mar 12, 2006 at 01:34:23PM +, Adam Rice wrote:
> python-minimal only exists in Ubuntu so that users can uninstall python
> without breaking the base system.

No, it exists so that Python can be used in Essential: yes applications in
Ubuntu.

> Ubuntu has an agreement with the python developers that a user will only
> get a system with python-minimal but without the full python base packages
> if they specifically request it.

We don't even offer such an option; the user would need to explicitly
uninstall the relevant packages (and cripple their Python installation) in
order to do so.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#356997: apt: internationalization of yes/no answers does not accept displayed answers

[Matt Zimmerman]

severity 356997 normal
thanks

On Wed, Mar 15, 2006 at 09:37:31AM +0100, Tim Krah wrote:
> Package: apt
> Version: 0.6.43.3
> Severity: important
> Tags: l10n
> 
> In the german locale (de_DE), apt asks:
> 
> Möchten Sie fortfahren [J/n]?
> (Do you want to continue [Y/n]?)
> 
> and eventually
> 
> Diese Pakete ohne Überprüfung installieren [j/N]?
> (Install these packages without verification [y/N]?)
> 
> 
> If you answer 'j' apt acts as if you'd answer 'n'!
> Only if you answer 'y' apt continues as expected.
> This normally does not occur in the first case, as 'y' is the predefined
> answer and you virtually allways only hit enter.
> But in the second case the predefined answer is 'n', so you have to tell
> apt that you really want to install these packages by answering 'j',
> what causes apt to cancel because it expects 'y'!
> 
> This is not clear to the user and is not what you would expect.

apt uses nl_langinfo(YESEXPR) to match affirmative and negative answers; it
shouldn't expect 'y'.  Can you provide more detail about your circumstances
which might help explain why that isn't working for you?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#350621: mipsel support for fdlibm.h

[Matt Zimmerman]

On Mon, Jan 30, 2006 at 07:41:33PM +, Thiemo Seufer wrote:
> On Mon, Jan 30, 2006 at 07:02:31PM +, Ian Jackson wrote:
> > -#if defined(linux) && (defined(__i386__) || defined(__x86_64__) || 
> > defined(__ia64))
> > +#if defined(linux) && (defined(__i386__) || defined(__x86_64__) || 
> > defined(__ia64) || defined(__MIPSEL))
> >  #define __LITTLE_ENDIAN
> >  #endif
> 
> This should probably be
> 
>   ... || (defined(__mips) && defined(__MIPSEL__)))

Isn't mips big-endian?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#351056: apt-get always install the same two packages during upgrade or dist-upgrade

[Matt Zimmerman]

reassign 351056 python-twisted
thanks

On Thu, Feb 02, 2006 at 03:33:23PM +0100, Przemek Swiderski wrote:
> Package: apt
> Version: 0.6.43.2
> Severity: important
> 
> 
> When i make upgrade two packages upgrades:
> 
>   ssc:/# apt-get upgrade
>   Reading package lists... Done
>   Building dependency tree... Done
>   The following packages will be upgraded:
>   python-twisted python2.3-twisted

This is a bug in python-twisted; the maintainer is already aware of it.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#357815: iax: not all source files are LGPLed

[Matt Zimmerman]

On Sun, Mar 19, 2006 at 06:21:04PM +0100, Julien BLACHE wrote:
> Package: iax
> Severity: important
> 
> Hi,
> 
> Not all the source files in the iax package are LGPLed. Half of the files
> in fact GPLed.
> 
> debian/copyright should reflect this.
> 
> (iaxmodem includes a copy of iax and got rejected for this very reason)

I am no longer using Asterisk and so have little use for iax; since you seem
to be actively working with this software, are you interested in adopting
iax?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#357815: iax: not all source files are LGPLed

[Matt Zimmerman]

On Sun, Mar 19, 2006 at 10:03:53PM +0100, Julien BLACHE wrote:
> Kilian Krause <[EMAIL PROTECTED]> wrote:
> 
> Hi,
> 
> > Go ahead. Once it's in SVN we all can add bits and pieces where needed,
> > so the important step is to import it. For me it largely makes sense to
> > have it in pkg-voip disregarding who's the primary developer working on
> > it.
> 
> OK, I'll prepare an upload this week.

Great, thanks for taking it over.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#375680: Ubuntu and Fedora patches reviewing

[Matt Zimmerman]

On Tue, Jun 27, 2006 at 05:19:19PM +0200, Aur??lien G??R??ME wrote:
> Package: yaboot
> Severity: wishlist
> 
> As Sven suggested, I post the current
> Ubuntu patch against Yaboot, because the URL
> 
> given in the Debian PTS does *not* seem to work reliably.

The patch archive was being regenerated.  This process seems to have been
started some hours ago, however, and I would not expect it to take this
long.  CCing Scott to check on it.

> Ubuntu feedback to Debian really sucks. :(

This isn't a very constructive comment; it would have been more appropriate
to report this (transient) problem to Ubuntu instead.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#375953: apt-listchanges: does not work anymore

[Matt Zimmerman]

severity 375953 grave
severity 375300 grave
merge 375953 375300
thanks

On Thu, Jun 29, 2006 at 09:55:35AM +0200, Didrik Pinte wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: important
> 
> 
> Hi,
> 
> Since monday, the 26th, my apt-listchanges does not seem to work
> anymore.
> 
> When using apt-get or aptitude, the output show it parses the changelogs
> but nothing is done by apt-listchanges.
> 
> I'm pretty sure the bug is not on the apt-listchanges side but I don't
> know where to report it else.
> 
> I've purged the apt-listchanges package from my laptop, and reinstalled
> it again. It didn't change anything to the behaviour. Nothing is done.
> 
> I can provide any debugging or logs needed to solve the problem
> 
> Many thanks in advance,
> 
> Didrik
> 
> -- System Information:
> Debian Release: testing/unstable
>   APT prefers unstable
>   APT policy: (990, 'unstable'), (500, 'testing')
> Architecture: i386 (i686)
> Shell:  /bin/sh linked to /bin/bash
> Kernel: Linux 2.6.17-1-686
> Locale: LANG=fr_BE, LC_CTYPE=fr_BE (charmap=ISO-8859-1)
> 
> Versions of packages apt-listchanges depends on:
> ii  apt   0.6.44.2   Advanced front-end for dpkg
> ii  debconf   1.5.2  Debian configuration management 
> sy
> ii  debianutils   2.16.1 Miscellaneous utilities specific 
> t
> ii  python2.3.5-11   An interactive high-level 
> object-o
> ii  python-apt0.6.18 Python interface to libapt-pkg
> ii  ucf   2.0012 Update Configuration File: 
> preserv
> 
> apt-listchanges recommends no packages.
> 
> -- debconf information:
>   apt-listchanges/confirm: false
> * apt-listchanges/email-address: did
> * apt-listchanges/which: both
> * apt-listchanges/frontend: mail
> * apt-listchanges/save-seen: true

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#372077: use NEWS.Debian

[Matt Zimmerman]

On Thu, Jun 29, 2006 at 09:09:31PM -0400, David Nusinow wrote:
> On Fri, Jun 23, 2006 at 04:01:38PM -0400, Joey Hess wrote:
> > One way to fix this would be to promote apt-listchanges to standard
> > priority so it's available everywhere and add a NEWS.Debian entry for
> > the issue.
> 
> I would love this personally. I feel like apt-listchanges should be
> standard, and to be honest, I'm sick of telling newbies in #debian to
> install it. That it would solve this bug is the cherry on top.

If it is important that it be displayed prior to the upgrade, and give the
user a chance to abort it, then debconf seems more approriate.  If it's
sufficient to notify the user after the fact that they need to verify their
system, then this would be better suited to something like update-notifier.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#313665: apt-listchanges: [INTL:de] German PO file corrections

[Matt Zimmerman]

On Mon, Jun 05, 2006 at 01:51:46PM +0200, Jens Seidel wrote:
> On Tue, Jun 14, 2005 at 11:27:52PM +0200, Jens Seidel wrote:
> > attached you will find an update of the German PO file de.po.
> 
> Matt, could you please fix this bug? It is open since one year and
> contains a patch!

I would very much appreciate an NMU, co-maintainer or new maintainer for
this package due to real-life constraints.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#318581: Processed: adjust bug severity.

[Matt Zimmerman]

severity 318581 normal
thanks

On Wed, Apr 19, 2006 at 08:18:50AM -0700, Debian Bug Tracking System wrote:
> Processing commands for [EMAIL PROTECTED]:
> 
> > severity 318581 grave
> Bug#318581: workaround: python2.3 gettext iconv to .po encoding -> current 
> locale
> Bug#238203: apt-listchanges: Polish localization in UTF-8, not ISO-8859-2 only
> Bug#309470: apt-listchanges: Invalid non-UTF8 character in French translation
> Severity set to `grave'.

Justification?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#318581: Processed: adjust bug severity.

[Matt Zimmerman]

On Thu, Apr 20, 2006 at 09:27:30AM +0900, Junichi Uekawa wrote:
> Hmm... after a night of sleep, this logic is flawed.
> Nevertheless, it's nice to have this fixed before etch.

Is etch not going to have python2.4?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#363378: flashplugin-nonfree: Ubuntu patches?

[Matt Zimmerman]

On Fri, Apr 21, 2006 at 03:26:00PM +0900, Kai Hendry wrote:
> Package: flashplugin-nonfree
> Version: 7.0.63.1-ubuntu1
> Followup-For: Bug #363378
> 
> This problem is evidently fixed in dapper.
> https://launchpad.net/distros/ubuntu/+source/flashplugin-nonfree/+bug/3204

That seems unlikely; the only change in that package relative to Debian is a
change in the priority of a debconf question.  It looks like a user stopped
having the problem and assumed it was due to an upgrade.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#362896: Severity silently changed without explanation

[Matt Zimmerman]

severity 362896 normal
thanks

On Fri, Apr 21, 2006 at 01:33:51PM +0200, Jonas Smedegaard wrote:
> severity 362896 important
> thanks
> 
> Please do not lower severity without at least explaining your
> reasoning. And please cc the bugreport itself, so as to make the bug
> submitter aware of the change.
> 
> I consider it an important issue when a tool running as root crashes.
> 
> I believe this particular bug is tied to "rude behaviour" of other
> packages, but it is no more than that: Debian policy do not mandate
> packages to not ship a NEWS.Debian directoy AFAIK.

A crash on invalid input is not an important bug, nor is a sub-optimal error
message.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#264985: on the status of checkrestart

[Matt Zimmerman]

On Thu, Apr 27, 2006 at 01:14:56PM -0400, Justin Pryzby wrote:
> I am still confused about the ".dpkg-new" check; could you please
> comment on it?

It was the right thing to do at the time in order to get the correct list of
files.

> If you like, I will work on patching checkrestart.  If so, let me know
> your preference for patching the present implementation, or starting
> from scratch, and trying to maintain a similar interface.

By all means; would you consider adopting the package?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#195701: gnucash: "Liabilities over time" report shows no data

[Matt Zimmerman]

On Thu, Apr 27, 2006 at 11:55:23AM -0700, Thomas Bushnell BSG wrote:
> I believe that this is a bug.
> 
> The "liability" account *is* marked with the type "Liability", and so
> it *should* show up in the report.  It should not matter what the
> parent account is.

I agree.  If the parent account type is used instead, it becomes impossible
to mix different types of accounts within a group, and I see no reason to
have such a restriction.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#369437: apt-get shouldn't say that A replaces B while A depends on B

[Matt Zimmerman]

severity 369437 minor
retitle 369437 "However the following packages replace it" is simple-minded
thanks

On Mon, May 29, 2006 at 08:49:32PM +0200, Vincent Lefevre wrote:
> Package: apt
> Version: 0.6.44.1
> Severity: normal
> 
> apt-get gives a confusing message:
> 
> ay:~> sudo apt-get install -t unstable sun-java5-bin
> Password:
> Reading package lists... Done
> Building dependency tree... Done
> Package sun-java5-bin is not available, but is referred to by another package.
> This may mean that the package is missing, has been obsoleted, or
> is only available from another source
> However the following packages replace it:
>   sun-java5-jre
> E: Package sun-java5-bin has no installation candidate
> 
> It shouldn't say that sun-java5-jre replaces sun-java5-bin because
> the user needs sun-java5-bin anyway (indeed sun-java5-jre depends
> on sun-java5-bin).

This feature looks at the Replaces field to guess that another package might
be meant to supersede the one that the user is looking for.  It should be
worded less authoritatively since this is at best a hint where to look for a
solution.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#376098: librrd0-dev: The package dependancies are broken and remove packages that I need+want

[Matt Zimmerman]

reassign 376098 libgd2
thanks

On Fri, Jun 30, 2006 at 11:00:41AM +0200, Ramon Bastiaans wrote:
> Package: librrd0-dev
> Severity: normal
> 
> 
> root# apt-get install librrd0-dev
> Reading Package Lists... Done
> Building Dependency Tree... Done
> The following extra packages will be installed:
>   libgd2-noxpm libgd2-noxpm-dev
> Suggested packages:
>   libgd-tools
> The following packages will be REMOVED:
>   libgd2-xpm php4-gd
> The following NEW packages will be installed:
>   libgd2-noxpm libgd2-noxpm-dev librrd0-dev
> 0 upgraded, 3 newly installed, 2 to remove and 0 not upgraded.
> Need to get 601kB of archives.
> After unpacking 942kB of additional disk space will be used.
> Do you want to continue? [Y/n]
> 
> Why is it removing php4-gd ? I need and want php4-gd! :)
> 
> What does GD have to do with RRDTool? Nothing..?

RRDTool uses libgd to create its graphics, and always has.  The problem here
is another unpleasant side effect of the fact that libgd2 has (in my
opinion, unnecessarily) been split up into -xpm and -noxpm versions.  One is
needed by php4-gd, the other by RRDtool, and only one can be installed at a
time.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#375300: diff for 2.59-0.3 NMU

[Matt Zimmerman]

On Sun, Jul 02, 2006 at 05:01:45PM +0200, Pierre HABOUZIT wrote:
> Hi,
> 
> Attached is the diff for my apt-listchanges 2.59-0.3 NMU.

Thanks.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#376777: apt-utils: apt-ftparchive fails to generate Contents files

[Matt Zimmerman]

On Tue, Jul 04, 2006 at 11:55:16PM +0100, James Troup wrote:
> Package: apt-utils
> Version: 0.6.44.2
> Severity: serious
> Tags: help
> 
> (Backported) apt-ftparchive on ftp-master currently fails to generate
> Contents files.  I've reproduced this locally on sid with vanilla
> apt-utils and a trivial apt.conf file.

Is the source identical to 0.6.44.2 in unstable?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#264985: checkrestart must be fixed or dropped

[Matt Zimmerman]

severity 264985 important
thanks

On Wed, Jul 12, 2006 at 12:38:26AM +0100, Sam Morris wrote:
> severity 264985 serious
> thanks
> 
> I'm upgrading this bug because checkrestart is currently useless at
> best, and a security problem at worst. It must be fixed or dropped.

It is not a security problem, and it doesn't make the package unusable.
I agree that it should be dropped if it isn't feasible to fix it.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#264985: checkrestart must be fixed or dropped

[Matt Zimmerman]

On Wed, Jul 12, 2006 at 01:10:49AM +0100, Sam Morris wrote:
> On Tue, 2006-07-11 at 17:02 -0700, Matt Zimmerman wrote:
> > > I'm upgrading this bug because checkrestart is currently useless at
> > > best, and a security problem at worst. It must be fixed or dropped.
> > 
> > It is not a security problem, and it doesn't make the package unusable.
> > I agree that it should be dropped if it isn't feasible to fix it.
> 
> I argue that it should be considered a security problem: it is possible
> for users to run it, and not realise that it doesn't work. The users may
> therefore not notice that they must restart a process in order to
> eliminate their exposure to a vulnerability (that was fixed by upgrading
> a library which that process makes use of).

This is a very tenuous argument; by this criteria, practically any
functionality bug could be considered a security problem ("the fonts in my
web browser are too small, therefore I can't read security advisories").

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#376777: apt-utils: apt-ftparchive fails to generate Contents files

[Matt Zimmerman]

On Wed, Jul 12, 2006 at 11:08:47PM +0100, James Troup wrote:
> Matt Zimmerman <[EMAIL PROTECTED]> writes:
> 
> > On Tue, Jul 04, 2006 at 11:55:16PM +0100, James Troup wrote:
> >> Package: apt-utils
> >> Version: 0.6.44.2
> >> Severity: serious
> >> Tags: help
> >> 
> >> (Backported) apt-ftparchive on ftp-master currently fails to generate
> >> Contents files.  I've reproduced this locally on sid with vanilla
> >> apt-utils and a trivial apt.conf file.
> >
> > Is the source identical to 0.6.44.2 in unstable?
> 
> Of the backport?  Yes.  But I don't understand the point of the
> question.  As I said, it's reproduceable "on sid with vanilla
> apt-utils", i.e. the 0.6.44.2 binaries from unstable running in a
> unstable chroot.

I wanted to be sure of what you were running so that when investigating the
problem, I was looking at the same source you were using.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#368462: alsa-utils depends on dummy package python-minimal

[Matt Zimmerman]

On Mon, May 22, 2006 at 02:04:23PM +0100, Adam Rice wrote:
> python-minimal is a dummy package which appears to exist only in a misguided
> attempt to be compatible with Ubuntu.

No package in Ubuntu depends on python-minimal except openoffice.org (which
I just noticed and consider a bug).

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#368695: apt-listchanges: Problem with the module apt_pkg : "ImportError: No module named apt_pkg"

[Matt Zimmerman]

severity 368695 normal
reassign 368695 python-apt
thanks

On Wed, May 24, 2006 at 10:04:31AM +0200, bdepardon wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: grave
> Justification: renders package unusable
> 
> When I try to use apt-listchanges it stops saying it can't import the
> module apt_pkg :
> Traceback (most recent call last):
>   File "/usr/bin/apt-listchanges", line 30, in ?
>   import apt_pkg
>   ImportError: No module named apt_pkg
>   
> I found that this module was meant to be included in the python-apt, I
> just installed it when I installed apt-listchanges.

That module is included in python-apt, and it is working for others.  Please
try to examine your system and find out what is wrong.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#317822: pathname conflicts are serious

[Matt Zimmerman]

On Tue, Apr 11, 2006 at 02:27:45AM +0200, Piotr Engelking wrote:
> On 08/04/06, Justin Pryzby <[EMAIL PROTECTED]> wrote:
> > The etch release policy states that:
> >
> > | http://release.debian.org/etch_rc_policy.txt
> > |   If two packages cannot be installed together, one must list
> > |   the other in its "Conflicts:" field.
> >
> > At least one of these packages must conflict with the other.
> 
> Conflicting is not enough. Etch RC Policy also states:
> 
>   Packages must not install programs in the default PATH with
>   different functionality with the same file name, even if they
>   Conflict:.

They both do more or less the same thing (though they are not fully
compatible), so I don't think this is a concern.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#379708: bogl-bterm: Graceful exit on SIGTERM

[Matt Zimmerman]

On Tue, Jul 25, 2006 at 10:10:58AM +0200, Petter Reinholdtsen wrote:
> diff -ur usplash-0.3-4/bogl/bterm.c bogl-0.1.18/bterm.c
> --- usplash-0.3-4/bogl/bterm.c  2006-06-20 12:11:13.0 +0200
> +++ bogl-0.1.18/bterm.c 2004-07-22 12:39:42.0 +0200
> @@ -64,7 +64,6 @@
> 
>  static int child_pid = 0;
>  static struct termios ttysave;
> -static int quit = 0;

This patch looks reversed.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#291642: Creates data directory under /usr/lib/doc-base breaking doc-central

[Matt Zimmerman]

On Fri, Jan 21, 2005 at 05:15:04PM -0800, Karl Hegbloom wrote:

> Package: doc-base
> Version: 0.7.20ubuntu5
> Severity: normal
> 
> 
> Please see Debian BTS bug #290846
> 
> Is it a Ubuntu problem, or a Debian Sid problem?

Please check before reporting the bug, or if in doubt, report it to Ubuntu,
so as not to bother Debian maintainers with inapplicable bugs.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#291447: acpi-support: asus-lock is just a copy of asus-wireless

[Matt Zimmerman]

On Mon, Jan 24, 2005 at 07:35:58PM -0800, Karl Hegbloom wrote:

> On Mon, 2005-01-24 at 00:02 +, Martin Michlmayr wrote:
> > * Karl Hegbloom <[EMAIL PROTECTED]> [2005-01-20 12:08]:
> > > Package: acpi-support
> > > Version: 0.13
> > 
> > This package does not seem to exist in Debian.
> 
> It must be Ubuntu specific.  I reported it with 'reportbug', and that
> must not be smart enough to report bugs to Ubuntu's 'bugzilla'.  Do you
> think there's a way to route these kind of reports to them?  What can be
> done?  It's difficult for the end-user to know who to report a bug to.

reportbug in Ubuntu is explicitly and brutally patched to submit bugs to
Ubuntu rather than to Debian, in order to avoid exactly this problem.  You
must be mixing packages from Debian and Ubuntu; perhaps you have Debian's
reportbug package installed.  Though, Debian's reportbug is now smart enough
to recognize that it is on an Ubuntu system and do the right thing (though I
haven't tested it personally), so I have no idea how you managed this.  At
any rate, yes, we already thought of this, and took measures to prevent this
kind of problem, but you have somehow circumvented them.

At any rate, as I already explained to you in a previous message, you should
report bugs to Ubuntu Bugzilla when in doubt.  We do not want Debian
developers' time to be wasted filtering out Ubuntu bugs.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#291447: acpi-support: asus-lock is just a copy of asus-wireless

[Matt Zimmerman]

On Mon, Jan 24, 2005 at 10:54:56PM -0800, Karl Hegbloom wrote:

> I have done nothing ot circumvent anything.  All I did was install
> 'reportbug' and use it to report a bug.  I do not have the Debian
> repositories in my sources.list; only the Ubuntu ones.
> 
> reportbug version is:  3.5ubuntu2

It looks like one of the patches may have been dropped in the process of a
manual merge.

> Certainly, I agree.  This should not have happened, but it's not my
> fault.  'reportbug' should either do the right thing or not be
> available.

You had already been told that Bugzilla was the proper place to report bugs.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#291447: acpi-support: asus-lock is just a copy of asus-wireless

[Matt Zimmerman]

On Tue, Jan 25, 2005 at 12:09:22AM -0800, Karl Hegbloom wrote:

> Sure, but how do I easily determine whether a bug belongs to Ubuntu
> base, or to universe?  If it's in universe, the bugzilla page says not
> to report them there.  So I thought that 'reportbug' would be able to
> look that up and do the right thing with the report.  If it cannot, then
> what's the point of having it in the distro?

I already explained that the situation with reportbug is a bug.  The version
in Warty behaves more reasonably; this was a bug introduced in the
development branch.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#283923: close?

[Matt Zimmerman]

On Thu, Jan 20, 2005 at 03:13:51PM -0500, Joey Hess wrote:

> AFAIK this bug is fixed and can be closed. All the tasks are
> successfully installing with the new apt.

The underlying bug still exists; it's just masked by increasing the argument
limits for dpkg.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#283923: close?

[Matt Zimmerman]

On Wed, Jan 26, 2005 at 08:25:59AM +0100, Christian Perrier wrote:
> Quoting Matt Zimmerman ([EMAIL PROTECTED]):
> > On Thu, Jan 20, 2005 at 03:13:51PM -0500, Joey Hess wrote:
> > 
> > > AFAIK this bug is fixed and can be closed. All the tasks are
> > > successfully installing with the new apt.
> > 
> > The underlying bug still exists; it's just masked by increasing the argument
> > limits for dpkg.
> 
> Sure, but this underlying bug is already recorded in the BTS, isn't
> it?

Yes, this bug is merged with it.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#292518: /etc/jackrc says flac isn't in Debian even though it is in woody

[Matt Zimmerman]

On Thu, Jan 27, 2005 at 03:59:56PM +, Kristinn Gylfason wrote:

> so this statement from /etc/jackrc: 
> 
> -
> # The default encoder. You can choose from 'oggenc'(default), 'xing',
> # 'mppenc', 'lame', 'l3enc', 'gogo', 'mp3enc' or 'flac' (only oggenc is
> # avaiable in Debian).
> #encoder:oggenc
> -
> 
> is a bit ambiguous or at least in danger of getting out of date ;-)

It is very curious indeed, considering that flac has been in Debian for
longer than jack. ;-)

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#283923: close?

[Matt Zimmerman]

On Thu, Jan 27, 2005 at 07:09:11AM +0100, Christian Perrier wrote:

> This is why I tended to agree with Joey. What I reported in #283923 is
> indeed fixed, so my opinion would be closing te bug...while keeping
> the real bug opened.
> 
> Anyway, if something similar to #283923 happens in the future it is
> quite likely to be reported against a completely different package as
> the problem is not obvious to identitify as pertaining to apt at first
> glance.
> 
> On the other hand, the advantage of keeping #283923 opened is keeping
> the long discussion in it, which finally concludes that the bug is in
> APT,  easy to find

If you feel it would be useful to unmerge this bug and close it, feel free.
It makes no real difference to me.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#292691: unable to verify signatures

[Matt Zimmerman]

reassign 292691 ftp.debian.org
retitle 292691 Release.gpg is 0 bytes
thanks

lftp ftp.us.debian.org:/debian/dists/unstable> ls
[...]
-rw-rw-r--1 1176 117634068 Jan 28 20:24 Release
-rw-r--r--   67 1176 11760 Jan 28 20:24 Release.gpg

The key expired yesterday.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#284081: New parted api, and possibility for parted 1.6.x (x>12) in debian/sarge ...

[Matt Zimmerman]

On Mon, Jan 31, 2005 at 12:14:35PM -0500, Branden Robinson wrote:

> Thanks for the update, the patch, and the ping.
> 
> Python-parted has not been actively worked on within Progeny for quite
> some time -- I think this is mainly due to our Debian "port" of Anaconda
> using Red Hat's version of Python-parted instead.
> 
> Nevertheless, I'm having a look at your patch right now.

Should Debian replace the Progeny python-parted with the Red Hat one, then?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#284081: New parted api, and possibility for parted 1.6.x (x>12) in debian/sarge ...

[Matt Zimmerman]

On Mon, Jan 31, 2005 at 05:13:04PM -0500, Jeff Licquia wrote:

> On Mon, 2005-01-31 at 17:05 -0500, Branden Robinson wrote:
> > Do you mean me or Matt when you say "you"?
> 
> Sorry.  I meant Matt.  But you can have them too, if you want them.
> 
> If both you (Branden) and Matt want them, then we've got a problem. :-)

I don't have any pressing need for them; I'm interested in having the
functionality, but I'm more interested in unmaintained software being
removed from the archive.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#310053: gzip: gzip hangs in futex(..., FUTEX_WAIT) indefinetely

[Matt Zimmerman]

On Sat, May 21, 2005 at 12:02:42PM +0200, Petter Reinholdtsen wrote:

> Package: gzip
> Version: 1.3.5-9
> Tags: patch
> 
> While looking at the ubuntu patches for gzip, I found some changes
> which are missing in the debian package.  These are the ubuntu
> changelog entries related to the fixes:

FWIW, I sent this patch to Bdale shortly after implementing the fix.

(CCing upstream this time as well)

-- 
 - mdz
--- Begin Message ---
To my knowledge, no one has experienced problems like this on Debian, but
it surfaced on a couple of Ubuntu systems.  The fix was folded into Warty,
but should probably go into Debian and upstream as well:

https://bugzilla.ubuntu.com/show_bug.cgi?id=1854

-- 
 - mdz
--- End Message ---

Bug#309470: apt-listchanges: Invalid non-UTF8 character in French translation

[Matt Zimmerman]

On Tue, May 17, 2005 at 03:27:43PM +0200, Rémi Denis-Courmont wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: minor
> Tags: l10n
> 
> 
> The french translation prints doesn't "Terminé" correctly on Unicode
> consoles. Looks like the translation was incorrectly made in Latin1.

Can you provide a patch?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#302548: apt-listchanges: Patch for this internationalization issue

[Matt Zimmerman]

tags 302548 pending
thanks

On Mon, May 16, 2005 at 04:16:11PM +0200, Frédéric Bothamy wrote:

> tags 302548 +patch
> thanks
> 
> Followup-For: Bug #302548
> Package: apt-listchanges
> Version: 2.59-0.2
> 
> Hello,
> 
> Here is a small patch to fix this internationalization issue:

Applied in my CVS, thanks.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#309262: apt-listchanges and quiet level

[Matt Zimmerman]

On Sun, May 15, 2005 at 09:48:23PM -0300, Adriano Nagelschmidt Rodrigues wrote:

> Package: apt-listchanges
> Version: 2.59-0.2
> 
> I'm running a script to upgrade my system, via cron, with stdout
> sent to /dev/null. I would like to receive the changelogs extracted by
> apt-listchanges via email.
> 
> However, if I run apt-get with quiet level < 2, I receive an extra
> email, due to this write to stderr:
> 
> /usr/lib/site-python/apt_listchanges.py:
> 
> def update_progress(self):
> if self.config.quiet > 1:
> return
> 
> if not hasattr(self,'message_printed'):
> self.message_printed = 1
> sys.stderr.write(_("Reading changelogs") + "...\n")
> 
> 
> If I run apt-get with quiet level >= 2, I don't receive anything, due
> to the last test below:
> 
> /usr/bin/apt-listchanges:
> 
> if config.quiet == 1:
> config.frontend = 'text'
> elif config.quiet >= 2:
> config.frontend = 'mail'
> 
> if config.quiet >= 2 or config.frontend == 'none':
> sys.exit(0)
> 
> 
> I think sys.exit() should be called only if config.quiet > 2. This
> way, it would be possible to receive only the changelogs when quiet
> level == 2.

quiet levels above two are not defined by apt, and I don't think that this
justifies creating new semantics there.

I think the right thing here is to remove "config.quiet >= 2" from the
second test, so that it only exits if config.frontend == 'none.'.  Let me
know if that change causes it to work the way you expect.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#309513: cricket: /etc/cron.d/cricket refers to non-existing file

[Matt Zimmerman]

retitle 309513 Comment in /etc/cron.d/cricket contains incorrect path to config 
tree
thanks

On Tue, May 17, 2005 at 08:51:10PM +0300, Jari Aalto wrote:
> Package: cricket
> Severity: normal
> 
> The /etc/cron.d/cricket  reads:
> 
>   # Note: if you change the collection interval here, you MUST edit
>   # /etc/cricket/Defaults to adjust the collection interval there as well.
>   # You may also have to rebuild any existing RRDs.
> 
> But there is no such file as /etc/cricket/Defaults installed.

The correct path is /etc/cricket/config/Defaults.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#308744: INTL:vi

[Matt Zimmerman]

tags 308744 pending
thanks

On Thu, May 12, 2005 at 02:32:11PM +0930, Clytie Siddall wrote:
> Package: apt-listchanges
> Version: 2.59-0.2
> Severity: minor
> Tags: l10n, patch
> 
> The Vietnamese translation for debconf: apt-listchanges

Committed to my CVS, thanks.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#298878: apt: lots of sources in sources.list --> gzip: stdin: Resource temporarily unavailable

[Matt Zimmerman]

This is the same as:

https://bugzilla.ubuntu.com/show_bug.cgi?id=1469

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#314385: Agreement?

[Matt Zimmerman]

On Mon, Nov 21, 2005 at 03:33:49AM -0800, Steve Langasek wrote:
> On Mon, Nov 21, 2005 at 11:45:05AM +0100, Loic Minier wrote:
> >  Anyway, your preferred option seems saner, and quite doable given the
> >  limited amount of API / ABI changes as listed by mdz.
> 
> Other symbols removed since 2.1.7, btw:
> 
> [lots]

I'm afraid I probably won't find time to do an exhaustive search, but I
think it would be worthwhile to directly test each package depending on
libfreetype6 and find the set of now-missing symbols which are important in
Debian.  Even if it isn't practical for us to restore 100% compatibility
with 2.1.7, we may be able to achieve something which is close enough, with
a relatively small deltia to 2.1.10.

> Even worse, though, is the number of symbols *added* -- 75 new symbols, vs.
> 32 removed.  Since the maintainer did not bump the shlibs when uploading,
> either, there's no way to determine which packages are using the new ABI
> except by looking at upload dates.

Anything which has started using new symbols from 2.1.10 is extremely likely
to still be API-compatible with previous versions, and can simply be
recompiled, no?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#340328: /usr/bin/ld: qemu-i386: Not enough room for program headers (allocated 8, need 9)

[Matt Zimmerman]

Package: binutils
Version: 2.16.1cvs20051117-1ubuntu1
Severity: normal

Attempting to build qemu 0.7.2-1ubuntu1 with binutils
2.16.1cvs20051117-1ubuntu1, I get:

gcc-3.4 -g -Wl,-shared -o qemu-i386 elfload.o main.o syscall.o mmap.o signal.o 
path.o osdep.o thunk.o vm86.o libqemu.a gdbstub.o   -lm -lrt
/usr/bin/ld: section .note.ABI-tag [2c58 -> 2c77] 
overlaps section .dynsym [2810 -> 836f]
/usr/bin/ld: qemu-i386: Not enough room for program headers (allocated 8, need 
9)
/usr/bin/ld: final link failed: Bad value
collect2: ld returned 1 exit status

backing down to 2.16.1-2ubuntu7, the same objects link OK.

(filing in the Debian BTS as requested)

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#333862: debian-policy: Policy forbids account creation

[Matt Zimmerman]

On Fri, Oct 14, 2005 at 10:07:04AM +0200, Marc Haber wrote:
> Package: debian-policy
> Version: 3.6.2.1
> Severity: normal
> 
> Policy 9.2.1 says:
> Packages other than base-passwd must not modify /etc/passwd,
> /etc/shadow, /etc/group or /etc/gshadow.
> 
> This makes, for example, the passwd package RC buggy since useradd
> modifies /etc/passwd. Also exim4 is RC buggy since its maintainer
> scripts modify /etc/passwd by calling adduser which in turn calls
> useradd which in turn modifies /etc/passwd while not belonging to
> base-passwd.
> 
> The section in the policy should say
> Packages other than base-passwd must not modify /etc/passwd,
> /etc/shadow, /etc/group or /etc/gshadow directly from their maintainer
> scripts.

The usage is consistent with other parts of policy; to talk about a package
which performs some action is different from a package which contains
programs which perform the action.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#251287: Status of this package?

[Matt Zimmerman]

On Sat, Oct 22, 2005 at 06:20:21PM +0200, Frank Lichtenheld wrote:
> So what's the current status of hercules? There was some discussion
> about the new upstream version in February but no further activity since
> then. Currently hercules is removed from testing because of it's
> FTBFS bug related to gcc 4.0.
> 
> On the other hand d-i people tell me that it is actually an important
> package to test and build d-i images for s390...
> 
> So what should happen with it? Orphaning it perhaps or at least asking
> for co-maintainers? Will the new upstream version be packaged or does
> it suck too much?

At the time I left it, it needed a maintainer with some patience to fight
their way through yahoogroups to have the usual conversation with upstream
about shared libraries and their versioning (and perhaps, meanwhile, hacking
around the problem in the package).  A few people have mailed me about it,
but so far no one has taken action on this front.

I'm happy to do what I can to assist with a transition if a new maintainer
can be found.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#335684: rrdtool graph FPE on alpha

[Matt Zimmerman]

On Tue, Oct 25, 2005 at 10:10:14PM +1000, Craig Small wrote:
> Package: rrdtool
> Version: 1.0.49-1
> Severity: important
> 
> rrdtool graph command gives a floating point exception on my alpha.
> None of my rrd files appear to work.
> 
> $ rrdtool graph /tmp/blah  DEF:ntest=/tmp/interface-6-1.rrd:data:AVERAGE 
> LINE1:ntest
> Floating point exception
> 
> I assume you will want more info, what more would you like? strace
> didn't show anything obviously wrong.

Was this working before?  When did it regress?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#335684: rrdtool graph FPE on alpha

[Matt Zimmerman]

On Wed, Oct 26, 2005 at 09:41:39AM +1000, [EMAIL PROTECTED] wrote:
> I'm not sure when this problem occured first. I know it was fine months
> ago. i just stared debugging why some graph didn't work and then noticed
> this fault.

Can you test with a pre-1.2 rrdtool?

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#330393: apt: Errors on Release download failure are baffling

[Matt Zimmerman]

On Tue, Sep 27, 2005 at 05:09:40PM -0400, Daniel Jacobowitz wrote:
> Package: apt
> Version: 0.6.41
> Severity: normal
> 
> I spent half an hour trying to figure out why Apt complained that a bunch of
> packages would be updated but were unauthenticated.  It turns out that the
> Release.gpg on the server was OK, but the Release file was unreadable
> (permission denied).  All apt-get update says after this is "Ign" without
> any hint as to why it's ignoring the file.
> 
> An error message, maybe even "could not fetch", would be nicer...

This is necessary in order to maintain compatibility, since Release has
traditionally been optional.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#330393: apt: Errors on Release download failure are baffling

[Matt Zimmerman]

On Tue, Sep 27, 2005 at 05:24:24PM -0400, Daniel Jacobowitz wrote:
> On Tue, Sep 27, 2005 at 02:21:22PM -0700, Matt Zimmerman wrote:
> > This is necessary in order to maintain compatibility, since Release has
> > traditionally been optional.
> 
> Does that preclude even a warning?  The file's there, and the error
> wasn't "not found".  Failing that, somewhere handy in the documentation
> to mention that "Ign" might mean "failed".

A warning would be reasonable at this stage in the game.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#331748: NMUs welcome

[Matt Zimmerman]

NMUs are welcome on these bugs (and other such transitional bugs in any of
my packages).

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#328485: 'apt-get update' leaves zero byte files in /var/lib/apt/lists/partial/

[Matt Zimmerman]

On Thu, Sep 15, 2005 at 05:50:59PM +0200, Bernd Schubert wrote:
> Package: apt
> Version: 0.6.41
> Severity: important
> 
> 
> We are running Sid within a chroot environment and from time to time I
> update the packages there. Since about apt-0.6.40 the 'apt-get update
> command' causes problems. It just leaves zero byte files in
> /var/lib/apt/lists/partial/, e.g.:

Perhaps your sources.list in the chroot has duplicate entries.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#332943: ubuntu reportbug crash

[Matt Zimmerman]

On Sun, Oct 09, 2005 at 07:14:09PM +0200, Antonio Biasio wrote:
> Package: reportbug
> Version: 3.17
> Severity: normal
> 
> 
> In ubuntu packages reportbug crash when send bug.

Please don't report Ubuntu bugs to the Debian bug tracking system.  Ubuntu
has its own bug tracking system at http://bugzilla.ubuntu.com.

Your bug report is also missing important information about what happened
when you ran reportbug.  If you decide to resubmit it, please describe in
detail what you did and what you saw.  The article at
http://www.chiark.greenend.org.uk/~sgtatham/bugs.html contains a useful
guide for reporting bugs.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#328813: very old package, should this be removed?

[Matt Zimmerman]

On Sat, Sep 17, 2005 at 02:57:05PM +0200, Marc 'HE' Brockschmidt wrote:
> Package: vmnet
> Version: 0.4-1
> Severity: serious
> 
> Hi,
> 
> During the Debian QA meeting hold during Sept. 09th till 11th, we
> decided that looking at packages that haven't been uploaded for a very
> long time could cover up some QA problems.
> 
> I've done this now and your package showed up on the list. I propose
> to remove it.

This package is essentially obsolete with the TUN/TAP support in Linux 2.4+.
Though I suspect it still has some users, I don't use it myself anymore and
have no objection to removing it from Debian.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#329405: installation-reports: Show details after automatic partition

[Matt Zimmerman]

On Wed, Sep 21, 2005 at 06:51:23PM +0300, Jari Aalto wrote:
> Package: installation-reports
> Severity: wishlist
> 
> I'm reporting this feature request based on the review of "Mad Penguin."
> I'm aware that Ubuntu has nothing to do wit Debian, but Ubuntu uses
> same installer, so the remarks apply. Please consider this comment:

Ubuntu uses a significantly modified version of the installer used in
Debian, so you cannot assume that such comments apply equally to Debian
without verification.  This is especially true when considering which
questions are asked and at what time.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#329405: installation-reports: Show details after automatic partition

[Matt Zimmerman]

On Wed, Sep 21, 2005 at 07:25:40PM +0100, Colin Watson wrote:
> On Wed, Sep 21, 2005 at 10:48:20AM -0700, Matt Zimmerman wrote:
> > On Wed, Sep 21, 2005 at 06:51:23PM +0300, Jari Aalto wrote:
> > > Package: installation-reports
> > > Severity: wishlist
> > > 
> > > I'm reporting this feature request based on the review of "Mad Penguin."
> > > I'm aware that Ubuntu has nothing to do wit Debian, but Ubuntu uses
> > > same installer, so the remarks apply. Please consider this comment:
> > 
> > Ubuntu uses a significantly modified version of the installer used in
> > Debian, so you cannot assume that such comments apply equally to Debian
> > without verification.  This is especially true when considering which
> > questions are asked and at what time.
> 
> As it happens, this bug applies perfectly well to Debian.

Thanks for confirming, but my warning still applies in general.  Reporting
bugs to Debian which haven't been observed in Debian is not a good practice.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#329156: CVE name

[Matt Zimmerman]

retitle 329156 [CAN-2005-0023] /usr/sbin/gnome-pty-helper: writes arbitrary 
utmp records
thanks

Use CAN-2005-0023 for this issue.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#339686: Fails where shadow passwords are disabled

[Matt Zimmerman]

Package: adduser
Version: 3.78ubuntu1
Severity: important
Tags: patch

Many chroot environments don't use shadow passwords, and adduser seems to
fail in those enviroments now.  The code in adduser attempts to cope with
this, but it's defeated by the check within the systemcall function.

Patch attached.

-- System Information:
Debian Release: testing/unstable
  APT prefers dapper
  APT policy: (500, 'dapper')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.12-9-k7
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)

Versions of packages adduser depends on:
ii  debconf [debconf-2.0]  1.4.59ubuntu1 Debian configuration management sy
ii  passwd 1:4.0.13-6ubuntu1 change and administer password and
ii  perl-base  5.8.7-8   The Pathologically Eclectic Rubbis

adduser recommends no packages.

-- debconf information excluded

-- 
 - mdz
diff -Nru /tmp/InEuVwVKcO/adduser-3.78/adduser 
/tmp/R5W8T7emGw/adduser-3.78ubuntu1/adduser
--- /tmp/InEuVwVKcO/adduser-3.78/adduser2005-11-15 14:44:39.0 
-0800
+++ /tmp/R5W8T7emGw/adduser-3.78ubuntu1/adduser 2005-11-17 14:45:06.0 
-0800
@@ -374,8 +374,8 @@
 &systemcall('/usr/sbin/useradd', '-d', $home_dir, '-g', $ingroup_name, 
'-s',
$shell, '-u', $new_uid, $new_name);
 print "/usr/bin/chage -M 9 $new_name\n" if ($verbose > 1);
-if (&systemcall('/usr/bin/chage', '-M', '9', $new_name)) {
-   if( ($?>>8) ne 15 ) {
+if (system("/usr/bin/chage -M 9 $new_name")) {
+   if( ($?>>8) != 15 ) {
&cleanup("$0: `/usr/bin/chage -M 9 $new_name' returned error 
code " . ($?>>8) . ".  Aborting.\n")
  if ($?>>8);
&cleanup("$0: `/usr/bin/chage -M 9 $new_name' exited from 
signal " . ($?&255) . ".  Aborting.\n");

Bug#314385: Freetype mess

[Matt Zimmerman]

A quick scan of the diff reveals at least the following API and ABI changes:

Functions:
FTC_Image_Cache_New -> FTC_ImageCache_New
FTC_Image_Cache_Lookup -> FTC_ImageCache_Lookup
FTC_SBit_Cache_New -> FTC_SBitCache_New
FTC_SBit_Cache_Lookup -> FTC_SBit_Cache_Lookup
FTC_Manager_Lookup_Face -> FTC_Manager_LookupFace
FTC_Manager_Lookup_Size -> FTC_Manager_LookupSize

Types:
FTC_Image_Desc -> FTC_ImageTypeRec
FTC_Image_Cache -> FTC_ImageCache
FTC_SBit -> FTC_SBitRec
FTC_SBit_Cache -> FTC_SBitCache

As a special bonus, FTC_ImageTypeRec, though it is said to obsolete
FTC_Image_Desc, has completely different members.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#314385: Freetype mess

[Matt Zimmerman]

On Thu, Nov 17, 2005 at 06:45:14PM -0800, Matt Zimmerman wrote:
> FTC_SBit_Cache_Lookup -> FTC_SBit_Cache_Lookup

This should be:

FTC_SBit_Cache_Lookup -> FTC_SBitCache_Lookup

of course.

-- 
 - mdz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]