Package: pipewire
Version: 0.3.44-1
Severity: minor
(I am running pipewire from Bookworm re-compiled for Bullseye, but I am
confident that the problem occurs in plain Bookworm, too.)
Is it intentional that the systemd user services for pipewire and pipewire-pulse
(and the related socket
Package: mailman3
Version: 3.3.3-1
Severity: minor
Hello!
I was receiving mysterious messages from the out runner:
Cannot connect to SMTP server 127.0.0.1 on port 25
even though the SMTP server was running fine. It turned out that the cause
of the failure is wrong permissions on the
Package: mupdf
Version: 1.14.0+ds1-4+deb10u2
Severity: normal
The man page of mupdf documents that we can send SIGHUP to mupdf to make it
reload the file. However, this is broken by the wrapper script provided by
Debian:
when the script receives a SIGHUP, it terminates instead of forwarding the
Package: texlive-binaries
Version: 2018.20181218.49446-1
Severity: minor
Hello!
I noticed that the pdftex man page is missing the "-synctex" option,
even though this option is described in "pdftex --help".
-- System Information:
Debian Release: 10.7
APT prefers stable-updates
APT policy:
Package: cups-daemon
Version: 2.2.10-6+deb10u3
Severity: normal
Apparantly, there is something fishy in cups around server host names.
I found several issues:
(1) When the client is configured to connect to the server using a host name
unknown to the server (the name is sent in the "Host:"
Package: mdadm
Version: 4.1-1
Severity: normal
Tags: patch upstream
Dear maintainers,
when mdadm is run from /etc/cron.daily/mdadm on one of my machines, it
deterministically segfaults.
The cron job runs:
mdadm --monitor --scan --oneshot
Surprisingly, when I run the same command from
Package: monitoring-plugins-basic
Version: 2.2-6
Severity: normal
When check_by_ssh times out, the ssh process still keeps running.
This is already fixed in upstream commit
7cafb0e84550035fe671662c293122be975065ca.
-- System Information:
Debian Release: 10.2
APT prefers stable-updates
APT
Package: cups-filters
Version: 1.21.6-5
Severity: important
When I use "driverless" to generate a PPD for my Xerox B215 printer,
I get definition of InputSlot which does not work.
In particular, the printer reports that it supports media source "tray-1".
This is translated to "Tray-1" by
Package: git
Version: 1:2.19.2-1~bpo9+1
Severity: normal
Hello, world!\n
After upgrade to Stretch, gitweb no longer finds the configuration file
"gitweb_config.perl" in the current directory. However, "man gitweb" still
mentions this as one of the possible locations of the config file (and
Package: pciutils
Version: 1:3.5.2-1
Severity: minor
Debian pciutils package includes the pcimodules program, which was never
a part of the upstream package and which was obsoleted years ago by both
udev and 'lspci -k'.
I think the time has come to remove it.
-- System Information:
Debian
Package: pciutils
Version: 1:3.5.2-1
Severity: minor
The Homepage field in debian/control lists a long obsolete homepage.
Please update it to http://mj.ucw.cz/sw/pciutils/, where the old page
redirects anyway.
-- System Information:
Debian Release: 9.3
APT prefers stable-updates
APT policy:
> This is surprising, because the file wasn't changed since 2009 and
> Jessie was released in 2015 ...
texlive-plain-extra 2014.20141024-1 from Jessie contains a version
with timestamp from 2009.
texlive-plain-extra 2016.20170123-5 from Stretch contains a version
dated 2016, which is broken.
Sorry for the confusion. It is broken in Stretch, the version in Jessie works.
I will produce a minimal example for Asymptote soon.
Package: texlive-plain-extra
Version: 2016.20170123-5
Severity: normal
Hello!
The package graphics-pln is broken in Debian Jessie. The cuplrit is that
the autopict.sty file (generated from ltpictur.dtx during package build)
contains only comments and \endinput and lacks the definitions of
Package: netpbm
Version: 2:10.0-15.2
Severity: normal
When anytopnm delegates its work to a conversion program (*topnm)
and this program fails, anytopnm still returns success. This confuses
other programs, for example asciiview (from the aview package).
-- System Information:
Debian Release: 8.6
Package: netpbm
Version: 2:10.0-15.2
Severity: normal
When I run anytopnm on a JPEG image with Exif data, it fails with:
> /tmp/anytopnm.wWyz74jE6o/atn.stdin: Not a TIFF or MDI file, bad magic number
> 1 (0xd8ff).
> tifftopnm: error opening TIFF file /tmp/anytopnm.wWyz74jE6o/atn.stdin
The
Package: libpaper1
Version: 1.1.24+nmu4
Severity: normal
Hello!
I noticed that the dimensions of the C4 paper are reported by
'paperconf -whm c4' as '229 mm 354 mm', but the correct values
are '229 mm 324 mm'.
The other ISO Cn formats are correct.
-- System Information:
Debian Release: 8.5
Hello!
> Are you sure?
> networking.service has
> > After=local-fs.target network-pre.target apparmor.service
> > systemd-sysctl.service
>
> This makes sure that networking.service, which deals with type auto
> interfaces, is started after systemd-sysctl.service.
On my system (Wheezy recently
Hello!
I think I have a couple of data points more for this bug.
First of all, disabling ipv6.conf.(all|default).accept_ra
in /etc/sysctl.conf does not work reliably for three reasons:
(1) With systemd, there is no dependency between sysctl and
networking, so there is a race condition. I just
Package: procps
Version: 2:3.3.9-9
Severity: normal
With sysvinit, the init script for procps ensures that sysctl's
are set before networking is started (via X-Start-Before: $network).
When run with systemd, procps.service does not have such dependency.
Since system startup is highly parallel,
Hello!
I suspect a kernel bug since the reboot froze (I had to press
the power button) and lspci -nn is now working.
I suspect either a kernel bug or a hardware bug. Reading the first
64 bytes of configuration space (which is what lspci does and the
strace confirms that) should not lock up the
Package: initramfs-tools
Version: 0.109.1
Severity: normal
When I try to use Debian initramfs to mount root over NFS on a machine with
multiple NICs, it never succeeds, because the initramfs configures all NICs
with the same IP address.
In more detail, the following happens:
o I have
Package: mime-support
Version: 3.52-1
Severity: normal
When I run `see' on a file, I get `Error: no read permission for file XXX',
although the file is perfectly readable.
This happens for example when the file in question is located in a directory
mounted by sshfs. Permissions reported by
Package: foomatic-db-engine
Version: 4.0.4-3
Severity: minor
The foomatic-getpjloptions does not work, since it tries to run /usr/bin/nc,
which nc resides in /bin/nc.
Also, all nc's error messages are redirected to /dev/null, so the user does
not get notified when it's impossible to run nc or
Package: inetutils-telnetd
Version: 2:1.6-3.1+squeeze1
Severity: normal
The telnetd manpage lists options, which do not exist in the current telnetd
(e.g., -u). On the other hand, it misses several interesting options, which
exist (e.g., --exec-login).
-- System Information:
Debian Release:
Package: stun
Version: 0.96.dfsg-5
Severity: normal
When I run /etc/init.d/stun stop and the daemon is already stopped,
the init script dies in the middle of a message, because start-stop-daemon
returns non-zero exit code.
This happens for example when I try to upgrade the package and the
daemon
Package: stun
Version: 0.96.dfsg-5
Severity: normal
When stund runs, it consumes a lot of CPU time, even though it receives
no packets.
A brief look at the source (stun.cxx, stunServerProcess) reveals that
select() is called with a timeout of 1ms. Since the only purpose of the
timeout is
Package: xaos
Version: 3.5+ds1-1
Severity: minor
When I run an application using desktop files (e.g., geeqie), I get the
following warning message:
Desktop file '/usr/share/applications/xaos.desktop' should not include
extension in Icon key: 'xaos.png'
Also, the `desktop-file-validate' utility
Package: fontforge
Version: 0.0.20100501-4
Severity: minor
When I run an application using desktop files (e.g., geeqie), I get the
following warning message:
Desktop file '/usr/share/applications/fontforge.desktop' should not include
extension in Icon key: 'ffanvil32.xpm'
Also, the
Package: gtkam
Version: 0.1.17-1
Severity: minor
When I run an application using desktop files (e.g., geeqie), I get the
following warning message:
Desktop file '/usr/share/applications/gtkam.desktop' should not include
extension in Icon key: 'gtkam-camera.png'
Also, the `desktop-file-validate'
Package: openssh-server
Version: 1:5.1p1-5
Severity: normal
When I configure PAM (in /etc/security/pam_env.conf) to supply a default value
for an environment variable that can be passed through a SSH connection, the
default always overrides the client's value.
Test case:
Clear
Package: gphoto2
Version: 2.4.0-1
Severity: normal
Hello!
When I use `gphoto2 --get-all-files' in Lenny, all downloaded files have
permissions 600, even though my umask is 022. Previous versions of gphoto2
did respect the umask properly.
-- System Information:
Debian Release: 5.0.4
APT
Package: krb5-admin-server
Version: 1.6.dfsg.4~beta1-5lenny2
Severity: normal
The kadmind(8) man page explicitly says that multiple realms are supported by
a single kadmind:
-r realm
specifies the default realm that kadmind will serve; if it is not
specified,
Package: cvs
Version: 1:1.12.13-12
Severity: important
When CVS detects an error while printing an error message, it reports
this situation to syslog with level LOG_EMERG, so the (rather useless)
message gets broadcasted to all logged-in users.
The culprit lies in the error() function in
This bug is not present in version 3.0.3 (Lenny), it's in 3.1.2 as
explained above by Julien.
This is not a bug in libpci, but rather a deficiency in its documentation
leading to improper usage in certain applications.
The pci_dev-base_addr[] array was always intended to contain the BAR
values
Hello!
SONAME in libpci.so.3.1.2 should be set to libpci.so.3.1 (not
libpci.so.3).
objdump -p libpci.so.3.1.2 | grep SONAME
SONAME libpci.so.3
The SONAME is perfectly correct, the library is designed to be forward
compatible by using versioned symbols in the same way as
Hello,
pciutils 3.1.2 fails to compile on mipsel.
could you please send me the version of asm/byteorder.h it fails with?
Have a nice fortnight
--
Martin `MJ' Mares m...@ucw.cz http://mj.ucw.cz/
Faculty of Math and Physics, Charles
Hello!
I believe it worked with either etch (1:2.2.4~pre4-1) or even sarge
(1:2.1.11-15). Sorry, but I am not sure. Will it help if I foreport
one of those to current sid and try again?
It probably does not depend on the version of pciutils at all, but it
would help if you could try the
Hello!
This problem is very likely caused by some PCI device hanging whenever
lspci attempts to read its configuration. This can happen due to a bug
in either the kernel or the hardware itself.
Let us try to find which device it is.
First of all, could you please send us the output of `ls -lR
I think that changing the format of the file (with other suffix) would
also be helpful, i.e. instead of using tab-indent I would explicitly
writing vendor id (ev. other implicit ids) in every line.
In this manner it is easier to grep for hardware, and also to merge
files from different
Hello!
Readable: I don't find that actual version is readable, in particular
with biggest vendor (i.e. I need a lot of scroll to see if I'm in the
right section).
This is however not solved by repeating the vendor ID at every line,
because people usually do not remember the vendor by the ID
Hello and thanks for the prompt response!
Martin, what is the pros of having this striped data available?
In many cases, the subsystem ID is needed to reliably identify a PCI card
as the manufacturers are used to make multiple cards with the same chip,
but wired differently.
Dropping this
Hello!
My point is that pci.ids has been stripped in both the udeb and the
normal binary packages for a number of debian versions of pciutils.
Yes, but only in testing. The Etch version is complete.
Have a nice fortnight
--
Martin `MJ' Mares
Hello!
Please consider applying Antony's patch.
Applied, changes approved.
(In the future, you can send patches to a mail robot at [EMAIL PROTECTED],
they are automatically analysed and queued for approval in the same way
as all submissions through the web interface at http://pci-ids.ucw.cz/
Dropping this information in the udeb is if course a good way of saving
space, but the full package should contain everything.
In the future (after Lenny), I would like to solve one more problem:
with the current rate of development of new hardware, the pci.ids file
is getting out of date very
volatile is up to the volatile maintainers, though updating it on
volatile won't make it usefull for installing new systems...
The ability to update the pci.ids from -volatile was meant as a small
icing at the top of the cake -- I see the real benefit in the possibility
of fetching a package
Package: pciutils
Version: 3.0.0-4
Severity: important
The pci.ids file provided by the Debian patch is seriously broken. It
contains no subsystem entries, no programming interface entries and no
comments including the top comment with version and copyright information.
As the upstream
Package: evince
Version: 0.4.0-5
Severity: normal
When evince is run from the shell with a file name given as an argument
and the file name contains a colon, an error message box with Unable to
open document // Unsupported operation appears. When I try to open the
same file from the GUI, it works
Package: zlib1g
Version: 1:1.2.3-13
Severity: normal
Tags: patch
When gzopen() is called on an empty file, gzgets() reports end of file
properly, but gzerror() returns the Z_BUF_ERROR. I think that the Debian
patch fixing problems with empty files solves the problem only partially
and that it is
Package: ulogd
Version: 1.23-8
Severity: normal
When the ipt_ULOG module is not available in the current kernel, the
ulogd startup script fails silently without even printing a newline
(start-stop-daemon returns with exit code 1, aborting the script,
but it has stdout and stderr redirected to
Package: ulogd
Version: 1.23-8
Severity: wishlist
It would be nice to have a package containing the header files needed
for compiling custom plugins.
-- System Information:
Debian Release: 4.0
APT prefers stable
APT policy: (500, 'stable')
Architecture: i386 (i686)
Shell: /bin/sh linked to
Package: gzip
Version: 1.3.5-15
Severity: normal
As already noted in Bug #407278, zless does not work on pipes when called
without arguments. However, this is not the whole story: zless actually
does not work on anything except regular files:
zless - -- fails to
Package: cpufreqd
Version: 2.2.1-2
Severity: normal
When /etc/init.d/cpufreqd tries to load governor modules, it runs
modprobe once with the list of all modules it wants. Unfortunately,
modprobe stops on the first module it's unable to load, which causes
it to fail silently when some of the
Package: ntp-doc
Version: 1:4.2.2.p4+dfsg-2
Severity: minor
The flash codes documented in ntpq manual do not match reality. For
example, TEST11 means `root distance exceeded', not any sort of
autokey problem.
-- System Information:
Debian Release: 4.0
APT prefers stable
APT policy: (500,
Package: libnet-ssleay-perl
Version: 1.30-1
Severity: normal
Tags: patch
When the underlying TCP connection gets broken and the SIGPIPE signal is
blocked, the ssl_write_all method loops forever instead of reporting the
error properly.
Below is my attempt to fix the problem.
-- System
Hi!
bird needs something like this because in case it daemonizes it still
keeps the fds open of the controlling tty which is bad in case you'd
like to log out which then will hang.
This is definitely worth fixing, but I think that closing all possible
fd's is an overkill, especially as
Hi Matt!
(I will take a look at the other ones soon, now just quickly...)
2.) The file at http://pciids.sourceforge.net/v2.2/pci.ids.gz; is not
gzipped. Likewise http://pciids.sourceforge.net/v2.2/pci.ids.bz2; is not
bzip2'd (although that's not an issue yet since we don't have bz2
Hi!
Hmm, I must have been on crack. I thought I tested with wget last night. I
know that they were appearing uncompressed with firefox, but that must just
be
firefox doing that. Sorry for the false alarm.
Yes, Firefox does that.
#1 and #3 from my email are still valid I think.
I will
Hello!
I disagree with the assertion that programs which rely on PAGE_SIZE are
broken -- if they are architecture specific, they are perfectly correct to
assume that PAGE_SIZE is a compile-time constant and there is just no valid
reason for trying to hide PAGE_SIZE from them.
PAGE_SIZE has been
This bug is assigned to bogl. I assume from your message you meant to
reach the linux-kernel-headers maintainers - try debian-glibc instead.
Sorry, got a bit confused, I will resend it.
Also upstream kernels do not export PAGE_SIZE for various
architectures. For instance ia64.
Agreed, but
Package: vim-common
Version: 1:6.3-071+1sarge1
Severity: wishlist
The /usr/share/vim/vim63/macros/ directory contains several *.info
files with AmigaOS icons, which are completely useless on Debian,
so it probably doesn't make any sense to distribute them.
-- System Information:
Debian Release:
If you apply the attached pciutils-warning.patch, it compiles and then
even works. :-)
Eh, that was silly :) Thanks!
Martin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Hello!
The attached patch is needed to make the pciutils package usable on
GNU/Hurd systems with the current Debian gnumach kernel packages.
I've merged the patch and cleaned it up considerably.
Could you please test that it still works? (It's in the public GIT tree
and also in the 2.2.4-pre3
As briefly explained on the hurd-devel-debian page[1], unconditional use
of PATH_MAX is a POSIX incompatibility.
Err, is there any sense in trying to compile pcimodules on the Hurd?
It's a program specific to Linux (and it isn't a part of the official
pciutils at all, anyway).
Hello!
I've tried that but the build fails (I attached the log). I also try to
move the #define before other preprocessor directives but that's not
better. I've also try to add the #define within the test case in
lib/i386-ports.c, but that's not sufficient either. So I guess that a
flag
Hello!
Aha. That makes sense; I suspect most PCI devices don't allow the OS to
write that field ;-) I didn't like the 'read it byte-by-byte' part of
the patch, so I changed it to pci_get_word().
Martin, now that we have a reason, what do you think to this patch?
Applied :)
(Available in
Hello!
I am not against this patch, but so far I don't understand the rationale for it.
Is there any case where the kernel reports a different class number than the one
reported by the device? And if so, it's important for any userland programs?
Martin
This bug is fixed in pciutils-2.2.1.
Martin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Fixed in pciutils-2.2.1.
Martin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Hello!
I would very much appreciate any changes to lspci output format to be
discussed in the linux-pci mailing list first. While I admit there
were some mistakes in the original format (which I've hopefully
fixed in 2.2.0 with minimal incompatibilities), changing the format
in just a single
Package: iproute
Version: 20041019-3
Severity: minor
The src option for ip route add is mentioned in the detailed list of
options in the man page, but not in the man page's introduction nor in
ip route help.
-- System Information:
Debian Release: testing/unstable
APT prefers stable
APT
Package: cdrecord
Version: 4:2.01+01a01-2
Severity: minor
/usr/share/doc/cdrecord/README.DVD.Debian mentions that there is no DVD
recording support in this package, but contrary to this, the package
seems to be distributed with the DVD recording patch applied.
-- System Information:
Debian
72 matches
Mail list logo
