Bug#961278: ip route json output invalid in Buster (fixed upstream)

2020-05-22 Thread Oskar Berggren
Package: iproute2
Version: 4.20.0-2
Tags: patch

ip route in Buster generates an invalid JSON if the route type has to
be printed,
eg. when detailed mode is active, or the type is different that unicast:

$ ip -d -j -p route show
[ {"unicast",
"dst": "192.168.122.0/24",
"dev": "virbr0",
"protocol": "kernel",
"scope": "link",
"prefsrc": "192.168.122.1",
"flags": [ "linkdown" ]
} ]

$ ip -j -p route show
[ {"unreachable",
"dst": "192.168.23.0/24",
"flags": [ ]
},{"prohibit",
"dst": "192.168.24.0/24",
"flags": [ ]
},{"blackhole",
"dst": "192.168.25.0/24",
"flags": [ ]
} ]

Fix it by printing the route type as the "type" attribute:

$ ip -d -j -p route show
[ {
"type": "unicast",
"dst": "default",
"gateway": "192.168.85.1",
"dev": "wlp3s0",
"protocol": "dhcp",
"scope": "global",
"metric": 600,
"flags": [ ]
} ]


Small upstream fix is available since March 2019:
https://git.kernel.org/pub/scm/network/iproute2/iproute2.git/commit/?id=073661773872709518d35d4d093f3a715281f21d

This bug renders the json output in Debian Buster unusable which is a
bad problem for my system that wants to parse it.

Best regards,
Oskar



Bug#933873: console-common: install-keymap utility missing from 0.7.90

2019-08-04 Thread Oskar Berggren
Source: console-common
Version: 0.7.90
Severity: important

Dear Maintainer,

The install-keymap utility that was present in 0.7.89 is now missing from
0.7.90.

I don't know if this was intentional or not, but the package description still
says it contains this utility, and it does still contain a lot of translation
data
for install-keymap. Also the changelog doesn't mention this. (But it does
mention
a rules cleanup. Maybe something broke there.)



-- System Information:
Debian Release: 10.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 4.19.0-5-amd64 (SMP w/1 CPU core)
Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled



Bug#717266: Wrong date on man page 'interfaces'

2013-07-18 Thread Oskar Berggren
Package: ifupdown
Version: 0.7.8
Severity: minor

The man page for 'interfaces' is still dated 2004-04-05. It has definitely
been updated since then, e.g. with the 'tunnel' method.

The old date is present also in 0.7.43.


Bug#661885: network-manager: Should not take IPv6 prefix length from DHCPv6 client

2012-03-27 Thread Oskar Berggren
This is reported in network manager upstream as
https://bugzilla.gnome.org/show_bug.cgi?id=656610



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



Bug#609125: vlc-nox postinst conifguration segfaults when regenerating plugins cache

2011-03-29 Thread Oskar Berggren
I believe this problem to be fixed.

Yesterday I upgraded a bunch of packages: several core X-packages that
had been waiting a long time, updated nvidia packages and latest
kernel package (this is on unstable). Currently I have the latest
version of every installed package.

With these packages updated, this crash no longer occurs. Because of
the large number of packages it's hard to tell exactly what fixed it
but given the stack traces above it might be the nvidia packages. This
time, those packages told me that I might have remnants of an old
manual nvidia installation, and that this might cause problems. I
asked the package installer to remove the left overs for me, and after
that the vlc-nox postinst worked correctly.

So the "fix" seems to be to install the latest version of the nvidia
packages and let it clean it up for you, or supposedly to manually
look around for files and links installed by nvidia-installer.

/Oskar



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



Bug#616725: Firebug 1.6 incompatible with unstable's current Iceweasel version (3.5)

2011-03-06 Thread Oskar Berggren
Package: xul-ext-firebug
Version: 1.6.2-1
Severity: grave

Firebug 1.6 is not compatible with Firefox 3.5, the version currently
available in unstable. The addin disables itself and is unusable.

IMHO Firebug 1.5.4 should be reuploaded to unstable to restore
functionality, and Firebug 1.6 or 1.7 packaged in experimental to go
with the experimental Firefox 4 packages.

/Oskar



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



Bug#613176: Please update to support new xserver ABI

2011-02-13 Thread Oskar Berggren
Package: nvidia-glx-legacy-173xx
Version: 173.14.27-2
Severity: grave

Package does not install with latest xorg.

nvidia-glx-legacy-173xx depends on xorg-video-abi-6.0 but this is no
longer provided by xserver-xorg-core version 2:1.9.4-1. Instead it
provides xorg-video-abi-8.0 and xorg-video-abi-8.

/Oskar



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



Bug#609125: vlc-nox postinst conifguration segfaults when regenerating plugins cache

2011-01-28 Thread Oskar Berggren
I'm seeing this too.

A more complete stack trace from vlc -vvv, if that helps:

Program received signal SIGSEGV, Segmentation fault.
0xb2ee645a in _nv00gl () from /usr/lib/tls/libnvidia-tls.so.1
(gdb) bt
#0  0xb2ee645a in _nv00gl () from /usr/lib/tls/libnvidia-tls.so.1
#1  0xb3c4fd45 in ?? () from /usr/lib/libGL.so.1
#2  0xb7ff0ce9 in ?? () from /lib/ld-linux.so.2
#3  0xb7ff4d79 in ?? () from /lib/ld-linux.so.2
#4  0xb7ff07f6 in ?? () from /lib/ld-linux.so.2
#5  0xb7ff45c6 in ?? () from /lib/ld-linux.so.2
#6  0xb7ea9c0b in ?? () from /lib/i686/cmov/libdl.so.2
#7  0xb7ff07f6 in ?? () from /lib/ld-linux.so.2
#8  0xb7eaa09c in ?? () from /lib/i686/cmov/libdl.so.2
#9  0xb7ea9b41 in dlopen () from /lib/i686/cmov/libdl.so.2
#10 0xb7f66929 in module_Load (p_this=0x804c224, psz_file=0x8180d90
"/usr/lib/vlc/plugins/video_output/libxcb_glx_plugin.so",
p_handle=0xb31c) at modules/os.c:188
#11 0xb7f60088 in AllocatePlugin (p_this=0x804c224,
psz_file=0xb3c5d3d4
"\203ì\004UWVS1Ûè^)ÿÿhIädzèÄE\023\004\203Ä\004\205Àt\t\200\070\060\017\225À\017¶Ø\205Û\017\205Ò")
at modules/modules.c:1015
[...]

This is with vlc-nox 1.1.3-1squeeze2. I skipped 1.1.3-1squeeze1.

dpkg -S /usr/lib/tls/libnvidia-tls.so.1
libgl1-nvidia-legacy-173xx-glx: /usr/lib/tls/libnvidia-tls.so.1

libgl1-nvidia-legacy-173xx-glx is 173.14.27-2.


/Oskar



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



Bug#602079: Replace Mono-provided Npgsql with official Npgsql

2010-11-01 Thread Oskar Berggren
Package: libmono-npgsql2.0-cil
Version: 2.6.7-3

The Npgsql version incorporated in Mono is Npgsql 1.0, though mono
sets the version to  either 1.0.5000.0 or 2.0.0.0 depending on Net 1.1
or 2.0. This is confusing since there is also an official Npgsql
2.0-series, now at 2.0.10, that is an entirely different version than
the one provided by Mono. Debian also lacks this much improved
version. (One problem that can occur is thinking that you don't need
to set ENCODING in your connection string, since real Npgsql 2.0
handles this properly anyway, but then having weird problems, because
Npgsql 1.0 was flawed in this regard.)

Please package Npgsql 2.0.10 for Debian. Perhaps Debian should also
stop providing the Mono-provided version entirely?

Npgsql web site:
http://npgsql.projects.postgresql.org/

I have also suggested this to be removed from Mono upstream - not sure
what will come of that:
https://bugzilla.novell.com/show_bug.cgi?id=650180


Regards
Oskar



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



Bug#571818: please ship API reference (and register it with doc-base)

2010-05-17 Thread Oskar Berggren
Any update on this? Can we please have this split soonish?

Regards,
Oskar



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



Bug#536206: Please package upstream version 173.14.20

2009-07-08 Thread Oskar Berggren
Package: nvidia-graphics-drivers-legacy-173xx
Version: 173.14.18-1

Please package upstream version 173.14.20 released 2009-07-01. This
should fix issues with 2.6.30 it seems.

Regards,
Oskar



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



Bug#391704: Can't access server.key (permission denied)

2006-10-08 Thread Oskar Berggren
Package: postgresql-8.1
Version: 8.1.4-7

Since upgrading to this version from 8.1.4-6 (I believe), postmaster
fails to start with the message 

   Starting PostgreSQL 8.1 database server: main* The PostgreSQL server
   failed to start.
   FATAL:  could not access private key file "server.key": Permission
   denied
 failed!


This would appear to be caused by the permissions on
the /etc/ssl/private directory. While the key file itself has group
ssl-cert, the directory is as follows:

   $ ls -dl /etc/ssl/private/ 
   drwx--x--- 2 root root 4096 Aug 10 21:32 /etc/ssl/private/


Not sure if this is a package bug, or some configuration error. I don't
remember having meddled with that directory however.


/Oskar




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



Bug#293822: Gtk::ListStore::clear() causes catastrophic failure

2006-08-23 Thread Oskar Berggren
This should be closed.

I'm using ListStore::clear() in lots of code without problems with
current versions of gtkmm - and have been doing so for several years.
Both in Debian sarge and etch.

Also, the original submitter hasn't responded after the initial
report. There is no point in keeping this open.

Regards,
Oskar




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



Bug#293832: Regarding Gtk::ListStore::Clear [was Re: Gtk::Toolbar::tools() member missing is expected, not a bug]

2006-08-23 Thread Oskar Berggren
Please disregard faulty subject line of my previous mail to this bug.

/Oskar

On Wed, 2006-08-23 at 10:46 +0200, Oskar Berggren wrote:
> This should be closed.
> 
> I'm using ListStore::clear() in lots of code without problems with
> current versions of gtkmm - and have been doing so for several years.
> Both in Debian sarge and etch.
> 
> Also, the original submitter hasn't responded after the initial
> report. There is no point in keeping this open.
> 
> Regards,
> Oskar
> 



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



Bug#293832: Gtk::Toolbar::tools() member missing is expected, not a bug

2006-08-23 Thread Oskar Berggren
This should be closed.

I'm using ListStore::clear() in lots of code without problems with
current versions of gtkmm - and have been doing so for several years.
Both in Debian sarge and etch.

Also, the original submitter hasn't responded after the initial
report. There is no point in keeping this open.

Regards,
Oskar




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



Bug#253363: Gtk::Toolbar::tools() member missing is expected, not a bug

2006-08-23 Thread Oskar Berggren
This should be closed, as it does not appear to be a bug.

The path in the submitters mail refers to file below gtkmm-2.0, not 2.4.

More importantly, the NEWS entry for gtkmm 2.3.5 explicitly mentions:

  - Toolbar
- Added append(), prepend(), insert() overloads which take
  slots, so that signal connecting requires less code.
- Removed old STL-style API, because the APIs do not 
  co-exist well together at the GTK+ level.
(Bryan Forbes)


"STL-style" includes the tools() member.


Current online documentation at gtkmm.org for versions above 2.4 does not
list the tools() member. Probably the submitter accessed outdated docs.

Regards,
Oskar




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



Bug#338589: grep linking with /usr/lib/libpcre.so.3 renders system unbootable

2005-11-11 Thread Oskar Berggren
Package: grep
Version: 2.5.1.ds2-3
Severity: critical

This version of grep links with /usr/lib/libpcre.so.3, however,
grep is required to run before local filesystems have been loaded.
If /usr is not on the /-partition, the system will not boot
correctly.

See e.g. /etc/rcS/S02mountvirtfs   (fails to load tmpfs)
and  /etc/rcS/S04udev  (fails because tmpfs isn't loaded)


/Oskar




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



Bug#296401: Font size for UML "normal" attributes not loaded correctly

2005-02-22 Thread Oskar Berggren
Package: dia
Version: 0.94.0-5

Reproducing:
  New document.
  Create a UML class.
  Open the properties dialog, change fontsize for "normal" attributes to
something other than 0.80.
  Save the file.(look at test file, the new value has
 been saved)
  Close the file, and reload it.
  Font size for "normal" for the class has been reset to 0.80.

So it seems that while the changed font size is written to disk, it isn't
used when the file is reopened.

Seeing this in all tested version of 0.94 as found in Debian during the
last year or so.


Upstream this is bug 157016:
   http://bugzilla.gnome.org/show_bug.cgi?id=157016
which has already been fixed.


I have looked in cvs and I believe that this is the fix:
http://cvs.gnome.org/viewcvs/dia/objects/UML/class.c?r1=1.68&r2=1.69


Please consider patching the Debian package with this fix as it is
a real nuisance for me. (Together with the fix for the font size
weirdness applied recently (debian bug #273625), it will be much
easier to use dia for my reports.)





Bug#277940: Dupe of #273625?

2005-02-22 Thread Oskar Berggren
Isn't this a dupe of #273625, which was recently fixed?

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=273625

/Oskar