Bug#869617: Acknowledgement (libutempter0: Helper executable multiply broken: can't add, can't del)

[Conrad Hughes]

Hi Dmitry,

Thanks for getting back to me.  Unfortunately I remember little of what
I learned while digging into this two years ago, so I don't know how to
recognise a properly configured pseudo-terminal master device or a
mis-configured operating system with a broken ptsname: I'm just running
Debian 10.3 with Gnome/Wayland.

All I was after was a program that could wrap gnome-terminal's shell
invocations, and add them to utmp, so:

  utmp_wrap /bin/bash -l

would use libutempter to create a utmp entry for each terminal window.
I attach example source that fails with libutempter as released, but
succeeds if it's patched as suggested above.

Are you saying that what I'm after is impossible because of security
concerns, or can you please suggest a means to make it work?

Best regards,
Conrad
#include 
#include 
#include 
#include 
#include 
#include 

int
main(int argc, char **argv)
{
  pid_t child;

  if(utempter_add_record(STDIN_FILENO, (char *) NULL) == 0) {
perror("Couldn't add record to utmp");
return EXIT_FAILURE;
  }

  if((child = fork()) == 0) {
execv(argv[1], argv + 2);
/*NOTREACHED*/
  }

  if(child < 0) {
perror("fork failed");
return EXIT_FAILURE;
  }

  if(waitpid(child, (int *) NULL, 0) != child) {
perror("waitpid failed");
return EXIT_FAILURE;
  }

  if(utempter_remove_record(STDIN_FILENO) == 0) {
perror("Couldn't remove record from utmp");
return EXIT_FAILURE;
  }

  return EXIT_SUCCESS;
}

Bug#941123: Acknowledgement (gdm3: action-middle-click-titlebar 'lower' is ignored: can't set window to lower on title bar middle-click)

[Conrad Hughes]

Urgh, too late I manage to crack Gnome's search engine.  This seems to
appear upstream in a few contexts, but I don't understand what the
consequences are for a fix in Debian:

  https://gitlab.gnome.org/GNOME/gnome-shell/issues/773
  Closed; says it's applications' fault!

  https://gitlab.gnome.org/GNOME/mutter/issues/602
  .. points to
  https://gitlab.gnome.org/GNOME/gtk/issues/539
  .. where the same guy says it's applications' fault.

  https://gitlab.gnome.org/GNOME/gtk/issues/1895
  .. where it looks as if the 'lower window' implementation has
  never been completed.

C.

Bug#933174: wayland: Brightness bar not displayed when adjusting screen brightness

[Conrad Hughes]

Sorry for the slow reply.  The laptop is an old (2012) Sony Vaio; I
attach the output of lspci and lshw below.

Conrad

lspci:
  00:02.0 VGA compatible controller: Intel Corporation 3rd Gen Core processor 
Graphics Controller (rev 09)

lshw:
  *-display
description: VGA compatible controller
product: 3rd Gen Core processor Graphics Controller
vendor: Intel Corporation
physical id: 2
bus info: pci@:00:02.0
version: 09
width: 64 bits
clock: 33MHz
capabilities: vga_controller bus_master cap_list rom
configuration: driver=i915 latency=0
resources: irq:28 memory:d700-d73f memory:c000-cfff 
ioport:9000(size=64) memory:c-d
*-usb:0
description: USB controller
product: 7 Series/C210 Series Chipset Family USB xHCI Host Controller
vendor: Intel Corporation
physical id: 14
bus info: pci@:00:14.0
version: 04
width: 64 bits
clock: 33MHz
capabilities: xhci bus_master cap_list
configuration: driver=xhci_hcd latency=0
resources: irq:27 memory:d740-d740

Bug#931749: [pkg-cryptsetup-devel] Bug#931749: cryptsetup: "no longer required" on encrypted system!

[Conrad Hughes]

Thanks for such a quick and helpful response Guilhem!

For anyone else reading, as described, solution for the time being is:

  apt-mark manual cryptsetup-initramfs

C.

Bug#380470: 11 years on, this would still be nice

[Conrad Hughes]

I'm yet another of those users who read the label on the tin and assumed
"lossless transformation of JPEG files" was just that.  Having a warning
about imperfect transformations and directing to the hidden depths of
the manpage where these details are explained would be really helpful.

Conrad

Bug#887627: [gnome-shell] No way to open (evolution) calendar events from quick calendar view

[Conrad Hughes]

Package: gnome-shell
Version: 3.22.3-3
Severity: normal

--- Please enter the report below this line. ---

When I click on the date/clock at top centre of my screen, I get a
quick calendar view with a list of current/upcoming events and an
overview of this month.  However, no amount of hovering, left-clicking
or right-clicking on those events will get me any more information or
open my calendar app (Evolution).  I'm sure that at some point in the
past, clicking on an event in this view would open the event in
Evolution, or there was at least a button here to open my calendar app.
 Is the current total inability to get to my calendar from the gnome-
shell quick view expected, correct behaviour?  If not, any idea why it
might not be working please?  Is there a log file that might offer
clues?

Hmm: minor correction - it's not a total inability: I've just
discovered that clicking on the word "Events" at the top of the left
pane opens Evolution (perhaps the possibility of this interaction could
be cued more visibly?), so gnome-shell knows Evolution's my calendar
and is capable of launching it.  But is it expected behaviour to not be
able to open individual events or obtain more information on them by
clicking/hovering?  They highlight when I hover, which hints that they
should have interactions..

It occurs to me that one final piece of information might be relevant:
my actual calendar data is all remotely hosted, via CalDAV, with the
connections set up in Evolution (not Gnome's "online accounts", since
that doesn't seem to offer CalDAV as an option).

It's not listed below, but I appear to be running evolution 3.22.6-
1+deb9u1, subtly different from evolution-data-server.

Best,
Conrad

--- System information. ---
Architecture: 
Kernel:   Linux 4.9.0-5-amd64

Debian Release: 9.3
  500 yakkety ppa.launchpad.net 
  500 stable-updates  deb.debian.org 
  500 stable  security.debian.org 
  500 stable  repository.spotify.com 
  500 stable  repo.vivaldi.com 
  500 stable  dl.google.com 
  500 stable  deb.opera.com 
  500 stable  deb.debian.org 
  100 stretch-backports deb.debian.org 

--- Package information. ---
Depends(Version) | Installed
-+-==
gir1.2-glib-2.0  (>= 1.45.3) | 1.50.0-1+b1
gir1.2-gtk-3.0 (>= 3.16) | 3.22.11-1
gir1.2-mutter-3.0(>= 3.22.1) | 3.22.3-2
gir1.2-networkmanager-1.0| 1.6.2-3
gir1.2-soup-2.4  (>= 2.40.1) | 2.56.0-2+deb9u1
gir1.2-telepathyglib-0.12| 0.24.1-1.1
dconf-gsettings-backend  | 0.26.0-2+b1
 OR gsettings-backend| 
libatk-bridge2.0-0(>= 2.5.3) | 2.22.0-2
libatk1.0-0  (>= 1.12.4) | 2.22.0-1
libc6  (>= 2.14) | 
libcairo2(>= 1.14.0) | 
libcanberra-gtk3-0 (>= 0.25) | 
libcanberra0(>= 0.2) | 
libcroco3 (>= 0.6.2) | 
libdbus-glib-1-2   (>= 0.78) | 
libecal-1.2-19 (>= 3.17) | 
libedataserver-1.2-22(>= 3.17.2) | 
libgcr-base-3-1   (>= 3.8.0) | 
libgdk-pixbuf2.0-0   (>= 2.22.0) | 
libgirepository-1.0-1 (>= 0.9.2) | 
libgjs0-libmozjs-24-0| 
libgjs0e (>= 1.46.0) | 
libglib2.0-0 (>= 2.45.3) | 
libgstreamer1.0-0 (>= 1.4.0) | 
libgtk-3-0   (>= 3.21.6) | 
libical2  (>= 2.0.0) | 
libicu57(>= 57.1-1~) | 
libjson-glib-1.0-0   (>= 0.13.2) | 
libmozjs-24-0| 
libmutter0i  (>= 3.21.0) | 
libnm-glib4 (>= 0.8.998) | 
libnm-util2 (>= 0.8.998) | 
libpango-1.0-0   (>= 1.14.0) | 
libpangocairo-1.0-0  (>= 1.14.0) | 
libpolkit-agent-1-0(>= 0.99) | 
libpolkit-gobject-1-0  (>= 0.94) | 
libpulse-mainloop-glib0  (>= 0.99.1) | 
libpulse0(>= 0.99.1) | 
libsecret-1-0   (>= 0.7) | 
libstartup-notification0   (>= 0.11) | 
libsystemd0  | 
libtelepathy-glib0   (>= 0.17.5) | 
libwayland-client0(>= 1.0.2) | 
libx11-6 | 
libxfixes3   | 
evolution-data-server(>= 3.17.2) | 
gir1.2-gdm-1.0   (>= 3.18.2) | 
gir1.2-accountsservice-1.0   | 
gir1.2-atspi-2.0 (>= 2.9.91) | 
gir1.2-caribou-1.0(>= 0.4.8) | 
gir1.2-freedesktop   | 
gir1.2-gdesktopenums-3.0   (>= 3.12) | 
gir1.2-gcr-3  (>= 3.7.5) | 

Bug#869617: Acknowledgement (libutempter0: Helper executable multiply broken: can't add, can't del)

[Conrad Hughes]

Sorry, the subject line is somewhat erroneous since I was still testing
things when I started the bug report: I think that logout/del works —
utmpdump doesn't show all the changes I'd expect after logout/del, but
/usr/bin/w does appear to show that logout has been enacted.  The
problem is just with add, I think.

Conrad

Bug#869617: Acknowledgement (libutempter0: Helper executable multiply broken: can't add, can't del)

[Conrad Hughes]

Oh dear.  And I now realise that this is submitted under a non-existent
release version of libutempter — that's the patched version of it I
created in order to test out the core idea in the fix, sorry.  On the
offchance that it's any use, I attach the patch below.

Conrad
Description: Use ttyname, not ptsname.
 Bugfix; see changelog.
 .
 libutempter (1.1.6-4) unstable; urgency=medium
 .
   * Use ttyname, not ptsname.
Author: Conrad J.C. Hughes (for Debian package stuff) 

---
The information above should follow the Patch Tagging Guidelines, please
checkout http://dep.debian.net/deps/dep3/ to learn about the format. Here
are templates for supplementary fields that you might want to add:

Origin: , 
Bug: 
Bug-Debian: https://bugs.debian.org/
Bug-Ubuntu: https://launchpad.net/bugs/
Forwarded: 
Reviewed-By: 
Last-Update: 2017-07-24

--- libutempter-1.1.6.orig/utempter.c
+++ libutempter-1.1.6/utempter.c
@@ -241,7 +241,7 @@ main(int argc, const char *argv[])
 		exit(EXIT_FAILURE);
 	}
 
-	device = ptsname(STDIN_FILENO);
+	device = ttyname(STDIN_FILENO);
 
 	if (!device)
 	{

Bug#869338: [gnome-terminal] Double-click no longer selects URLs; workaround doesn't seem to work

[Conrad Hughes]

Package: gnome-terminal
Version: 3.22.2-1
Severity: normal

--- Please enter the report below this line. ---

[I'm submitting this partially for documentation so that others can
find the solution if there is one.]

The pattern for double-click-select-words in gnome-terminal used to
select most URLs.  Now it excludes ':', so fails to select URLs by
excluding the protocol part.  This has been the subject of numerous
debates elsewhere, including upstream, and regardless of the rights or
wrongs there appears to have been a documented workaround. 
Unfortunately that does not appear to work on gnome-terminal in
stretch.

In particular, there apparently used to be gconf settings which allowed
sufficiently-motivated users to restore old behaviour through some
command line work, by setting "word-char-exceptions" on their gnome-
terminal profile.  My attempts to do this have failed; indeed that
setting doesn't appear to exist any more.  Any suggestions as to an
alternative?

Upstream issue related to this bug:

  https://bugzilla.gnome.org/show_bug.cgi?id=730632

Ubuntu discussions of same:

  https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1463072
  https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1501250

Supposed fix:

  https://muug.ca/pipermail/roundtable/2015-December/004433.html

Personally, while I can appreciate the rigid logic of the new approach,
I don't think usability == rigid logic; the old behaviour which
recognised most URLs as words was useful.  I think it would have made
more sense to explicitly make word selection recognise whole URLs when
they were detected, rather than prevent that behaviour.  That aside,
any guidance as to how to restore the old "bad" behaviour for those
recidivists who appreciate it would be most gratefully received.

--- System information. ---
Architecture: 
Kernel:   Linux 4.9.0-3-amd64

Debian Release: 9.0
  500 yakkety ppa.launchpad.net 
  500 stable  repo.vivaldi.com 
  500 stable  dl.google.com 
  500 stable  deb.opera.com 
  500 stable  deb.debian.org 
  100 stretch-backports deb.debian.org 

--- Package information. ---
Depends (Version) | Installed
=-+-
libatk1.0-0   (>= 1.12.4) | 2.22.0-1
libc6(>= 2.9) | 
libdconf1 (>= 0.14.0) | 
libglib2.0-0  (>= 2.42.0) | 
libgtk-3-0  (>= 3.20) | 
libnautilus-extension1a   (>= 3.21.92-3~) | 
libpango-1.0-0(>= 1.14.0) | 
libuuid1(>= 2.16) | 
libvte-2.91-0(>= 0.45.90) | 
libx11-6  | 
dconf-gsettings-backend   | 
 OR gsettings-backend | 
gsettings-desktop-schemas  (>= 0.1.0) | 
gnome-terminal-data (>= 3.22) | 
gnome-terminal-data (<< 3.23) | 


Recommends(Version) | Installed
===-+-===
default-dbus-session-bus| 
 OR dbus-session-bus| 
gvfs| 1.30.4-1
yelp| 3.22.0-1


Package's Suggests field is empty.

Bug#802919: unison: synchronization incompatibility when built with Ocaml versions pre/post-4.02

[Conrad Hughes]

Hi Ralf,

Ralf> The problem is that any software used to build packages in a
Ralf> distribution (sid, in this case) must also be in the distribution.
Ralf> Hence, providing builds of unison produced with different ocaml
Ralf> versions implies having packages for different ocaml versions
Ralf> in sid.

That's kindof what I thought, but if I understand correctly, unison in
stretch must have been built with a different version of ocaml than the
ocaml that's actually in stretch, which confusingly contradicts this?
Stretch's unison is only compatible with other pre-ocaml-4.02 unisons,
which suggests it was built with ocaml-4.01 — but stretch's ocaml is
4.02.3-9?  This would seem to be confirmed by the fact that I had to
downgrade ocaml to 4.01 on one of my (non-Debian, more recently updated)
servers when building unison there, so that unison on my jessie and
stretch machines could talk to unison on that server.

Sorry if I'm being slow or missing the point here.

If I've gotten that right, then the current version of unison on stretch
is built with the "wrong" ocaml (i.e. not stretch's ocaml).  The idea
that it should be built with stretch's ocaml would then argue that
technically unison on stretch needs to be updated to an ocaml-4.02
build.  When that update rolls out, unison on stretch will become
incompatible with everything that it currently works with (except
itself, of course).  So a warning release note would seem appropriate
when that "upgrade" happens..?

It would be nice if we could persuade upstream to at least give a more
helpful error message when the incompatibility arises.

Conrad

Bug#802919: unison: synchronization incompatibility when built with Ocaml versions pre/post-4.02

[Conrad Hughes]

Hi Stéphane,

Thanks for your response.

By this:

> > Possibly some pressure should be put on upstream to (1) catch the
> > relevant exception and flag the ocaml version dependency,

I meant that since this is such a non-obvious failure, which should
never happen normally (I've never seen a unison data stream corrupted in
probably fifteen years of using it), that it might be reasonable and
helpful to catch the exception caused by the serialisation change and to
add a message saying "This exception may be because the two versions of
unison were compiled with different versions of OCaml", with a reference
to some relevant documentation (e.g. bug report or release notes).

--

I'm unfamiliar with how the Debian release process works internally, and
minimising the work involved in packaging unison is of course essential
— you're already saving all Debian users a huge amount of work by
packaging it, so thank you for that.

At the moment because you're building with ocaml-4.01, the published
unison is backwards compatible but not forwards compatible; can you
continue to build unison with 4.01 now that Debian's on ocaml-4.02, or
is that going to become difficult as well?

My initial thought process was that producing two conflicting ocaml
builds, say ocaml_pre_402-4.01.0 and ocaml-4.04.2 (or whatever the
actual versions are), which could *not* be installed simultaneously,
would have been the easiest route to automatically building multiple
versions of unison on some kind of automated build farm (including the
ocaml version in unison package names), but that's probably because I
don't understand what the actual release process is like.

So if two ocaml versions is not a practical way of proceeding, do you
think there's any reasonable way of producing pre- and post-402 releases
of unison for Debian in the future?

If not, then would it be possible to include a release note (the kind
that you get shown and emailed when you upgrade) warning users about the
incompatibility?  I realise that arguably this note should not yet have
been sent since Debian's unison is still backwards compatible; perhaps
you already have it planned for the -4.02 changeover..

Regards,
Conrad

Bug#866795: gnome-contacts: No CardDAV support?

[Conrad Hughes]

> Afaik, you couldn't directly add CardDAV URLs in jessie either.

Perhaps not under gnome-contacts, but on jessie the official calendar
app seemed to be evolution, insofar as when I said "open calendar" from
the top bar to get at my contacts, evolution is what started up, and
that supported CardDAV.

> What you can use is gnome-online-accounts, same as in jessie.

Oddly my remaining jessie machine doesn't have gnome-online-accounts
installed, but handles CardDAV fine via evolution anyway.  On stretch
gnome-online-accounts doesn't present a WebDAV option, and trying the
ownCloud option I have no idea what to type since it just asks for a
server, not a URL.  Previous attempts at entering a URL there failed.

C.

Bug#866794: gnome-calendar: No CalDAV support?

[Conrad Hughes]

> As you noted, evolution is still available and it is not likely to
> disappear any time soon.

Yep; my emails are in here in non-chronological order, took me a while
to discover the evolution workaround.  It's a bit counterintuitive to
have to use what's ostensibly an email client to add calendars though.

> I only use it to connect to my Google account, but there is an option
> under Calendar Icon > Calendar Settings > Add > From Web.  Does that
> work for you?

No: as I enter the URL, the border of the text box swirls green as if
it's checking the URL as I type, but before I reach the end of the
address the swirling stops, and when I do finish the URL, the "add"
button is greyed out so I can't continue.  The address is of the form:

  https://hostname:/some/path/calendarname/

.. and works fine when entered via the evolution workaround.

C.

Bug#866795: Acknowledgement (gnome-contacts: No CardDAV support?)

[Conrad Hughes]

Workaround: even though evolution no longer seems to be the official
contacts/calendar app, it can still be used to set up CardDAV accounts.

C.

Bug#866794: Acknowledgement (gnome-calendar: No CalDAV support?)

[Conrad Hughes]

Hmm,

Submitted bug #866795 too (contacts & CardDAV also broken), but suspect
that's a duplicate of this, and that the real problem is with
gnome-online-accounts, and is ultimately an unsolved upstream problem as
described here:

  "Add separate components for CalDAV and CardDAV accounts"
  https://bugzilla.gnome.org/show_bug.cgi?id=720519

.. if that's correct though, isn't that a pretty major regression?
Dropping support for CalDAV/CardDAV is certainly something that should
have been mentioned in the stretch release notes.

C.

Bug#866794: Acknowledgement (gnome-calendar: No CalDAV support?)

[Conrad Hughes]

Workaround: even though evolution no longer seems to be the official
calendar app, it can still be used to set up CalDAV account connections,
which then turn up in the gnome-calendar feed.

C.

Bug#866028: gnome-control-center: "Typing" tab missing from Keyboard settings after jessie → stretch upgrade?

[Conrad Hughes]

Thanks very much for the rapid and helpful response!

Seems as if something might still be wrong though since searching for
neither "cursor" nor "blink" from All Settings gives Universal Access as
a result.

Conrad

Bug#866031: Acknowledgement (xserver-xorg-input-synaptics: Breaks xserver-xorg-input-libinput on jessie → stretch upgrade)

[Conrad Hughes]

[Perhaps worth noting that this is after wiping my home directory, so
 should be a "clean" Gnome desktop environment created by the upgraded
 system]

Regards,
Conrad

Bug#866031: xserver-xorg-input-synaptics: Breaks xserver-xorg-input-libinput on jessie → stretch upgrade

[Conrad Hughes]

Package: xserver-xorg-input-synaptics
Version: 1.9.0-1+b1
Severity: normal
Tags: l10n

Dear Maintainer,

I've just upgraded from jessie to stretch on a laptop with a "SynPS/2 Synaptics
TouchPad" (/proc/bus/input/devices).  After upgrading, tap-to-click no longer
worked, and I had to use physical buttons for pointer clicks; further, under
Mouse & Touchpad settings, the only options were primary button, mouse speed
and natural scrolling: no option to enable tap-to-click.  Various sites
suggested various remedies (including Gnome Tweak Tool, which didn't work), but
the solution that worked for me was to uninstall/purge xserver-xorg-input-
synaptics: with that gone, after logging out and in, I had a much longer list
of Mouse & Touchpad settings, including tap-to-click.

It seems that xserver-xorg-input-synaptics sabotages xserver-xorg-input-
libinput.  From some sources I understand that -synaptics may be obsolete and
-libinput the future; if so then perhaps the upgrade process is at fault, as I
had both of these installed after the stretch upgrade.

Regards,
Conrad



-- Package-specific info:
X server symlink status:

lrwxrwxrwx 1 root root 13 Sep 23  2012 /etc/X11/X -> /usr/bin/Xorg
-rwxr-xr-x 1 root root 274 Mar  3 14:41 /usr/bin/Xorg

VGA-compatible devices on PCI bus:
--
00:02.0 VGA compatible controller [0300]: Intel Corporation 3rd Gen Core 
processor Graphics Controller [8086:0166] (rev 09)

/etc/X11/xorg.conf does not exist.

Contents of /etc/X11/xorg.conf.d:
-
total 4
-rw-r--r-- 1 root root 2700 Nov 17  2012 synaptics.conf

KMS configuration files:

/etc/modprobe.d/i915-kms.conf:
  options i915 modeset=1
/etc/modprobe.d/radeon-kms.conf:
  options radeon modeset=1

Kernel version (/proc/version):
---
Linux version 4.9.0-3-amd64 (debian-ker...@lists.debian.org) (gcc version 6.3.0 
20170516 (Debian 6.3.0-18) ) #1 SMP Debian 4.9.30-2+deb9u1 (2017-06-18)

Xorg X server log files on system:
--
-rw-r--r-- 1 root   root  32124 Feb 13  2016 /var/log/Xorg.1.log
-rw-r--r-- 1 root   root  30874 Jun 26 16:33 /var/log/Xorg.0.log
-rw-r--r-- 1 conrad users 32357 Jun 26 19:05 
/home/conrad/.local/share/xorg/Xorg.0.log

Contents of most recent Xorg X server log file 
(/home/conrad/.local/share/xorg/Xorg.0.log):
---
[   664.405] (--) Log file renamed from 
"/home/conrad/.local/share/xorg/Xorg.pid-8216.log" to 
"/home/conrad/.local/share/xorg/Xorg.0.log"
[   664.405] 
X.Org X Server 1.19.2
Release Date: 2017-03-02
[   664.405] X Protocol Version 11, Revision 0
[   664.405] Build Operating System: Linux 3.16.0-4-amd64 x86_64 Debian
[   664.405] Current Operating System: Linux waif 4.9.0-3-amd64 #1 SMP Debian 
4.9.30-2+deb9u1 (2017-06-18) x86_64
[   664.405] Kernel command line: BOOT_IMAGE=/vmlinuz-4.9.0-3-amd64 
root=/dev/mapper/waif-root ro quiet
[   664.405] Build Date: 03 March 2017  03:14:41PM
[   664.405] xorg-server 2:1.19.2-1 (https://www.debian.org/support) 
[   664.405] Current version of pixman: 0.34.0
[   664.405]Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
[   664.405] Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   664.405] (==) Log file: "/home/conrad/.local/share/xorg/Xorg.0.log", Time: 
Mon Jun 26 19:05:45 2017
[   664.405] (==) Using config directory: "/etc/X11/xorg.conf.d"
[   664.405] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[   664.406] (==) No Layout section.  Using the first Screen section.
[   664.406] (==) No screen section available. Using defaults.
[   664.406] (**) |-->Screen "Default Screen Section" (0)
[   664.406] (**) |   |-->Monitor ""
[   664.406] (==) No monitor specified for screen "Default Screen Section".
Using a default monitor configuration.
[   664.406] (==) Automatically adding devices
[   664.406] (==) Automatically enabling devices
[   664.406] (==) Automatically adding GPU devices
[   664.406] (==) Max clients allowed: 256, resource mask: 0x1f
[   664.406] (WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
[   664.406]Entry deleted from font path.
[   664.406] (==) FontPath set to:
/usr/share/fonts/X11/misc,
/usr/share/fonts/X11/100dpi/:unscaled,
/usr/share/fonts/X11/75dpi/:unscaled,
/usr/share/fonts/X11/Type1,
/usr/share/fonts/X11/100dpi,
/usr/share/fonts/X11/75dpi,
built-ins
[   664.406] (==) ModulePath set to "/usr/lib/xorg/modules"
[   664.406] (II) The server relies on udev to provide the list of input 
devices.
If no devices become available, reconfigure udev or disable 

Bug#866028: gnome-control-center: "Typing" tab missing from Keyboard settings after jessie → stretch upgrade?

[Conrad Hughes]

Package: gnome-control-center
Version: 1:3.22.2-3
Severity: normal

Dear Maintainer,

I've just upgraded from jessie to stretch.  After wiping my old home directory
to get a fresh Gnome environment, I found my gnome-terminal cursor flashing
again.  Went to Keyboard settings to disable this, but selecting Keyboard from
All Settings in gnome-control-centre just gives me a list of shortcuts: where
there used to be two tabs, Typing and Shortcuts, now there are none.  I'm
assuming the missing tab is a bug since when I search All Settings for "blink"
it still brings up the "Keyboard" section as having something relevant.  Am I
missing a crucial package?

Regards,
Conrad



-- System Information:
Debian Release: 9.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/8 CPU cores)
Locale: LANG=en_GB.utf8, LC_CTYPE=en_GB.utf8 (charmap=UTF-8), 
LANGUAGE=en_GB.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages gnome-control-center depends on:
ii  accountsservice0.6.43-1
ii  apg2.2.3.dfsg.1-4+b1
ii  colord 1.3.3-2
ii  desktop-file-utils 0.23-1
ii  gnome-control-center-data  1:3.22.2-3
ii  gnome-desktop3-data3.22.2-1
ii  gnome-settings-daemon  3.22.2-2
ii  gsettings-desktop-schemas  3.22.0-1
ii  libaccountsservice00.6.43-1
ii  libatk1.0-02.22.0-1
ii  libc6  2.24-11+deb9u1
ii  libcairo-gobject2  1.14.8-1
ii  libcairo2  1.14.8-1
ii  libcanberra-gtk3-0 0.30-3
ii  libcanberra0   0.30-3
ii  libcheese-gtk253.22.1-1+b1
ii  libcheese8 3.22.1-1+b1
ii  libclutter-1.0-0   1.26.0+dfsg-3
ii  libclutter-gtk-1.0-0   1.8.2-2
ii  libcolord-gtk1 0.1.26-1.1
ii  libcolord2 1.3.3-2
ii  libcups2   2.2.1-8
ii  libfontconfig1 2.11.0-6.7+b1
ii  libgdk-pixbuf2.0-0 2.36.5-2
ii  libglib2.0-0   2.50.3-2
ii  libgnome-bluetooth13   3.20.1-1
ii  libgnome-desktop-3-12  3.22.2-1
ii  libgoa-1.0-0b  3.22.5-1
ii  libgoa-backend-1.0-1   3.22.5-1
ii  libgrilo-0.3-0 0.3.2-2
ii  libgtk-3-0 3.22.11-1
ii  libgtop-2.0-10 2.34.2-1
ii  libgudev-1.0-0 230-3
ii  libibus-1.0-5  1.5.14-3
ii  libkrb5-3  1.15-1
ii  libmm-glib01.6.4-1
ii  libnm0 1.6.2-3
ii  libnma01.4.4-1
ii  libpango-1.0-0 1.40.5-1
ii  libpangocairo-1.0-01.40.5-1
ii  libpolkit-gobject-1-0  0.105-18
ii  libpulse-mainloop-glib010.0-1
ii  libpulse0  10.0-1
ii  libpwquality1  1.3.0-1+b1
ii  libsmbclient   2:4.5.8+dfsg-2
ii  libsoup2.4-1   2.56.0-2
ii  libupower-glib30.99.4-4+b1
ii  libwacom2  0.22-1+b1
ii  libx11-6   2:1.6.4-3
ii  libxi6 2:1.7.9-1
ii  libxml22.9.4+dfsg1-2.2

Versions of packages gnome-control-center recommends:
ii  cracklib-runtime  2.9.2-5
ii  cups-pk-helper0.2.6-1+b1
ii  gkbd-capplet  3.22.0.1-1+b1
ii  gnome-online-accounts 3.22.5-1
ii  gnome-user-guide  3.22.0-1
ii  gnome-user-share  3.18.3-1+b1
ii  iso-codes 3.75-1
ii  libcanberra-pulse 0.30-3
ii  libnss-myhostname 232-25
ii  mousetweaks   3.12.0-1+b1
ii  network-manager-gnome 1.4.4-1
ii  policykit-1   0.105-18
ii  pulseaudio-module-bluetooth   10.0-1
ii  realmd0.16.3-1
ii  rygel 0.32.1-3
ii  rygel-tracker 0.32.1-3
ii  system-config-printer-common  1.5.7-3

Versions of packages gnome-control-center suggests:
ii  gstreamer1.0-pulseaudio  1.10.4-1
ii  libcanberra-gtk-module   0.30-3
ii  libcanberra-gtk3-module  0.30-3
ii  x11-xserver-utils7.7+7+b1

-- no debconf information

Bug#623605: #623605 - gnome-system-monitor: Network usage drops on mouseover

[Conrad Hughes]

Hi Pedro,

I'm afraid I've no idea how to reproduce this any more: there appears to
be no tooltip on mouseover in 3.14.1-1, so I guess you could say the
problem has gone since the feature has gone..

Regards,
Conrad

Bug#845530: evince: Search doesn't navigate (by index or 'next') to found results

[Conrad Hughes]

Jason> Does it work if you view the document in continuous mode?  There
Jason> was a bug in evince https://bugzilla.gnome.org/730252.

That's the bug.  Thanks for taking the time to point it out and sorry
for wasting your time — I did look for similar bugs on Debian but it's
not always obvious where/how to look upstream!

Best regards,
Conrad

Bug#800018: closed by Gianfranco Costamagna <locutusofb...@debian.org> (Bug#800018: fixed in s3cmd 1.6.0-2)

[Conrad Hughes]

Hi Gianfranco,

That fixes the problem, thank you!

One minor note for upstream if you're reading though: the error given
now for unreadable files in a sync is

  's3://bucket/file' -> './local-file'  [140 of 140]
  WARNING: Remote file ''. S3Error: 403 (Forbidden)

.. seems odd to refer to "Remote file ''" when it looks as if s3cmd
knows what the file is called.

Best regards,
Conrad

Bug#800018: closed by Gianfranco Costamagna <locutusofb...@debian.org> (Bug#800018: fixed in s3cmd 1.6.0-2)

[Conrad Hughes]

Smashing — thanks very much Gianfranco and Matt; I'll try it out as soon
as I see it hit unstable.

Regards,
Conrad

Bug#800018: s3cmd: sync leaves empty temp file after permission denied error

[Conrad Hughes]

Hi Gianfranco,

Thanks for the fast response.

> can you please try 1.5.2 from Stretch?

I tried 1.5.2-3~bpo8+1 from jessie-backports; hope that's good enough.

It's a substantial regression: not only does it still leave empty temp
files, but it terminates on the first permission error, so I can no
longer sync at all (there are some files there that I don't have
permission to read, but many that I do).  The '--ignore-failed-copy'
flag, which is the only "ignore" in the documentation, doesn't prevent
this behaviour either.

The error reported has changed too: where previously I got a "403
Forbidden", now I get this:

  s3://[blahblahblah] ->   [1 of 134]
  ERROR: S3 error: Unknown error

Regards,
Conrad

Bug#800018: s3cmd: sync leaves empty temp file after permission denied error

[Conrad Hughes]

Hi Gianfranco,

Thanks again for such a rapid turnaround!  So 1.6.0-1 returns to the
behaviour of 1.5.0~rc1-2: it can run the whole sync despite the 403
errors, but it still leaves an empty temporary file for every failure..

Regards,
Conrad

Bug#761671: gnome-terminal: Keypress lagging by one character

[Conrad Hughes]

Pedro Sorin and Conrad, could you please still reproduce this issue
Pedro with newer version of libvte-2.91-0 and gnome-terminal ?

It gradually faded away for me shortly after I reported the bug: I can't
remember encountering it recently at all.  Currently on:

  Desired=Unknown/Install/Remove/Purge/Hold
  | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
  |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
  ||/ Name   Version  Architecture Description
  +++-==---=
  ii  libvte-2.90-9  1:0.36.3-1   amd64Terminal emulator widget for GTK+
  ii  libvte-2.90-co 1:0.36.3-1   all  Terminal emulator widget for GTK+
  ii  libvte-2.91-0  0.38.1-1 amd64Terminal emulator widget for GTK+
  ii  libvte-2.91-co 0.38.1-1 all  Terminal emulator widget for GTK+
  ii  libvte-common  1:0.28.2-5   all  Terminal emulator widget for GTK+
  ii  libvte91:0.28.2-5   amd64Terminal emulator widget for GTK+
  ii  gnome-terminal 3.14.1-1 amd64GNOME terminal emulator applicati
  ii  gnome-terminal 3.14.1-1 all  Data files for the GNOME terminal

.. I think the barest hint of it may still be present, in that I still
notice when starting vim that terminal control sequences sometimes show
up half-completed on screen as text (so it's a bit of a mess at the
bottom of the screen, and the cursor doesn't move to the top until you
type i, but things clear up once you do that).

Pedro If so, is the same behavior as
Pedro https://bugzilla.gnome.org/show_bug.cgi?id=730763 like Egmont reported ?

They could amount to the same thing, but I can't test this.  I did
hypothesise that it could be display lag rather than input lag in my
original report.  I never got a chance to thoroughly test this though.

Regards,
Conrad


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

Bug#747916: Blocker

[Conrad Hughes]

I'd agree, this is something of a blocker: you can walk away from the
system while it's upgrading, and it'll still appear to be upgrading
every time you come back until the end of time: it's not at all obvious
that you have to open the console log and interact with the upgrade
there for it to complete.

Latest example trigger is the postgresql-9.4 9.4~beta3-3 update.

Either there should be a dependency on the relevant libs, or the console
should be forced always open whenever they're not available (or ideally
opened and highlighted in flashing neon if it's waiting for user input).

Conrad


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

Bug#761671: Not just keypress: terminal output, and mouse events too

[Conrad Hughes]

Oh, for what it's worth, here's what reportbug shows for gnome-terminal
for me...

Conrad

Debian Release: jessie/sid
  APT prefers testing-updates
  APT policy: (500, 'testing-updates'), (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 3.16-2-amd64 (SMP w/8 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages gnome-terminal depends on:
ii  dconf-gsettings-backend [gsettings-backend]  0.22.0-1
ii  gconf-service3.2.6-3
ii  gnome-terminal-data  3.14.0-1
ii  gsettings-desktop-schemas3.14.0-1
ii  libatk1.0-0  2.14.0-1
ii  libc62.19-11
ii  libcairo-gobject21.12.16-5
ii  libcairo21.12.16-5
ii  libdconf10.22.0-1
ii  libgconf-2-4 3.2.6-3
ii  libgdk-pixbuf2.0-0   2.30.8-1+b1
ii  libglib2.0-0 2.42.0-2
ii  libgtk-3-0   3.14.1-1
ii  libnautilus-extension1a  3.14.0-1
ii  libpango-1.0-0   1.36.8-2
ii  libpangocairo-1.0-0  1.36.8-2
ii  libuuid1 2.20.1-5.11
ii  libvte-2.91-00.38.0-1
ii  libx11-6 2:1.6.2-3

Versions of packages gnome-terminal recommends:
ii  dbus-x11  1.8.8-2
ii  gvfs  1.22.0-1+b1
ii  yelp  3.14.0-1


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

Bug#761671: Not just keypress: terminal output, and mouse events too

[Conrad Hughes]

I can confirm this.  It makes Gnome unusable if you're a text terminal
type of person.  That said, I think it may be deeper than
gnome-terminal, but you'll be a better judge of that.

I first encountered this issue after dist-upgrading to jessie at the
start of September.  Symptoms:

  - Input events appear to be processed only on receipt of the next
event, so if you type ['d', 'e', backspace, 'f', space], you'll see
the following:

- After d: [nothing]
- After e: d
- After backspace: de
- After f: d
- After space: df

  - Interactions with mouse clicks throughout the GUI suffered the same
problem.

  - Terminal output suffers interruptions too: for example, you could
start vim and only half of the first screen of text would be
displayed.  You'd see the rest after your first keypress.  It
sometimes even stops half way through terminal control codes, so you
get a ^[-type half-escape sequence, which turns into whatever the
right things is when you hit the next key.

  - In fact, it's possible that *output* is the sole problem (maybe
all events are processed immediately, but the consequences just
aren't shown).  I haven't yet been able to discern whether this is
the case though; it just occurred to me as a possibility, and I'll
keep an eye out for evidence now.

Since the computer became completely unusable (imagine using vi in the
above circumstances), I wiped it and tried a fresh install of jessie
from scratch.

Everything seemed good, but of course at that point we were still on
xfce-by-default.  Now with the recent update back to Gnome, the problem
has reappeared, although it is much less frequent: it doesn't happen
*all the time* (which was the case in September; now it happens in fits
every hundred keypresses), and I haven't witnessed mouse problems yet,
but I only started using the - um - upgrade this morning.

In fact, since I'd been using xfce-terminal when I first logged back in
under Gnome, I first witnessed its reappearance while using
xfce-terminal under Gnome, not gnome-terminal - having switched to
gnome-terminal the problem is visible in both.  This is why I suggest
above that the problem may be deeper than gnome-terminal.

I've tried swapping between USB and PS/2 keyboards, and that doesn't
affect the problem.  No sign of it either in the pure xfce environment.

I'd be very grateful for any help, as this pretty much stops me from
being able to use jessie at all.

Best regards,
Conrad


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

Bug#766044: Acknowledgement (gnome-terminal: Custom text/background colour should offer colours from currently selected palette scheme)

[Conrad Hughes]

Minor mea culpa on this: while the main thrust of this report stands,
don't get distracted by the specifics about Solarized and base02: I
thought Gnome's choice of background colour was wrong because my editor
was defaulting to a brighter (and incorrect) background colour, and I
assumed the problem was with the Gnome scheme rather than with the
editor's use of the Gnome palette.

I still think that showing colours from the currently selected palette
in the text/background picker would be appropriate, as described in the
original report.

Regards,
Conrad


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

Bug#748211: os-prober: Dangerous failure during kernel upgrade (probed NFS?)

[Conrad Hughes]

Ben I think this is a bug in GRUB (which invokes it during kernel
Ben update) rather than os-prober.

Ok cool — thanks for the fast reply...  Does this mean I should resubmit
this for grub, or can you pass it over?

Regards,
Conrad


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

Bug#686955: grub-installer: Grub fails to install on dm-crypt on LVM on SATA RAID 0

[Conrad Hughes]

Package: grub-installer
Version: 1.78
Severity: important
Tags: d-i

Dear Maintainer,

I've been trying to install testing (wheezy) on a laptop with SATA RAID 0 
(trying
to preserve dual boot with an existing Windows install).  Used manual 
partitioning,
having booted installer with dmraid=true kernel flag.  Set up a small 
partition
for /boot, and then a large dm-crypt partition, on which I created an LVM 
containing
swap and root.  Install failed when trying to install Grub (I think it 
tried to
install to hda1 or something like that).  However, the Debian wiki on SATA RAID 
says
to expect failure, and fix this from rescue mode.  So, reboot to rescue mode:
installer identifies the encrypted partition, mounts it and finds the encrypted
/dev/mapper/waif-root partition, so start a shell there.  The wiki page says to 
run
grub-install /dev/dm, which fails with a no such disk error; a comment 
suggested
using the base /dev/mapper name for the RAID array, which appears to be the 
right
thing: grub-install /dev/mapper/isw_feefgadah_Volume0 doesn't complain about 
the
disk, but dies with Autodetection of a filesystem of /dev/mapper/waif-root 
failed
(waif is the name of my volume group).

Some online comments suggest that this is a problem with /boot/grub/device.map.
Here's mine:

  (hd0) /dev/disk/by-id/usb-LG_USB_DRIVE_44C4435EF3D60051-0:0
  (hd1) /dev/disk/by-id/ata-MZ-RPC512T_0SO_S0Z6NE0C601503
  (hd2) /dev/disk/by-id/ata-MZ-RPC512T_0SO_S0Z6NE5C601503
  (hd3) /dev/mapper/isw_feefgadah_Volume0

 hd0 is the USB install disk; hd1 and hd2 the real hard drives; hd3 the RAID
array.  This is exactly what grub-mkdevicemap produces.  If I manually add this
line:

  (hd4) /dev/mapper/dm-6_crypt

 pointing to the encrypted volume containing the LVM, then the above 
grub-install
 command executes without complaint.  However, when I reboot it can't find its 
disc
and confronts me with a grub  prompt.  I can get the install to work without
encryption, but this is undesirable for obvious reasons.  Have you any 
suggestions
please?

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-3-amd64 (SMP w/8 CPU cores)
Locale: LANG=C.UTF-8, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash


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

Bug#686955: grub-installer: Grub fails to install on dm-crypt on LVM on SATA RAID 0

[Conrad Hughes]

Dear Lennart,

Thanks very much for such a fast response --- I'm well aware that dmraid
is not recommended, but it would be really handy if I could keep it
for the Windows dual boot.  Your suggestion that mdadm supports Intel
RST sounds really interesting --- but I can't see anything that
documents how you activate/use it in the Debian installer.

It does appear possibly to just be enabled by default, in that if I run
through the wheezy installer *without* adding the dmraid=true kernel
boot parameter, then it recognises a Linux Software Raid array on
/dev/md126; with some false starts I've managed to partition this as
before (small /boot, then a large dm-crypt partition on which I put LVM,
then splitting the LVM into swap and root partitions).  Again, grub
tries to install to /dev/hda1 and fails, but this time when I reboot to
the wheezy rescue mode, things are much more complicated.  During disk
discovery, it doesn't recognise the RAID array until I tell the
installer to look for it, at which point it recognises /dev/md126 but
doesn't do anything with it.  I had to manually start a shell and invoke

  cryptsetup luksOpen /dev/md126p7 dm-6_crypt

then restart the disk discovery process, finally being given an option
to use /dev/mapper/waif-root for my root partition.  Then, however, I
ran into trouble with grub: update-grub complained of file not found
repeatedly, and grub-install failed to install because it couldn't
find my LVM root volume.  /boot/grub/device.map at this point contains
just three drives, my USB install disk, and the base hard drives of the
machine's RAID array (as hd0, hd1, hd2 respectively).  If I add a bogus
(hd3) entry pointing at the raid drive, /dev/md126, then update-grub
stops complaining file not found, and even successfully finds the
Windows partition.  If I then add a bogus (hd4) entry pointing at the
dm-crypt partition I mounted, /dev/mapper/dm-6_crypt, then grub-install
stops complaining about not finding /dev/mapper/waif-root and claims to
have successfully executed.

The net result though, is exactly the same --- on reboot I get

  Welcome to GRUB!

  error: no such disk.
  Entering rescue mode...
  grub rescue

I'm assuming I need some additional magic to make grub assemble the raid
array, find the boot partition, decrypt the dm-crypt partition, and open
the root volume...

Conrad


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

Bug#686955: grub-installer: Grub fails to install on dm-crypt on LVM on SATA RAID 0

[Conrad Hughes]

 # cat /boot/grub/device.map 
 (hd0)   /dev/disk/by-id/md-uuid-5847a5f9:19773d99:f13783a1:b611c764

Ok, so doing something similar on my machine again has the effect of
successfully locating the RAID drive, but it still fails to find the
encrypted LVM (Auto-detection of a filesystem of /dev/mapper/waif-root
failed).  Do you have any cryptdevice lines in your /etc/default/grub,
or /boot/grub/grub.cfg, or some customisation in /etc/grub.d?

Conrad


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

Bug#686955: grub-installer: Grub fails to install on dm-crypt on LVM on SATA RAID 0

[Conrad Hughes Conrad Hughes]

Lennart I have no crypt at all and am not interested in getting my life that
Lennart complicated. :)

Oh, oh well.  I can get it to work without crypto; it's the crypto that
breaks things.

Lennart I don't see how grub could deal with crypt.  Does it do that?

On my non-RAID machines, yes.  I can't see what the difference between
them is unfortunately.

Conrad


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

Bug#596958: Bug, not wishlist?

[Conrad Hughes]

I think this looks more like a bug than wishlist: the list of
file_patterns[] in src/sj-prefs.c supports leading zeros with %dN/%tN,
and the *only* pattern which uses the no-leading-zero form %dn is
Number - Title --- leading me to think that %dn is an oversight?  The
fix is a one character edit changing %dn to %dN AFAICS:

  {N_(Number - Title), %dn - %tt},
  {N_(Track Title), %tt},
  {N_(Track Artist - Track Title), %ta - %tt},
  {N_(Track Artist (sortable) - Track Title), %ts - %tt},
  {N_(Number. Track Artist - Track Title), %dN. %ta - %tt},
  /* {N_(Number. Track Artist (sortable) - Track Title), %tN. %ts - %tt}, */
  {N_(Number-Track Artist-Track Title (lowercase)), %dN-%tA-%tT},
  /* {N_(Number-Track Artist (sortable)-Track Title (lowercase)), %tN-%tS-%tT
}, */

I'd be very grateful for either this fix or addition of a leading-zeros
configuration.

Regards,
Conrad



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

Bug#613670: linux-image-2.6.32-bpo.5-686: EDID checksum failure kills X while switched away using KVM switch

[Conrad Hughes]

Hi Jonathan, Julien,

Thanks for getting back to me on this.  Unfortunately I've since
switched to a DisplayPort monitor which my DVI KVM switch can't switch,
so can't easily check this.  I'll try to find time at some point though.

Regards,
Conrad



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

Bug#623604: xserver-xorg: AllowEmptyInput kills keyboard mouse

[Conrad Hughes]

Cyril Agreed. Conrad, can you please check udev+/run's status on your end?

Not sure what you mean by this, sorry..?

Conrad



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

Bug#623604: xserver-xorg: AllowEmptyInput kills keyboard mouse

[Conrad Hughes]

Hi Julien,

Julien Please submit the dmesg output when running in that default
Julien configuration (no xorg.conf, no nvidia packages).

Hrm.  I've done a number of apt-get upgrades since last week, and cannot
now replicate the problem.  Removing nvidia (to run nouveau) and
removing xorg.conf, everything now just works, unlike from that fresh
install.  If you still want to see an old /var/log/dmesg from prior to
my fix, let me know.  Otherwise I guess this should be closed.  Sorry
for seemingly wasting your time :(

Conrad



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

Bug#623605: gnome-system-monitor: Network usage drops on mouseover

[Conrad Hughes]

Package: gnome-system-monitor
Version: 2.28.1-1
Severity: normal


Waving the mouse over the network graph (in order to bring up the tool
tip with actual bandwidth figures) causes the graph to plummet:

  1) Open preferences and tick network to bring up the network usage
 graph.
  2) Start a long down/upload.
  3) Watch the steady high network usage graph.
  4) Move the pointer over the network graph to see the Receiving/
 Sending tooltip, then off the graph and back on again; repeat.
  5) Watch the graph drop from high usage to (seemingly) very low.

This weirdness is visible on all Debian systems I have access to, not
just wheezy.

Regards,
Conrad

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.38-2-amd64 (SMP w/8 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages gnome-system-monitor depends on:
ii  gconf2   2.28.1-6GNOME configuration database syste
ii  libc62.11.2-11   Embedded GNU C Library: Shared lib
ii  libcairo21.10.2-6The Cairo 2D vector graphics libra
ii  libgcc1  1:4.6.0-2   GCC support library
ii  libgconf2-4  2.28.1-6GNOME configuration database syste
ii  libglib2.0-0 2.28.4-1The GLib library of C routines
ii  libglibmm-2.4-1c2a   2.24.2-1C++ wrapper for the GLib toolkit (
ii  libgtk2.0-0  2.24.3-1~sid1   The GTK+ graphical user interface 
ii  libgtkmm-2.4-1c2a1:2.20.3-1  C++ wrappers for GTK+ (shared libr
ii  libgtop2-7   2.28.1-1gtop system monitoring library (sh
ii  librsvg2-2   2.32.1-1SAX-based renderer library for SVG
ii  libsigc++-2.0-0c2a   2.2.9-1 type-safe Signal Framework for C++
ii  libstdc++6   4.6.0-2 The GNU Standard C++ Library v3
ii  libwnck222.30.4-3Window Navigator Construction Kit 
ii  libxml2  2.7.8.dfsg-2+b1 GNOME XML library

Versions of packages gnome-system-monitor recommends:
ii  gvfs   1.6.4-3   userspace virtual filesystem - ser
ii  libgksu2-0 2.0.13~pre1-4 library providing su and sudo func

gnome-system-monitor suggests no packages.

-- debconf-show failed



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

Bug#581346: Patch with different defaults

[Conrad Hughes]

Attached is a patch which (seems to work for me) changes the defaults to
two-page, non-continuous, best fit view, as a workaround for those who
prefer those settings.  Install as follows (you'll need a chunk of disc
space):

  As root:
apt-get build-dep evince
  As you:
apt-get source evince
cd evince-2.30.3
patch -p1  ../evince-defaults.patch
debuild
(ignore the debsign failed error)
  As root:
dpkg -i ../evince_2.30.3-2cjch_amd64.deb 
../libevince2_2.30.3-2cjch_amd64.deb

Afterward you can remove all the files and directories created by the
above steps.

Conrad
diff -c -r evince-2.30.3/debian/changelog evince-2.30.3-2cjch/debian/changelog
*** evince-2.30.3/debian/changelog	2011-01-10 18:04:29.0 +
--- evince-2.30.3-2cjch/debian/changelog	2011-04-12 19:08:10.825269403 +0100
***
*** 1,3 
--- 1,9 
+ evince (2.30.3-2cjch) unstable; urgency=low
+ 
+   * Change defaults two dual page, best fit, non-continuous.
+ 
+  -- Conrad Hughes con...@beastie.xrad.org  Tue, 12 Apr 2011 19:07:35 +0100
+ 
  evince (2.30.3-2) unstable; urgency=medium
  
* Fix PostScript capitalization. Closes: #591872.
diff -c -r evince-2.30.3/libview/ev-document-model.c evince-2.30.3-2cjch/libview/ev-document-model.c
*** evince-2.30.3/libview/ev-document-model.c	2010-04-05 10:21:59.0 +0100
--- evince-2.30.3-2cjch/libview/ev-document-model.c	2011-04-12 19:01:29.0 +0100
***
*** 223,243 
  			Sizing Mode,
  			Current sizing mode,
  			EV_TYPE_SIZING_MODE,
! 			EV_SIZING_FIT_WIDTH,
  			G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_CONTINUOUS,
  	 g_param_spec_boolean (continuous,
  			   Continuous,
  			   Whether document is displayed in continuous mode,
! 			   TRUE,
  			   G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_DUAL_PAGE,
  	 g_param_spec_boolean (dual-page,
  			   Dual Page,
  			   Whether document is displayed in dual page mode,
! 			   FALSE,
  			   G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_FULLSCREEN,
--- 223,243 
  			Sizing Mode,
  			Current sizing mode,
  			EV_TYPE_SIZING_MODE,
! 			EV_SIZING_BEST_FIT,
  			G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_CONTINUOUS,
  	 g_param_spec_boolean (continuous,
  			   Continuous,
  			   Whether document is displayed in continuous mode,
! 			   FALSE,
  			   G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_DUAL_PAGE,
  	 g_param_spec_boolean (dual-page,
  			   Dual Page,
  			   Whether document is displayed in dual page mode,
! 			   TRUE,
  			   G_PARAM_READWRITE));
  	g_object_class_install_property (g_object_class,
  	 PROP_FULLSCREEN,
***
*** 264,271 
  {
  	model-page = -1;
  	model-scale = 1.;
! 	model-sizing_mode = EV_SIZING_FIT_WIDTH;
! 	model-continuous = TRUE;
  	model-inverted_colors = FALSE;
  	model-min_scale = 0.;
  	model-max_scale = G_MAXDOUBLE;
--- 264,272 
  {
  	model-page = -1;
  	model-scale = 1.;
! 	model-sizing_mode = EV_SIZING_BEST_FIT;
! 	model-dual_page = TRUE;
! 	model-continuous = FALSE;
  	model-inverted_colors = FALSE;
  	model-min_scale = 0.;
  	model-max_scale = G_MAXDOUBLE;
Only in evince-2.30.3-2cjch/libview: .ev-document-model.c.swp
diff -c -r evince-2.30.3/.pc/applied-patches evince-2.30.3-2cjch/.pc/applied-patches
*** evince-2.30.3/.pc/applied-patches	2011-04-12 19:12:56.504269072 +0100
--- evince-2.30.3-2cjch/.pc/applied-patches	2011-04-12 19:08:23.776266817 +0100
***
*** 1 
--- 1,3 
  01_dvi_security.patch
+ debian-changes-2.30.3-2
+ debian-changes-2.30.3-2cjch
Only in evince-2.30.3-2cjch/.pc: debian-changes-2.30.3-2
Only in evince-2.30.3-2cjch/.pc: debian-changes-2.30.3-2cjch
diff -c -r evince-2.30.3/shell/ev-window.c evince-2.30.3-2cjch/shell/ev-window.c
*** evince-2.30.3/shell/ev-window.c	2010-06-24 09:19:29.0 +0100
--- evince-2.30.3-2cjch/shell/ev-window.c	2011-04-12 19:01:29.0 +0100
***
*** 5217,5226 
  	  G_CALLBACK (ev_window_view_sidebar_cb), TRUE },
  { ViewContinuous, EV_STOCK_VIEW_CONTINUOUS, N_(_Continuous), NULL,
  	  N_(Show the entire document),
! 	  G_CALLBACK (ev_window_cmd_continuous), TRUE },
  { ViewDual, EV_STOCK_VIEW_DUAL, N_(_Dual), NULL,
  	  N_(Show two pages at once),
! 	  G_CALLBACK (ev_window_cmd_dual), FALSE },
  { ViewFullscreen, GTK_STOCK_FULLSCREEN, N_(_Fullscreen), F11,
N_(Expand the window to fill the screen),
G_CALLBACK (ev_window_cmd_view_fullscreen) },
--- 5217,5226 
  	  G_CALLBACK (ev_window_view_sidebar_cb), TRUE },
  { ViewContinuous

Bug#613670: Acknowledgement (linux-image-2.6.32-bpo.5-686: EDID checksum failure kills X while switched away using KVM switch)

[Conrad Hughes]

After some further experimentation it seems that it might be the actual
moment of switching the KVM switch back to my squeeze machine that X
dies: through ssh I could see processes attached to my login right until
I switched.

Conrad



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

Bug#613670: linux-image-2.6.32-bpo.5-686: EDID checksum failure kills X while switched away using KVM switch

[Conrad Hughes]

Package: linux-2.6
Version: 2.6.32-30~bpo50+1
Severity: important


Hello,

This problem may well reside in X or gnome, but unfortunately they
aren't reporting anything, while the kernel (drm module) is.

I'm using a keyboard-video-mouse (KVM) switch to switch between a number
of computers; before squeeze, this worked brilliantly.  Since upgrading
to squeeze, a few minutes after I switch to another computer, my squeeze
install kills X completely.  All I'm seeing is the following in
/var/log/syslog:

  Feb 16 12:00:11 mouse kernel: [826528.890150] 
  Feb 16 12:00:12 mouse kernel: [826528.992844] [drm:edid_is_valid] *ERROR* 
EDID checksum is invalid, remainder is 130
  Feb 16 12:00:12 mouse kernel: [826528.992851] [drm:edid_is_valid] *ERROR* Raw 
EDID:
  Feb 16 12:00:12 mouse kernel: [826528.992857] 300 ff ff ff ff ff ff 00 ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992863] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992869] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992874] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992880] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992885] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992891] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  
  Feb 16 12:00:12 mouse kernel: [826528.992896] 3ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff  

... the above repeats eight times at 0.1 second intervals, and is
followed by this (the messages appear to come from drivers/gpu/drm/
drm_edid.c):

  Feb 16 12:00:12 mouse kernel: [826528.992901] 
  Feb 16 12:00:12 mouse kernel: [826528.992907] i915 :00:02.0: DVI-D-1: 
EDID invalid.

I'm getting nothing in /var/log/Xorg.0.log or /var/log/gdm/:0.log.1.
Xorg.0.log does record successful identification of the monitor though
(I *'ed out its Serial#):

  (II) intel(0): EDID for output DVI1
  (II) intel(0): Manufacturer: HWP  Model: 2615  Serial#: 
  [..]
  (II) intel(0): Monitor name: hp L2335

All processes associated with my login vanish at this point, and when I
switch back to the computer running squeeze, the monitor detects no
signal and goes into standby.  I can recover by hitting control-alt-F1
(this does not bring the display back), logging in blind as root and
running '/etc/init.d/gdm restart'.

My hardware comprises a Mac mini:

  00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 
943/940GML Express Integrated Graphics Controller (rev 03)

.. connected to an Avocent Switchview DVI connected to an HP L2335
monitor.

A similar problem has turned up in Microsoft Windows 7, where continuous
checking of EDID results in the OS switching to default monitor
resolution after the user switches away (or even turns off their monitor
in some cases).  Microsoft apparently blame this on obsolete KVM
hardware and AFAICS have not offered a facility to disable the
continuous checking; I hope there might be a means of disabling the
check in Debian though, or remembering the last valid EDID rather than
failing catastrophically on receipt of an invalid one?  The consequences
are severe, as my entire login is just killed stone dead: all processes
gone, all work lost.

  
http://social.answers.microsoft.com/Forums/en-US/w7hardware/thread/84f41660-1933-4109-9b13-1ea8a1c27be7

Regards,
Conrad

-- Package-specific info:
** Version:
Linux version 2.6.32-bpo.5-686 (Debian 2.6.32-30~bpo50+1) 
(norb...@tretkowski.de) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #1 SMP Tue Jan 
18 23:27:36 UTC 2011

** Command line:
BOOT_IMAGE=/vmlinuz-2.6.32-bpo.5-686 root=/dev/mapper/mouse-root ro

** Not tainted

** Kernel log:
[828620.852437] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852443] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852448] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852455] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852460] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852465] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[828620.852470] 
[828620.852476] i915 :00:02.0: DVI-D-1: EDID invalid.
[831256.449906] opera[5403]: segfault at 6e ip 086eac5b sp bfc426d0 error 6 in 
opera[8048000+f41000]
[831709.853927] [drm:edid_is_valid] *ERROR* EDID checksum is invalid, remainder 
is 130
[831709.853936] [drm:edid_is_valid] *ERROR* Raw EDID:
[831709.853943] 300 ff ff ff ff ff ff 00 ff ff ff ff ff ff ff ff  

[831709.853949] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  

[831709.853954] 3ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 

Bug#489613: openafs-modules-source: attempt to reinsert module results in Cannot allocate memory

[Conrad Hughes]

FWIW, my Debian stable machine just started exhibiting this problem
after months of perfect operation; persisted across reboots too.
Switching to the 1.4.12.1+dfsg-2~bpo50+1 openafs suite in
lenny-backports appears to have resolved it.

uname -a:
  Linux ... 2.6.26-2-686 #1 SMP Mon Jun 21 05:58:44 UTC 2010 i686 GNU/Linux

Thanks for the suggestion!

Conrad

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.




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

Bug#340453: crack-md5: Crack wrapper script breaks on almost all common usage.

[Conrad Hughes]

Package: crack-md5
Version: 5.0a-8
Severity: important

/usr/sbin/Crack is broken in several ways:

  - It does the right thing with shadow passwords *only* if you run
*exactly* Crack /etc/passwd, so for example the command line
recommended on Crack's manpage, Crack -nice 10 /etc/passwd does
nothing.

  - It cds into /usr/share/Crack before running the real Crack, so if
you've generated your own merged passwd/shadow file fixedpasswd
(say) because Crack -nice 10 /etc/passwd doesn't work, then Crack
fixedpasswd won't work because Crack will cd away from the
directory in which fixedpasswd exists and then (I think) load
/etc/passwd as some kind of default so Crack-Reporter comments
(disorientingly, since you've fixed them in fixedpasswd) that it's
ignoring all your shadowed passwords.

At least a big note to the effect that ful pathnames are required for
the password files would be useful - in the SYNOPSIS of the manpage for
example.  Better would be to extend /usr/sbin/Crack so that it does the
right thing under more circumstances, including the cases documented in
the manpage, and so that it notices when it's been given a relative
pathname to the password file and either complains or fixes it up (for
example by temporarily copying it to /var/run/Crack)...

-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (990, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.14-2-686
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)

Versions of packages crack-md5 depends on:
ii  crack-common  5.0a-8 Password guessing program
ii  libc6 2.3.5-8GNU C Library: Shared libraries an

crack-md5 recommends no packages.

-- no debconf information


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

Bug#317817: Left alt vs right alt!?

[Conrad Hughes]

So interestingly the latest unstable sawfish fixes the problem for me
for my left alt key, but not the right one.  I needed to reset the key
binding for Cycle windows using grab key (it picked up M-TAB, think
it used to be W-TAB or something else beforehand).  But the cycle-window
display still sticks with my right alt key.  If I xmodmap code 113
(right alt) to be Alt_L/Meta_L instead of Alt_R/Meta_R the problem goes
away, but that's a workaround, not a solution.  This arises no matter
which Alt key I use to grab the key in the sawfish configurator, and
doesn't seem to be affected by the setting of Modifier key(s) used for
default shortcuts either.  Only workaround seems to be making right alt
pretend to be left alt.

I tried looking at the patch which fixes the left alt problem, but that
*seems* to be simply compiling with fewer optimisations
(-fnostrict-aliasing switch to compiler), which doesn't leave me with
any clue as to why right alt should behave differently to left...

Any ideas?

(I'm running a fully up-to-date unstable Debian btw)

Conrad


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

Bug#324850: Acknowledgement (grip: Grip crashes when no CD is present (including after auto-eject at end of rip))

[Conrad Hughes]

On further investigation this problem (and the spew of kernel complaints
regarding sg) only seems to manifest with ide-scsi, which I think is now
deprecated.  Grip doesn't crash with straight ide.  Rips a bit more
slowly though.  So you might want to ignore this bug.

Conrad


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

Bug#324850: grip: Grip crashes when no CD is present (including after auto-eject at end of rip)

[Conrad Hughes]

Package: grip
Version: 3.3.1-4
Severity: important


Grip crashes when no CD is present; if I start it with no CD in the
drive:

  grip --verbose --debug

.. I get the following output:

  Using config file [.grip]
  Drive status is 4

.. and then a crash (window pops up saying `The Application grip has
quit unexpectedly').

Since this bug also triggers immediately Grip ejects a CD, it can wreck
the encoding of the last track or two (if you have auto-eject after
rip on, it ejects before encoding has finished and then crashes).

I'm also getting continuous complaints on syslog as follows:

  Aug 24 13:27:11 synaesthesia kernel: sg_write: data in/out 30576/30576 bytes 
for SCSI command 0xbe--guessing data in;
  Aug 24 13:27:11 synaesthesia kernel:program grip not setting count and/or 
reply_len properly
  Aug 24 13:27:16 synaesthesia kernel: printk: 145 messages suppressed.

I can't get gdb to make sense of the crash when I build a debug version
though I'm afraid.

Conrad

-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (990, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.12-1-686
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)

Versions of packages grip depends on:
ii  libart-2.0-2  2.3.17-1   Library of functions for 2D graphi
ii  libatk1.0-0   1.10.1-2   The ATK accessibility toolkit
ii  libaudiofile0 0.2.6-6Open-source version of SGI's audio
ii  libbonobo2-0  2.10.0-1   Bonobo CORBA interfaces library
ii  libbonoboui2-02.10.0-1   The Bonobo UI library
ii  libc6 2.3.5-4GNU C Library: Shared libraries an
ii  libcdparanoia03a9.8-11   Shared libraries for cdparanoia (r
ii  libcurl3  7.14.0-5   Multi-protocol file transfer libra
ii  libesd-alsa0 [libesd0]0.2.36-1   Enlightened Sound Daemon (ALSA) - 
ii  libfontconfig12.3.2-1generic font configuration library
ii  libfreetype6  2.1.10-1   FreeType 2 font engine, shared lib
ii  libgcc1   1:4.0.1-6  GCC support library
ii  libgconf2-4   2.10.1-1   GNOME configuration database syste
ii  libgcrypt11   1.2.1-4LGPL Crypto library - runtime libr
ii  libglib2.0-0  2.8.0-1The GLib library of C routines
ii  libgnome-keyring0 0.4.3-1GNOME keyring services library
ii  libgnome2-0   2.10.1-1   The GNOME 2 library - runtime file
ii  libgnomecanvas2-0 2.10.2-2   A powerful object-oriented display
ii  libgnomeui-0  2.10.1-1   The GNOME 2 libraries (User Interf
ii  libgnomevfs2-02.10.1-5   The GNOME virtual file-system libr
ii  libgnutls11   1.0.16-13.1GNU TLS library - runtime library
ii  libgpg-error0 1.1-4  library for common error values an
ii  libgtk2.0-0   2.6.9-1The GTK+ graphical user interface 
ii  libice6   6.8.2.dfsg.1-5 Inter-Client Exchange library
ii  libid3-3.8.3c23.8.3-4.2  Library for manipulating ID3v1 and
ii  libidn11  0.5.18-1   GNU libidn library, implementation
ii  libjpeg62 6b-10  The Independent JPEG Group's JPEG 
ii  libncurses5   5.4-9  Shared libraries for terminal hand
ii  liborbit2 1:2.12.2-3 libraries for ORBit2 - a CORBA ORB
ii  libpango1.0-0 1.8.2-1Layout and rendering of internatio
ii  libpopt0  1.7-5  lib for parsing cmdline parameters
ii  libsm66.8.2.dfsg.1-5 X Window System Session Management
ii  libstdc++64.0.1-6The GNU Standard C++ Library v3
ii  libtasn1-20.2.13-1   Manage ASN.1 structures (runtime)
ii  libvte4   1:0.11.13-4Terminal emulator widget for GTK+ 
ii  libx11-6  6.8.2.dfsg.1-5 X Window System protocol client li
ii  libxft2   2.1.7-1FreeType-based font drawing librar
ii  libxml2   2.6.20-1   GNOME XML library
ii  libxrender1   1:0.9.0-2  X Rendering Extension client libra
ii  xlibs 6.8.2.dfsg.1-5 X Window System client libraries m
ii  zlib1g1:1.2.3-3  compression library - runtime

Versions of packages grip recommends:
ii  vorbis-tools  1.0.1-1.4  Several Ogg Vorbis Tools

-- no debconf information


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

Bug#324853: grip: Error in cdda tests in configure.in

[Conrad Hughes]

Package: grip
Version: 3.3.1-4
Severity: important
Justification: fails to build from source


The cdda tests in grip's configure script fail because cdda_paranoia.h
won't compile unless cdda_interface.h has already been #included.  The
fix isn't desperately difficult - replace the original check line:

  AC_CHECK_HEADERS(cdda_interface.h cdda_paranoia.h)

.. with something like this:

  AC_CHECK_HEADERS(cdda_interface.h)
  AC_CHECK_HEADERS([cdda_paranoia.h], [], [], [#if HAVE_CDDA_INTERFACE_H
#include cdda_interface.h
#endif ])

.. I expect that it's probably necessary to do the same thing for the
cdda/cdda_* test as well, but my cdda headers are in /usr/include so
that's not tickled on my box.

Conrad

-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (990, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.12-1-686
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)

Versions of packages grip depends on:
ii  libart-2.0-2  2.3.17-1   Library of functions for 2D graphi
ii  libatk1.0-0   1.10.1-2   The ATK accessibility toolkit
ii  libaudiofile0 0.2.6-6Open-source version of SGI's audio
ii  libbonobo2-0  2.10.0-1   Bonobo CORBA interfaces library
ii  libbonoboui2-02.10.0-1   The Bonobo UI library
ii  libc6 2.3.5-4GNU C Library: Shared libraries an
ii  libcdparanoia03a9.8-11   Shared libraries for cdparanoia (r
ii  libcurl3  7.14.0-5   Multi-protocol file transfer libra
ii  libesd-alsa0 [libesd0]0.2.36-1   Enlightened Sound Daemon (ALSA) - 
ii  libfontconfig12.3.2-1generic font configuration library
ii  libfreetype6  2.1.10-1   FreeType 2 font engine, shared lib
ii  libgcc1   1:4.0.1-6  GCC support library
ii  libgconf2-4   2.10.1-1   GNOME configuration database syste
ii  libgcrypt11   1.2.1-4LGPL Crypto library - runtime libr
ii  libglib2.0-0  2.8.0-1The GLib library of C routines
ii  libgnome-keyring0 0.4.3-1GNOME keyring services library
ii  libgnome2-0   2.10.1-1   The GNOME 2 library - runtime file
ii  libgnomecanvas2-0 2.10.2-2   A powerful object-oriented display
ii  libgnomeui-0  2.10.1-1   The GNOME 2 libraries (User Interf
ii  libgnomevfs2-02.10.1-5   The GNOME virtual file-system libr
ii  libgnutls11   1.0.16-13.1GNU TLS library - runtime library
ii  libgpg-error0 1.1-4  library for common error values an
ii  libgtk2.0-0   2.6.9-1The GTK+ graphical user interface 
ii  libice6   6.8.2.dfsg.1-5 Inter-Client Exchange library
ii  libid3-3.8.3c23.8.3-4.2  Library for manipulating ID3v1 and
ii  libidn11  0.5.18-1   GNU libidn library, implementation
ii  libjpeg62 6b-10  The Independent JPEG Group's JPEG 
ii  libncurses5   5.4-9  Shared libraries for terminal hand
ii  liborbit2 1:2.12.2-3 libraries for ORBit2 - a CORBA ORB
ii  libpango1.0-0 1.8.2-1Layout and rendering of internatio
ii  libpopt0  1.7-5  lib for parsing cmdline parameters
ii  libsm66.8.2.dfsg.1-5 X Window System Session Management
ii  libstdc++64.0.1-6The GNU Standard C++ Library v3
ii  libtasn1-20.2.13-1   Manage ASN.1 structures (runtime)
ii  libvte4   1:0.11.13-4Terminal emulator widget for GTK+ 
ii  libx11-6  6.8.2.dfsg.1-5 X Window System protocol client li
ii  libxft2   2.1.7-1FreeType-based font drawing librar
ii  libxml2   2.6.20-1   GNOME XML library
ii  libxrender1   1:0.9.0-2  X Rendering Extension client libra
ii  xlibs 6.8.2.dfsg.1-5 X Window System client libraries m
ii  zlib1g1:1.2.3-3  compression library - runtime

Versions of packages grip recommends:
ii  vorbis-tools  1.0.1-1.4  Several Ogg Vorbis Tools

-- no debconf information


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

Bug#302480: nvidia-kernel-source: cpu does not support PAT, aborting.. - what?

[Conrad Hughes]

Randall I don't think it's anything you have to worry about. 

Grand - thanks for the rapid response and for maintaining this package,
it makes use of this driver much easier than before!

Conrad


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

Bug#302729: kernel-image-2.6.8-2-386: PS/2 mouse support broken with KVM switch

[Conrad Hughes]

Package: kernel-image-2.6.8-2-386
Version: 2.6.8-15
Severity: important


For all kernels more recent than 2.6.6-2-386, one of my machines (quite
an old ALi M1541-based one) treats all mouse movement as random noise
when my Intellimouse Explorer 3.0 is connected to my machine through my
I/O Gear CS-182 MiniView Plus KVM (keyboard/video/mouse) switch - any
mouse motion produces pointer motion all over the screen (both under
gpm/curses and X) and all kinds of button clicks - selects, pastes, etc.
Completely unusable.  The problem is not present if I connect the mouse
directly to the computer, nor did it ever manifest under kernel 2.4.  It
also doesn't seem to be an issue for the other machine on the switch,
which is a more recent nForce2-based box running kernel 2.6.9-2-k7.

This problem has been reported extensively elsewhere on kernel.org and
in discussion fora for other distributions, so feel free to ignore - my
main reason for writing is that after some experimentation (since I now
need a more bleeding-edge kernel than 2.6.6) I seem to have a fairly
trivial hack to work around the problem and since I've not seen it
documented elsewhere I thought it'd be worth writing down.

Firstly, although I've not seen this written down anywhere, it seems to
me that the problem actually seems to come in two forms: the first form
is where people encounter problems (loss of synchronisation, chaotic
pointer movement etc.) whenever they switch using the KVM; the second
form is a complete failure to initialise - i.e. the mouse/pointer seems
to be messed up from the moment the machine boots.  My workaround is for
the second form - problems from boot time - and consists of repeatedly
reloading psmouse with increasing levels of protocol support before
starting the display manager - most easily accomplished by creating
something like /etc/rc2.d/S14psmouse-restart (for example) as follows:

  #!/bin/sh

  rmmod psmouse
  modprobe psmouse proto=imps
  rmmod psmouse
  modprobe psmouse proto=exps

Missing either stage of this incremental approach to loading psmouse
will leave me with the original problem (so suggestions along the lines
of adding kernel boot parameters psmouse.proto=bare don't work for
me).  I don't pretend to understand what's going on, I'll just say that
this works for me and hope that it might provide some insight into
what's happening for someone who does understand, and in the meantime
possibly help others get the full feature set from their mouse...

Conrad

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (900, 'unstable')
Architecture: i386 (i586)
Kernel: Linux 2.6.8-2-386
Locale: LANG=en_GB, LC_CTYPE=en_GB (charmap=ISO-8859-1)

Versions of packages kernel-image-2.6.8-2-386 depends on:
ii  coreutils [fileutils] 5.2.1-2The GNU core utilities
ii  fileutils 5.2.1-2The GNU file management utilities 
ii  initrd-tools  0.1.77 tools to create initrd image for p
ii  module-init-tools 3.2-pre1-2 tools for managing Linux kernel mo

-- no debconf information


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

Bug#302480: nvidia-kernel-source: cpu does not support PAT, aborting.. - what?

[Conrad Hughes]

Package: nvidia-kernel-source
Version: 1.0.7167-1
Severity: normal

Just upgraded to 1.0.7167 from 1.0.6629, and as I modprobe the freshly
built nvidia module I get this error:

  NVRM: loading NVIDIA Linux x86 NVIDIA Kernel Module  1.0-7167  Fri Feb 25 
09:08:22 PST 2005
  NVRM: cpu does not support PAT, aborting..

.. the weird thing is that whatever's being aborted doesn't seem to be
critical, as it seems that the module is still being loaded, and I can
still run X anyway.  I can find no documentation for the error, nor can
I find the phrase using google or anything describing its meaning in the
module source.

The problem seems to stem from debian/patches/04_minion, which refers to
a website (http://www.minion.de/files/) which notes that all the patches
it contained were hopelessly outdated.

What's going on?  Should I care?  If not, it'd at least be nice for this
abort message to indicate that it's not something to worry about...

My CPU is an AMD K6-3+ - here's /proc/cpuinfo:

  processor : 0
  vendor_id : AuthenticAMD
  cpu family: 5
  model : 13
  model name: AMD-K6(tm)-III Processor
  stepping  : 0
  cpu MHz   : 551.135
  cache size: 256 KB
  fdiv_bug  : no
  hlt_bug   : no
  f00f_bug  : no
  coma_bug  : no
  fpu   : yes
  fpu_exception : yes
  cpuid level   : 1
  wp: yes
  flags : fpu vme de pse tsc msr mce cx8 pge mmx syscall 3dnowext 3dnow 
k6_mtrr
  bogomips  : 1089.53

.. and my graphics card is a GeForce 256 - here's lspci's comment:

  :01:00.0 VGA compatible controller: nVidia Corporation NV10DDR [GeForce 
256 DDR] (rev 10)

I get the same problem under kernel 2.6.6-2-386 BTW.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (900, 'unstable')
Architecture: i386 (i586)
Kernel: Linux 2.6.8-2-386
Locale: LANG=en_GB, LC_CTYPE=en_GB (charmap=ISO-8859-1)

Versions of packages nvidia-kernel-source depends on:
ii  debhelper 4.2.32 helper programs for debian/rules
ii  dpatch2.0.10 patch maintenance system for Debia
ii  make  3.80-9 The GNU version of the make util
ii  sed   4.1.4-2The GNU sed stream editor

-- no debconf information


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

Bug#299057: mpack: Numbering for duplicate filenames isn't reset with multiple attachments

[Conrad Hughes]

Package: mpack
Version: 1.6-1
Severity: normal


Hi,

When running munpack on a file with multiple attachments, more than one
of which produces a duplicate filename (and hence a .1, .2, etc.
suffix), the counter for the .N suffixes isn't reset with each
attachment, so the second duplicate is named foo.2 and the third
bar.3 - even if foo.1, bar.1 and bar.2 don't exist.

Reproduction is fairly easy - just take a document with two attachments
and run munpack twice.  You'll get attachment_a and attachment_b
(say) on the first run, and attachment_a.1 and attachment_b.2 ond
the second run - the new attachment_b.2 should really be
attachment_b.1.

Conrad

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (990, 'unstable'), (400, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-686
Locale: LANG=en_GB, LC_CTYPE=en_GB (charmap=ISO-8859-1)

Versions of packages mpack depends on:
ii  libc6   2.3.2.ds1-20 GNU C Library: Shared libraries an

-- no debconf information


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

Bug#299057: mpack: Numbering for duplicate filenames isn't reset with multiple attachments

[Conrad Hughes]

Hi Michelle,

Michelle If I must unpack something like an E-mail, make changes to one
Michelle or more Attachments and mpack it, the unpacked Attachments
Michelle should be mumeroted to get the same resultat as the
Michelle originating file.

Either I'm not understanding you or I didn't make myself clear; perhaps
this will help..

I've attached two files, foo.txt and bar.txt, to this message (I
hope).  If you run this email through munpack you'll get two files
called foo.txt and bar.txt, which is correct.

However, if I then run munpack *again* with foo.txt and bar.txt
*still there*, munpack sees that creating foo.txt and bar.txt will
overwrite existing files.  This is good.  Its response is to change the
names.  It creates a new file called foo.txt.1 (which is good), and
another new file called bar.txt.2 (which I think is a mistake - that
should be bar.txt.1, not 2).  In shell commands:

  % munpack ~/.Mail/drafts/5
  foo.txt (text/plain)
  bar.txt (text/plain)
  % ls
  bar.txt  foo.desc  foo.txt
  % munpack ~/.Mail/drafts/5
  foo.txt.1 (text/plain)
  bar.txt.2 (text/plain)
# ^--- Should be 1!
  % ls
  bar.txt  bar.txt.2  foo.desc  foo.txt  foo.txt.1  foo.txt.desc
#  ^-- Should be 1!

... bar.txt.2 should be bar.txt.1.  It's only a small thing, but
still seems like a bug to me.

Conrad
foo
bar

Bug#294951: libtest-class-perl: Dependency on libdevel-symdump-perl not listed.

[Conrad Hughes]

Package: libtest-class-perl
Version: 0.10-1
Severity: important

It seems that libtest-class-perl depends on libdevel-symdump-perl, but
this isn't mentioned in its dependencies so you can successfully
install it yet be unable to use it as a result.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (900, 'unstable')
Architecture: i386 (i586)
Kernel: Linux 2.6.6-2-386
Locale: LANG=en_GB, LC_CTYPE=en_GB (charmap=ISO-8859-1)

Versions of packages libtest-class-perl depends on:
ii  libtest-builder-tester-perl   1.01-1 Helper testing library for Test::B
ii  libtest-differences-perl  0.47-2 Test string and data structure dif
ii  libtest-exception-perl0.20-2 Test functions for exception based
ii  perl  5.8.4-6Larry Wall's Practical Extraction 

-- no debconf information


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