Bug#801487: introducing xserver-xorg-legacy without telling anybody?

[Harald Dunkel]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Package: xorg-server
Version: 2:1.17.2-3

Please either depend or recommend xserver-xorg-legacy. It
is *extremely* painful if you login next morning after the
upgrade and xinit doesn't work anymore and you have no web
interface to find out WTH has been changed.

Sorry to say, but this was *highly* annoying. Please don't
break things by default.


Thanx
Harri
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJWGgKaAAoJEAqeKp5m04HLRHkH/AsZ+CNc6FjbdJKKn1iwBvVc
cLVNpE1bZlx+3JZ0WJSa+ZgajdAZUMTqcRfNkjfe606Cvs1Xnc4VrL7lpwV1dUN5
K+TUs/U1M7QemVMJxVBPdzMd7ouKKmac0BeCfeZa3H+wNNm0s5uQbSMJJAojTIXk
5p7hVzLPworLx1zF9Md+t2R2moFvIMC9BqZODZTIqWn5eQQ3+xy2aXtFDDMf1vm8
c8Xim3vAjBmfwa2ZSR42moNeqdwboLJDkRauRmI44gpewHiBOadGGUsZkI4pp/03
+uFgpcEJ5Ui37tF1ct7OqK6vHRy5NT/Rja4WgWX1t0lI0v0w7YUekTU3LTWit9w=
=B3wl
-END PGP SIGNATURE-

Bug#801401: cannot start X from the console command line

[Julien Cristau]

On Fri, Oct  9, 2015 at 18:26:01 +0200, Giuseppe Bilotta wrote:

> Package: xserver-xorg
> Version: 1:7.7+12
> Severity: important
> 
> I normally boot to console and then manually launch X if/when I need it.
> With the latest update to Xorg, trying to start X fails with the error
> 
>  (EE) xf86OpenConsole: Cannot open /dev/tty0 (No such file or directory)
> 
> Interestingly, the error is about /dev/tty0 regardless of whether I try
> to start it from the first VT or from a different one (see attached
> logs).
> 
How exactly are you starting X?  'startx' is supposed to do the right
thing.

Cheers,
Julien


signature.asc
Description: PGP signature

Bug#801518: xserver-xorg: Display freezes with two X servers, switching TTYs no longer possible

[Simon Ruderich]

Package: xserver-xorg
Version: 1:7.7+12
Severity: important

Hello,

I use the following commands to start two X servers (the
`runuser` command is only used to spawn a new systemd session):

runuser --login --command='/usr/bin/startx -- :0 -nolisten tcp -novtswitch 
vt7' user1
runuser --login --command='/usr/bin/startx -- :1 -nolisten tcp -novtswitch 
vt8' user2

This worked fine in 1:7.7+9, but since 1:7.7+12 this setup causes
a freeze once I switch from the first X on vt7 to vt8. Now I
can't switch back to either vt7 or any other TTY and also can't
enter anything in the TTY. The system isn't locked up though,
e.g. sound is still working. The logs contain no additional
information (the log below is from 1:7.7+9 though), except the
following (which is always displayed when switching TTYs also in
the working version).

[ 19275.504] (II) AIGLX: Suspending AIGLX clients for VT switch
[ 19281.595] (II) AIGLX: Resuming AIGLX clients after VT switch

I've tried it with the compat wrapper which uses a setuid binary.
I've no idea how to debug this issue. Some pointers would be very
appreciated.

Regards
Simon

-- Package-specific info:
VGA-compatible devices on PCI bus:
--
00:02.0 VGA compatible controller [0300]: Intel Corporation Broadwell-U 
Integrated Graphics [8086:1616] (rev 09)

Xorg X server configuration file status:

-rw-r--r-- 1 root root 240 Oct  4 04:16 /etc/X11/xorg.conf

Contents of /etc/X11/xorg.conf:
---
# Try to prevent tearing for Intel graphics.
Section "Device"
Identifier  "Intel Graphics"
Driver  "intel"
Option  "AccelMethod"   "sna"
Option  "TearFree"  "true"
EndSection

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

/etc/modprobe.d contains no KMS configuration files.

Kernel version (/proc/version):
---
Linux version 4.2.0-1-amd64 (debian-ker...@lists.debian.org) (gcc version 4.9.3 
(Debian 4.9.3-4) ) #1 SMP Debian 4.2.3-1 (2015-10-06)

Contents of most recent Xorg X server log file (/var/log/Xorg.1.log):
-
[   246.944] 
X.Org X Server 1.17.1
Release Date: 2015-02-10
[   246.944] X Protocol Version 11, Revision 0
[   246.944] Build Operating System: Linux 3.16.0-4-amd64 x86_64 Debian
[   246.944] Current Operating System: Linux blood-stain-child 4.2.0-1-amd64 #1 
SMP Debian 4.2.3-1 (2015-10-06) x86_64
[   246.944] Kernel command line: BOOT_IMAGE=/vmlinuz-4.2.0-1-amd64 
root=/dev/mapper/blood--stain--child-root ro intel_iommu=on,igfx_off
[   246.944] Build Date: 04 May 2015  11:22:06PM
[   246.944] xorg-server 2:1.17.1-2 (http://www.debian.org/support) 
[   246.944] Current version of pixman: 0.33.2
[   246.944]Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
[   246.944] Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   246.944] (==) Log file: "/var/log/Xorg.1.log", Time: Fri Oct  9 23:04:44 
2015
[   246.944] (==) Using config file: "/etc/X11/xorg.conf"
[   246.944] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[   246.944] (==) No Layout section.  Using the first Screen section.
[   246.944] (==) No screen section available. Using defaults.
[   246.944] (**) |-->Screen "Default Screen Section" (0)
[   246.944] (**) |   |-->Monitor ""
[   246.944] (==) No device specified for screen "Default Screen Section".
Using the first device section listed.
[   246.944] (**) |   |-->Device "Intel Graphics"
[   246.944] (==) No monitor specified for screen "Default Screen Section".
Using a default monitor configuration.
[   246.944] (==) Automatically adding devices
[   246.944] (==) Automatically enabling devices
[   246.944] (==) Automatically adding GPU devices
[   246.944] (WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
[   246.944]Entry deleted from font path.
[   246.944] (==) 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
[   246.944] (==) ModulePath set to "/usr/lib/xorg/modules"
[   246.944] (II) The server relies on udev to provide the list of input 
devices.
If no devices become available, reconfigure udev or disable 
AutoAddDevices.
[   246.944] (II) Loader magic: 0x55a80ebf3d80
[   246.944] (II) Module ABI versions:
[   246.944]X.Org ANSI C Emulation: 0.4
[   246.944]X.Org Video Driver: 19.0
[   246.944]X.Org XInput driver : 21.0
[   246.944]X.Org Server Extension : 9.0
[   246.945] (II) xfree86: Adding drm device (/dev/dri/card0)
[   246.945] 

Processed: reassign 801348 to xserver-xorg-video-intel

[Debian Bug Tracking System]

Processing commands for cont...@bugs.debian.org:

> # could also be a kernel issue
> reassign 801348 xserver-xorg-video-intel
Bug #801348 [xserver-xorg] xserver-xorg: X server controls wrong backlight 
device
Bug reassigned from package 'xserver-xorg' to 'xserver-xorg-video-intel'.
No longer marked as found in versions xorg/1:7.7+7.
Ignoring request to alter fixed versions of bug #801348 to the same values 
previously set
> thanks
Stopping processing here.

Please contact me if you need assistance.
-- 
801348: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=801348
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#786873: More info on Chrome-graphics

[Andreas Glaeser]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


see a new report there:

801...@bugs.debian.org
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAlYad84ACgkQ5+rBHyUt5wtgZwCcDitk0YeGr5sYrldIJEupWrnm
luAAoKNtre1wLJ/irjJ8KcSQMUZcylyP
=yECd
-END PGP SIGNATURE-

Bug#801487: introducing xserver-xorg-legacy without telling anybody?

[Julien Cristau]

On Sun, Oct 11, 2015 at 08:33:03 +0200, Harald Dunkel wrote:

> Package: xorg-server
> Version: 2:1.17.2-3
> 
> Please either depend or recommend xserver-xorg-legacy. It
> is *extremely* painful if you login next morning after the
> upgrade and xinit doesn't work anymore and you have no web
> interface to find out WTH has been changed.
> 
> Sorry to say, but this was *highly* annoying. Please don't
> break things by default.
> 
We didn't.  Login managers should still work, and startx should still
work.  But you didn't provide any details as to what broke...

Cheers,
Julien


signature.asc
Description: PGP signature

Processed: tagging 801487

[Debian Bug Tracking System]

Processing commands for cont...@bugs.debian.org:

> tags 801487 + moreinfo
Bug #801487 [xorg-server] introducing xserver-xorg-legacy without telling 
anybody?
Added tag(s) moreinfo.
> thanks
Stopping processing here.

Please contact me if you need assistance.
-- 
801487: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=801487
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#798097: Restarting logind kills Xserver

[Michael Biebl]

Control: severity -1 important

On Sat, 05 Sep 2015 17:40:32 +0200 Michael Biebl  wrote:
> Package: xserver-xorg-core
> Version: 2:1.17.2-2
> Severity: serious
> 
> I installed gdm3 from experimental which brought xserver-xorg-core from
> experimental along with it.
> This version has support for logind enabled, which apparently is
> required to run gdm successfully.
> 
> There is one particular issue though: Atm, we do restart systemd-logind
> on upgrades in systemd.postinst. This kills your running X session
> though, due to [1].
> 
> I'm not yet sure, how to address this, i.e. if we can revert this Xorg
> patch and find a solution in X or stop restarting systemd-logind as part
> of the upgrade process.
> 
> I've decided to make this bug RC, since we need to fix that one way or
> another.

We dropped the logind restart in systemd_226-4 for now and the
xserver-xorg-core package gained a  Breaks against older systemd versions.

This is not a real fix yet, but it should be sufficient to not be RC.

Thus downgrading to important. We still should change both logind and
Xorg so logind can be restarted safely without wreaking havoc.

-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?



signature.asc
Description: OpenPGP digital signature

Processed: Re: Restarting logind kills Xserver

[Debian Bug Tracking System]

Processing control commands:

> severity -1 important
Bug #798097 [xserver-xorg-core] Restarting logind kills Xserver
Severity set to 'important' from 'serious'

-- 
798097: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=798097
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#801518: xserver-xorg: Display freezes with two X servers, switching TTYs no longer possible

[Simon Ruderich]

On Sun, Oct 11, 2015 at 07:56:29PM +0200, Julien Cristau wrote:
> I guess you probably need to configure the compat wrapper to not drop
> root privileges, if you want X to be allowed to switch VTs.

I did use needs_root_rights=yes option in Xwrapper.config. Do I
need anything else?

How would this work without the compatibility wrapper and thus
without root permissions?

Is it expected that I can't enter anything after the (failed) the
TTY switch?

Regards
Simon
-- 
+ privacy is necessary
+ using gnupg http://gnupg.org
+ public key id: 0x92FEFDB7E44C32F9


signature.asc
Description: PGP signature

Bug#801524: Seg Fault - failure loading gnome in virtualbox VMs

[jnqnfe]

Control: reassign -1 virtualbox-guest-x11 5.0.6-dfsg-1
Control: retitle -1 Dependency on new pkg xserver-xorg-legacy needed
Control: severity -1 grave


On Sun, 2015-10-11 at 19:57 +0200, Julien Cristau wrote:
> On Sun, Oct 11, 2015 at 18:13:31 +0100, jnqnfe wrote:
> 
> > Package: xserver-xorg
> > Version: 1:7.7+12
> > Priority: Important
> > 
> > Since installing updates last night in some VMs, I can no longer
> > load
> > gnome in those VMs.
> > 
> > They load to a terminal login (normally would auto-load gnome),
> > which
> > is flashing on and off the screen rapidly, with text input to login
> > being very difficult/impossible. This  flashing stops after about a
> > minute, leaving you at a stable/usable login prompt. After login,
> > startx fails to load gnome. The xorg log (below) lists a segfault.
> > 
> > Edit: Actually, while retrieving copies of the log files, I ran
> > startx
> > as root, which worked fine, it just doesn't load as the normal
> > user.
> > 
> try installing xserver-xorg-legacy.
> 
> Cheers,
> Julien

Yes, that solves it. Thanks. Moving to virtualbox...

Increasing severity partly due to concern that flashing login prompt
may negatively affect people with epilepsy.

Processed: Re: Bug#801524: Seg Fault - failure loading gnome in virtualbox VMs

[Debian Bug Tracking System]

Processing control commands:

> reassign -1 virtualbox-guest-x11 5.0.6-dfsg-1
Bug #801524 [xserver-xorg] Seg Fault - failure loading gnome in virtualbox VMs
Bug reassigned from package 'xserver-xorg' to 'virtualbox-guest-x11'.
No longer marked as found in versions xorg/1:7.7+12.
Ignoring request to alter fixed versions of bug #801524 to the same values 
previously set
Bug #801524 [virtualbox-guest-x11] Seg Fault - failure loading gnome in 
virtualbox VMs
Marked as found in versions virtualbox/5.0.6-dfsg-1.
> retitle -1 Dependency on new pkg xserver-xorg-legacy needed
Bug #801524 [virtualbox-guest-x11] Seg Fault - failure loading gnome in 
virtualbox VMs
Changed Bug title to 'Dependency on new pkg xserver-xorg-legacy needed' from 
'Seg Fault - failure loading gnome in virtualbox VMs'
> severity -1 grave
Bug #801524 [virtualbox-guest-x11] Dependency on new pkg xserver-xorg-legacy 
needed
Severity set to 'grave' from 'important'

-- 
801524: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=801524
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#801524: Seg Fault - failure loading gnome in virtualbox VMs

[jnqnfe]

Package: xserver-xorg
Version: 1:7.7+12
Priority: Important

Since installing updates last night in some VMs, I can no longer load
gnome in those VMs.

They load to a terminal login (normally would auto-load gnome), which
is flashing on and off the screen rapidly, with text input to login
being very difficult/impossible. This  flashing stops after about a
minute, leaving you at a stable/usable login prompt. After login,
startx fails to load gnome. The xorg log (below) lists a segfault.

Edit: Actually, while retrieving copies of the log files, I ran startx
as root, which worked fine, it just doesn't load as the normal user.

VMs are sid installs, fully up to date, with a handful of core gnome
packages from experimental (gdm3, gnome-shell, gnome-session, mutter,
gnome-setting-daemon, nautilus).

Things were working fine the previous day, prior to the latest updates.
The updates installed resulting in the breakage are as follows:

Install: libsctp1:amd64 (1.0.16+dfsg-3, automatic)
Upgrade: libpython2.7-stdlib:amd64 (2.7.10-4, 2.7.10-5), libpangomm-
1.4-1v5:amd64 (2.36.0-2+b1, 2.38.1-1), librsvg2-2:amd64 (2.40.10-1,
2.40.11-1), os-prober:amd64 (1.67, 1.68), xorg:amd64 (7.7+9, 7.7+12),
xserver-xorg-input-all:amd64 (7.7+9, 7.7+12), xwayland:amd64 (1.17.2-
1.1, 1.17.2-3), libpam-systemd:amd64 (226-4, 227-2), python2.7:amd64
(2.7.10-4, 2.7.10-5), xserver-xorg-core:amd64 (1.17.2-1.1, 1.17.2-3),
openjdk-7-jdk:amd64 (7u85-2.6.1-4, 7u85-2.6.1-5), udev:amd64 (226-4,
227-2), gir1.2-gcr-3:amd64 (3.16.0-1, 3.18.0-1), xserver-xorg-video-
all:amd64 (7.7+9, 7.7+12), xserver-common:amd64 (1.17.2-1.1, 1.17.2-3), 
w3m:amd64 (0.5.3-24, 0.5.3-25), librsvg2-bin:amd64 (2.40.10-1, 2.40.11-
1), baobab:amd64 (3.18.0-1, 3.18.1-1), openjdk-7-jre-headless:amd64
(7u85-2.6.1-4, 7u85-2.6.1-5), libudev1:amd64 (226-4, 227-2),
binutils:amd64 (2.25.1-5, 2.25.1-6), libmm-glib0:amd64 (1.4.10-1,
1.4.12-1), gcr:amd64 (3.16.0-1, 3.18.0-1), libpython2.7:amd64 (2.7.10-
4, 2.7.10-5), librsvg2-common:amd64 (2.40.10-1, 2.40.11-1), xserver-
xephyr:amd64 (1.17.2-1.1, 1.17.2-3), libgck-1-0:amd64 (3.16.0-1,
3.18.0-1), libencode-locale-perl:amd64 (1.03-1, 1.05-1), systemd-
sysv:amd64 (226-4, 227-2), libcairomm-1.0-1v5:amd64 (1.10.0-1.2,
1.12.0-1), openjdk-7-jre:amd64 (7u85-2.6.1-4, 7u85-2.6.1-5),
modemmanager:amd64 (1.4.10-1, 1.4.12-1), systemd:amd64 (226-4, 227-2),
libjetty8-java:amd64 (8.1.17-2, 8.1.18-1), libsigc++-2.0-0v5:amd64
(2.6.1-1, 2.6.1-2), python2.7-minimal:amd64 (2.7.10-4, 2.7.10-5),
xserver-xorg:amd64 (7.7+9, 7.7+12), libgtkmm-3.0-1v5:amd64 (3.16.0-
2+b1, 3.18.0-1), libnss-myhostname:amd64 (226-4, 227-2), gir1.2-gck-
1:amd64 (3.16.0-1, 3.18.0-1), libsystemd0:amd64 (226-4, 227-2), x11-
common:amd64 (7.7+9, 7.7+12), libgcr-base-3-1:amd64 (3.16.0-1, 3.18.0-
1), libpython2.7-minimal:amd64 (2.7.10-4, 2.7.10-5), libgcr-3-
common:amd64 (3.16.0-1, 3.18.0-1), libgcr-ui-3-1:amd64 (3.16.0-1,
3.18.0-1)

Upgrade: gir1.2-gdm3:amd64 (3.14.2-2, 3.17.92-1), gdm3:amd64 (3.14.2-2, 
3.17.92-1), libgdm1:amd64 (3.14.2-2, 3.17.92-1)

Edit: Just FYI, the latest updates from a few moments ago, updating
mutter and gdm3 to 3.18 stable versions along with a few other updates
made no difference.

Xorg.0.log file:

[   546.135] 
X.Org X Server 1.17.2
Release Date: 2015-06-16
[   546.136] X Protocol Version 11, Revision 0
[   546.137] Build Operating System: Linux 4.2.0-1-amd64 x86_64 Debian
[   546.138] Current Operating System: Linux debian 4.2.0-1-amd64 #1
SMP Debian 4.2.3-1 (2015-10-06) x86_64
[   546.138] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.2.0-1-
amd64 root=UUID=b58dcb99-f1c3-4e69-9817-8375e226945d ro quiet
[   546.139] Build Date: 06 October 2015  07:27:47AM
[   546.139] xorg-server 2:1.17.2-3 (http://www.debian.org/support) 
[   546.139] Current version of pixman: 0.33.2
[   546.140]    Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
[   546.140] Markers: (--) probed, (**) from config file, (==) default
setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   546.143] (==) Log file: "/home/test/.local/share/xorg/Xorg.0.log",
Time: Sun Oct 11 18:00:34 2015
[   546.143] (==) Using system config directory
"/usr/share/X11/xorg.conf.d"
[   546.144] (==) No Layout section.  Using the first Screen section.
[   546.144] (==) No screen section available. Using defaults.
[   546.144] (**) |-->Screen "Default Screen Section" (0)
[   546.144] (**) |   |-->Monitor ""
[   546.144] (==) No monitor specified for screen "Default Screen
Section".
Using a default monitor configuration.
[   546.144] (==) Automatically adding devices
[   546.144] (==) Automatically enabling devices
[   546.144] (==) Automatically adding GPU devices
[   546.144] (WW) The directory "/usr/share/fonts/X11/cyrillic" does
not exist.
[   546.144]    Entry deleted from font path.
[   546.144] (==) FontPath set to:

Bug#801529: xinit: startx fails when output is redirected

[Akkana Peck]

Package: xinit
Version: 1.3.4-1
Severity: normal

Dear Maintainer,

As of my most recent sid upgrade, startx stopped working.
The problem turned out to be that it no longer supports redirection.

I was logging in to tty1 and running this in my .zlogin:

startx >& $HOME/.xsession-errors

As of the upgrade to the "run as user" X server, X fails to start.
It pauses for 5-10 seconds, then exits, leaving the terminal
settings wrong so I have to run reset.

.local/share/xorg/Xorg.0.log says:

[  1000.714] (++) using VT number 1

[  1000.714] (EE) 
Fatal server error:
[  1000.714] (EE) xf86OpenConsole: VT_ACTIVATE failed: Operation not permitted
[  1000.714] (EE) 

Running simply startx, without the redirect, works. But of course I
can no longer check output from programs run inside X.

Some comments on the web suggest that startx -- -keeptty will allow
redirection, but man xinit warns:
This  option is only useful when debugging the server.
Not all platforms support (or can use) this option.
so it seems like that might not be a good long-term option.

Redhat claims to have a fix for the redirection problem:
https://bugzilla.redhat.com/show_bug.cgi?id=1177513#c5
I can't verify whether their fix actually fixes the issue (and I'm
not sure how to find the actual code change involved in the fix,
or whether it was pushed upstream, though I'm guessing not since
that was back in January so Debian hopefully would have it by now).

I haven't been able to find any other way besides redirection to get
stdout and stderr for programs run inside X. Is there a new and
improved way that would make redirection unnecessary?


-- System Information:
Debian Release: stretch/sid
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)

Kernel: Linux 3.14-2-686-pae (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages xinit depends on:
ii  coreutils   8.23-4
ii  libc6   2.19-22
ii  libx11-62:1.6.3-1
ii  x11-common  1:7.7+12
ii  xauth   1:1.0.9-1

Versions of packages xinit recommends:
ii  gnome-terminal [x-terminal-emulator]  3.18.0-1
ii  openbox [x-window-manager]3.6.1-1
ii  xserver-xorg [xserver]1:7.7+12
ii  xterm [x-terminal-emulator]   320-1

xinit suggests no packages.

-- no debconf information

Bug#801518: xserver-xorg: Display freezes with two X servers, switching TTYs no longer possible

[Julien Cristau]

On Sun, Oct 11, 2015 at 17:57:12 +0200, Simon Ruderich wrote:

> Package: xserver-xorg
> Version: 1:7.7+12
> Severity: important
> 
> Hello,
> 
> I use the following commands to start two X servers (the
> `runuser` command is only used to spawn a new systemd session):
> 
> runuser --login --command='/usr/bin/startx -- :0 -nolisten tcp 
> -novtswitch vt7' user1
> runuser --login --command='/usr/bin/startx -- :1 -nolisten tcp 
> -novtswitch vt8' user2
> 
> This worked fine in 1:7.7+9, but since 1:7.7+12 this setup causes
> a freeze once I switch from the first X on vt7 to vt8. Now I
> can't switch back to either vt7 or any other TTY and also can't
> enter anything in the TTY. The system isn't locked up though,
> e.g. sound is still working. The logs contain no additional
> information (the log below is from 1:7.7+9 though), except the
> following (which is always displayed when switching TTYs also in
> the working version).
> 
> [ 19275.504] (II) AIGLX: Suspending AIGLX clients for VT switch
> [ 19281.595] (II) AIGLX: Resuming AIGLX clients after VT switch
> 
> I've tried it with the compat wrapper which uses a setuid binary.
> I've no idea how to debug this issue. Some pointers would be very
> appreciated.
> 
I guess you probably need to configure the compat wrapper to not drop
root privileges, if you want X to be allowed to switch VTs.

Cheers,
Julien


signature.asc
Description: PGP signature

Bug#801524: Seg Fault - failure loading gnome in virtualbox VMs

[Julien Cristau]

On Sun, Oct 11, 2015 at 18:13:31 +0100, jnqnfe wrote:

> Package: xserver-xorg
> Version: 1:7.7+12
> Priority: Important
> 
> Since installing updates last night in some VMs, I can no longer load
> gnome in those VMs.
> 
> They load to a terminal login (normally would auto-load gnome), which
> is flashing on and off the screen rapidly, with text input to login
> being very difficult/impossible. This  flashing stops after about a
> minute, leaving you at a stable/usable login prompt. After login,
> startx fails to load gnome. The xorg log (below) lists a segfault.
> 
> Edit: Actually, while retrieving copies of the log files, I ran startx
> as root, which worked fine, it just doesn't load as the normal user.
> 
try installing xserver-xorg-legacy.

Cheers,
Julien


signature.asc
Description: PGP signature

Bug#801401: cannot start X from the console command line

[Kelly Clowers]

I just hit this bug as well.

I have a bash script that sets the WM I choose (from when I was trying
different ones - now I always choose "awesome"), and then does "exec
xinit". Ultimately xinitrc is running the command "awesome" which
starts Awesome WM. No idea what Awesome does internally, but it has
always worked.

The /dev/tty* files all exist, but while /dev/tty has perms of
rw-rw-rw-, /dev/tty1 has rw--- and the others including tty0 have
rw--w

I have no idea if that is normal. I find if I chmod them to 666, I no
longer get the unable to open tty0 message, but get a blank screen
that accepts no input. I have to use the reset button for soft
restart, and then examining the logs show a message when it tries to
modeset "Permission denied [13]"

Both udev and xorg were upgraded at the same time, I first downgraded
1.17.2-3 to 1.17.2-2 (and 7.7+12 to 7.7+11), the problem persisted. So
I downgraded udev from 227 to 226 (and I know for udev that 226 is
what I was on before), but the problem persisted.

I don't know what version of xorg I had before (although I shouldn't
be that far behind), so I will try going back further next. Although
snapshot.debian.org says that 1.17.2-2 was seen on 2015-08-21, So that
really should have been the one I was on, no way I was that out of
date...

Any advice is appreciated, if there are any tests I can do, let me know...

Kelly Clowers

Bug#801401: cannot start X from the console command line

[Bill Brelsford]

I have the same problem, since upgrading from 1:7.7+9 to 1:7.7+12.
I also start X through a script that does "exec xinit".

Bill

Bug#801401: cannot start X from the console command line

[Kelly Clowers]

So, whatever TTY you are logged in on gets 600 perms, while the others
get 620 (write on group).

If I manually change that TTY from 600 to 620, that clears out the
(EE) xf86OpenConsole: Cannot open /dev/ttyX error

But it doesn't let it run (from xinit or startx). However it does run
from root. Cranking xorg log verbosity up and comparing, the only
relevant error seems to be "xf86EnableIOPorts: failed to set IOPL for
I/O (Operation not permitted)"

I think these permission errors are stemming from this Xorg not
running as root (it is not suid, and /usr/bin/X is no longer a suid
wrapper to Xorg, but a simple symlink to it). To not run as root, it
makes use of logind. I don't know if this bug then is truly on Xorg,
or if it is on the logind side of the fence or what. So far I haven't
found enough info to run tests on what logind is doing.


Thanks,
Kelly Clowers