Hot answers tagged

3

Reset the user by moving everything in /home/faultyuser to a backup directory like this: mkdir /home/faultyuser/BACKUP mv /home/faultyuser/* /home/faultyuser/BACKUP mv /home/faultyuser/.* /home/faultyuser/BACKUP cp -v /etc/skel/.[a-z]* /home/faultyuser/ chown faultyuser:faultyuser /home/faultyuser/.* and try to login. When this works, move all you need ...


2

The most natural way would be to kill the session manager process for that session. Killing the session manager ends the session. On Linux, you can use ps xeww | grep DISPLAY=:1 to list processes that have DISPLAY=:1 in their environment. Many other Unix variants have a way to do this, but the options to ps vary. Beware that this can return false positives ...


2

I face the same issue. The magic trick is in the order of the keys. You have to: press and hold Ctrl press and release Shift release Ctrl If you release the Ctrl before Shift, or use any other order, you will face the issue. Pressing those keys together at the same time makes the order random. BTW. It doesn't depend on the WM at all. It seems to be ...


1

Putting a script like the one below in /etc/cron.hourlyshould be able to accomplish your goal. The $? grabs the exit status and if it not equal to 0 then it executes the command, otherwise it does nothing. #!/bin/bash email=user@localhost address=192.168.0.2 ping -c 3 "${address}" > /dev/null 2>&1 if [ $? -ne 0 ]; then echo "Pi is down" | ...


1

I run unit tests of my GUI application using QTestLib on a headless continuous integration server by setting up an X virtual framebuffer. Xvfb :1 & PID=$! DISPLAY=:1 make check kill $PID


1

Sorry for answering my own question, but as no other responses were published, I've decided to do it. It seems, that I have found a working solution: I start the X server with the -sharevts flag: X -config displaylink.conf -sharevts :2 I start my applications: DISPLAY=2: x-window-manager DISPLAY=2: xterm I start the vnc server: x11vnc ...


1

Sugar tries to run xdg-user-dirs-update: File "/usr/libexec/sugar-runner/xinitrc", line 77, in _setup_xdg_user_dirs subprocess.check_call("xdg-user-dirs-update") which you don't have. Install the package xdg-user-dirs which includes it. /edit: For your new error message you'll have to wait until the Sugar guys fix that, they already have an open pull ...


1

You probably meant to run: Xvfb :1 -screen 0 1024x768x16 & where the :1 is an argument on its own and means use the default network connection +1, i.e. you will need to export DISPLAY=:1 in the environment to connect to this display. Also, the X11 server may run under the name X instead of Xorg.


1

Without Xorg on server you can't initiate -X forwarding from your client. If you are interested in editing files, the easiest way is to mount the server to your local computer using sshfs(1). You can consult manual page for more info about this.


1

I am assuming you want X11 forwarding through PuTTy on a Windows operating system. Please see the answer to the following related question: http://superuser.com/questions/119792/how-to-use-x11-forwarding-with-putty


1

Most simple X11 applications just need you to set the environment variable DISPLAY, so just export DISPLAY=:0 before running the command may be enough.


1

I've just found out that this due to limitation of my hardware and not an issue with the driver. Firing up the nVIDIA proprietary, I've got the message: MetaMode 1 of Screen 0 has more than 2 active display devices. and after some googling around, I found a reference for nVidia cards older than the Kepler architecture (released March 22, 2012) that they are ...


1

Nothing is immediately obvious to me in this file. But here's a generic debugging tip which should lead you to the problem. Just below the #!/bin/sh shebang line, add the following lines: exec >~/"xinitrc-$DISPLAY.log" 2>&1 set -x set -x turns on the shell's trace mode, so every command will be printed just before it's executed. The exec line ...


1

So after installing other things to fix the drivers the crash message went away and the fix ended up being adding Modes "1280x1024" to the SubSection in the Screen section in xorg.conf


1

I don't know if that works, but instead of using xrandr --output CRT1 --mode "1280x1024_60.00" you should try xrandr -d:0 --output CRT1 --mode "1280x1024_60.00"


1

An approach that's often convenient is to mount the files, and then access them like you would access ordinary local files. For a server that you access through FTP, you can use CurlFtpFS. mkdir theserver curlftpfs theserver.example.com theserver You'll need to pass the username and password to curlftpfs, either on the command line (which is unsafe as ...


1

If you are OK with running periodically a script, you can utilize wget -N, see: http://superuser.com/questions/283481/how-do-i-properly-set-wget-to-download-only-new-files You can create a script like: cd /var/download_here/ wget -N ftp://example.com/new.txt --ftp-user="my_user" --ftp-password="my_password" And add crontab entry like: */5 * * * * ...


1

Try using the -host-cursor option. Edit: per the Xephyr man page, this option reuses the host X session's cursor. Theoretically this is not supposed to produce a significant performance improvement. In practice, I've found that Xephyr's cursor is very slow on lower-spec machines, and this option handily fixes that issue every time. It would be invoked by ...



Only top voted, non community-wiki answers of a minimum length are eligible