| xautolock(1) | General Commands Manual | xautolock(1) |
xautolock - fire up programs in case of user inactivity under X
This man page applies to xautolock version 2.2.
Xautolock monitors the user activity on an X Window display. If none is detected within mins minutes, a program is started as specified by the -locker option. Xautolock will typically be used to lock the screen (hence its primary name) but it really doesn't care what program you make it start. For this reason, xautolock does not interfere with the default X screen saver, unless the -resetsaver option is used. This implies that it is the job of the locker or the user to take the appropriate actions if the default screen saver is to be disabled. The only real assumption made by xautolock is that a new countdown starts as soon as the locker exits.
In the presence of the -notify option, a warning signal will be issued margin seconds before starting the locker. Warning signals come in two kinds:
You can tell xautolock to take special actions when you move the mouse into one of the corners of the display and leave it there, by using the -corners, -cornerdelay, -cornerredelay and -cornersize options. This works as follows:
The xxxx argument to the -corners option must consist of exactly 4 characters from the following set: '0', '+', '-'. Each one of these specifies what xautolock should do when the mouse enters a small square area located in each of the corners of the screen. The corners are considered in the following order: top left, top right, bottom left, bottom right. A '0' indicates that xautolock should ignore the corner. A '+' indicates that xautolock should start the locker after secs or altsecs seconds (see below for the difference between both), unless the mouse is moved or keyboard input is received. A '-' indicates that xautolock should not start the locker at all. The pixels argument specifies the size in pixels of the corner areas.
Most users of the -corners option want the locker to activate within a very short time interval after they move the mouse into a '+' corner. This can be achieved by specifying a small value for the -cornerdelay option. However, if the mouse is subsequently left where it is, xautolock will almost immediately start a new locker right after the user quits the current one. To prevent this from happening, the -cornerredelay option can be used to specify the time-out interval to be used if and only if the mouse is sitting in a `+' corner and has not been moved since the previous locker exited.
A running xautolock process can be disabled (unless if the -secure option has been specified), in which case it will not attempt to start the locker. To disable an already running xautolock process, use the -disable option. To re-enable it, use -enable. To toggle it between both states, use -toggle. Using this method is preferable to using sending it SIGSTOP and SIGCONT signals, because while disabled xautolock will still be emptying its event queue.
A running xautolock process can also be told to exit (unless if the -secure option has been specified). To do this, use the -exit option.
The -killtime and -killer options allow, amongst other things, to implement an additional automatic logout, on top of the automatic screen locking. In the presence of one or both of these options, a secondary timeout will be triggered killmins after starting the locker (unless user activity is detected in the mean time). Upon expiration of this secondary timer, the killer program is run. Note that, despite the name of the options, xautolock really doesn't care what the killer does in reality. If it doesn't (indirectly) cause xautolock to get killed, and assuming that no user activity is detected, the secondary trigger will periodically expire every killmins minutes for as long as the locker runs.
In combination with -killtime and -killer, the -secure option allows system administrators to enforce xautolock as a part of their security procedures, and to prevent people from locking shared displays for an excessive amount of time. One way to achieve this is to start xautolock (using -secure and optionally -killtime and -killer) from within XDM's Xsession file in such a way that the session automatically ends if xautolock itself is killed.
By default xautolock closes stdout and stderr. This prevents the locker from writing error messages to these files in case you manually lock your display. The -nocloseout, -nocloseerr and -noclose options cause xautolock to not close stdout and/or stderr. On some platforms users of xnlock will need to use -nocloseout, in order to make xnlock's witty sayings show up. These options can also be used for debugging cases in which locker invocation is not successful.
Xautolock is capable of managing multi-headed displays.
See the xset man page for more information about managing the X screen saver.
Resources can be specified in your ~/.Xresources or ~/.Xdefaults file (whichever your system uses) and merged via the xrdb(1) command. They can be specified either for class Xautolock, or for whatever name your xautolock program has been given. This can be useful in case xautolock is to be used for other purposes than simply locking the screen. For example: if you have two copies of xautolock, one called "xmonitor", and one called "xlogout", then both will honour the following:
In addition, "xmonitor" will honour:
while "xlogout" will honour:
Each command line option takes precedence over the corresponding (default) resource specification.
The -disable, -enable, -toggle, -exit, -locknow, -unlocknow, and -restart options depend on access to the X server to do their work. This implies that they will be suspended in case some other application has grabbed the server all for itself.
If, when creating a window, an application waits for more than 30 seconds before selecting KeyPress events on non-leaf windows, xautolock may interfere with the event propagation mechanism. This effect is theoretical and has never been observed in real life. It can only occur in case xautolock has been compiled without support for both the Xidle and the MIT ScreenSaver extensions, or in case the X server does not support these extensions.
xautolock does not always properly handle the secure keyboard mode of terminal emulators like xterm, since that mode will prevent xautolock from noticing the keyboard events occurring on the terminal. Therefore, xautolock sometimes thinks that there is no keyboard activity while in reality there is. This can only occur in case xautolock has been compiled without support for both the Xidle and the MIT ScreenSaver extensions, or in case the X server does not support these extensions.
xautolock does not check whether notifier and/or locker are available.
The xautolock resources have dummy resource classes.
Copyright 1990, 1992-1999, 2001-2002, 2004, 2007 by Stefan De Troch and Michel Eyckmans.
Versions 2.0 and above of xautolock are available under version 2 of the GNU GPL. Earlier versions are available under other conditions. For more information, see the License file.
Xautolock was conceived, written, and performed by:
Michel Eyckmans (MCE)
Stefan De Troch
Please send queries for help, feature suggestions, bug reports, etc. to mce@scarlet.be.
Kris Croes
| December 28, 2007 |