Web Development
I have received some feedback from subscribers that suggests that the term "headless server" is not as widely known as I believed. I apologize.
Many environments, such as mainframe machines and dedicated servers, do not support a display, keyboard, or mouse. This is a headless server.
In terms of old systems, this is a true definition of
'headless', but in the Java on Unix environment,
'headless' has come to mean not having a graphical
context available. So any java.awt object cannot be
built.
Running a graphical context on a machine is a waste of
resource [Windows rant rant] and often is avoided. The
keyboard/mouse/display definition does not apply to
the concept of headless as defined by Sun in Java.
[To my knowledge]
by freddio
Headless environments *can* exist even on machines which support keyboards and monitors. For example, if you were to start a Tomcat engine via Telnet on Solaris or AIX (which may or may not have a "head"), there would NOT be a graphical context (i.e., XWindows session) and a headless approach is warranted. This also may apply to UNIX startup scripts which are typically run as root and not necessarily in a graphical context (X session).
Also, you did not mention a cleaner approach Xvfb (XWindows Virtual Frame Buffer). See below for details
http://www.xfree86.org/4.0.1/Xvfb.1.html
Hey freddio.
While I've not used Xfvb, it would be a cleaner
approach if non-java applications also needed a fake
graphical context. It comes down to whether the
definition of cleaner involves installing a Java library
[clean to me] or messing around with X systems
[traditionally a pain].
As a Java developer, having to build Xvfb from source,
or installing an eTeks jar, the cleaner solution is clearly
the Java solution.
This discussion is closed to new replies.
Sort by Trending