From: Aron Roberts (aron@socrates.berkeley.edu)
Date: Thu Sep 26 2002 - 16:35:54 PDT
Graham Patterson writes:
>The idea of running a web server on every workstation (PC or Mac) makes me
>very nervous from an administration and security viewpoint.
True. However, from a security standpoint alone, as the W3C's "World
Wide Web Security FAQ" <http://www.w3.org/Security/Faq/wwwsf1.html> noted:
"The safest Web site is a bare-bones Macintosh running [the
Classic Mac OS and] a bare-bones Web server", such as WebSTAR.
Even WebSTAR, however, can potentially be vulnerable, depending on
how the server's own features are configured and which third-party
server add-ons are installed:
http://www.w3.org/Security/Faq/wwwsf3.html#Q20
Graham also noted, to Ilan:
>I'm surprised that the network is down with the workstations still running
>often enough to make this necessary, but you know your environment.
This is an important point, and thanks, Graham, for raising it. If
the VLSB network(s) used by the Bioscience & Natural Resource Library
are offline more than on rare occasions, this issue should be brought
to the attention of Communication and Network Services (CNS) via the
Trouble Desk <http://www.net.berkeley.edu/trouble/>, who will take
prompt action to investigate this problem.
As a gratuitous aside, another common purpose for running full-fledged
Web servers on hosts which are disconnected from the Internet is as a
backup for presentations or demonstrations, or as a means of working or
presenting at locations where no network connectivity is available.
Once again, running a local Web server is only necessary if a
Web site depends on certain features that cannot be replicated by
simply walking the tree of source files -- or of server-generated Web
pages -- and copying these to local disk storage for viewing by a
browser.
Ford Chiang wrote:
>Another possibility is to get a copy of httrack (it's free). using it, you
>can download a copy of your web site. As long as it's mostly static web
>pages it works great (it'll download all the files off the site). ...
>all the links are adjusted so you can browse from the hard drive or CD.
httrack looks terrific, thanks for telling us about this, Ford!
For the Mac OS, Web Devil, a shareware program with an lurid logo
befitting its name :-), can perform much the same type of task:
http://www.chaoticsoftware.com/ProductPages/WebDevil.html
Another possibility is 'curl', a command line tool which can retrieve
content at arbitrary URLs via HTTP and many other protocols, as well.
There are some tools based on curl (and/or its associated programming
library, libcurl) which can be used to mirror generated content on Web
sites:
http://curl.haxx.se/curlprograms.html
As its Web site's home page notes, "Curl comes bundled with a large
variety of Linux, *BSD and Mac OS X distributions."
Aron Roberts
Workstation Software Support Group
-----------------------------------------------------------------------
The following was automatically added to this message by the list server:
Webnet information is available at <URL:http://webnet.berkeley.edu/>.
This archive was generated by hypermail 2b29 : Thu Sep 26 2002 - 16:37:35 PDT