[Mailman-Users] really dumb question on getting started; extracting archive

[Mark Sapiro]

Holmes, Deb wrote:

>How do you unzip the .tar and .gz files prior to install?


In a command shell on your unix web server type

man gunzip

and

man tar

or

tar --help

for documentation.

Depending on your specific tar, you should be able to do

tar -xf mailman-2.1.10b1.tgz

or

gunzip -c mailman-2.1.10b1.tgz | tar -xf -



>For most
>applications that I'm running on my unix based web server, my process is to
>unzip on my Windows XP box prior to uploading to the web server.  I know
>basically nothing about unix except how to post files and write simple .cgi
>code. I'm wanting to install mailman in conjunction with a WordPress plug-in
>and am pretty much expecting that between WordPress and the WP mailman plug
>in I won't need to know very much about the unix stuff underneath.  But the
>archive format has me stumped before I even get out the gate.  


I suspect you'll find you need to know a lot more than you think.

It is not recommended to unzip/untar Mailman on Windows first. The
tarball contains subordinate tarballs for the Python email library and
the Japanese and Korean codecs. If these get expanded on Windows,
Mailman's make will probably fail.


>I searched the FAQ on "extract" and "unzip" and found nothing on point.


This is so fundamental that no one expects that someone without this
much *nix knowledge would try to install Mailman.


>I
>searched the archives for this mailing list and found 3 other people with
>the same question but none of the answers were helpful.  
>
>One person suggested http://www.gzip.org, but that web page states that the
>solution for unzipping on a windows machine is to use winzip:


As I said, don't unzip on Windows. Put the tarball on the *nix machine
and unzip and untar it there.


>Winzip says the file format of the files downloaded from SourceForge isn't
>valid.  Specifically the error when I try to extract mailman-2.1.10b1.tgz.gz
>is "invalid directory"


The file on SourceForge is 2.1.10b1.tgz. Possibly your download process
added an extranous .gz extention because of it's content-type, but it
should be named 2.1.10b1.tgz, not 2.1.10b1.tgz.gz.


>And when I try to extract mailman-2.1.10b1.tar I get "Error reading header
>after processing 0 entries"


And where did mailman-2.1.10b1.tar come from? The failed extract of
2.1.10b1.tgz.gz?  I suspect it's empty.


>Another respondent suggested the GNU installation manual , but those
>instructions go from this:
>2.2 Create the installation directory 
>
>You are now ready to configure and install the Mailman software. 
>
>To this 
>
>SECTION0013100000000000000003.1 Run configure
>
>Usually you can just cd to the directory you unpacked the Mailman source
>tarball into, and run configure with no arguments:
>
>Without ever stopping to say how to unpack the file.   
>
>I'm sure it's obvious to those with passing familiarity with unix, but if
>its that easy it couldn't hurt to enumerate the steps could it?


It's a hurdle. If you can't get past that one, I think you'll have more
issues later. It takes more than a "passing familiarity with unix" to
install software - not just Mailman - any software.

That said, we will try to help if you ask, but we probably can't teach
you *nix administration from the ground up.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan