[comp.windows.x] X11R4 on the 3B2 - A report from the front

aj4640@dean1.usma.edu (Robertson John MAJ) (10/08/90)

I'd like to share my experiences so far in trying to get clients up
and running on an AT&T 3B2.

The 3B2 runs System V 3.2.  While the 3B2 is NOT one of target
machines listed in the X11R4 documents, this version of the operating
system IS listed (with an AT&T 6386 as the target machine.)

I ran into the following BIG problems:

1.  The "make world" generated a flood of errors.  Some of them were
"invalid syntax" in a number of the generated makefiles.  Our
preprocessor chokes on many of the source files with messages like
"too many defines" and "no space." Furthermore a number of the source
files look for headers in a directory called /usr/include/lanman
which, as Murphy would have it, is nowhere to be found on any of our
machines.

2.  The "make install" also ran into problems.  Obvious ones were
linking failures casued by the non-existence of certain .o files.  The
preprocessor had lots of problems here, too.

3.  Surprisingly some of the clients and most of the demos were
actually installed, apparently because they only needed only one of
the libraries that actually went together.  Unfortunately, none of
these has yet been run because I can't quite figure out what magic
words to put in the Xconnections file.  Documentation for this, if it
exists, has so far eluded me.

Several calls for help were kindly answered by readers of
comp.windows.x, but no one was able to answer the critical question
regarding the contents of the Xconnections file.

As much as I'd like to get clients running on the 3B2, I'm afraid the
detective work needed to solve all the puzzles will take more time
than I have.  Nevertheless, I hate to give up on a challenge and if
anyone knows the whereabouts of the magic talisman, I'd be happy to
follow it up and report back to the net.

Jack