[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Redhat 5.0 problems/solutions
From: |
Paul Johnson |
Subject: |
Redhat 5.0 problems/solutions |
Date: |
Fri, 5 Dec 1997 21:14:58 -0600 () |
I upgraded to redhat 5, my system wouldn't start, I
had to repartition a drive and install fresh. I'm not
sure it was worth the amusement....
Anyway, I tried to post earlier today and, for reasons
uknown, my posts didn't go, so Im trying again.
Redhat 5 is based on a new C library and also many of its
features are built on tk/tcl 8.0. I tried to compile libtclobjc
against that new library and got some errors, so I installed
tk/tcl 7.6 4.2 in the /usr/local directory and tried to make
things go.
However, the rudest surprise was that the swarm package
would not make. I kept seeing errors about mistakes in stdlib.h.
After a long time, I figured out that the make was looking
first in the /swarm/include directory for header files, and
those header files are "stale." they dont work with the libraries
I have. SO, one by one, I renamed those libraries, such as
stdlib.h, math.h, one of the ieee785.h. Finally swarm made
without errors because it found better header files. Note,
I don't think you can just delete all of them because some
are found nowhere else on the pc.
I could not get applications to make, however, because of a
lot of TK/TCL/BLT errors after an archive install. I can
list the errors, perhaps they are just a result of some minor
mistake I made, not due to redhat 5.0 at all. I think the
compiler was finding the files in /usr/local, at least according to the
compiler messages, but the weird thing is the error messages follow
all the way back to the /usr/src files from which the packages
were made and installed. (Yes, when I installed libtclobjc, I
told it to use the tcl/tk in the /usr/local/lib directory, and
I think blt found them on its own).
A binary install, however, allowed programs to make and I think
it works, but I had to leave the office before I had a chance
to reconfigure the tk/tcl environment variables to point at swarm.
I'd appreciate some of the list's usual expert opinion on
what a person ought to do in this situtation. I guess I can
use the binary, even though it plays havoc on the debugger. But
I wish I understood better why the header file issue arises
and how to control the tk/tcl problems. I've read somewhere a
rather detailed treatment of how a person can keep both the
old and new version on the same machine, but I can't find it
anywhere now.
Gosh, I love linux. If you hear of anyone who needs to hire
someone who can wipe out an operating system with a single
swipe, I'm the guy.
Paul E. Johnson address@hidden
Dept. of Political Science Office: (913) 864-9086
University of Kansas FAX: (913) 864-5700
Lawrence, Kansas 66045 Home: (913) 842-9916
==================================
Swarm-Support is for discussion of the technical details of the day
to day usage of Swarm. For list administration needs (esp.
[un]subscribing), please send a message to <address@hidden>
with "help" in the body of the message.
==================================
- Popup windows ?, Sven N. Thommesen, 1997/12/04
- Redhat 5.0 problems/solutions,
Paul Johnson <=
- Re: Redhat 5.0 problems/solutions, Marcus G. Daniels, 1997/12/06
- Re: Redhat 5.0 problems/solutions, Paul E. Johnson, 1997/12/06
- Redhat 5.0 libtclobjc problems: here are the error messages, Paul E. Johnson, 1997/12/06
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Marcus G. Daniels, 1997/12/06
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Paul Johnson, 1997/12/07
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Benedikt Stefansson, 1997/12/07
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Doug Donalson;, 1997/12/07
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Marcus G. Daniels, 1997/12/07
- Re: Redhat 5.0 libtclobjc problems: here are the error messages, Paul Johnson, 1997/12/08
- Re: Redhat 5.0 problems/solutions, Marcus G. Daniels, 1997/12/06