Very Computer

I already have:

# ls -log /usr/lib/libc.so.5
lrwxrwxrwx 1 root 14 Dec  7 17:57 /usr/lib/libc.so.5 -> libc.so.5.4.46
# ls -log /usr/lib/libc.so.5.4.46
-rwxr-xr-x    1 root      1839459 Dec  7 17:56 /usr/lib/libc.so.5.4.46
also:
# pwd
/usr/i486-linux-libc5/lib
# ls -l libc.so*
lrwxrwxrwx 1 root 14 Jan 27  1999 libc.so.5 -> libc.so.5.3.12
-rwxr-xr-x    1 root     root       699832 Apr 15  1999 libc.so.5.3.12

so I linked the /usr/lib/libc.so.5 to the
/usr/i486-linux-libc5/lib/libc.so.5.3.12, to see if the older version
was required, ran wingz, still got a core dump....

willing to try most anything...

--

Your original post mentioned libc.so.6, but the ldd output shows both
libc.so.5 and libc.so.6. That seems a bit odd to me, but I won't claim
deep understanding of how/whether different libc.so versions could work
together. Have you verified the glibc-2.* rpm installation?

I just did a google search on '__getopt_clean_environment', and there
are a number of hits for bug reports on problems similar to yours. One
makes it look like it's a very bad idea to mix shared libraries
compiled with libc5 with others compiled with libc6. I suspect (but
have no way of confirming) that this may be the cause of the problems
you're having. Certainly what I read in these bug reports makes it seem
like both libc.so.5 and libc.so.6 dependency in one program is a bad
thing.

Are you working with a clean RH6.2 machine? Or was it upgraded from a
previous RH install? I'm fishing for possibilities. I don't have any
concrete solutions to offer you, other than to wonder if you might be
better off with a different spreadsheet program.

Sent via Deja.com
http://www.deja.com/