frequent “memory has been corrupted” errors

Guest wrote on Oct 13, 2005

Hi Gabe,
I've tried the binaries here on a Fedora Core 3 system using the
same kernel version, and I get the same failure but only when the
'stacksize' is something other than 'unlimited'. I assumed from your
earlier mail that you're using bash and not tcsh? Since it happens
with everything I agree it's probably not data-related, but is there
some reason you're not using the CL? Do the tasks work when you DO
use the CL or is it the same problem? If things run in the CL but not
outside then I think that somehow the stacksize isn't being reset
properly, if the CL fails as well then is this a 64-bit version of
the OS or hardware?
Lastly, the motd header is the most recent version but you
should check that the dates on the binaries are similarly from Jul04
to rule out old binaries causing trouble. You might also try logging
in as the 'iraf' user to try the same tasks, if it works for IRAF then
something in your environment is causing it. Hope this helps.

Cheers,
-Mike

Post generated using Mail2Forum (http://www.mail2forum.com)

---

Last post on Oct 13, 2005