> Whenever I am running Ethereal (Linux), I cannot read ethereal trace
> file larger thant 3000 packets or so. Ethereal crashes with a
> Seqmentation fault. Is there a know issue with this?
No, it's not - for a capture file that's too large, it would probably
crash with an error inside the GLib memory allocation routines, which
panic if an attempt to allocate memory fails. However, "too large" is
likely to be larger than 3000 packets, I suspect.
However, the problem doesn't necessarily have anything to do with the
*size* of the capture file; it may have to do with particular packets in
the capture - there might, for example, be a device on your network
periodically transmitting packets that trigger a bug in one of
Ethereal's dissectors, and a small trace might be the result of a
capture that didn't run long enough to catch one of those packets while
a larger trace might be the result of a capture that *did* run long
enough to catch one of them.
(I.e., the most likely cause of a "some captures cause Ethereal to
crash, other captures don't" problem is a bug in a dissector, so that
captures containing certain packets cause Ethereal to crash.)
If you could
1) use GDB to get a stack trace of the crash;
2) if possible, send us the trace file that causes the crash (if
it was from a live capture, there will probably be a trace file
in "/var/tmp" with a name beginning with "ether")
that would make it much more likely that we'd be able to debug the
problem. (See, for example, the "How to Report a Bug" section at the
end of the README file for Ethereal.)