> "print cf.count" gives me ...
>
> "$1 = 2".
>
> And truly, at this point, I don't recall how far along it went before
> it coredumped.
The fact that "cf.count" is only 2 suggests that it didn't get very far
at all - "count" starts out as zero, and gets bumped in
"wtap_dispatch_cb()" after reading a packet, which says that it's the
third packet on which it blew up. That's actually a good sign, in a
perverse sense - it means that it can probably be reproduced with a
small bit chopped off the front of the capture file.
I can probably cook up a quick program to slice out the first N packets
of a "libpcap" capture, if such a program doesn't already exist
("tcpslice" slices by time stamp, not by packet count), so, if your
capture files aren't private, you could slice off the first 5 of them,
say, and, if the program still blows up on that one, send the resulting
capture file to one or more of us and we could take a look at it.
Powered by MHonArc 2.6.10