On 28/10/2020 11:46, Mostaf Faridi wrote:
Hi,
I use nfdump 1.6

That must be very old then.  The current version is 1.6.20.  I'd start by upgrading to that.


Sometimes I see this error
Kernel:pid 75614 (nfdump),uid 0:exited on signal 11 (core dumped)

Kernel:pid 76732 (nfdump), uid 0:exited on signal 10 (core dumped)

I want know which packet make core dump?

You can use gdb to read the core dump and tell you exactly what line of code crashed, and inspect all the variables, which will let you work out the code path and hence the packet.

Alternatively, you could capture a load of packets at the same time (e.g. with tcpdump -w file.pcap ...), and then replay them to nfdump (nfdump -r file.pcap) until you find the offending packet.

It might not even be a bad packet. signal 11 (SEGV) can also be caused by hardware errors, e.g. bad RAM.  But if you only see this with nfdump, and not when doing something else which stresses the CPU (e.g. compiling a kernel), then it's probably nfdump.


_______________________________________________
Nfdump-discuss mailing list
Nfdump-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfdump-discuss

Reply via email to