Re: possible regression handling packet fragmentation in 14.0 with tftp/pxe

Gerrit Kühn <gerrit.kuehn@aei.mpg.de> writes:
> Any idea what the "bad length 1460 > 1392" message on the 13.3 system
> means (and why everything is still working)?

Off the top of my head, it means the packet length according to its
header is longer than the capture length but the MF flag is not set.
Historically, it meant you needed to increase tcpdump's snaplen, but
these days it defaults to 256 kB so it shouldn't matter.  It might be a
bug in bpf, or an interaction between pf and bpf.

> As this appears to be different behaviour on 13.3 and 14.0, I had hoped
> this might already be sufficient to ring a bell for someone here reading
> this (like "oh, yes, there were changes in pf that cause different
> handling of fragmented udp packets").

There were changes in tftpd, but I don't know if that's relevant.

DES
-- 
Dag-Erling Smørgrav - des@FreeBSD.org