On Wed, Nov 22, 2006 at 11:08:51PM +0100, Anders Broman wrote:
> Hi,
> If the corresponding type 2 message with the bytecode isn't in the trace
> Wireshark can't decode the message.
>
> If you still think there is a problem we'd need the trace to be able to
> locate the problem.
cristian: hi anders. some more questions:
1. is wireshark always using 64K dms? is this somehow configurable?
(I saw cases when wireshark was able to decompress but the peer not due
to lower dms; configuring dms in wireshark could help)
2. any specific compartment mapping? if yes what about sms?
3. I am not an wireshark developer and I cannot figure out if
showing exactly which sigcomp message from the trace has caused a NACK would
be feasable. (when a lot of NACKs are there it is really difficult to
figure out which message has caused it)
thanks a lot!
bye now!
cristian
