On 11/29/2012 01:00 PM, Claudio Carbone wrote:
> Did a stupid thing obviously: modifying the message length is not the
> right way.
> Now I'm fiddling with Header.cpp and trying to at first print out the
> necessary values that get screwed, then maybe I'll understand something.

I had a problem on a Gumstix that had the same endianess as my X86 
machine. I do not remember the exact error message but the problem was 
that I had an int8 in a service call and on the ARM I think that took 1 
byte but on the X86 it took 4 byte. So changing all my int8 to int32 
solved the problem.

/Tommy Persson
Linköping University
Sweden