Hi there,
Unfortunately don't have time to look into it further at the moment, but:
Has there been a change to the packet re-assembly logic since 20/08/09?
The reason I ask is that I'm working on a dissector and svn up'd today to
apply some fix-ups to it. After applying the changes I noticed the re-
assembly wasn't working properly anymore, specifically instead of resuming
dissection at the point where the sub-dissector finished dissection is
resuming at the next packet.
I did 'svn up -r {2009-08-20}' and it worked properly again so something
appears to have changed since then.
I'm not familiar enough with where reassembly happens to look into it
further at the moment, so I guess if no-one pipes up I can file a bug later
in the week with more helpful information.
Thanks,
Robert