A more flexible 'knob' would be cool, something like combining the 'cos'
bits in the labels with the decision to decremening or not. This will give
the management station the power to 'trace' even the regular trace packet
can't do.
No, no, no, no, no. You are confusing TTL expiration with
"visibility".
You have a trade off between size of forwarding loop and
TTL decrement. The longer it takes to decrement a TTL to 0
in a forwarding loop, the worse the loop's effects will be.
What you want -- and I think you are very strange for wanting it,
by the way, but I will supply you a solution anyway -- is a knob
that modifies the ICMP ttl exceeded message that is sent back
to the source.
Sean.