James Carlson wrote:

> And I'm not buying any "performance" argument here.  We're talking
> about the delivery of errors.  Applications aren't built to deliver
> bulk data via error reports.  If performance (or perhaps more
> accurately DoS risk) is really a concern, then we should just throttle
> the rate at which we send any messages at all, rather than trying to
> trim down the byte count to save a couple of cycles.  That
> optimization needs to be made in the right place, and being capable of
> sending zillions of uselessly tiny ICMP errors per second is not
> something to strive for.


Note that I am *not* saying that increasing the ICMP return
byte size is not good.  But I don't think the suggestion on
using rate is good enough.  The bandwidth consumed is the
number ICMP messages sent per second (rate) times the ICMP
packet size.  To some sys admins, just controlling the rate is
not good enough for them to handle DoS situation.


-- 

                                                K. Poon.
                                                kacheong.poon at sun.com


Reply via email to