Re: New draft possibilities

[der Mouse <mouse%Rodents.Montreal.QC.CA@localhost>]

>>>> [UDP-over-ssh]
>>> For example, I would expect it to work fine for DNS traffic [...]
>> But what happens when DNS times out and sends a retry, while TCP has
>> also queued it's own retry?
> A reasonable implementation of UDP-over-SSH would notice the TCP
> congestion (in the Unix world, by write() on the TCP socket returning
> EWOULDBLOCK) and drop incoming UDP until it cleared.

If that's your idea of TCP congestion pushed back to userland, you need
to get your hands dirty with code more[%]. :-)  Most kernels buffer
some 10-20 K of data before that happens, and that can hold a lot more
than a few DNS retries.

Yes, eventually buffers fill up, but you need either crippling
underlying congestion or a comparative lot of upper-layer traffic to
push it clear back to userland.

[%] Yes, I know you _do_ get your hands dirty with code.  That's why
    the smiley.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse%rodents.montreal.qc.ca@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B