In message <[EMAIL PROTECTED]> you write:
> Hello!
>
> > - "reject" sends RSTs in response to SYN+no ACK TCP frames
> > - "reject" sends ICMP port unreachables to all other TCP frames
>
> It will work and sounds reasonable.
>
> It smells bad, though. Introducing such strong tricks only
> to workaround minor bugs in some OSes... I do not know, where
> we will finish then. 8)
It's a horrible, ugly hack: please provide proof that you've
unsuccessfully petitioned the broken vendors before trying to
introduce it.
> The case with NT is the most spectacular. Seems, they have at least
> two independant teams. One introduces bugs, another invents workarounds.
> Silly bugs are followed by ugly workarounds. 8)
I agree with Alexey: this is exactly what we should be avoiding. If
you want to generate a RST in response to a locally-destined packet,
it's easy for most cases. There's even an example of doing exactly
this in libfw.
Rusty.
--
Hacking time.
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
