I posted a patch to fix this (about 2.0.30). Alan Cox was the only one
who responded to the patch however it has never been included.
I'll dig up the patch if anyone is interested.
I added it as a compile time option when building the kernel.
ps: I've been using it on my/our network since 2.0.30 (currently 2.0.34)
and has not caused any problems.
> >
> I had one of those a while back, and eventually traced it, but never
> really understood it.
>
> There's some ICMP thingy to get the address mask from a router or sommat.
> If you read net/ipv4/icmp.c there are all sorts of comments about why it
> should/shoudn't be implemented, and when we should reply etc. etc.
> It ends up with:
> * All these rules are so bizarre, that I removed kernel addrmask
> * support at all. It is wrong, it is obsolete, nobody uses it in
> * any case. --ANK
It is used by tkined/scotty snmp package to "discover" an IP network.
> *
> * Furthermore you can do it with a usermode address agent program
> * anyway...
> */
>
> static void icmp_address(struct icmphdr *icmph, struct sk_buff *skb, int
> len)
> {
> if (net_ratelimit())
> printk(KERN_DEBUG "a guy asks for address mask. Who is
> it?\n");
> }
>
> If anyone has a more complete explanation of what these packets actually
> are, I would be interested.
>
> HTH.
>
> --
> Mike <rickettm@ox.compsoc.net>
>
> What color is a chameleon on a mirror?
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu