This standard is now considered obsolete. All-zero and all-1 subnet masks
are prefectly acceptable these days.

They should know that.

I don't think there needs to be a sysctl variable for this.

Guenther Grau wrote:
>
>Hi,
>
>at work I am "forced" to work with hp-ux ( which is, IMHO, one of the
>worst unix implementations I ever worked with. But I don't want to
>talk about this right now :-). Starting from version 10.x they forbid
>the following netmask: 255.255.255.128 for a class C IP-Address (eg.
>192.13.13.13).
>When I called hp-support they said that this netmask was illegal, citing
>rfc-1122 (well, they also offered me a workaround, because other
>customers had complained about this as well :-). Reading through
>rfc-1122 I found the following passage (page 30):
>
>...
>            IP addresses are not permitted to have the value 0 or -1 for
>            any of the <Host-number>, <Network-number>, or <Subnet-
>            number> fields (except in the special cases listed above).
>            This implies that each of these fields will be at least two
>            bits long.
>
>where -1 means all bits one and the special cases are just broadcast and
>initialisation addresses.
>
>Can anybody explain why this was done?
>
>And my second question: Do we comply to this standard?
>Although I can see no reason for the requirement, we should comply
>and offer a variable which can be set by sysctl to turn this behaviour
>on or off.
>
>  Guenther
>


-- 
Matthias Urlichs
noris network GmbH