Re: NT 3.51 dhcp client and server ip address

[Shawn Mamros <[e-mail]>]

>Yeah - looks like we blew it by overloading the "identifier" field to also
>act as the "return address" field.
>
>We can either fix it or leave it:
>
>  * Fixing will delay the advancement to Draft Standard, as it would
>    represent a fairly basic change to the behavior of clients and
>    servers.

"Fixing" it also means breaking thousands upon thousands of client
implementations already out there (not just Redmond's, either...).

>  * Leaving it leaves the problem with multiple identifiers for a single
>    server - as pointed out by Ted - unfixed.

How much of a problem is this really?  The server implementation I was
responsible for had to handle multiple interfaces, and it did so just
fine.  The server kept a list of all of its own IP addresses.  When a
DHCPREQUEST comes in with a Server Identifier option specified, we ran
down the list and compared it - if it was on the list, it's ours; if
not, it isn't.  Unless there's a DHCP server with hundreds of interfaces
out there, running down the list shouldn't take all that long...

-Shawn Mamros
E-mail to: [e-mail]