[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: DS MIB - please review this list...
- To: Juergen Schoenwaelder <schoenw@ibr.cs.tu-bs.de>
- Subject: Re: DS MIB - please review this list...
- From: Brian E Carpenter <brian@hursley.ibm.com>
- Date: Tue, 27 Feb 2001 15:00:34 -0600
- CC: fred@cisco.com, fenner@research.att.com, daniele@zk3.dec.com, haberman@nortelnetworks.com, sar@epilogue.com, mibs@ops.ietf.org, khchan@nortelnetworks.com, nichols@packetdesign.com, andrew@allegronetworks.com
- Delivery-date: Tue, 27 Feb 2001 15:07:26 -0800
- Envelope-to: mibs-data@psg.com
- Organization: IBM
Juergen Schoenwaelder wrote:
...
> BTW, are you really absolutely sure that the two addresses in the
> filer will never ever be of different types? Did you think through all
> the various transition mechanisms and the various "any" addresses that
> are possible?
Yes. The rules for diffserv and tunnels mean that at any point where
you apply traffic conditioning, source and destination are in the
same address family. And you're not going to put a traffic conditioner
in the *middle* of a NAT-PT (an IPv6/IPv4 translator); it will be at
one side or the other.
However as I said before, I think we'd better accept the conventional
wisdom, even though it's weak computer science IMHO.
Brian