Re: Fwd: [ipv6mib] So, where were we?

[Juergen Schoenwaelder <schoenw@ibr.cs.tu-bs.de>]

>>>>> Michael MacFaden writes:

Michael> What has been fooling mgmt applications and users for years
Michael> now is using out of date full standard mib modules such as
Michael> ipRouteTable.

I agree. But I continue to believe that implementing the current
IP-FORWARDING-MIB on boxes that support more complex FIBs is in the
same way fooling mgmt applications. We obviously disagree on this
since probably our estimation how many boxes support simple FIBs and
how many boxes support more complex FIBs diverge.

Michael> I have yet to see a "use case" that states exactly what
Michael> operators need beyond the tuple I listed before (prefix, who
Michael> installed/age, nexthop).

Operators want to have interface where they can ask questions such as
"to which hop is a packet with desctination X forwarded"? If the MIB
does not tell the mgmt application the whole story, the answer
produced will be useless for them. (How we solve the inherent
efficiency issue is another problem, but perhaps EOS can help.)

Michael> Please read draft-ietf-snmpconf-bcp-10.tt section 3.13.2
Michael> "Supporting Multiple instances of a MIB module".

Michael> This issue has an existing viable solution that has worked
Michael> for RFC1850 and others where the mib module defines a single
Michael> instance but the device can support multiple instances. I
Michael> would prefer the solution to having everyone pay the price of
Michael> some complex index structure when 99% of the time a device
Michael> supports a FIB.

See, we do not seem to agree on the 99% case.

/js

-- 
Juergen Schoenwaelder    <http://www.informatik.uni-osnabrueck.de/schoenw/>