Re: [secdir] secdir reviewofdraft-ietf-isms-transport-security-model-12

[Wes Hardaker <[e-mail]>]

>>>>> On Fri, 8 May 2009 23:43:46 +0200, Juergen Schoenwaelder <[e-mail]> said:

>> The idea of Models in SNMP is to be able to mix and match.  In
>> practice, this has not worked - USM with sshTM will not function,
>> regardless of whether it is secure or not.

JS> Not sure I understand why. Can you explain?

The way it's worded right now in the SSH document is that there must be
a tmStateReference for outgoing messages or the SSHTM will drop the
message:

   1.  If tmStateReference does not refer to a cache containing values
       for tmTransportDomain, tmTransportAddress, tmSecurityName,
       tmRequestedSecurityLevel, and tmSameSecurity, then increment the
       sshtmSessionInvalidCaches counter, discard the message and return
       the error indication in the statusInformation.  Processing of
       this message stops.

USM won't generate a tmStateReference so it can't be used as an
upper-level security protocol.  This was discussed (I'm fairly certain)
on the list at some point and it was decided this was ok as doing
something to try and recover at that point would really be adding words
and complexity for a situation that didn't really need to be supported.

[that being said, note that there isn't a MUST drop it in the text]
-- 
Wes Hardaker
Cobham Analytic Solutions