RE: SNMP over Ethernet

["Wijnen, Bert (Bert)" <bwijnen@lucent.com>]

Inline
> 
> In line. 
> 
> > Inline
> > 
> > > -----Original Message-----
> > > From: owner-mreview@ops.ietf.org 
> > [mailto:owner-mreview@ops.ietf.org]On
> > > Behalf Of Romascanu, Dan (Dan)
> > > Sent: Wednesday, November 16, 2005 02:34
> > > To: Mreview (E-mail)
> > > Subject: SNMP over Ethernet
> > > 
> > > 
> > > Please help me with a homework. The IEEE 802.1 WG is discussing the 
> > > option on of using SNMP over Ethernet (don't ask). I see that this 
> > > mapping originally defined by RFC 1089 was not included later in the 
> > > list run by RFC 3417? Does anything prevent doing RFC 1089 
> > > encapsulation for SNMPv3?
> > > 
> > 
> > I donot immediately see why SNMPv3 packets could not be sent 
> > with an RFC1089 encapsulation. (not that the status of RFC1089 is
> > UNKNOWN by the way).
> 
> I saw this - what does it mean, BTW? Was those never put on standards
> track, was it informational, or because it was during 
> pre-history these categories were not used then?
> 

Nope. it is just unknown (or so I think).
I see that RFC1041 and RFC1043 is PS. And 1042 is STD. 
But maybe those classifications were added later. 
I can go try to figure out if we find it important.

> > 
> > Note that RFC1449 also did not include 1089 encapsulation
> > (as far as I can tell from a quickj scan). So it seems it already
> > was not too important anymore around early 90s.
> 
> Paradoxally it may become now again. The main good reason is that the
> 802.1 folks are looking for a OAM-like protocol that would run atop of a
> layer 2 transport and not mandate an IP layer. SNMP came across as a
> candidate, seems to meet quite well the requirements and it's 
> there, no need to invent something new. 
> 

See Juergens posting. I tend to agree with his "better do a new short
RFC that obsoletes 1089 and adds transport mapping definitions".

Bert

> > Bert
> 
> Thanks and Regards,
> 
> Dan
> > > 
> > 
>