Re: [Isms] Why SNMPv3? [WG Review: Integrated Security Model for SNMP (isms)]

> On Thu, 23 Sep 2004 16:08:34 +0200 Juergen wrote:
> JS> This goes back to a rather old discussion whether there are 
> environments
> JS> where SNMP writes are used, how big these environments might be (I 
> leave
> JS> it to you do define the metric) and whether the IETF should worry 
> about
> JS> these environments. There are two camps, one arguing that SNMP 
> writes
> JS> are just an illusion since nobody actually uses them.  The other 
> camp
> JS> claims that SNMP writes are a reality (or believes they will 
> become a
> JS> reality once all the security issues have been solved).
>
> SNMP writes are a reality. I personally have developed at least 3 SNMP 
> agents
> with write support that interface with existing proprietary control 
> mechanisms.

	Being a reality and actually being used are two different things.
I too have developed numerous modules that are writable, but not many
people use them. :P  Like SNMP writes, Betamax is a reality for only
the few that use it, but for the majority there are 1,000 reasons why
they use something else. *)

	--Tom


> JS> To this end, there was never a real investigation to sort this 
> out. It
> JS> would be more than interesting to collect packet traces from 
> real-world
> JS> networks (and not just ISP IP networks since SNMP seems to be used 
> quite
> JS> a bit in telco networks, cable networks, enterprise networks 
> largely
> JS> based on IEEE 802 LANs, ATM networks, ...)
>
> All of the agents were for clients in the cable industry who were 
> adding SNMP
> support under pressure from their customers.
>
> All used SNMP v1/v2c, relaying on a non-public network for access 
> control.
>
>
> -- 
> Robert Story; NET-SNMP Junkie
> <irc://irc.freenode.net/#net-snmp>   <http://www.net-snmp.org/>
>
> You are lost in a twisty maze of little standards, all different.
>