Ethereal-dev: RE: [Ethereal-dev] OID Names/Comments ...
Guy,
True both products are concerned with SNMP, but libsmi would solve the ssl
issue.
libsmi doesn't require ssl, since it is more accurately described as an ASN1
parser. This is what ethereal wants, rather than the full blown SNMP
capability that is net-snmp.
Regards,
Andrew Hood
A distributed system is one in which the failure of a computer you
didn't even know existed can render your own computer unusable. --
Leslie Lamport, as quoted in CACM, June 1992
> -----Original Message-----
> From: Guy Harris [mailto:gharris@xxxxxxxxx]
> Sent: Thursday, 5 September 2002 10:45
> To: HOOD, Andy
> Cc: ethereal-dev@xxxxxxxxxxxx; Richard Sharpe
> Subject: Re: [Ethereal-dev] OID Names/Comments ...
>
>
> On Thu, Sep 05, 2002 at 10:38:00AM +1000, HOOD, Andy wrote:
> > I haven't looked but net-snmp probably lets you retrieve
> the description.
>
> Yes, but it knows stuff about SNMP, so it doesn't meet at least one of
> my preferred criteria:
>
> Ideally, that code would know *nothing whatsoever* about SNMP,
> so that it doesn't drag in a pile of other libraries not needed
> for the purpose of interpreting ASN.1 descriptions (e.g., the
> crypto library that the UCD SNMP library requires on some
> platforms; that requirement causes all sorts of pain for people
> building from source on those platforms, as the configure script
> doesn't automatically try to find that library).
>
> The fact that you have to use "--with-ssl=" causes periodic complaints
> to ethereal-users and ethereal-dev.
>
> I don't know whether it meets another criterion:
>
> I say "information base" rather than "management information
> base" or "MIB" as I'd like it to be able to read COPS policy
> information base information (see RFC 3084) as well, so the COPS
> dissector could use it rather than having some knowledge
> hard-wired into it.
>
> > http://www.ibr.cs.tu-bs.de/projects/libsmi/ certainly does.
>
> It also, as far as I know, knows about SNMP; I don't know whether it
> could handle non-SNMP IB information. If not, it has the
> same problems.
>
<<application/ms-tnef>>
