Re: BUG in SNMPv1 Traps

Juergen Schoenwaelder (schoenw@cs.utwente.nl)
Wed, 3 Jul 1996 21:50:51 +0200

Graeme McKerrell <graemem@pdd.3com.com> wrote:

> 06 07 2b 06 01 04 01 8d 25 enterprise RUBBISH OID iso.6.1.4.1.141.37

> The enterprise OID is rubbish!! but the generic and specific values are
> correct.

Sorry, but your decoding is also broken. :-) The last sub-identifier
is contained in two bytes since the highest bit is turned on. See the
BER encoding rules for details. However, you are correct that this
OID is not semantically correct. That's why I changed it in 2.1.0.

> The error is in specifying generic=6 and specific=0
> for a generic SNMPv1 trap the generic integer should be set to
> the required trap number.

This works correctly in my version. Do you have any additional MIB
definitions loaded which probably re-define coldStart with a different
OID? Scotty extracts the generic/specific fields by checking the OID
and if your coldStart does not resolve to 1.3.6.1.6.3.1.1.5.1 scotty
might have problems.

Juergen