Thanks for the clarification, David. You are quite correct - I was thinking/talking in terms of the full OID of an entry in a table (i.e. a column instance). And specifically one indexed by a single integer.
The final subidentifier of the OID of a MIB definition cannot be 0, regardless of whether this is a column or scalar object (or an internal framework definition). And it's not strictly instance OIDs *ending* in 0 that are problematic - rather OIDs where the list of instance subidentifier(s) is exactly the single value 0. So for tables indexed by anything other than a single integer (strings, OIDs, multiple indexes), there's no problem about the final element of that index instance being 0. It's just where the *whole* of the index instance sequence is precisely "0" that this issue arises. Dave _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
