AW: [AGENT++] please guide me about the MIB definition and implemention of a subagent

Eggers, Henning Henning.Eggers at plath.de
Tue Jan 30 10:16:49 CET 2007


Hi Leo Lei,
if understand your setup right, you only need one table with a double index, one for the row, one to identify the subagent.

Should look like this, where the index is the combination of basicNo and basicID:

infoTable OBJECT-TYPE
	SYNTAX SEQUENCE OF InfoEntry
	MAX-ACCESS not-accessible
	STATUS current
	DESCRIPTION
		""
  ::= { demoMIB 1 }


infoEntry OBJECT-TYPE
	SYNTAX InfoEntry
	MAX-ACCESS not-accessible
	STATUS current
	DESCRIPTION
		""
	INDEX {
		basicNo,
		basicID }
 ::= { infoTable 1 }


InfoEntry ::= SEQUENCE {

	basicNo Unsigned32,
	basicID Unsigned32,
	infoIP  IpAddress }


Remember, that index columns are "not-accessible", so the table will have just one data column (which is perfectly fine!).

Regards,
Henning

> -----Ursprüngliche Nachricht-----
> Von: agentpp-bounces at agentpp.org 
> [mailto:agentpp-bounces at agentpp.org] Im Auftrag von Leo Lei
> Gesendet: Dienstag, 30. Januar 2007 07:21
> An: agentpp
> Betreff: [AGENT++] please guide me about the MIB definition 
> and implemention of a subagent
> 
> hi list:
> 	in my case,   one master agent for many subagents 
> implementing the SAME MIB.
> 	i have some question about the MIB definition 
> implemented by many subagents.
> 
> 	the MIB (i assumed that the OID is 1.3.6.1.4.1.1111, 
> named demoMIB) have two conceptual
> table, one is named basicTable with two columnar object, 
> basicNo(index object) and basicID(
> uniquely identifing one subagent), another is named 
> infoTable(using basciID as index object) with 
> one columnar object, infoIp(ipv4 address for this subagent).
> 
> 	so, for one subagent, one row for basicTable, many rows 
> for infoTable. when many subagents
> connected and registered at the same master agent. can the 
> snmp manager station get the rows of infoTable
> of some specified subagent(identified by basicID) by 
> requesting the master agent?  if not, can i use 
> another reasonable MIB definition for my specifical case? 
> 
> 	thank you!
> 
> 	
>  				
> --------------
> Leo Lei
> 2007-01-30
> 
> 
> _______________________________________________
> AGENTPP mailing list
> AGENTPP at agentpp.org
> http://lists.agentpp.org/mailman/listinfo/agentpp
> 
> 




More information about the AGENTPP mailing list