Re: LC MIB draft
"tom.petch" <[email protected]>
| Newsgroups | gmane.ietf.forces |
|---|---|
| Message-ID | <02df01c6c685$5be01540$0601a8c0@pc6> |
----- Original Message ----- From: "Jin Rong" <[email protected]> To: <[email protected]> Sent: Tuesday, August 22, 2006 6:02 AM Subject: Re: LC MIB draft > > There are two notifications defined in draft-ietf-forces-mib-04: > > forcesAssociationEntryUp NOTIFICATION-TYPE > OBJECTS { > forcesAssociationRunningProtocolVersion > } > STATUS current > DESCRIPTION > "This notification is generated as soon > as an association enters the UP state." > ::= { forcesMibNotifications 1 } > > > forcesAssociationEntryDown NOTIFICATION-TYPE > OBJECTS { > forcesAssociationRunningProtocolVersion, > forcesAssociationTimeUp, > forcesAssociationTimeDown, > forcesAssociationHBMsgSent, > forcesAssociationHBMsgReceived, > forcesAssociationOtherMsgSent, > forcesAssociationOtherMsgReceived } > STATUS current > DESCRIPTION > "This notification is generated as soon > as an association leaves the UP state." > ::= { forcesMibNotifications 2 } > > > I think the "OBJECTS" should also include "forcesAssociationCEID" and > "forcesAssociationFEID" It does:-) All objects in an SNMP PDU consist of fully qualified OID and value. Whereever the object comes from a table, then the fully qualified OID will contain all the index values so in this case, the manager will receive the relevant index values forcesAssociationCEID and forcesAssociationFEID many times. This is a difference between SMIv1 and SMIv2. In SMIv1, table indices were included as separate objects generating unnecessary redundancy; in SMIv2, they are made not-accessible to eliminate that waste. Tom Petch <snip>