Re: Doubt about VACM.
"David T. Perkins" <[email protected]>
| Newsgroups | gmane.ietf.snmpv3 |
|---|---|
| Message-ID | <[email protected]> |
HI, Randy's advice is correct. And here is another example.... Consider objects in the bridge MIB module, found in RFC 1493. This MIB module models a device as supporting a single instance of the bridging function. Currently, there are many vendors that provide devices that support more than one instance of bridging in a device. One "easy" way to support this is with contexts, which are simply additional indexing! For each new instance of the bridging function that is created, a new entry is added to table vacmContextTable. How this is done is implementation specific. At 10:32 AM 12/26/2002 -0800, Randy Presuhn wrote: >Hi - > >> From: "parasuraman" <[email protected]> >> To: <[email protected]> >> Sent: Thursday, December 26, 2002 2:44 AM >> Subject: Doubt about VACM. >... >> Could any one of you explain me, the exact meaning of the above >> paragraph. >... > >The table reports the membership of the set of contexts recognized >by the SNMP entity. If the membership of that set is dynamic, then >the table will necessarily be dynamic. VACM (now RFC 3415) does >not define an interface for management to modify the membership of >the set. > >Concrete example: let's say we have an instance of some MIB for each >card in a chassis. When a card in such an environment is inserted or >withdrawn, the set of contexts known to the SNMP entity could change. > >Randy Regards, /david t. perkins