[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: InetAddressType and InetAddress in INET-ADDRESS-MIB
Hi!
Bert> Frank... it is nice to see that we have a compiler that does check
Bert> rules as documented in RFCs. However, can you also tell us if
Bert> libsmi uses the rules for some usefull purpose?
It does not yet.
However, Christian Moeller <cmoeller@ibr.cs.tu-bs.de> is currently
working on a libsmi-based agent code generator. I could imagine that
he will include support to allow an instrumented device/service/whatever
to pass an AF-independent struct sockaddr (is this the right struct?)
to the compiler-generated stub if the MIB uses InetAddressType/InetAddress
appropriately.
-frank
>> ----------
>> From: Frank Strauss[SMTP:strauss@ibr.cs.tu-bs.de]
>> Sent: Thursday, March 08, 2001 12:06 AM
>> To: Wijnen, Bert (Bert)
>> Cc: mibs@ops.ietf.org
>> Subject: Re: InetAddressType and InetAddress in INET-ADDRESS-MIB
>>
>> Hi!
>>
Bert> The current I-D draft-ietf-ops-rfc2851-update-00.txt contains this
>> TC:
>>
Bert> [...] In other words, the object
>> identifiers
Bert> for the InetAddressType object and the InetAddress
>> object
Bert> MUST have the same length and the last sub-identifier of
Bert> the InetAddressType object MUST be 1 less than the last
Bert> sub-identifier of the InetAddress object.
>>
Bert> [...] So the question that I want to see answered is:
>>
Bert> - How many tools/apps/code-generators do we know about that
Bert> indeed make use of this rule
Bert> - How does this rule help any such tools
Bert> - How important are those tools so that they justify this extra
Bert> MUST rule?
>>
>> This does not answer your questions, however, it's related:
>> Libsmi *checks* for this rule and warns about violations.
>> An example (the only one I found from the current I-Ds) from
>> draft-nadeau-mpls-vpn-mib-00.txt:
>>
>> ./MPLS-VPN-MIB:398: `InetAddress' object should have an accompanied
>> preceding `InetAdressType' object
>> ./MPLS-VPN-MIB:406: `InetAddress' object should have an accompanied
>> preceding `InetAdressType' object
>> ./MPLS-VPN-MIB:416: `InetAddress' object should have an accompanied
>> preceding `InetAdressType' object
>>
>> -frank
>>