Re: AgentX WG: New work items or close?

At 4/6/2002:05:02 PM, Mark Ellison wrote:

Hi Mark,

>I support your list of possible AgentX work items.  

Ok...good...we'll have to see if this feedback from
you and the comments from Wes, thus far, can be
turned into sufficient momentum to keep the WG open.
It's too bad that Wes, Dave Shield, and John Naylon
from the NET-SNMP list can come back soon with some
more specifics re the kinds of things they'd like
the WG to work on.

>I'm wondering if the original agenda item, "(optional) a programmatic
>interface to the services offered by that protocol" is officially
>dropped.  I don't know this is necessary.  Possibly others will have a
>different perspective?

Well, it has not been officially dropped.  On the
other hand, there has not been a lot of push for
completing that work item coming from the WG.  When
you say that you "don't know this necessary", do you
mean you don't know whether it's necessary to drop
it or retain it?

To some extent, having the NET-SNMP and AgentPP groups
worked out their API implementations may be a good
thing here:  Do any AgentX developers who may have
used either or both of those APIs see a need to
standardize this technology in the IETF?  If you
have an opinion on that, I would appreciate hearing
it soon.

>Also, with respect to work in the EOS and SMIng WG-  I think it best not
>to take on any dditional AgentX work items until the EOS and SMIng work
>product progresses beyond internet-draft.

Yeah, I agree...good point.

Cheers,

BobN

>"Natale, Robert C (Bob)" wrote:
>> 
>> Hi folks,
>> 
>> With the advancement of the AgentX specs to
>> Draft Standard status some time ago, we now
>> need to decide whether we want to take on any
>> new work items as an WG or ask the IESG to
>> close the WG until the time comes to review
>> the specs' qualifications for Full Standard
>> status.
>> 
>> The possible work items I know of that we
>> could incorporate into a revised charter
>> include:
>> 
>>    - Mechanisms for supporting the SNMPv3
>>      security features in AgentX...this has
>>      been discussed on the list in the past
>>      under the "security credentials" and
>>      "isAccessAllowed()" threads.
>> 
>>    - Process mechanisms for ensuring that
>>      IETF standard MIBs are written in such
>>      a way as to avoid constructs that
>>      prevent or seriously impede standard
>>      AgentX support (e.g., avoiding the use
>>      of scalar objects).  This might take
>>      the form of a new boilerplate section
>>      for MIBs called "AgentX Considerations",
>>      akin to the "Security Considerations"
>>      section currently required.  For vendor
>>      MIBs, this process would encourage
>>      AgentX supportability.
>> 
>>    - Any potential revisions to the specs
>>      that have surfaced since the implementation
>>      survey that we did prior to going to
>>      Draft Standard status that we will need
>>      to address for Full Standard status...
>>      might as well air these at the earliest
>>      possible point and come up with solutions.
>> 
>>    - Any other suggestions from the floor...?
>> 
>> However, we should undertake these new work items
>> *only* if there is sufficient interest in actually
>> working on them (writing, editing, reviewing,
>> commenting, coding, testing, etc.) among a reasonable
>> subset of the group.  Otherwise, we should ask the
>> IESG to close the WG.
>> 
>> So, if you are interested in working on any of
>> the above items in this forum, please post your
>> comments to the list within the next several days.
>> I'll aim for summarizing the response by the end
>> of next week (Fri 4/12) and making a recommendation
>> to the ADs for their consideration.
>> 
>> I have bcc:'d the "AgentPP" and "NET-SNMP" lists
>> on this notice since they have more AgentX traffic
>> than this list -- a good thing at this point! --
>> but please post all responses back to this list
>> alias only.
>> 
>> Thanks,
>> 
>> BobN
>> Chair