Let's  take as an example  the setting of ifAdminStatus of all interfaces of a 
particular device to down.   SETBULK can work in this example as all data types 
and values will be the same for all instances for the ifAdminStatus column but 
special considerations need to be taken on the last PDU.  The last PDU will 
likely over extend the target column object making it fail as the next column 
can be read only for example.

As an alternative, SET operations using multiple variable bindings in one PDU 
can be used but it is necessary to keep in mind the maximum UDP message size. 
There is also additional  flexibility using this method as each binding can be 
unrelated.

-Andy

On Jul 8, 2013, at 12:44 PM, Tom Sanders <toms.sand...@gmail.com> wrote:

> Assume that a user wants to create 100 vlans. Instead of using 100 set 
> operations, cant we use one SETBULK operation since in this case its only the 
> index that is different - all other parameters are the same.
> 
> Toms
> 
> 
> On 8 July 2013 22:04, Mike MacFaden <m...@zimbra.vmware.com> wrote:
>> 
>> On Jul 8, 2013, at 9:15 AM, Tom Sanders <toms.sand...@gmail.com> wrote:
>> 
>>> I hear you when you say that operators might want to use NETCONF.
>> 
>> 
>> Operators did more than say they might, they wore t-shirts that read:
>> 
>> "It's the CLI stupid"  
>> 
>> IIRC that was during NANOG 19 conference plus or minus.
>> 
>> 
>>> Regardless of this, i would like to understand the rational (if anybody 
>>> knows) of only having a GETBULK PDU and not a SETBULK PDU?
>> 
>> First of all sets require a layer of security and that took a bit longer to 
>> accomplish, not until 2002/SNMPv3.
>> Secondly  what would a set bulk do differently than a SET command?
>> 
>> Thirdly ….  http://tools.ietf.org/html/rfc3512
>> 
>> Mike
> 
> 
> 
> -- 
> Toms.
> _______________________________________________
> OPSAWG mailing list
> OPSAWG@ietf.org
> https://www.ietf.org/mailman/listinfo/opsawg
_______________________________________________
OPSAWG mailing list
OPSAWG@ietf.org
https://www.ietf.org/mailman/listinfo/opsawg

Reply via email to