On 18 Jul 2011 at 15:56, Robert Story wrote:
> On Mon, 18 Jul 2011 10:28:58 +0200 Giuseppe wrote:
> GM> What do you think? Are my thoughts correct?
>
> You are correct, except for the part about needing non-volatile storage. If
> you have non-volatile storage, you could certainly do it that way. But there
> are many, many MIBs that don't use non-volatile storage. If the device is
> switched off, then you simple report what the API gives to you.
The part about non-volatile storage was referred to the scenario
with *two* variables (mute and volume level), but only one function
(volume_set()).
Suppose the following sequence of events:
snmpset ... volumeLevel = 30 --> volume_set(30);
snmpset ... mute = 1 --> volume_set(-1);
Now the GET request of volumeLevel variable returns 30, the request
of mute returns 1 (the mute is active). The agent has to save
internally the value 30, because it can't be returned by the
function volume_set(), that now returns only -1.
Now suppose the device is turned off and on again, and the user
sends GET requests of mute and volumeLevel variables. Which value
should the agent return for volumeLevel, if the value 30 wasn't
permanently saved in a non-volatile memory? It can't be deduced by
the volume_set() API function that returns just -1.
Where my arguments are wrong?
------------------------------------------------------------------------------ Magic Quadrant for Content-Aware Data Loss Prevention Research study explores the data loss prevention market. Includes in-depth analysis on the changes within the DLP market, and the criteria used to evaluate the strengths and weaknesses of these DLP solutions. http://www.accelacomm.com/jaw/sfnl/114/51385063/
_______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
