Re: [snmpsim-users] Multiple variation modules for a single OID?

[Forman, A. <And...@Te...>]

Hi Ilya,

Currently I'm considering whether I'd need to.  One possible case is where an OID is writable (writecache) and I also need to send out a notify (notification).

I completely agree with how very complex this could get, making the super-module look like the best option.

Thanks!
Andrew

Andrew B. Forman
Tectura
Tel:        +1 425 749 4446
Mobile:  +1 206 335 9414

From: Ilya Etingof [mailto:il...@gl...]
Sent: Wednesday, September 25, 2013 12:03 PM
To: Forman, Andrew
Cc: snm...@li...
Subject: Re: [snmpsim-users] Multiple variation modules for a single OID?

Hi Andrew,

That's right, multiple variation modules for a single OID are not currently implemented. The reason is that the logic of modules configuration and chaining their invocation looks overly complex to use. Your suggestion on a super-set variation module may be a good way of isolating these complexities from the base case of a single module per OID.

But what's your use case? What kind of variation you need? Maybe it would be easier in the end to create another single variation module that does the logic you need... What do you think?

-ilya

On 09/25/13 20:54, Forman, Andrew wrote:
Would it be possible to have multiple variation modules trigger for a single OID?

Looking in snmpsimd SnmprecRecord.evaluateValue I know this isn't supported today.

My thought wasthat the .pysnmp entry would look like:
                [OID]|[type]:variation1,variation2|[settings]
Further, I'd think that the settings would be shared across both modules, so any parameters of the same name would apply to both.

Alternately, perhaps it would be better to create a super-set variation module that would then proxy the calls out to the other modules?

Thanks,
Andrew