[SNMP4J] Wrong digest is a right digest

[Frank Fock]

Hi Chris,

You do not need to wait for the next release. Creating a Snmp instance
for each request will also be really bad with the next version. It will
always result in a bad performance, because you create a Thread,
several objects, and open a port!

The problems generated with changing the USM (by creating a new Snmp
instance) "on the fly" without incrementing the boot counter will be
removed in the next release, but that's all.

More inline:

chk-world at gmx.de wrote:

>
>So one thread is using or altering the USM of another thread? The USM uses a
>  
>
Yes.

>different engine ID and therefor the digest is wrong? 
>  
>
No. When one thread replaces the USM, the user(s) and engine time cache 
created by
the other thread are dropped.

>There is another side effect when creating a new SNMP instance for each
>query: I first thought it would be a memory hole but then I found that there
>are a lot of threads created which do not finish.
>
>  
>
They finish when you call Snmp.close().

>So I guess I'll have to wait for the 1.0.4c ;-)
>
>  
>
See above.

>Why doesn't anyone else got this problem with multiple agents? Does someone
>have an idea of an usable or even proofed architecture of a management
>application using snmp4j?
>
>  
>
Do not create Snmp instances for each request but reuse them.
ThreadPool can be also helpful here.

Best regards,
Frank