[AGENT++] Problem with set request on a Celeron machine

[Frank Fock]

Hi Patrice,

Please check your agent's instrumentation code for uninitialized data.
This may cause the different behavior on a different processor
(and compiler). Another problem could be a compiler bug in conjunction
with some loop optimizations. And of course, there could be also a bug
in AGENT++, but I do not expect such a "global" bug.

Best regards,
Frank

Patrice Bentolila wrote:

> Hi to all,
> I've got some odd result on set request.
> I've got a sub agentX which works well on a RH 7.3 / AMD K6 400 
> machine (get, set, trap).
> The same agent on a celeron and same OS works well on get/trap requests.
> However, In set, it returns Timeout/Commit Fail.
> But after a get request on the same OID,
> I get the result as is set request succeeded.
> Does someone have an idea?
> Best Regards.
> Patrice.
>
> _________________________________________________________________
> MSN Messenger : dialoguez en temps réel avec vos amis ! 
> http://g.msn.fr/FR1001/866
>
> _______________________________________________
> AGENTPP mailing list
> AGENTPP at agentpp.org
> http://agentpp.org/mailman/listinfo/agentpp
>