[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RFC 1905 Section 4.2.5



HI,

On the following...

At 10:45 AM 8/30/2001 +0200, you wrote:
>>>>>> sidney antommarchi writes:
>Sidney> Giving the user feedback for row creation problems is not
>Sidney> possible via variables, tables, columns, or traps without
>Sidney> correlation problems.  In other words, protocol support is
>Sidney> needed.  As such, is it possible to have all new PDUs be
>Sidney> extended to include an errTxt (DisplayString)?  This could
>Sidney> give the user the ability to augment the explanation given in
>Sidney> error-status.
>
>The textual message is only useful for debugging and simplistic
>tools. What is really needed for more substantial management
>applications are operation specific error codes that signal exceptions
>in a machine readable format. In fact, you probably want to have both
>- a machine readable operation specific error code and a human
>readable error message.
>
>/js
I agree with you that this problem can, and has, been solved (i hope).
(And it can be solved running over a connectionless transport.)
The point is that it is an issue to be addressed, and just using
a connection oriented transport does not magically solve the
problem.

Regards,
/david t. perkins