[c-nsp] Software forced crash on 2651XM/12.3(6e)

[Rodney Dunn]

On Thu, Sep 28, 2006 at 07:48:31PM +0200, Asbjorn Hojmark - Lists wrote:
> >> Not to pick on you Rodney.....but I'd like to better 
> >> understand why this statement gets uttered so much. For those
> >> of us in large enterprises (maybe the smaller ones too), code
> >> upgrades are rarely permitted on whim.
> 
> > You hear it because it's a good choice for a lot of customers.
> > Honestly as crazy as it may sound more customers "want" to
> > upgrade to see if it fixes a problem than there are that do
> > not.
> 
> Honestly, I've frequently seen that attitued... within TAC. And
> they, apparently, get it from... the BUs.

Not totally true. Would the BU like to see a customer on later code
to eliminate the possibility that they waste time helping debug a problem
they already fixed which prevents them from spending time fixing new valid
bugs...sure they would. 

> 
> More often that not, you get the "Try to upgrade. It might fix
> it"-answer, and if you complain about that, you get the "But if
> this turns out to be a bug, I'll never get a fix for that version
> from the BU"-answer or maybe even the "I'll never get a developer
> convinced to troubleshoot that version"-answer.

All are valid depending on how "old" it is. We always push back on the
BU's to work on any bug we can't figure out. But it's a balancing act
between wasting time and resources on something we may have already fixed
vs. fixing new issue.

The golden nugget is a lab recreate by the customer or TAC. Those are hard
but once you have that you have the power to figure it all out.
I encourage customers to think about how to recreate a problem while they
complain about the impact of the problem.

> 
> > At the end of the day it's up to the customer what they do.
> 
> "Otherwise I can't help you".

Not true. We can always help.

> 
> -A