[c-nsp] RAM thing

[Phil Mayers]

On 17/02/2014 15:47, Saku Ytti wrote:
> On (2014-02-17 12:24 +0000), Phil Mayers wrote:
>
>> So nothing has changed except we know about it. For anyone who
>> assumed devices could fail at any time, this isn't *that* worrying.
>> For anyone who assumed devices would run forever, this should be a
>> wake-up call - it was never true, and will likely never be so ;o)
>
> Should we expect devices to be build so that broken memory is detected and
> reported to operator? Or is it OK that broken memory is undetected and mangles
> packets without us being aware?

As someone else has pointed out, the Cisco description is of a sudden 
hard failure triggered by a power cycle, not some kind of progressive 
degradation AFAICT.

Do you have information to the contrary?