Manuel Bouyer <bouyer@antioche.eu.org> writes:

> On Thu, Feb 07, 2002 at 02:50:48PM -0600, Greg Oster wrote:

>> I don't get that either... Was wd2 on the controller card that
>> failed?  Maybe it managed to zero the component labels on you??
>> Seems unlikely, but that's the only explanation I can think of so
>> far...
> 
> Or maybe it's just because the controller failed, the mod counter on this
> comonnent didn't get updated ?
> 
> If this is the drive which was on the failed controller, then the disk
> did really fail at one time, and RAIDFrame remembers of that. The only way
> to recover from this is to initiate a reconstruct.

Seems reasonable. However there is another data point:

The machine was up and running happily. Clean shutdown. I changed a
number of cables (from ribbon to round to improve airflow). All kinds
of IDE problems on reboot, so I of course suspected the new cables. I
switched cables again, still problems, bogus interrupts from the
Promise controller, confusion, etc.

My point is that between the clean shutdown and now the raid was
*never* mounted as a live filesystem, because the machine never
reached multiuser until the Promise controller was removed.

However, I can live with the reconstruct. That's the point with
raid. I do have a problem with the zeroed label on a raid set with two
out of three components intact.

Johan