[c-nsp] "common" causes for 6500/7600 FIBDISABLE?

[Kevin Graham]

Having been recently hit by CSCsl62851 and/or CSCsu95171, which I suspect
should be cross-referenced (yes, I was behind where we should have been
in SRC), are there any "common" causes for a FIBDISABLE (in which PFC/DFC
is effectively unloaded)? On an RSP720, this very neatly left both
EGP/IGP mostly functional while killing any pratical forwarding-plane
activity.

I had thought this had come up a year or so here in the context of SXF,
but again my search-fu is weak.

Whatever the cause, it would seem that the responsible behavior in this
condition would be to trigger a crash. CSCsm53392 certainly suggests as
much (applying to DFC's), though TAC asserts that there are too many 
cases where "this would end up in a reboot cycle requiring manual 
intervention" to be proper (my own inclination being that this
degenerate case is still desirable) to allow an admin to "correct" the
configuration that caused it.

I've been trying to think of cases where a fibdisable would occur outside
of a bug-condition and the only thing that comes to mind is a FIB
overflow (which as discussed extensively here, is at least purportedly
handled gracefully now). 

Are there cases I'm not thinking of that make non-bug conditions for this
this so much more common that destruction of the forwarding plane is
desirable over a crash? Though I'm somewhat inclined for a bit of EEM
to try to ensure a reload, the lack of an existing knob is vexing...