Help with T6340/B6000 NEM issue

Hi,
we had an unexpected outage on our B6000/T6340 blades tonight.

The core problem is that the hxge interfaces have swapped 0 & 1 
so that hxge0 is now hxge1 and vice verse.  I had a look at swapping the 
interfaces in /etc/path_to_inst and tried that and while it brought
up the interfaces correctly then they are not on the right port.

If I run up the system with the correct interface and mapped properly
in path_to_inst I get this:

Booting to milestone "svc:/milestone/multi-user:default".
hxge0: DL_ATTACH_REQ failed: DL_SYSERR (errno 22)
hxge0: DL_BIND_REQ failed: DL_OUTSTATE
hxge0: DL_PHYS_ADDR_REQ failed: DL_OUTSTATE
hxge0: DL_UNBIND_REQ failed: DL_OUTSTATE
hxge16000: DL_ATTACH_REQ failed: DL_SYSERR (errno 22)
hxge16000: DL_BIND_REQ failed: DL_OUTSTATE
hxge16000: DL_PHYS_ADDR_REQ failed: DL_OUTSTATE
hxge16000: DL_UNBIND_REQ failed: DL_OUTSTATE
Failed to plumb IPv4 interface(s): hxge0


If I swap the 0 & 1 for hxge in path_to_inst, the system boots and 
configures the network, but on the port that is not configured.

How do I fix this issue so that interface swapping does not occur?
I thought Sun solved these sorts of problems in the 1990's?

Can someone please help resolve this issue?  I have been called out 
cold on this and need resolution!  The problem has affected all of the 
blades in the chassis.....

-- 
Rachel Polanskis                 Kingswood, Greater Western Sydney, Australia
[email protected]                http://www.zeta.org.au/~grove/grove.html
 	The more an answer costs, the more respect it carries.