Automatic System recovery of SPARC T server

Hello

Would you tell me if public doc is available about Automatic System recovery of SPARC T servers ?

What I want to know is, for example,

1. When a core / a thread is failing in POST.
2. When a DIMM is failing in POST.
3. When whole CMP is failing in POST.

I believe POST/OBP will manage the failure during self test as much as they can but
I'm really interested in how the failure in POST is managed and how the system will
bring up with limitted resources.

Thank you very much for your help.
Tomonori-