Journal sequence number - impact of reset in middle of *CURCHAIN

Part 1:

Can someone identify, or provide me with a good RTFM reference (already
read through Backup & Recovery, but not sure if I caught them all), all of
the circumstances by which the journal sequence number can be reset to 1? 
These are the ones I know about:
 - CHGJRN SEQOPT(*RESET)
 - IPL (when not in CC recovery mode and receivers are managed 
         by the system [MNGRCV(*SYSTEM)] )

I am unclear on whether a reset happens when a system-managed receiver
reaches the threshold that causes a new receiver to be created.
 
Part 2:

Is it conceivable that a *CURCHAIN set of receivers may hold duplicate
sequence numbers?  A scenario involving multiple CHGJRN SEQOPT(*RESET)
within a short period of time, would not this cause multiple sequence
numbers of 1?  If so, what is the impact of using RCVJRNE that starts
"receiving" entries that precede the last reset?  In order to CMA, should
I be specifying the starting journal receiver by name?  (Currently, I am
using *CURCHAIN and a starting sequence number that is stored in a data
area that is updated after each entry is processed by the RCVJRNE exit
program.)  I think all this boils down to this question:  If I have
multiple occurences of sequence number 15 (for example) in the journal's
*CURCHAIN, what happens when I do a RCVJRNE RCVRNG(*CURCHAIN) FROMENT(15)?

TIA, GA

__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/