Unigueness of RID; changing RID

[Nick Urbanik <nick.urbanik@optusnet.com.au>]

Dear Folks,

I am trying to improve my understanding of the RID before making many
large deployments of syncrepl.

My understanding is that the replica ID (RID) is unique within one
level of [provider] --> [consumer], [consumer],... relationship.

Here, an arrow --> represents replication of one directory tree from
   provider to consumers, and
commas represent consumers at the same level, all replicating from the
   same provider, and the
square brackets [...] represent one machine.

1. If there is a relationship like this, where at least one machine
   acts simultaneously as consumer and provider
   [provider] --> [consumer+provider] --> [consumer], [consumer],...

   does the RID need to be unique within all these consumers at all
   levels in the propagation of replication?

2. What are the consequences of changing the RID on a consumer?
   Would this inevitably require a dump and restore?
   Is the RID stored in the data?  Where is it stored, besides in the
   consumer's syncrepl configuration?
--
Nick Urbanik http://nicku.org 808-71011 nick.urbanik@optusnet.com.au
GPG: 7FFA CDC7 5A77 0558 DC7A 790A 16DF EC5B BB9D 2C24  ID: BB9D2C24
I disclaim, therefore I am.