Best practice for choosing RID

[Peter Mogensen <apm@mutex.dk>]

Hi,

I'm moving more and more towards loading new databases via ldapadd to 
cn=config.
All databases are replicated between two mirrormode servers - including 
cn=config.

From my unserstanding I need to syncrepl directives for each database.
For cn=config it could be:

syncrepl rid=1
   provider=ldaps://<SERVER1>/
   searchbase="cn=config"
   type=refreshAndPersist
   retry="60 +"
   scope=sub
   schemachecking=on
   bindmethod=sasl
   binddn="cn=config"
   saslmech="EXTERNAL"
   tls_cert=/etc/ldap/ssl/config.crt
   tls_key=/etc/ldap/ssl/config.nopass.key
   tls_cacert=/etc/ldap/ssl/ca.crt
   tls_cipher_suite="NULL-SHA"

syncrepl rid=2
   provider=ldaps://<SERVER2>/
   searchbase="cn=config"
   type=refreshAndPersist
   retry="60 +"
   scope=sub
   schemachecking=on
   bindmethod=sasl
   binddn="cn=config"
   saslmech="EXTERNAL"
   tls_cert=/etc/ldap/ssl/config.crt
   tls_key=/etc/ldap/ssl/config.nopass.key
   tls_cacert=/etc/ldap/ssl/ca.crt
   tls_cipher_suite="NULL-SHA"

The two directives are needed since the configuration will be replicated 
and the two servers have to contain the same configuration. As long as 
Server-2 can recognize it self as <SERVER2> it will ignore the directive 
 pointing to it self - correct?

So cn=config will use RID 1 and RID 2.

When I add a new database it will also have to have 2 syncrepl 
directives, but will have to use RID 3 and RID 4 - correct?

Now, I would like not to have to specify the RID in the LDIF loaded to 
create a new database. In principle I would not know how many other 
replications were in place on the server I load it into, so it would be 
a bad thing to hardwire the RID into the LDIF for the new database.

Are there any mechanism/best practice to do chose new RIDs automatically?

/Peter