clustermode rdb

repicated database = ring of atomically updated database of which one node is the master.

rdb communication is like the heartbeat.

rdb units: bcom (block configuration and operations manager

: vldb ( volume location database)

: vifmgr

: management

related to components: M-host CSM N-Blade D-Blade

====

RDB cannot be backed up!!!

if you lose all nodes you will have no RDB anymore....

RDB is viewable in systemshell /mroot/etc/cluster_config/rdb

/mroot contains .snapshot.

You could rsync the .snapshot to another node and have the rdb

overthere. (compared to SUNCLUSTER ccr)