Hi all, hope someone might be able to help with this question.
I've configured two 389-DS servers each on a Centos 6.7 server and successfully
configured MMR between them - all working well with 'always keep in synch'
configured.
However, I'm uncertain how to configure the failback behaviour for my LDAP client -
specifically:
I configure my client to bind to instance 1 and failover to instance 2 if instance 1
becomes unavailable
If after failover, instance 1 is brought back up some time later, then If I configure my
client to automatically fail back to instance 1 when it detects it as available is there a
risk that LDAP operations will commence against instance 1 before replication has
completed from instance 2 to instance 1?
If so, are there any ways to avoid this so that reading/writing stale data can be
avoided.
Many thanks,
Bernie
Show replies by thread