Data loss considerations

The cluster services might lose data under certain circumstances.

Distributed session cache

• The policy data, used to indicate the first port that is available for use by the cluster, is changed.
• The policy data that defines the masters is changed.

Configuration database

An appliance that is operating as a single node cluster fails. In this situation, we must rely on snapshot information to restore the configuration database.

Internal runtime database

• An appliance that is operating as a single node cluster fails. In this situation, there is no recovery possible.
• The primary master fails, and no secondary master is configured.
• The maximum size of the internal runtime database is adjusted such the new maximum size is smaller than the existing database.

Parent topic: Cluster support