Retention Limits

The metadata required to identify the rows that have changed between checkpoints is stored in journals. These journals are cleaned up as checkpoints are copied to mirror grids.

Limiting the Size of Journals

A limit on the size of the journals can be specified when creating the data grid by using the checkpoint_journal_max_bytes setting. When the journal exceeds this size, ActiveSpaces begins deleting journal rows, starting with the oldest available. The journals may be disabled entirely by setting the checkpoint_journals option to disabled. If journals are disabled or if the maximum size is set too low, ActiveSpaces is forced to rely on bulk mirroring, which results in more network traffic between data grids. For more information about bulk mirroring, see "Bulk Mirroring" in TIBCO ActiveSpaces® - Enterprise Edition Concepts.

Limiting the Size of Checkpoints

You can also delete both manual and periodic checkpoints by setting the checkpoint_retention_limit property. By default, it is 0, which means that the older checkpoints are not removed. This setting ensures that the number of checkpoints do not keep growing and consuming disk space.