If you choose the all synchronous or hybrid method, you must select a protection mode. A protection mode specifies what HiRDB does when synchronous copy to the remote site fails. The table below shows the criteria for selecting a protection mode.

Note that you specify the selected protection mode in the pd_rise_fence_level operand.

Table 2-1 Criteria for selecting a protection mode

Protection mode	What HiRDB does when synchronous copy fails	Advantage	Disadvantage
data	Stops updating at the main site (updating of the volume containing the file for which synchronous copy failed).	Integrity is always maintained between the main site and the remote site.	An error at the remote site impacts part, or all, of the main site. Most critically, when a link failure occurs between the main site and the remote site, none of the volumes at the main site can be updated. In some cases, this results in the HiRDB system at the main site terminating abnormally.
never	Continues updating at the main site.	Transactions at the main site continue even when a synchronous copy error occurs.	The following may occur until the error is eliminated and integrity can be restored between the main site and the remote site. The HiRDB system at the remote site cannot be restarted. Some data may be lost during site switchover. Furthermore, because the remote site may not be able to detect that a failure has occurred, the integrity of the applicable paired logical volume groups must be monitored and guaranteed.

Reference note

Asynchronous copy always operates in the never protection mode.