Rematerializing the primary database for failback in ASE 12.5.1 or later

Normally, the primary database is the source of all data and transactions replicated in the Replication Server Heterogeneous Replication Options system. Rematerializing the primary database is required only as part of a failback procedure, to restore normal system operations after a failover.

After a failover event, the standby database becomes the “active” database in the system, and the primary database must be rematerialized from the standby database to restore the normal operating condition, and resume replication from the primary database to the standby database.

When you rematerialize a primary database, the primary database is the target, and the standby database is the source in the materialization procedure.

Table 2-4 provides a checklist of the snapshot rematerialization tasks for a primary (target) database in ASE version 12.5.1 or later. This checklist assumes the following:

The primary (target) ASE is already installed and configured identically to the standby (source) ASE.

The primary (target) database exists in the primary (target) ASE, and its database options and devices are configured identically to the standby (source) database.

All server logins defined in the standby (source) ASE exist in the primary (target) ASE, with identical suid values and names, and all server roles defined in the standby (source) ASE exist in the primary (target) ASE.

**Table 2-4: Rematerializing a primary database in ASE 12.5.1 or later**
Task	Description
1	Quiesce the standby (source) database to suspend all update activity and generate a manifest file for the standby (source) database. The standby (source) database remains suspended during failback, until after the materialization procedure is complete for the primary (target) database.
2	Use the disk replication system to copy a snapshot of all standby (source) database data and log devices to the primary (target) database devices.
3	In the primary (target) ASE, mount the primary (target) database devices (created in step 3) and bring the primary (target) database online.

After you complete the primary database rematerialization, you must perform additional tasks to complete the failback procedure. See the Mirror Activator Administration Guide.

Quiesce the standby (source) database to suspend all update activity, and generate a manifest file for the standby (source) database.

Log in to the standby (source) ASE with a system administrator user role, and execute the following command:
```
quiesce database MA_setup hold sdb
for external dump to sdb_manifest
```
where:
- MA_setup is a user-defined tag that identifies the database.
- sdb is the name of the standby (source) database.
- sdb_manifest is the name of the manifest file.
The primary (target) ASE uses the manifest file to mount the devices created by the disk replication system in the following step.
Use the disk replication system facilities to do the following:
- Capture a snapshot (or point-in-time) image of all standby (source) database data and log devices
- Transfer the snapshot to the primary (target) devices at the primary (target) site
While the standby (source) database is suspended during failback, you must also:
- Transfer the snapshot of the standby (source) database log devices to the mirror log devices
- Configure the disk replication system for synchronous replication from the primary (target) log devices to the mirror log devices (to prepare for normal system operation after failback)
See the documentation provided by your disk replication system vendor for more information about the procedures in this step.
Mount the primary (target) devices in the primary (target) ASE to recover the primary (target) database, and then bring it online.
1. Log in to the primary (target) ASE with a system administrator user role, and execute the following command:
```
mount database all from sdb_manifest
with listonly
```
  where sdb_manifest is the manifest file created by the quiesce command at the standby (source) database.
  
  The mount command with listonly option returns the device paths specified at the standby (source) ASE for all standby (source) database data and log devices.
  
  If necessary, invoke the mount command to “remap” the device paths to the primary (target) devices in the primary (target) ASE. For example:
```
mount database all from sdb_manifest 
using "pdb_path" = "sdb_data"
```
  where:
  - sdb_manifest is the manifest file created by the quiesce command at the standby (source) database.
  - pdb_path is the path to the primary (target) database data device.
  - sdb_data is the device name of the standby (source) database data device specified in the standby (source) ASE.
  When you invoke mount, ASE performs all required supporting activities, including adding database devices and activating them, creating the catalog entries for the new database, and recovering the database.
2. After the primary (target) ASE completes the mount processing, use the following command to bring the primary (target) database online:
```
online database pdb
```
  where pdb is the name of the primary (target) database.
The names of the primary database and standby database must be the same.