Full System Recovery: MySQL Agent
A full system recovery of a MySQL client involves the following phases:
Preparing for a Full System Recovery
Before you begin a disaster recovery, make sure to plan for disaster recovery.
Building a Standby Server for Disaster Recovery
You can build a standby server for quick recovery and ensure that the standby database is always kept in a near ready state. If you plan to adopt this method follow the steps described in Replication Using Warm Database Restore.
As a best practice, we recommend that you group databases into multiple subclients.
- Add larger databases into separate subclients.
- Add small databases together into one or more subclients. This is important for the following reasons:
- During Disaster recovery, when you have to rebuild the entire instance quickly, you can restore all of the subclients together.
- During Backup failures, the backup restarts from the beginning of the database instead of from the beginning of the entire instance. This ensures that large database backups are not affected by the restart of a smaller database.
- Once the subclients are created, schedule frequent backups for dynamic data and regular backup schedules for static data in the MySQL Server.
Full System Recovery
To recover the MySQL client in the event of a disaster, perform the following tasks:
The system-generated MySQL database can become lost or corrupted. When that happens, the MySQL Server and the user-created databases are unusable, even though they are still present. In this case, you must restore the latest backup of the MySQL system database using the appropriate File System iDataAgent.
After you restore the MySQL system database, make sure that the file permissions are the same as when the database was backed up. Otherwise, the MySQL service does not come up.