Fast recovery after a checkpoint

Fast recovery returns the database server to a consistent state as part of shared-memory initialization. All committed transactions are restored, and all uncommitted transactions are rolled back.

Fast recovery occurs in the following steps:
  1. The database server uses the data in the physical log to return all disk pages to their condition at the time of the most recent checkpoint. This point is known as physical consistency.
  2. The database server locates the most recent checkpoint record in the logical-log files.
  3. The database server rolls forward all logical-log records written after the most recent checkpoint record.
  4. The database server rolls back all uncommitted transactions. Some XA transactions might be unresolved until the XA resource manager is available.

Copyright© 2018 HCL Technologies Limited