Recovery algorithms are techniques to ensure database consistency, transaction atomicity, and durability despite failures.

The key primitives that used in recovery algorithms are UNDO and REDO.

• UNDO: The process of removing the effects of an incomplete or aborted transaction.
• REDO: The process of re-instating the effects of a committed transaction for durability.

1. Failure Classsification

<aside> 💡 Storage Types

• Volatile Storage
• Non-volatile Storage
• Stable-Storage → A non-existent form of non-volatile storage that survives all possible failures scenarios. </aside>

1.1 Transaction Failures

A transaction reaches an error and must be aborted

• Logical Errors → Transaction cannot complete due to some internal error condition (e.g., integrity constraint violation).
• Internal State Errors → DBMS must terminate an active transaction due to an error condition

1.2 System Failures

Unintented failures in hardware or software that must also be accounted for in crash recovery protocols.

• Software Failure → There is a problem with the DBMS implementation (e.g., uncaught divide-by-zero exception) and the system has to halt.
• Hardware Failure → The computer hosting the DBMS crashes (e.g., power plug gets pulled).
  • Fail-stop Assumption: Non-volatile storage contents are assumed to not be corrupted by system crash.

1.3 Storage Media Failure

Non-repairable failures that occur when the physical storage machine is damaged.

When the storage media fails, the DBMS must be restored from an archived version.

• Non-Repairable Hardware Failure → A head crash or similar disk failure destroys all or parts of non-volatile storage.
  • Destruction is assumed to be detectable.