Step 1: Effect of crash during recovery.
When a database crashes during recovery and no checkpointing is used, the recovery process must restart from the beginning using the log.
Step 2: Undo and redo lists.
Since there is no checkpoint to record progress, the system cannot determine which undo or redo actions were completed before the second crash. Hence, the same undo and redo lists are reconstructed and applied again.
Step 3: Evaluation of options.
Statement (A) is correct because recovery restarts with the same undo and redo lists.
Statement (B) is incorrect since recovery is still possible.
Statement (C) is incorrect because operations may be repeated without checkpointing.
Statement (D) is incorrect because recovery mechanisms ensure consistency.
Let \( r_i(z) \) and \( w_i(z) \) denote read and write operations respectively on a data item \( z \) by a transaction \( T_i \). Consider the following two schedules.
\[ \begin{aligned} S_1 &: r_1(x)\; r_1(y)\; r_2(x)\; r_2(y)\; w_2(y)\; w_1(x) \\ S_2 &: r_1(x)\; r_2(x)\; r_2(y)\; w_2(y)\; r_1(y)\; w_1(x) \end{aligned} \] Which one of the following options is correct?