Restoring Data on the Standby Node
standby_page_repair
Parameter description: Specifies whether to automatically repair the page after a fault is detected on the replay target page during replay on the standby node. The current version supports three types of rectification: CRC failure, uninitialized page, and page LSN error. Bad blocks of VM, Xlog, Clog, MOT, and UNDO metadata cannot be repaired, and LSN errors of FSM cannot be repaired. UNDO records can be repaired only when CRC fails. Temporary tables, global temporary tables, and unlogged tables do not involve primary and standby restoration. The page file-level loss cannot be repaired.
Parameter type: Boolean.
Unit: none
Value range:
- on: The standby node automatically detects and repairs pages during replay.
- off: The standby node does not automatically detect and repair pages during replay.
Default value: on

If a large number of bad blocks exist due to a standby node exception, the replay performance of the standby node is affected, and Xlogs may be stacked on the standby node. In addition, an exception log is printed each time the Xlog is replayed to the bad block.
To ensure the consistency between the primary and standby nodes, if the standby node instance has unrepaired bad blocks, the checkpoint needs to be stopped. As a result, Xlogs may be stacked, and the replay on the bad blocks will be suspended, which may cause inconsistency between the primary and standby nodes. If a bad block exists when the standby node is restarted or promoted to primary, the instance exits abnormally. Before the instance exits, the Xlog chain on the bad block is collected, which may prolong the RTO.
Setting method: This is a SIGHUP parameter. Set it based on instructions provided in Table 1.
Setting suggestion: Retain the default value.
Risks and impacts of improper settings: If automatic repair is disabled, bad blocks will be stored in the bad block table once bad blocks are detected, causing the adverse impact described in NOTICE.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot