Cluster Storage File Replay Log for Fast Failover Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In clustered storage systems, when a storage system fails and undergoes a failover, the rapid error condition of the partner system can lead to data unavailability for hours or days due to incomplete parity recomputation, especially if storage devices are missing or inoperative, causing significant downtime.
Innovation Solution
The system implements a method where the partner storage system stores file system log information in a file replay log within its root volume and transmits this information back to the original storage system upon re-initialization, allowing the disks to be brought to a consistent state without requiring a full file system check.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the partner storage system performs parity recomputation to recover data after a failure, then data reliability is improved, but the time required for recovery increases significantly when storage devices are missing or inoperative
Solution Approach 1:
The partner storage system performs preliminary actions by storing file system log information in a file replay log during the takeover period. This log contains the necessary information to reconstruct the file system state, enabling the original storage system to resume operations without performing a full parity recomputation or file system check, thus reducing data unavailability time while maintaining data reliability
Solution Approach 2:
The patent creates a copy of the file system log information in the file replay log on the partner storage system. This copy allows the original storage system to restore its state using the replicated log data, avoiding the need to wait for complete parity recomputation or perform extensive file system checks, thereby reducing the time loss during recovery
2Reliability
If the storage system performs a full file system check after re-initialization, then data consistency is ensured, but the time required for consistency checking increases significantly
Solution Approach 1:
The partner storage system performs preliminary action by maintaining and storing file system log information in the file replay log during operation. When the original storage system re-initializes, this pre-stored log information allows it to quickly restore file system consistency without performing a time-consuming full file system check, thus ensuring data consistency with minimal time loss
Solution Approach 2:
The file replay log provides feedback information about file system operations to the original storage system during re-initialization. This feedback mechanism enables the system to understand the state of file system logs and restore consistency efficiently, avoiding the need for extensive checking while maintaining reliable data consistency
Data Source
AI summary
A system and method for storing and retrieving file system log information in a cluster computer system is provided. In response to an aborted takeover operation, a surviving storage system stores a set of file replay log information on storage devices associated with the surviving storage system. When the failed storage system re-initializes, it retrieves the file replay log information from the surviving storage system and utilizes the file replay log information to render its disks consistent.