Concurrent Primary Standby Database Upgrade via NFS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for upgrading primary and standby databases require significant downtime, as they need to be taken offline sequentially to maintain data integrity, which can lead to unacceptable inaccessibility and risks during the upgrade process.
Innovation Solution
Concurrent upgrading of primary and standby databases is achieved by mounting both to an NFS mount point, generating and copying level-zero and incremental backups, and synchronizing metadata to minimize downtime and ensure data consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If databases are upgraded sequentially one at a time to maintain data integrity, then data reliability is improved, but system unavailability increases
Solution Approach 1:
The upgrade process is segmented into distinct phases: pre-upgrade backup, concurrent upgrade execution, and post-upgrade validation. The standby database is segmented from the primary database during the upgrade process, allowing independent upgrade of each while maintaining data integrity through controlled cutover points.
Solution Approach 2:
A backup of the standby database is created before the upgrade process begins. This preliminary action ensures that if any issues arise during the concurrent upgrade, the system can rollback to the backup state, maintaining data reliability while enabling concurrent upgrade operations.
2Manufacturing precision
If databases are taken offline for upgrade to ensure data consistency, then manufacturing precision (data consistency) is improved, but productivity decreases
Solution Approach 1:
The upgrade operations for both primary and standby databases are merged into a single concurrent execution window. By combining the upgrade processes and using shared backup resources, the system achieves data consistency without requiring sequential offline upgrades, thus maintaining productivity.
Solution Approach 2:
The database systems continue to serve production workloads during the upgrade process. The standby database remains operational and synchronized with the primary database, ensuring continuous data availability and system productivity throughout the concurrent upgrade operation.
3Reliability
If standby database remains synchronized with primary database during upgrade, then reliability is improved, but device complexity increases
Solution Approach 1:
The standby database acts as an intermediary during the upgrade process. It receives and applies changes from the primary database, then serves as the new primary after cutover. This intermediary role simplifies the overall upgrade process by providing a stable target for concurrent operations and reducing direct complexity in the primary database upgrade.
Data Source
AI summary
Methods and systems are provided for concurrently upgrading a primary database and a standby database that is synchronized with the primary database, while still protecting the stored data from the risk of hardware or other failure during the upgrade process. The standby database is mounted to an NFS (Network File System) located on a database access node. The upgraded primary and standby databases are built and mounted to the NFS mount point. A level-zero backup and one or more incremental backups of the deprecated standby database are generated. Each backup may be concurrently transferred to the upgraded databases via the mount point. Upon generation of a final incremental backup, the primary database is frozen and the tablespace metadata to transferred to the upgraded databases via the mount point. The upgraded primary database may be started upon importing of the tablespace metadata.