Redis: Disaster Recovery Plan

Since Redis functions as a cache for a PostgreSQL database and does not include critical data, the disaster recovery plan prioritizes service availability and quick recovery.

Architecture

Primary-Standby Nodes: Redis operates with a primary node and a standby replica for high availability. Standby nodes  automatically replace the primary node in the case of a failure, ensuring your data stays available.

Common Failure Scenarios and Recovery Steps

1. Primary Node Failure:

Impact:

• The cluster remains operational.

• Standby node is automatically promoted to replace it

Recovery Steps:

• Follow DigitalOcean instruction to fix the issue

2. Region Failure:

Impact:

• The cluster becomes unavailable.

Recovery Steps:

• Deploy Redis cluster in another DigitalOcean region

• Set new username/password in the application variables