Master Redis Monitoring, Migration, and Cluster Management with Prometheus and CacheCloud

Basic Monitoring Command: INFO

Redis provides the INFO command, which returns rich instance monitoring data and serves as the foundation for most Redis monitoring tools. The command can accept a section argument to return specific categories of information, which are grouped into five major sections.

Key parameters that operators should monitor include stat , commandstat , cpu , and memory for command execution counts, CPU usage, and memory consumption, as well as persistence when RDB or AOF is enabled, and replication for master‑slave sync status.

Prometheus‑compatible Redis‑exporter Monitoring

Prometheus is an open‑source monitoring and alerting framework that scrapes metrics from monitored systems and visualizes them with Grafana . Exporters are plugins that expose metrics in a format Prometheus understands.

The Redis‑exporter collects data from the INFO command and presents it to Prometheus for visualization and alerting. It supports Redis versions 2.0 through 6.0 and can also monitor key‑value sizes and collection element counts via the

check-keys

option.

Lightweight alternatives such as redis‑stat (Ruby) and Redis Live (Python) also visualize INFO data, useful for custom monitoring tool development.

Data Migration Tool: Redis‑shake

Redis‑shake is a data migration tool developed by Alibaba Cloud for synchronizing Redis data between instances. It starts a simulated Redis process that performs full‑volume sync from the source instance, transfers the RDB file, and then streams incremental commands to the target.

The tool supports single‑instance migration, cluster‑to‑cluster migration, and migration involving proxy‑based sharding solutions such as Codis. It also works across on‑premise and cloud Redis deployments, facilitating cloud migration scenarios.

Data Consistency Check Tool: Redis‑full‑check

Redis‑full‑check compares data between source and target instances to verify consistency. It operates in multiple rounds, first collecting all keys, then comparing values, and storing mismatches in a SQLite database. Subsequent rounds only re‑check previously mismatched keys.

The tool offers three comparison modes controlled by the

comparemode

parameter:

KeyOutline : compare only key existence.

ValueOutline : compare only value length.

FullValue : compare key existence, value length, and full value.

Users can set the number of comparison rounds with the

comparetimes

option, balancing thoroughness against impact on live traffic.

Cluster Management Tool: CacheCloud

CacheCloud is a cloud platform developed by Sohu for Redis operations, providing automated deployment and management of master‑slave, sentinel, and Redis Cluster architectures. It offers five core operational actions: instance offline, instance online, add replica, manual failover, and configuration management.

Beyond operations, CacheCloud collects and visualizes real‑time INFO metrics, stores historical data such as memory usage, client connections, and key counts, enabling administrators to analyze incidents with contextual performance data.

Overall, CacheCloud serves as a unified portal for managing Redis instances, streamlining routine maintenance and monitoring tasks.