Overview

Keeping Rayls Sovereign running well depends on effective monitoring. Tracking metrics such as CPU usage, memory consumption and transaction throughput lets you detect and resolve issues before they escalate. This page introduces monitoring for Rayls Sovereign and points to the fuller instructions in the Rayls Infrastructure section.

Monitoring Tools and Best Practices

The Rayls Sovereign infrastructure supports a range of monitoring tools and techniques. Below are the recommended practices for setting up your monitoring environment.

• Use Prometheus and Grafana: Together these collect and visualise metrics from the Sovereign ledger. The ledger exposes its metrics endpoint when it is started with the metrics flags, and the metrics worth alerting on include the current block number, the pending transaction pool depth, memory use and goroutine count.
• Log Management: Extract and analyse logs from Rayls components to detect errors or unusual behaviour.
• Database Monitoring: The Relayer keeps its state in PostgreSQL, and that is the database to watch for cross-institution message delivery. The Sovereign ledger holds its consensus and execution state in the node's own embedded store rather than in a separate database server. MongoDB appears only where an institution has deployed the optional Rayls Backend, which uses it for wallet and custody records.

For a full guide to setting up monitoring and integrating tools such as Grafana and Prometheus, refer to the Rayls Infrastructure section of the Rayls product documentation.

Explore Infrastructure Monitoring

Key Metrics to Monitor

To maintain the health of Rayls Sovereign, monitor the following metrics:

• CPU and Memory Usage: Confirm that Rayls Sovereign and its supporting components are neither under- nor over-utilised.
• Disk Space: Track storage usage, particularly the ledger's own data directory and the Relayer's PostgreSQL volume.
• Transaction Throughput: Monitor the number of transactions processed per second to identify bottlenecks or performance issues.
• Error Rates: Watch the error logs to catch failed transactions or problems with the Relayer.

For more detail on which metrics to monitor for each component, see the Rayls Infrastructure section.