Log aggregation collects logs from multiple systems or nodes in a centralized store, making them searchable and easier to correlate. With a small ScyllaDB cluster and one application instance, operators can inspect logs manually; as systems grow, centralized aggregation becomes critical. section 11.1.5
A practical monitoring setup can use log aggregation to:
- Query and correlate log entries from multiple ScyllaDB nodes and application instances. section 11.4
- Pair logs with application traces to follow a request across services and identify where it spends time. section 11.1.5
- Use Grafana Loki when adopting the Prometheus and Grafana ecosystem, or select an existing organizational tool. section 11.1.5
Tracing a growing cluster
When an application begins reporting errors, centralized logs let an operator search application and ScyllaDB entries together. Traces can then connect increased latency to a particular endpoint or database partition. section 11.1.5
