Monitoring stack

As explained in ScyllaDB in Action

A monitoring stack is a collection of software components that collects, stores, analyzes, and presents operational system measurements.

A monitoring stack combines components that collect, analyze, visualize, and alert on operational measurements. For ScyllaDB, Prometheus ingests node metrics, Grafana displays dashboards, and Alertmanager sends notifications when configured thresholds are crossed. section 11.1 section 11.4

Figure 11.1 Prometheus pulls metrics from your database nodes, which are used by the downstream components to provide visualizations and alert notifications.

Prometheus pulls metrics from database nodes; downstream components use those metrics for dashboards and alert notifications. section 11.1

A practical monitoring stack provides:

• Prometheus targets configured database hosts, scrapes their metrics endpoints, and supports filtering, transformation, and aggregation with PromQL. section 11.1.1 section 11.4
• Grafana reads Prometheus data to provide dashboards, visualizations, and preconfigured performance views. section 11.1 section 11.4
• Alertmanager receives alerts, routes and filters them with labels, and includes diagnostic details through annotations. section 11.4
• Production monitoring can extend beyond metrics with log aggregation and application traces that correlate activity across nodes. section 11.4

Deploying a Scylla monitoring stack

Download and unpack the monitoring files, configure Prometheus with the scylla-reviews hosts, generate Prometheus and Grafana configuration, create dashboards for Scylla 5.4, and run the three containers with docker compose up --detach. The resulting containers provide alerting, dashboards, and metric ingestion. section 11.1.1

How ScyllaDB in Action covers it

How deep it goes

Explains component flow, configuration, dashboards, PromQL, recording rules, alerts, and deployment, but stops short of detailed incident remediation.

Examples

Uses Scylla nodes, the scylla-reviews cluster, Docker Compose, generated Grafana dashboards, and cassandra-stress load tests.

What it assumes

Assumes a Scylla cluster and Docker-based deployment, while introducing the monitoring components and their connections.

What you can do after

Operate/configure a Docker-based monitoring stack and diagnose cluster behavior through metrics, dashboards, alerts, logs, and traces.

How it connects

Connects Prometheus, Grafana, Alertmanager, metrics, PromQL, recording rules, log aggregation, traces, and load testing.

liveBookmanning.com

Read ScyllaDB in Action on liveBook Get the book at manning.com

Related concepts

• Prometheushas part · Prometheus is an open-source monitoring and alerting system that collects, stores, and queries time-series metrics.
• Alertmanagerhas part · Alertmanager is a Prometheus component that groups, routes, silences, suppresses, and delivers alerts to notification channels.
• Grafanahas part · Grafana is an open-source observability and visualization platform for querying, exploring, and displaying data from multiple sources and configuring alerts.
• alertingproduces · Alerting is the generation and delivery of notifications when monitored conditions or events meet defined criteria or require attention or action.
• Dockerconfigured by · Docker is a platform and ecosystem for building, distributing, and running applications in containers.
• cluster topologymeasures · Cluster topology is the arrangement, connectivity, roles, and partition relationships of nodes within a computing cluster.

This explainer was generated by Manning's knowledge graph from the text of ScyllaDB in Action. The section links point to the parts of the book it draws on.