Introduction
Graylog is an open-source log management platform that centralizes logs from servers, applications, and network devices into a searchable index. It combines collection, parsing, alerting, and dashboarding in a single tool designed for operations and security teams who need fast answers from high-volume log streams.
What Graylog Does
- Ingests log data via Syslog, GELF, Beats, and raw TCP/UDP inputs at thousands of messages per second
- Provides a full-text search interface backed by OpenSearch or Elasticsearch for sub-second log queries
- Parses and enriches incoming messages with extractors, pipeline rules, and lookup tables
- Triggers alerts based on search conditions with notification to Slack, PagerDuty, email, or HTTP endpoints
- Renders real-time dashboards with aggregation widgets, heatmaps, and statistical summaries
Architecture Overview
Graylog runs three core components: the Graylog server (Java), MongoDB for configuration and metadata storage, and OpenSearch (or Elasticsearch) for log indexing. Inputs receive messages over the network and hand them to a processing pipeline that applies extractors, pipeline rules, and stream routing. Messages are then written to time-based OpenSearch indices. The web UI is built into the server process and communicates via a REST API that is also available for automation.
Self-Hosting & Configuration
- Deploy with Docker Compose for evaluation or use the official Linux packages (DEB/RPM) for production
- Configure inputs in the web UI to receive Syslog (RFC 5424), GELF, or Beats protocol data
- Define streams to route messages by source, severity, or content into separate index sets with independent retention policies
- Write pipeline rules in Graylog's processing language to parse, enrich, and transform messages on ingest
- Set index rotation by time or size and configure retention to automatically close and delete old indices
Key Features
- Content packs bundle inputs, extractors, dashboards, and pipeline rules into shareable, importable packages
- Decorators modify search results on the fly without changing stored data, adding context like GeoIP fields
- Correlation engine groups related events across multiple log sources for security investigation workflows
- Sidecar manages Filebeat and Winlogbeat agents centrally from the Graylog UI
- Role-based access control restricts stream and dashboard visibility per team or user
Comparison with Similar Tools
- Grafana Loki — log aggregation optimized for labels, not full-text indexing; Graylog offers richer search and built-in parsing
- Elasticsearch + Kibana (ELK) — flexible but requires assembling ingest, UI, and alerting separately; Graylog integrates all three
- Splunk — enterprise leader in log analytics; Graylog provides similar core features under an open-source license
- Fluentd / Fluent Bit — log shippers and processors; Graylog is the destination that indexes, searches, and visualizes
- SigNoz — APM-focused observability; Graylog specializes in log management with deeper parsing and alerting features
FAQ
Q: What is the difference between Graylog Open and Graylog Enterprise? A: Graylog Open (formerly Open Source) covers core log management. Enterprise adds features like audit logging, archiving, and the correlation engine under a commercial license.
Q: Can Graylog replace the ELK stack? A: For log management, yes. Graylog uses OpenSearch or Elasticsearch as its indexing backend but replaces Logstash with its own inputs and pipelines, and replaces Kibana with its built-in UI.
Q: How much storage does Graylog need? A: Storage depends on message volume and retention. A typical production setup indexes 50-100 GB per day and retains 30-90 days of hot data, with older indices optionally archived to cold storage.
Q: Does Graylog support Kubernetes logs? A: Yes. Deploy Fluent Bit or Filebeat as a DaemonSet to forward container logs to Graylog via GELF or Beats inputs.