Jaeger - Distributed Tracing
Basic Information
- Company/Brand: Jaeger (CNCF Graduated Project)
- Founder: Yuri Shkuro (Uber Technologies)
- Country/Region: Global Open Source Community (Originated from Uber)
- Official Website: https://www.jaegertracing.io/
- GitHub: https://github.com/jaegertracing/jaeger
- Type: Open Source Distributed Tracing Platform
- Founded: 2015 (Internal at Uber), Joined CNCF in 2017
- Status: CNCF's 7th Graduated Project (Graduated in October 2019), Celebrating its 10th Anniversary
Product Description
Jaeger is an open-source distributed tracing platform hosted by CNCF, focusing on providing robust, scalable, and pluggable tracing storage and query engines. Jaeger v2, released at the end of 2024, places OpenTelemetry at its core, allowing Jaeger to focus on its core mission while delegating instrumentation and data collection to OTel. Jaeger v1 will be fully deprecated by January 2026.
Core Features/Characteristics
- Distributed Tracing: End-to-end request tracing and latency analysis
- Root Cause Analysis: Locate performance bottlenecks and error root causes through tracing data
- Service Dependency Graph: Visualize call relationships between microservices
- Adaptive Sampling: Intelligently control tracing volume to avoid data overload
- Jaeger v2 Architecture: Built on OpenTelemetry Collector
- Multiple Storage Backends: Cassandra, Elasticsearch, Kafka, Badger, ClickHouse (Official support coming soon)
- Native OTLP Support: Native support for OpenTelemetry Protocol
- Rich UI: Tracing visualization, comparison, and analysis
Business Model
- Completely Free and Open Source: Apache 2.0 License
- CNCF Support: Hosted by Cloud Native Computing Foundation
- Commercial Hosting: Multiple vendors offer Jaeger hosting services
- Grafana Tempo (Compatible with Jaeger queries)
- Elastic APM (Supports Jaeger data)
Deployment Methods
- All-in-one single process (Development/Testing)
- Distributed deployment (Production environment)
- Kubernetes Operator
- Docker/Docker Compose
- Helm Chart
Target Users
- Microservices architecture operations teams
- DevOps and SRE engineers
- Developers needing tracing analysis
- Kubernetes users
- CNCF/OpenTelemetry ecosystem users
Competitive Advantages
- CNCF Graduated Project with a mature community
- Jaeger v2 is future-oriented with OpenTelemetry
- Adaptive sampling intelligently controls data volume
- Componentized architecture for independent scaling
- Native OTLP support
- Official support for ClickHouse as a storage backend coming soon
Comparison with Competitors
| Dimension | Jaeger | Zipkin | Tempo |
|---|---|---|---|
| CNCF Status | Graduated | Non-CNCF | Non-CNCF |
| OTel Integration | Native (v2 core) | Requires Exporter | Native |
| Architecture | Componentized and Scalable | Monolithic | Grafana Integrated |
| Sampling | Adaptive Sampling | Fixed Sampling | Basic Sampling |
| Scale | Large-scale Deployment | Small to Medium Scale | Large-scale Deployment |
| Storage | Multiple Backends | Multiple Backends | Object Storage |
Relationship with OpenClaw Ecosystem
Jaeger is the core choice for distributed tracing in the OpenClaw ecosystem. When OpenClaw's AI agent call chains involve multiple microservices, Jaeger provides end-to-end request tracing, latency analysis, and root cause localization. The deep integration of Jaeger v2 with OpenTelemetry ensures seamless integration with OpenClaw's observability architecture. The maturity and stability of the CNCF Graduated Project provide confidence for production environments.
External References
Learn more from these authoritative sources: