Distributed Tracing

Kind: mechanism

Layer: Observability

Record: architecture:distributed-tracing

Severity: contextual

Scope: distributed system, service mesh

Canonical: Ontology

A mechanism that propagates a trace context across service calls and records each call as a timed span of one trace.

Listed in Architecture principles, after Causation ID and before SLO/SLI.

Requires

• Trace Context Propagation

Reinforces

• Observability
• Causality

Enables

• Latency/Failure Root Cause Analysis

Conflicts with

• Opaque Distributed Calls

In tension with

• Overhead/Sampling

Tensions

• Distributed Tracing / Overhead/Sampling

Violated by

• Opaque Distributed Calls

Severity

• contextual

Category

• Observability / Auditability / Traceability

Refactors

• Error Handling
• Traceability

Reinforced by

• Correlation ID

Linked from

• Causality / Ordering / Distributed Time
• Error Handling / Resilience
• Observability / Auditability / Traceability
• Observability / Auditability / Traceability
• Severity levels
• The resolutions