Service mesh

Intent

Use infrastructure proxies to provide service-to-service networking features (mTLS, routing, retries, telemetry) consistently.

Use when

• You need consistent networking/security controls across many services and languages.
• You want standardized telemetry and traffic management (canaries, routing rules).

Avoid / watch-outs

• A mesh doesn’t remove the need to design correct app semantics (idempotency, time budgets, error mapping).
• Operational complexity is high; ensure you can operate and debug the mesh itself.

Skill mapping

• architecture: decide what belongs in mesh vs app and document failure semantics.
• observability: correlate app telemetry with proxy/mesh telemetry.
• resilience: ensure retries/timeouts align across layers to avoid retry storms.