📜 Diagram for Lesson 8: Activity Retry + Timeout

Visualizing Resilient Activity Execution

Retry patterns and timeout strategies for resilient activity execution

Activity Retry Flow with Exponential Backoff

sequenceDiagram participant W as Workflow participant TS as Temporal Server participant A as Activity participant ES as External Service Note over W,ES: Resilient Activity Execution with Retries W->>TS: Schedule Activity with RetryOptions Note over TS: StartToCloseTimeout: 2min

MaxAttempts: 3

InitialInterval: 1s TS->>A: Execute Activity (Attempt 1) A->>ES: Call External Service ES-->>A: Timeout/Failure A-->>TS: Activity Failed Note over TS: Wait 1s (InitialInterval) TS->>A: Execute Activity (Attempt 2) A->>ES: Call External Service ES-->>A: Timeout/Failure A-->>TS: Activity Failed Note over TS: Wait 2s (Backoff x2) TS->>A: Execute Activity (Attempt 3) A->>ES: Call External Service ES->>A: Success Response A->>TS: Activity Completed TS->>W: Activity Result

Retry Strategy by Activity Type

graph TD A[Activity Type] --> B{Operation Category} B -->|Critical| C[Payment Processing] B -->|Standard| D[Validation] B -->|Long-Running| E[File Processing] C --> F[MaxAttempts: 5

LongTimeout

Conservative Backoff] D --> G[MaxAttempts: 3

MediumTimeout

Standard Backoff] E --> H[MaxAttempts: 10

Heartbeat Required

Aggressive Backoff] style C fill:#ffebee style D fill:#e8f5e8 style E fill:#e3f2fd style F fill:#ffcdd2 style G fill:#c8e6c9 style H fill:#bbdefb

💡 Key Insights from the Diagram

Retry Flow Characteristics:

✅ Automatic retry handling by Temporal Server
✅ Exponential backoff prevents service overload
✅ Configurable max attempts prevents infinite retries
✅ Success on final attempt shows resilience in action

Strategy Differentiation:

🔴 Critical activities (payments) → Conservative, fewer attempts
🟢 Standard activities (validation) → Balanced approach
🔵 Long-running activities (file processing) → More attempts, heartbeats

Timeout and Heartbeat Strategy

Timeout Hierarchy:

ScheduleToStartTimeout → Maximum queue wait time
StartToCloseTimeout → Single execution time limit
ScheduleToCloseTimeout → Total time including all retries
HeartbeatTimeout → Progress reporting frequency

Benefits:

Prevents stuck workflows through proper timeout configuration
Enables progress monitoring through heartbeat reporting
Balances resilience with resource efficiency

🚀 Production Resilience

This diagram demonstrates production-ready patterns for:

✅ Service failure recovery through smart retry policies
✅ Resource protection via timeout boundaries
✅ System stability through exponential backoff
✅ Operational visibility via heartbeat monitoring

Building fault-tolerant distributed systems! 🎉

Diagram