Flow Studio
Timing and Retry Info
The timing section of the Node Inspector displays when the node started, when it completed, how long it took, and how many automatic retries were used. These metrics help diagnose slow nodes and retry-heavy executions.
Timing Fields
| Field | Source Event | Format |
|---|---|---|
| Started At | NodeExecutionStarted.startedAt | Local time: HH:MM:SS.mmm |
| Completed At | NodeExecutionCompleted.completedAt or NodeExecutionFailed.failedAt | Local time: HH:MM:SS.mmm |
| Duration | NodeExecutionCompleted.durationMs | Human-friendly (e.g., "342ms", "1m 2s") |
| Retry Count | NodeExecutionCompleted.retryCount | Integer. 0 = no retries, N = N additional attempts |
Duration Includes Retry Time
The durationMs value reported in the NodeExecutionCompleted event covers the total wall-clock time from the first attempt to the final result — including all retry delays. If a node was configured with a 5-second retry backoff and required 3 retries, the durationMs will be significantly larger than the actual execution time of one attempt.
// Example: HTTP node with 3 retries, 2s backoff each
// Attempt 1: 1s execution → fail
// Wait 2s
// Attempt 2: 1s execution → fail
// Wait 2s
// Attempt 3: 1s execution → success
// Total durationMs ≈ 9000ms (3 × 1s execution + 2 × 2s backoff)
// retryCount = 2Using Timing for Performance Diagnosis
- Sort the Node List by Duration (slowest first) to see which nodes are bottlenecks
- Check the retry count — high retry counts indicate instability in upstream services
- Compare startedAt of consecutive nodes — gaps indicate queuing or scheduling delays in the Process Engine