Skip to main content

X-Ray: Sampling Rules

🛠️ Step-by-Step Custom Sampling Rule From Console​

  • Step 1: Locate the X-Ray Settings Hub

    • Go to CloudWatch ──► scroll all the way down to Settings on the left-hand navigation column panel.
    • Look under X-Ray traces and click Sampling rules to display the rule ledger grid.

  • Step 2: Evaluate the Default Fallback Guardrail

    • Inspect the pre-provisioned Default baseline rule sitting at the bottom of the stack:
    • Priority: Locked at 10000 (the absolute lowest priority ceiling; all custom rules step ahead of it).
    • Criteria: Explicitly wildcarded (*) to intercept every ambient method, URL path, and service.
    • The Adjustment Limitation: While you can edit the Reservoir size and Fixed Rate percent threshold values on the default rule, you cannot modify its matching criteria parameters, chief.

  • Step 3: Provision a Custom Sampling Rule

    • Click Create sampling rule at the top of the pane.
    • Rule Name: Input your identifier string (e.g., DemoSampling).
    • Rule Priority: Assign a value between **1 and 9999**.
    • The Priority Order Inversion Rule: The lowest priority number carries the highest evaluation weight! If you assign a rule a priority of 5000, the X-Ray engine evaluates and matches it long before it ever hits the default fallback rule block.

🎯 Fine-Tuning Your Scoped Ingestion Criteria​

Instead of blanket-sampling your whole system, you can constrain your custom rule's scope by passing explicit matching values directly into the console filter criteria forms:

  • Service Name: Target a distinct app workspace module identifier (e.g., MYSERVICE).
  • HTTP Method: Target specific operational request methods (e.g., typing POST to monitor data writes closely, while ignoring noisy polling traffic patterns).
  • URL Path: Isolate target endpoint routes using standard wildcard pattern tags (e.g., /api/checkout/* ensures your high-value billing transactions are monitored closely).

📊 Evaluation Logic Flow Matrix​

When requests hit your instrumented application gateway, the background client engine loops through your rule criteria paths according to this exact structural logic chain:

Incoming API Payload Evaluation⟶Loop Rules Ascending via Priority Number (1→9999)\text{Incoming API Payload Evaluation} \longrightarrow \text{Loop Rules Ascending via Priority Number }(1 \rightarrow 9999)

First Pattern Match Winning Condition=Borrow from Reservoir Size  ∨  Apply Fixed Rate % Threshold  ⟹  Stop Evaluation\text{First Pattern Match Winning Condition} = \text{Borrow from Reservoir Size} \;\lor\; \text{Apply Fixed Rate \% Threshold} \implies \text{Stop Evaluation}

tip

The Zero-Downtime Pipeline Lock: Once you click save on your custom rule configurations inside the console dashboard, the local X-Ray Daemons actively pull down the updated JSON schemas via background polling loops on the fly. No code changes, no production build restarts—the new sampling velocity applies instantly.