INFRASTRUCTURE LEARNINGINFRASTRUCTURE LEARNING
インフラ大全インフラ大全IT INFRASTRUCTURE / LEARNING
Infrastructure learning guide
Find a lesson or lab/

Esc to close · Search terms stay on your device

LEARNING日本語

Topic lessonTopic: AWS and SQL Server telemetry and audit

Choose metrics, logs and traces for AWS and SQL Server monitoring

Break a report of slowness into scale, failed operations and time spent waiting. Choose the next evidence source for an API that connects to SQL Server.

What you will learn

Choose the next evidence source for an API that connects to SQL Server.

Prerequisite

This exercise uses fictional evidence and makes no changes in a live environment.

Use the signal that answers the question

Metrics describe trends, logs capture individual outcomes and traces show intervals within a request. Audit records help identify who changed a setting.

Investigate slow API requests
Investigate slow API requests1: Trend — Fix the time window — Request. 2: Request — Match identifiers — Interval. 3: Interval — Test a hypothesis — Change123TrendRequestIntervalChangeInvestigate slow API requests1: Trend — Fix the time window — Request. 2: Request — Match identifiers — Interval. 3: Interval — Test a hypothesis — Change123TrendRequestIntervalChange
  1. Trend → RequestFix the time window
  2. Request → IntervalMatch identifiers
  3. Interval → ChangeTest a hypothesis
Trend
Identify affected intervals
Request
Find the failing operation
Interval
Measure database-related waits
Change
Check the audit record

A synthetic order API is slow for ten minutes

Compare connection failures, retries and change records in a common window.
SignalObservationNext decision
Metricsp95 rises from 180 to 900 msFix the affected window
Application logsConnection retries on the requestIdentify the request
Trace740 ms before database workInspect each interval
Audit recordA configuration changedReconcile time and request conditions

Timing alone cannot tell you which change caused the difference. Match a representative request and its conditions.

Evidence for the decision

Define coverage Record collection scope, aggregation interval and unobserved requests.

Match the request Use request IDs or trace IDs to distinguish separate attempts.

Measure after correction Compare success, waits and retries under equivalent traffic and conditions.

Check your decision

Choose an option to see the result and explanation.

What first identifies the trend over several minutes?
Read the answer and reason

Answer: Latency metrics at matching intervals · Use an observation with a consistent time basis.

What separates waiting intervals in one request?
Read the answer and reason

Answer: Trace intervals · Use a measurement of its constituent work.

What supports a connection to a configuration change?
Read the answer and reason

Answer: Audit target, request IDs and reproduction conditions · Match more than timestamps.

Check, then explain in your own words

  1. State a measurable question.
  2. Choose the appropriate signal.
  3. Correlate the request and actual change.
Check your understanding: Do untraced requests count as successful?

Use logs and metrics and record the gaps.

Signal coverage and workload outcomes are separate.

Sources and conditions checked

The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.