Search

Lessons (17)

• Capacity Planning, SLOs and Error Budgets

  … An error budget is how much failure that promise permits. Example: checkout API Question Clear answer --- --- How fast …

• Timeouts, Retries & Backpressure

  … age, saturation, rejected work, dependency latency, retries, and error-budget burn. A concrete overload plan Imagine image processing …

• Observability: Logs, Metrics & Traces

  … Budget burned → focus on reliability. Alerting that doesn't ruin your sleep - ✅ Alert on symptoms users feel (error …

• 10,000 Users Do Not Need 10,000 Database Connections

  … Set a fleet-wide connection budget, bounded wait queue and short transactions. Tune from database saturation and query …

• AWS: Start Here

  … Add a budget alert before experimenting. Stop or delete test resources when finished. First commands bash aws configure …

• Putting AI Systems into Production

  … Monitor aggregate quality signals and operational errors. User feedback can reveal problems, but it is not a substitute …

• Observability and Incident Response

  … alert: "checkout success SLO will exhaust its monthly error budget in 45 minutes." Noisy alert: "one CPU sample …

• AWS Interview Questions: Clear Answers

  … load balancer errors, latency, healthy targets, and database pressure. I compare the time with traffic and deployment changes …

• Cost Optimization and the Well-Architected Framework

  … spend by service, account, tag and usage type Budgets Alerts and actions when spend or usage exceeds a …

• Testing, Preconditions, and Policy Checks

  Use validation, lifecycle conditions, Terraform tests, and policy checks to catch different classes of errors. Different checks answer …

• Event-Driven Patterns, Step Functions and Reliable Lambdas

  … 30, "Retry": [ { "ErrorEquals": [ "Lambda.ServiceException", "Lambda.TooManyRequestsException" ], "IntervalSeconds": 2, "MaxAttempts": 3, "BackoffRate": 2.0 } ], "Catch": [ { "ErrorEquals": [ "States.ALL …

• Concurrency, Resilience & Distributed Workflows

  … Do not retry validation errors or a request that may be non-idempotent unless it has an idempotency …

• CloudWatch: Know What Your App Is Doing

  … That is a 10% server-error rate. An alarm based on error rate can be more informative than …

• Backend Design Scenarios

  … Retries without timeouts and budgets amplify an outage. The dependency policy should be visible in traces so you …

• AWS Databases: Choose the Right One

  … A timeout suggests network/listening issues; an authentication error suggests credentials or database permissions. These require different fixes …

• L6 Production Debugging Casebook

  … Put p50, p95, p99, throughput, error rate, pod CPU, GC pauses, pool wait, database locks, query latency and …

• Scaling, Load Balancing & Rate Limiting

  … Retry only safe/idempotent requests, with a strict budget. - Monitor healthy target count, 4xx/5xx rates, p50/p95 …