Search

Lessons (34)

• Brokers, Replication & Durability

  … Brokers, leaders and followers, in-sync replicas, min.insync.replicas, leader election and KRaft, drawn step by step …

• Database Internals, Replication and Scaling

  … Route confirmation reads to the writer or use a supported replication-position barrier. Replica reads are suitable only …

• Replication, Partitioning & Sharding

  Scale reads with replicas, split data safely with partitions and shards, and plan for consistency, hot keys and …

• DynamoDB Modeling, RDS Scaling and Data Resilience

  … Do not read-your-own-write from a replica without handling lag. - Aurora offers a reader endpoint that …

• Testing, Preconditions, and Policy Checks

  … Further reading: [Terraform tests](https://developer.hashicorp.com/terraform/language/tests), [custom conditions](https://developer.hashicorp.com/terraform …

• Deployments: Keep Your App Running

  … 3000 readinessProbe: httpGet: path: /health/ready port: 3000 The label app: shop-api appears twice on purpose: - The …

• Kafka Producers

  … acks=all (with min.insync.replicas=2) P- L: send batch L- L: write to log F1- L …

• Storage Engines & Database Operations

  … replicas provide recoverability](/img/database/storage-operations.svg) Mechanism Strength Cost to remember --- --- --- B-tree Point/range reads …

• Multi-Region and Disaster Recovery

  … The standby has asynchronous replicas that are 20 seconds behind and customers retry pending purchases. What must be …

• Variables, Locals, Outputs, and Validation

  … Further reading: [Input variables](https://developer.hashicorp.com/terraform/language/values/variables), [outputs](https://developer.hashicorp.com/terraform …

• Internals, Upgrades, Resilience and Interview Questions

  … 0 , correct readiness probes, graceful SIGTERM handling, a preStop delay, PDBs and enough replicas across zones. For extra …

• Backend Design Scenarios

  … Add read replicas when the primary is healthy but read traffic is the constraint. Route only replica-safe …

• Scaling, Load Balancing & Rate Limiting

  … Copy the data to replicas and send reads there. mermaid flowchart LR App -- writes P[(Primary)] P -- replication …

• Containers at Scale: ECS Deep Dive and EKS

  … Start a new replica, wait for readiness, route traffic, then drain the old replica. Handle termination by stopping …

• Your Practical Kubernetes Workflow

  … Add a readiness check, CPU/memory requests, and a memory limit. - Keep at least two replicas for a …

• Resources, QoS and Scheduling Control

  … than drop below 2 ready pods. Do not set minAvailable equal to replicas , or drains will block forever …

• GitOps, Progressive Delivery and Observability

  … Useful cluster alerts: - Pods in CrashLoopBackOff or not ready for more than 10 minutes. - Node NotReady , disk or …

• AWS Databases: Choose the Right One

  … A read replica can handle selected read workloads, but replication lag means a new write may not be …

• SQL vs NoSQL Databases

  … SQL databases can scale out too (read replicas, Citus, Vitess, CockroachDB), but it's more work. ACID vs …

• ACID, Transactions & Locks

  … Dirty read Read data another transaction may roll back Read committed or stronger Non-repeatable read Same row …

• Health Checks and Scaling

  Tell Kubernetes when an app is alive and ready, then let it add copies when traffic grows. Health …

• CloudWatch: Know What Your App Is Doing

  … Further reading: [Logs Insights fields](https://docs.aws.amazon.com/AmazonCloudWatch/latest/logs/CWL AnalyzeLogData-discoverable-fields.html …

• Environments, Workspaces, and Multiple Accounts

  … Which control stops a dev CI job from reading production state? Further reading: [CLI workspaces](https://developer.hashicorp …

• Capacity, Availability, Latency and CAP

  … reads/day, with a peak ten times the daily average. A network partition separates two database replicas during …

• StatefulSets, DaemonSets, Jobs and CronJobs

  … db-headless a headless Service gives each pod its own DNS record replicas: 3 selector: matchLabels: { app: db …

• Kubernetes: Start Here

  … v1 kind: Deployment metadata: name: shop-api spec: replicas: 3 selector: matchLabels: app: shop-api template: metadata: labels …

• AWS Interview Questions: Clear Answers

  … The traditional RDS Multi-AZ DB instance standby does not serve application reads. A read replica serves read …

• Collaborative Document Editing & Real-Time Sync

  … the sequencer publishes to a few relay nodes , and each relay serves thousands of read-only sockets. - Presence …

• Resilience, Disaster Recovery and Multi-Region Design

  … a second; promote the secondary) RDS Cross-region read replica or automated backup replication EBS / EC2 Snapshots copied …

• Practical Project: Deploy a Real CRM App on EKS

  … 512Mi } readinessProbe: receive traffic only when ready httpGet: { path: /healthz, port: 3060 } initialDelaySeconds: 20 periodSeconds: 10 livenessProbe: restart …

• Cross-Service Data: How to Join Without a Database JOIN

  … Build a dedicated read model For dashboards, search and complex screens, create a read model designed specifically for …

• L6 Production Debugging Casebook

  … Validate that new writes and reads agree, then deploy code that reads only new column while still dual …

• A WhatsApp OTP Is Only Single Use If You Make It So

  … for "expired" and "wrong". - Trusting a cache or read replica for the "unused" check. The consume must hit …

• Caching & Redis

  … Then consider: - Read/write ratio: cache longer when reads dominate and writes are rare. - Cost of stale data …