⚡ 179 Live Scenarios
🎯 STAR Method Answers📋 60s Elevator Pitches
Amazon Web Services (AWS) represents the cloud backbone for global digital enterprises, requiring DevOps engineers and Cloud Architects to demonstrate profound mastery over high-availability infrastructure, distributed networking, and zero-trust security postures. Today’s senior cloud interviews bypass basic compute provisioning to challenge candidates on real-world edge cases: resolving split-brain scenarios across Multi-AZ RDS clusters, diagnosing asynchronous cross-region replication lag in Aurora Global Databases, and untangling AWS IAM evaluation logic involving SCPs, Permission Boundaries, and Session Policies. Engineering managers want to see how you orchestrate AWS Transit Gateways across hundreds of AWS Organizations accounts, optimize CloudFront caching hierarchies to slash origin egress expenses, and design resilient event-driven architectures leveraging SQS Dead Letter Queues and DynamoDB Streams. These curated AWS scenario questions prepare you for Staff and Principal Cloud Architect rounds by detailing practical production runbooks, AWS CLI remediation commands, and rigorous disaster recovery frameworks designed to achieve aggressive RTO and RPO objectives under strict compliance constraints.
The 4-Stage AWS Hiring Loop & Behavioral Interview Blueprint
Complete breakdown of the 1-hour recruiter screening, 1-hour technical screen, hiring manager leadership evaluation, and the 4-hour onsite loop with the Bar Raiser.
Get 1 AWS & Cloud Architecture interview question in your inbox every week
Preparing for a cloud or platform interview? Subscribe to get our weekly deep-dive scenario plus instant access to the Top 50 Kubernetes Interview Questions & Incident Runbooks PDF.
Triage runbook for handling sudden 100% CPU utilization on production EC2 instances: telemetry triage, identifying culprit processes, handling legit vs maliciou...
Systematic OSI and AWS-layer troubleshooting methodology when an EC2 instance shows 'Running' but SSH connection fails, covering network timeouts, con...
Production runbook for isolating Application Load Balancer (ALB) 5xx errors: distinguishing ELB-generated vs Target-generated codes, debugging 502/503/504, and ...
Production playbook for migrating mission-critical enterprise workloads from on-premises to AWS with near-zero downtime: discovery, Direct Connect hybrid networ...
Practical portfolio walkthrough of enterprise Microsoft Azure services utilized in modern production DevOps: Key Vault, Azure Container Registry (ACR), Managed ...
End-to-end operational guide for Azure Kubernetes Service (AKS): control plane vs node pools, Azure CNI vs Kubenet, GitOps deployment with Azure DevOps / GitHub...
Block Public Access settings — go to S3 → Block Public Access → Enable all four settings. This overrides any bucket/object ACLs and policies that grant public a...
S3 automatically partitions by prefix — requests are distributed across S3 partitions. More unique prefixes = better parallelism. Add a hash/timestamp prefix to...
A permission boundary is a policy attached to an IAM entity that sets the maximum permissions the entity can ever have — even if an identity policy grants more....
EC2 launch type — you manage the EC2 instances (the container hosts). You're responsible for patching, scaling the cluster, choosing instance types. More c...
Production triage runbook for AWS ECS tasks stopping with exit code 137 (OOMKilled): inspecting ECS StoppedReason, distinguishing task hard limit from container...
CloudWatch Logs — stores raw log data (text). Your app, Lambda, ECS, VPC Flow Logs all write here. You can search/query with CloudWatch Logs Insights....
CloudTrail — records API calls made to AWS. Who did what, when, from where. "Audit trail." Example: who deleted that S3 bucket? Who changed that Secur...
Static website hosting requires public read. Either make bucket public (then enable static hosting) or use CloudFront with Origin Access Control (OAC) — better....
A multi-region active-active architecture runs full application stacks in two or more AWS Regions that all serve live traffic, with global routing in front and ...
Check CloudFormation events in console for the failure reason. Common: IAM permissions, resource limit, invalid property value. Fix the template, then redeploy....
To implement blue-green deployments on ECS, run the new task set (green) beside the live one (blue) behind the same Application Load Balancer, test it on a sepa...
Scale out consumer EC2/ECS instances. Use ASG scaled on ApproximateNumberOfMessagesVisible CloudWatch metric. Or move to Lambda consumer (auto-scales with queue...
When an SNS topic notification is not being received, work the path in order: was the message published, is the subscription confirmed, does a filter policy dro...
The difference between SQS and SNS is delivery model: SQS is a pull-based message queue where one consumer group processes each message, while SNS is a push-bas...
Each ECS task or Lambda function has its own IAM role with only the permissions it needs. Use task IAM roles for ECS, execution roles for Lambda. Never share ro...
When CloudFront is serving stale content after you updated S3, the cause is almost always caching, not S3: S3 gives strong read-after-write consistency, so the ...
When a Route 53 health check is failing while the endpoint looks fine from your laptop, the usual reason is that the checkers cannot reach or satisfy it the way...
WAF inspects HTTP requests before they reach the ALB. Rules: AWS Managed Rule Groups (OWASP top 10, bot control), IP rate limiting (prevent DDoS), geo-blocking,...
To fix a Lambda cold start with a big ML model, load the model once in the init phase (outside the handler), shrink and cache it, and pay to keep environments w...
ElastiCache Redis — shared in-memory store that all instances can access. Never store sessions in local EC2 memory (breaks when an instance is replaced) or in c...
Parameter Store = config and non-sensitive parameters. Free tier available. Secrets Manager = specifically for secrets. Auto-rotation built in. Charges per secr...
Use a no-lock migration approach: add new column (no lock), backfill data in batches, add new index concurrently, switch app to use new column, drop old column ...
Spot capacity issue or that AZ/region is out of that instance type. Mitigate: use multiple instance types in the ASG (mixed instances policy), use multiple AZs,...
You cannot toggle encryption on an existing unencrypted RDS instance; instead you snapshot it, copy the snapshot with a KMS key to produce an encrypted snapshot...
Elastic Beanstalk console → Environment → Configuration → Software → Environment properties. Or via .ebextensions files in your code. Or aws elasticbeanstalk up...
Request a limit increase via AWS Service Quotas. Or redesign to use fewer VPCs with more subnets. Or use a shared VPC (Resource Access Manager) that other accou...
EC2 Auto Scaling scale in vs scale out decisions come from CloudWatch alarms and scaling policies: when a metric crosses a threshold the group adds instances (s...
If an SQS consumer processes the same message twice, treat it as expected behavior to design for, not a bug to eliminate: standard queues guarantee at-least-onc...
Enable S3 Versioning + MFA Delete. Enable S3 Object Lock (WORM). Use a bucket policy with Deny for s3:DeleteBucket. Enable AWS Config rule that alerts on deleti...
Options by RPO/RTO: Backup & Restore (hours RPO/RTO, cheapest) → Pilot Light (critical infra always on, warm data, minutes to hours) → Warm Standby (scaled-...
Architectural review of AWS GuardDuty: intelligent threat detection using machine learning and threat intelligence feeds, multi-account organization setup, and ...
Isolate: change security group to block all outbound. 2) Take a snapshot (forensics). 3) Check VPC Flow Logs for the outbound connections. 4) Check GuardDuty fi...
KMS generates and stores encryption keys. You never handle raw key material. AWS-managed keys: automatic rotation, free, no management needed — use for basic en...
Amazon EventBridge is a serverless event bus that routes JSON events from AWS services, your applications and SaaS partners to targets using content-based rules...
Data consistency (cross-region replication has latency), data sovereignty (some data can't leave specific regions), cost (cross-region data transfer fees),...
Check ECS task security group allows ALB security group on the container port. Check the health check path returns 200 on that port. Check task is fully started...
To implement infrastructure drift detection, run an automated comparison between declared state (Terraform, CloudFormation) and real resources on a schedule, al...
Implement exponential backoff with jitter in API retry logic. Use AWS SDK built-in retry (most SDKs have this). Reduce polling frequency. Request limit increase...
Amazon Inspector is AWS's managed vulnerability management service: it continuously scans EC2 instances, container images in ECR and Lambda functions for k...
Each AZ is a physically separate data center (separate power, cooling, networking). AZs in same region connected via low-latency links. Design: deploy in min 2 ...
SCP is a guardrail for entire AWS accounts in an Organization. It restricts what IAM policies in those accounts can allow. If SCP doesn't allow an action, ...
In each account: create CloudTrail and send to S3 in the security account. Update the security account S3 bucket policy to allow PutObject from all org accounts...
Cache the secret in application memory (most secrets don't change frequently). AWS Secrets Manager SDK supports caching. Or use Parameter Store (cheaper) f...
PrivateLink: exposes a specific service (not a whole network) from one VPC to another. Traffic goes through AWS backbone. Supports cross-account and even cross-...
Parallelize independent resources (CFN does this automatically). Use nested stacks to update only changed stacks. Use ChangeSets to preview changes before apply...
Use API Gateway WebSocket API (scales automatically, no infra to manage). Each connection triggers Lambda functions for connect/disconnect/message. Store connec...
Historically, in S3, the AWS account that uploads the object retains explicit ownership and full control of that object, even if the bucket itself belongs to a ...
A CNAME (Canonical Name) essentially maps one domain to another domain. However, the strict global DNS protocol absolutely forbids a CNAME from being placed at ...
Creating a VPC Endpoint simply keeps the traffic on the AWS private backbone; it does not secure it inherently. An attacker could still run aws s3 cp secrets.tx...
By default, CloudTrail only records Management Events (Control Plane actions). These include creating infrastructure (CreateBucket, RunInstances, UpdateSecurity...
You would append a Condition block to their overarching IAM Policy (or a global SCP) that heavily restricts authentication based on their explicit IP address....
Engineering blueprint for designing an enterprise multi-cloud substrate across AWS and GCP covering private BGP interconnects, federated OIDC workload identitie...
Deep architectural trade-off comparison between Multi-AZ and Multi-Region deployments in AWS: synchronous vs asynchronous replication, network latency, data con...
Enterprise AWS IAM security architecture: replacing static credentials with OIDC role assumption, scoping resource ARNs, enforcing IAM Permission Boundaries to ...
Strategic framework for designing disaster recovery across the 4 AWS DR tiers (Backup & Restore, Pilot Light, Warm Standby, Multi-Site Active/Active) while ...
Root cause isolation and troubleshooting procedure when an Azure DevOps CI/CD pipeline reports successful stage execution, but the deployed web application or A...
Architectural decision matrix for sharing specific S3 objects across AWS accounts: evaluating Resource-based Bucket Policies, Identity-based IAM Policies, and t...
Complete architectural journey of an HTTP request hitting an Amazon CloudFront distribution: Anycast routing, Edge Point of Presence (PoP), cache key evaluation...
Enterprise architectural blueprint for migrating 500 Terabytes of live object data between two distinct AWS accounts securely, with automated checksum verificat...
Step-by-step production runbook for eliminating compromised JSON service account keys in Google Kubernetes Engine (GKE) by binding Kubernetes Service Accounts d...
Architectural playbook for deploying a centralized Google Cloud Shared VPC across Host and Service Projects, ensuring security isolation, subnetwork IAM delegat...
Production implementation of Google Cloud Armor security policies, rate-limiting rules, Adaptive Protection ML anomaly detection, and OWASP Top 10 mitigation on...
Engineering a high-availability Google Cloud SQL architecture across regional availability zones with Private Services Access, automated failover testing, and c...
Engineering a resilient 99.99% hybrid connectivity topology between enterprise datacenters and GCP using Cloud Interconnect with automated BGP MED metric failov...
Deep architectural comparison between GKE Autopilot and GKE Standard, evaluating SLA guarantees, daemonset constraints, per-pod billing mechanics, and migration...
Engineering analysis for choosing between Google Cloud Run serverless container runtime and GKE, modeling cold starts, concurrency, gRPC streaming, and financia...
Production guide to implementing Customer-Managed Encryption Keys (CMEK) across Cloud Storage, BigQuery, and GKE with automated 90-day key rotation and disaster...
Root cause analysis and mitigation runbook for SNAT port exhaustion on Google Cloud NAT gateways, preventing outbound packet drops and third-party API timeout c...
Deploying declarative multi-cluster governance across hybrid GKE and on-premises clusters using Anthos Config Sync (RootSync/RepoSync) and Gatekeeper Policy Con...
Engineering production Service Level Objectives (SLOs) and Google Cloud Monitoring Query Language (MQL) multi-window multi-burn-rate alerting to eliminate alert...
Enterprise FinOps framework for mitigating unbudgeted BigQuery on-demand analysis spikes by transitioning to BigQuery Editions, capacity slot reservations, BI E...
Design pattern for ensuring strictly idempotent, at-least-once message processing in high-throughput GCP Cloud Functions Gen 2 triggered by Eventarc and Cloud P...
Comprehensive architectural guide for deploying Google Cloud Multi-Cluster Ingress (MCI) and Multi-Cluster Services (MCS) to deliver global anycast HTTP(S) rout...
Runbook for replacing default public Cloud Build workers with Private Worker Pools peering into private VPC networks, enabling secure access to private GKE clus...
Architectural runbook for designing and rolling out Google Cloud VPC Service Controls (VPC-SC) security perimeters with dry-run audit logging, ingress/egress ru...
Production guide for provisioning multi-writer ReadWriteMany (RWX) storage in GKE using Google Cloud Filestore Enterprise, tuning NFS client mount options, and ...
Architectural runbook for migrating Azure Kubernetes Service (AKS) clusters from legacy iptables-based Azure Network Policies to Azure CNI Powered by Cilium, un...
Step-by-step production migration from deprecated aad-pod-identity (NMI/MIC) to Azure AD / Entra ID Workload Identity using OIDC federation and Kubernetes Servi...
Architectural decision framework and deployment guide for migrating enterprise multi-region networking from traditional custom Hub-Spoke VNet peering to Azure V...
Engineering a globally distributed active-active data tier using Azure Cosmos DB Multi-Region Multi-Write, conflict resolution policies (LWW vs Stored Procedure...
Engineering a centralized, scalable Azure Private Endpoint and Private DNS Zone resolution architecture across multi-subscription hub-and-spoke networks with on...
Production guide for deploying and hardening Application Gateway Ingress Controller (AGIC) on AKS with WAF v2 CRS 3.2 rule tuning, cookie-based affinity, and SS...
Production guide for synchronizing Azure Key Vault secrets, keys, and certificates into AKS pods using the Secrets Store CSI Driver with auto-rotation and Kuber...
Architecting a secure, low-latency global delivery tier using Azure Front Door Premium, Anycast edge routing, WAF rules, and Private Link origin connectivity to...
Enterprise framework for implementing automated guardrails, compliance enforcement, and automatic remediation using Azure Policy, Management Group hierarchies, ...
Engineering a resilient cost-optimization architecture on AKS utilizing Azure Spot Node Pools with eviction rate metrics, Kubernetes taints/tolerations, and aut...
Step-by-step engineering runbook for reducing Azure Log Analytics ingestion costs by 65% using Data Collection Rules (DCR), custom log filtering, and ContainerL...
Designing a fault-tolerant hybrid cloud topology where an active ExpressRoute circuit automatically fails over to an encrypted Site-to-Site VPN with BGP route m...
Engineering a responsive event-driven autoscaling pipeline on AKS using Kubernetes Event-driven Autoscaling (KEDA) triggered by Azure Service Bus queue depths a...
Runbook for eliminating exposed public jump-box VMs by deploying Azure Bastion Developer / Standard tier, tunneling SSH/RDP via Azure CLI, and provisioning Bast...
Deep architectural comparison and decision matrix between Azure Front Door (Layer 7 Anycast reverse proxy) and Azure Traffic Manager (Layer 4 DNS-based routing)...
Engineering guide for refactoring unmaintainable, multi-thousand-line ARM JSON templates into clean, modular, reusable Azure Bicep modules with CI/CD What-If va...
Engineering a continuous reliability validation framework using Azure Chaos Studio to inject pod failures, network latency, and node outages into AKS clusters, ...
Architectural design for establishing a high-throughput, private, sub-5ms cross-cloud backbone between AWS VPCs and GCP VPCs using cloud exchange routers (Megap...
Architecting an enterprise GitOps deployment control plane with a centralized Argo CD management cluster orchestrating microservice lifecycles across AWS EKS, G...
Engineering a zero-static-credential cross-cloud identity federation pipeline enabling Google Kubernetes Engine (GKE) workloads to assume AWS IAM roles using Go...
Engineering a FinOps framework to audit, visualize, and eliminate catastrophic multi-cloud data egress fees across AWS ($0.09/GB), Azure ($0.087/GB), and GCP us...
Engineering a vendor-neutral distributed tracing architecture across AWS Lambda, GKE microservices, and Azure Functions using the OpenTelemetry Collector and Gr...
Designing a redundant, vendor-agnostic global DNS steering architecture using dual-provider Anycast DNS (NS1 / Cloudflare) with Real User Monitoring (RUM) laten...
Architecting a centralized, cloud-agnostic secrets management platform with HashiCorp Vault and Consul across AWS, Azure, and GCP featuring auto-unseal via clou...
Production guide for engineering a secure cross-cloud service mesh connecting AWS EKS and Azure AKS using Istio Multi-Primary on different networks with East-We...
Architectural breakdown of RTO and RPO in AWS: comparing Backup & Restore, Pilot Light, Warm Standby, and Multi-Region Active-Active across latency, replica...
Root cause detection and architectural remediation runbook for resolving concurrency throttling and cold-start failures on Azure Functions under production load...
Deep troubleshooting runbook for an EC2 instance that becomes unresponsive during specific afternoon hours with normal CPU/memory metrics but rising disk queue ...
Diagnostic guide to identify and eliminate latency bottlenecks in AWS Auto Scaling Groups when new compute capacity fails to launch quickly during traffic surge...
Master AWS iam roles vs iam policies: key architectural differences, identity vs resource policies, trust policies, STS AssumeRole, and IRSA in Kubernetes....
Master aws vpc subnets nat gateway internet gateway architecture: CIDR planning, public vs private subnets, Internet Gateway routing, NAT Gateway egress, and Mu...
Key incident runbooks, interview talking points, and architecture tradeoffs.
EC2 CPU suddenly reaches 100% — how would you troubleshoot?
When a production EC2 instance hits 100% CPU, my priority is two-fold: stop customer impact immediately (mitigation) while capturing telemetry to pinpoint the exact culprit process (root cause analysis).
Key Architectural Takeaway: Never reboot an instance blindly. Capture thread dumps and top process telemetry first so you don't lose the smoking gun, and ensure auto-scaling and CPU alarms prevent single-node exhaustion.
EC2 is running but SSH isn't working — what would you check?
When SSH fails while EC2 shows 'Running', the very first step is to observe the exact error message: 'Connection timed out' means a networking/firewall block, whereas 'Connection refused' or 'Permission denied' means you reached the OS but sshd or auth failed.
Key Architectural Takeaway: Categorize the error immediately: 'Timed out' is AWS network/SG; 'Connection refused' is sshd/port; 'Permission denied' is key/username. Always have AWS SSM Session Manager enabled as a zero-SSH out-of-band management backdoor.
ALB starts returning 5xx errors — how would you identify the root cause?
When an ALB starts throwing 5xx errors, my immediate first goal is to differentiate between errors generated by the ALB itself versus errors generated by backend application targets.
Key Architectural Takeaway: Always split ELB 5xx from Target 5xx in CloudWatch first. If it's ELB 502, check Keep-Alive timeout mismatches; if 503, check Target Group healthy host count; if 504, check target processing latency.
AWS app works internally but not internet — how would you troubleshoot?
When an AWS app works internally but not internet, the compute instance and container runtime are verified healthy. The root cause is isolated strictly along the outside-in ingress network boundary between the public internet and the AWS VPC.
Key Architectural Takeaway: When an AWS app works internally but not internet, the compute tier is healthy. Methodically trace outside-in: Route 53 public DNS -> Internet Gateway route (0.0.0.0/0) -> Public Subnet ALB -> Security Group (80/443 from 0.0.0.0/0) -> Stateless NACLs (inbound 80/443 + outbound ephemeral 1024-65535) -> AWS WAF. Use VPC Reachability Analyzer to mathematically pinpoint the blocking rule in seconds.
Scenario: Your organization needs to migrate a large production workload from on-premises infrastructure to AWS/Azure with minimal downtime and no major service disruption. Explain your migration strategy, dependency mapping, networking, data replication, security, IaC, testing, cutover, rollback, and post-migration optimization.
Migrating enterprise workloads with near-zero downtime requires decoupling the migration into distinct phases: hybrid network foundation, continuous data synchronization with Change Data Capture (CDC), and a low-risk DNS/router cutover with an instant rollback mechanism.
Key Architectural Takeaway: Near-zero downtime migration is achieved by pre-syncing data with AWS DMS CDC (Change Data Capture) over Direct Connect, lowering DNS TTL to 60s, and configuring reverse CDC back to on-prem as an instant safety rollback net.
🌐
Explore Related DevOps & Cloud Domains
Cross-train across interconnected systems for senior and staff infrastructure rounds.
Master AWS & Cloud Architecture & Platform Engineering In Production
Preparing for a cloud or platform interview? Subscribe to get our weekly deep-dive scenario plus instant access to the Top 50 Kubernetes Interview Questions & Incident Runbooks PDF.