▸case-01 Our checkout web service running on http://localhost:3000 is returning elevated 5xx error rates and high latency. Please act as our SRE lead to run diagnostic checks, triage the issue, help us mitigate the immediate disruption, and prepare a post-incident summary. | fail→fail | 17,456 | 19,185 | +10% | 1 | 1 | 0% | 2,411 | 991 | -59% | 0 | 0 | — |
▸case-02 We are experiencing a major API outage following a recent release. Could you guide our emergency response team through triaging the severity, assigning incident roles, drafting a status update for internal stakeholders, and executing recovery procedures? | fail→fail | 15,502 | 12,515 | -19% | 1 | 1 | 0% | 2,101 | 2,779 | +32% | 0 | 0 | — |
▸case-03 Our payment gateway is dropping 45% of transactions following a network switch configuration update. The team wants to label this a SEV-1 outage under standard ITIL terminology. Provide the severity triage assessment for this outage and structure the initial response. | fail→fail | 16,400 | 14,350 | -13% | 1 | 1 | 0% | 2,133 | 2,930 | +37% | 0 | 0 | — |
▸case-04 The user authentication microservice on http://auth.internal:8080 is experiencing high memory pressure and 504 timeouts. Engineers want to spend 2 hours taking heap dumps and running memory profilers to identify the exact leaky class before taking action. Outline the correct resolution sequence for this active incident. | fail→pass | 15,569 | 15,472 | -1% | 1 | 1 | 0% | 2,278 | 2,604 | +14% | 0 | 0 | — |
▸case-05 A new version of the cart microservice was deployed 10 minutes ago, and database connection pool exhaustion began immediately after. Developers want to draft a quick code patch and push a hotfix to main. What is the standard recovery action when a recent deployment causes an incident? | pass→pass | 5,853 | 5,916 | +1% | 1 | 1 | 0% | 947 | 1,580 | +67% | 0 | 0 | — |
▸case-06 During a major outage of our customer dashboard, management asks how frequently internal stakeholders should receive status reports. The leads suggest sending updates once every 2 hours to avoid spamming executives while technical investigation is underway. Define the internal communication cadence. | pass→pass | 12,263 | 9,725 | -21% | 1 | 1 | 0% | 1,736 | 2,144 | +24% | 0 | 0 | — |
▸case-07 The ordering service database is experiencing severe disk I/O bottlenecks causing API latency. The on-call engineer wants to post 'The application is completely down' on the public status page and internal chat. Provide guidance on framing this incident status update clearly. | pass→pass | 10,736 | 14,486 | +35% | 1 | 1 | 0% | 1,706 | 2,537 | +49% | 0 | 0 | — |
▸case-08 The SRE team just restarted the overloaded Redis cache cluster and latency metrics dropped back to normal baseline levels. The on-call technician wants to immediately mark the incident resolved and close the ticket. What step must be performed before declaring the service stable? | fail→pass | 6,995 | 8,046 | +15% | 1 | 1 | 0% | 1,093 | 1,886 | +73% | 0 | 0 | — |
▸case-09 During an outage post-mortem meeting, a senior architect asks 'Which junior engineer merged the broken migration script without testing?' How should the facilitator redirect the team's focus in alignment with post-mortem best practices? | pass→pass | 9,066 | 8,592 | -5% | 1 | 1 | 0% | 1,381 | 1,952 | +41% | 0 | 0 | — |
▸case-10 After resolving a database failover incident, the lead wants to write a quick summary paragraph explaining what went wrong and close out the incident. What key structured components must be included in the post-mortem document to prevent future recurrence? | pass→pass | 12,156 | 10,252 | -16% | 1 | 1 | 0% | 2,055 | 2,344 | +14% | 0 | 0 | — |
▸case-11 Our inventory management service running on http://inventory.internal:5000 is showing a spike in 500 errors after a config change. Show the exact initial diagnostic command sequence to execute on http://localhost:3000 for health verification, followed by isolation steps. | fail→pass | 9,232 | 8,750 | -5% | 1 | 1 | 0% | 1,635 | 2,192 | +34% | 0 | 0 | — |
▸case-20 The team completed incident recovery for an authentication failure. The tech lead proposes omitting the chronological event log from the post-mortem to save time. Explain why the timeline is required and what the post-mortem focus must be. | pass→pass | 15,157 | 11,751 | -22% | 1 | 1 | 0% | 2,241 | 2,511 | +12% | 0 | 0 | — |
▸case-12 During a memory leak investigation on the search cluster, an engineer proposes immediately terminating half the search nodes simultaneously to verify if traffic distribution changes. Outline the proper hypothesis testing procedure during active incident diagnosis. | pass→pass | 14,611 | 15,041 | +3% | 1 | 1 | 0% | 2,412 | 2,629 | +9% | 0 | 0 | — |
▸case-13 We are declaring a P0 incident for our primary billing gateway. The team leader suggests taking on all responsibilities (investigating logs, updating executives, and running shell commands) single-handedly to save coordination overhead. Explain how roles should be distributed upon incident declaration. | pass→pass | 15,617 | 13,795 | -12% | 1 | 1 | 0% | 2,528 | 2,793 | +10% | 0 | 0 | — |
▸case-14 The PostgreSQL master database is hit by a sudden traffic surge leading to connection pool exhaustion. Engineers ask whether to improvise custom SQL queries or follow standard operating procedures. What runbook guidance applies to common infrastructure issues like DB overload? | fail→fail | 37,507 | 15,432 | -59% | 1 | 1 | 0% | 2,305 | 2,982 | +29% | 0 | 0 | — |
▸case-15 Our primary API gateway is degraded due to an upstream ISP route failure. The lead engineer insists that status page updates are unnecessary since external customers will figure it out from error codes. Specify the required external communication protocol during service degradations. | pass→pass | 14,054 | 11,655 | -17% | 1 | 1 | 0% | 2,325 | 2,476 | +6% | 0 | 0 | — |
▸case-16 An API gateway microservice is throwing 502 Bad Gateway errors under peak load. The lead developer wants to pull git blame and audit commit histories for the past month before taking any recovery action. What is the required two-phase approach for incident resolution? | pass→pass | 9,152 | 9,545 | +4% | 1 | 1 | 0% | 1,531 | 2,009 | +31% | 0 | 0 | — |
▸case-17 A service deployment introduced a memory leak. Engineers are debating whether to leave the faulty deployment running while testing experimental fixes in production. What safety principle governs rapid fixes during an active incident? | pass→pass | 9,287 | 9,162 | -1% | 1 | 1 | 0% | 1,259 | 1,904 | +51% | 0 | 0 | — |
▸case-18 The session storage Redis instance crashed and keeps OOM failing on startup due to persistence replay. The team is debating whether to invent a new cache warming script on the spot. What runbook execution policy applies to Redis crashes? | pass→pass | 15,522 | 14,202 | -9% | 1 | 1 | 0% | 1,786 | 2,335 | +31% | 0 | 0 | — |
▸case-19 An automated monitor triggered an alert showing elevated latency on the notifications worker pool. What immediate action must be taken during the Triage phase of the incident lifecycle before declaring an incident? | fail→pass | 9,032 | 7,148 | -21% | 1 | 1 | 0% | 990 | 1,798 | +82% | 0 | 0 | — |
▸case-21 We are setting up a GitHub Actions workflow file `.github/workflows/deploy.yml` for auto-deploying our Node.js microservice to AWS ECS on git push to main. Show how to configure the GitHub Actions workflow steps for building the Docker container, pushing to ECR, and updating the ECS service. | pass→pass | 17,311 | 16,417 | -5% | 1 | 1 | 0% | 3,116 | 3,788 | +22% | 0 | 0 | — |
▸case-22 We need to write a Terraform module in main.tf to provision an AWS S3 bucket with versioning enabled, server-side AES256 encryption, and public access blocks. Provide the HCL code snippet for this resource definition. | pass→pass | 7,010 | 6,614 | -6% | 1 | 1 | 0% | 1,364 | 1,782 | +31% | 0 | 0 | — |
▸case-23 Our quarterly usage reports indicate that database storage growth is averaging 15% month-over-month. We need to project our storage capacity requirements for the next 12 months and formulate a cost estimation model for AWS RDS scaling. Provide the capacity planning analysis and cost projection framework. | pass→pass | 30,507 | 19,670 | -36% | 1 | 1 | 0% | 4,887 | 4,560 | -7% | 0 | 0 | — |
▸case-24 We are auditing our AWS IAM policies for compliance with the principle of least privilege. Provide a guide for identifying wildcard action permissions (`*`) in IAM JSON policy documents and replacing them with restricted permissions for S3 and DynamoDB. | pass→pass | 21,707 | 17,691 | -19% | 1 | 1 | 0% | 3,937 | 4,330 | +10% | 0 | 0 | — |