Deploying API implementations to CloudHub11% of the exam
Failure Domains and Disaster Recovery
Match classic CloudHub recovery designs to tunnel, gateway, site and regional failures, and distinguish application restart settings from infrastructure recovery.
Identify and avoid single points of failure in CloudHub deployments of API implementations.
~7 min read · 4 practice questions
What the exam expects
- Identify worker, availability-zone, region, tunnel, customer-gateway and customer-site failure boundaries.
- Match redundant paths to the component that must survive failure.
- Assign cross-region application, traffic, state and private-network responsibilities.
- Distinguish unresponsive-application recovery from impaired-worker replacement.
HA keeps a system accessible through component failure; DR restores service after a wider disruption; a failure domain groups components that can fail together; RTO describes tolerated recovery time; RPO describes tolerated data loss; a customer gateway terminates the customer's VPN connection (HA and DR, VPN HA). This lesson concerns classic CloudHub deployments.
Sources for this section (2)
Key facts
Within-region HA
- Detail
- Multiple workers are distributed across availability zones.
Regional DR
- Detail
- Customers own the cross-region application and dependency strategy.
Tunnel maintenance
- Detail
- Configure both supplied tunnels.
Gateway failure
- Detail
- Separate endpoints, public IPs and VPN connections are required.
Site failure
- Detail
- Put customer gateways in different data centers.
Regional state
- Detail
- Object Store v2 does not provide cross-region failover.
Private recovery network
- Detail
- Provide an Anypoint VPC in each required region.
Impaired worker
- Detail
- Replacement attempts are independent of the application-restart checkbox.
Unresponsive application
- Detail
- With automatic restart disabled, notifications and logs require manual follow-up.
- When
monitoringEnabledis changed through the API, which lifecycle follow-up is required? - Why is one failed readiness probe insufficient to expect the documented consecutive-failure warning?
- When planning around scheduled worker maintenance, where can the actual restart date be inspected?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the MuleSoft Platform Architect pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 47 lessons of the MuleSoft Platform Architect guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (4 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Choose guide-only, practice-only, or save $4 with both