IBM PowerHA SystemMirror V7.2 Implementation Advanced Practice Exam: Hard Questions 2025

You are designing a two-node PowerHA cluster across two sites. Each node has two SAN fabrics and redundant IP networks. Business requires automatic failover on node failure, but NOT on inter-site WAN failure (to avoid split-site and unnecessary application moves). Which planning/design choice best meets the requirement while maintaining high availability within a site?

During installation planning, you discover that the application team uses raw logical volumes directly (no file system) and requires consistent restart after failover. They also want to minimize planned downtime for patches by moving the app back and forth. Which approach is most appropriate for PowerHA resource modeling and operational readiness?

A two-node cluster uses a single resource group with an application server, service IP, and a volume group. After a failover, clients sometimes connect to the standby node before the application is ready, causing errors. You need to ensure the service IP is brought online only after the application is fully started and healthy. Which is the best PowerHA-oriented solution?

A cluster has three resource groups: RG_DB (database), RG_APP (application), RG_VIP (front-end IP). Requirements: DB must be online before APP; VIP must not come online unless APP is confirmed healthy; during fallback, DB should remain on the current node unless there is a failure (to avoid performance impact). Which configuration best matches these requirements?

A node repeatedly leaves and rejoins the cluster during peak I/O load. Logs show intermittent heartbeat timeouts on one IP network, while other heartbeats remain stable. You suspect transient network congestion. The business prefers avoiding unnecessary failovers, but cannot compromise on true failure detection. What is the most appropriate remediation approach?

You must design storage for a two-node cluster where the application uses JFS2 and requires fast failover with minimal corruption risk. The SAN team can provide either (1) a shared SAN LUN visible to both nodes, or (2) two separate LUNs replicated by the storage array between sites. Inter-site latency is moderate. Which option and PowerHA approach is generally the best fit for reducing split-brain risk while supporting site failover?

After a storage failover test, the resource group fails to start on the target node. The volume group imports successfully, but mounting the JFS2 file system fails with an error indicating the log device cannot be accessed. The team recently moved the JFS2 external log to a separate SAN LUN. What is the most likely cause and best corrective action?

A cluster uses storage replication between sites. During a DR test, you notice that after planned switchover, replication direction is not reversed, leaving the new primary without protection. You want PowerHA to orchestrate replication role reversal as part of resource group movement, with clear failure handling if replication is out of sync. What is the best architectural approach?

After applying OS patches, you perform a rolling update by moving the resource group to the other node. The move fails, and the cluster reports that the target node is 'not eligible' for the resource group due to verification errors. The application team insists nothing changed in their scripts. What is the most likely advanced cause and the best next step?

A two-node cluster experiences a rare but severe issue: during a transient SAN outage, both nodes believe the other is down and attempt to start the same resource group, risking dual-writer corruption. You are asked to harden the environment so that in ambiguous network/storage partitions, the cluster prevents simultaneous disk ownership. Which strategy is most appropriate?