Question 910
A team has corrected this HPE compute solution issue: A cluster of compute nodes depends on a single path to shared storage and drops VM I/O when a switch is upgraded. Which continuing practice would best prevent the same problem?
Explanation
Correct Answer
Require path failover validation before production compute acceptance. This recurring assurance measure is tied to the diagnosed risk: The storage fabric lacks the validated multipath and fault-domain design. For the active incident, the corrective action remains: Provision supported redundant host-to-storage paths with correct policy and zoning.
Incorrect Answers
Coordinated maintenance windows and stop-on-failure controls were not considered. The option addresses that risk, not the one responsible for the present issue. The storage fabric lacks the validated multipath and fault-domain design. Require path failover validation before production compute acceptance.
The prevention measure is relevant to this condition: The workload may be bottlenecked before the accelerator rather than constrained by GPU capacity. Here the recurring weakness is: The storage fabric lacks the validated multipath and fault-domain design. Require path failover validation before production compute acceptance.
The compute node configuration may not qualify for the selected virtualization release. The selected safeguard helps with that exposure, but it leaves the recurrence risk here unaddressed. The storage fabric lacks the validated multipath and fault-domain design. Require path failover validation before production compute acceptance.