Confirm the scope
Determine whether latency affects one VM, one datastore, one ESXi host or multiple hosts. Compare read and write latency across the affected objects.
Check datastore capacity
Review free space and snapshot growth. Low capacity can cause unexpected storage behavior and can make VM operations slower.
Check ESXi storage paths
Review VMkernel storage paths, path state and recent APD or PDL events. Compare the affected host with a healthy host using the same datastore.
Check queue and contention
Look for high device latency, queue pressure and competing workloads. A single busy workload can affect several VMs sharing the same datastore.
Review backend storage
Coordinate with the storage team when the issue spans hosts. Check array performance, controller health, replication activity and maintenance windows.
Validate recovery
Confirm latency returns to the normal range and verify affected VMs have no continuing storage errors in vCenter or ESXi logs.
Explore more TechRunbook guides
Browse practical troubleshooting guides, PowerShell scripts, checklists and infrastructure resources for your environment.
Browse all articles →