Virtual Machines Hang or Experience Poor Performance on vSAN ESA During Large Amount of Unmap Activity
Article ID: 388241
Updated On:
Products
Issue/Introduction
Virtual machines running on VMware vSAN ESA (8.x & 9.x) may experience severe latency, hanging, or high host CPU utilization following a high volume of guest OS TRIM or Unmap operations. This performance degradation occurs because excessive space reclamation requests overload the vSAN Distributed Object Manager (DOM) queues, leading to storage backend contention.
- Any script initiated from the Guest OS takes significantly longer to complete than expected (e.g. 12 hours while previously the same script completed within approximately 3 hours).
- After issuing TRIM or Unmap operations from a guest operating system, virtual machines on the same ESXi host experience hanging, lockup, extreme latency, or other undesired impact.
- This is often but not always seen with a sudden increase in CPU utilization for the host the guest is running on.
Environment
VMware vSAN ESA 8.x & 9.x
VMware OSA 8.X & 9.X
Cause
The Guest Operating System generates a high volume of SCSI UNMAP (TRIM) commands within a short period. This excessive number of space reclamation requests overloads the vSAN Distributed Object Manager (DOM) queues. As a result, backend contention increases, leading to delays in servicing regular read and write operations for workloads such as database activity and CPU contention on host.
Resolution
Open a support request with Broadcom and provide the captured vSAN native trace logs for further analysis.
Broadcom Support can analyze the trace data and determine appropriate advanced configuration changes or tuning parameters to control the rate of background UNMAP operations and prevent storage queue contention.
Additional Information
Feedback
