When a vSAN cluster is undergoing a rolling upgrade, where some hosts have been upgraded to ESXi 9.x while others remain on ESXi 8.x, the upgraded ESXi 9.x hosts trigger network alerts visible on the summary dashboards.
Host or Cluster Summary Page Alerts:
Navigating to the Summary page of the affected ESXi 9.x host or the vSAN Cluster displays the following critical alert:
High pNic error rate detected. Check the host's vSAN performance view for details
vSAN Performance View Alerts:
Navigating to Monitor -> vSAN -> Performance on the affected ESXi 9.x host displays specific timestamped alerts such as:
(where vmnicX represents the vSAN uplink adapters, e.g., vmnic2, vmnic3...)High pNic RX Generic error rate detected on vmnicX on <DATE>High pNic TX Generic error rate detected on vmnicX on <DATE>
Performance Metrics Breakdown:
Navigating to Monitor -> vSAN -> Performance -> PHYSICAL ADAPTERS shows that the "pNIC Packet Error Rate" is not 0.
Navigating to Monitor -> vSAN -> Performance -> HOST NETWORK shows that the "vSAN Host Packet Discard Rate" is not 0.
VMware vSphere ESXi 8.x
VMware vSphere ESXi 9.x
Mixed-version vSAN clusters undergoing rolling upgrades
This alert is caused by the temporary co-existence of different major ESXi versions (8.x and 9.x) within the same vSAN cluster during a rolling upgrade.
The version disparity between heterogeneous nodes affects the cluster communication metrics, leading to the pNIC alerts on the Summary page. These alerts are transitional and will clear automatically once all cluster nodes are upgraded to a unified version.
To resolve this issue.