Rx errors and drops reported in hostd.log on Cisco VIC nenic interfaces

Products

• VMware vSphere ESXi

Issue/Introduction

An ESXi 8.0.3 host reports Receive (Rx) errors and drops on physical network interfaces (vmnics) using the nenic driver. The hostd logs may trigger alerts indicating that Rx error counters are greater than zero.

Symptoms:

• hostd logs show "Error stats for Pnic" and "droppedRx" on one or more vmnics.

• esxcli network nic stats get shows non-zero values for Rx Drop or Total Err.

• Performance issues may be reported, but the error ratio is extremely low (e.g., < 0.01%).

Environment

VMware vSphere ESXi

Cause

Nominal accumulation of standard network noise over the ESXi host's uptime period.

Resolution

No action is required if the error rate is negligible relative to total traffic.

1. Review the total packet count versus error count using the following command: esxcli network nic stats get -n vmnicX

2. Calculate the error percentage. If the rate is below 0.01%, the errors are considered standard network noise and are not indicative of a hardware or driver fault.

3. Verify the driver and firmware versions against the Broadcom Compatibility Guide.

4. Monitor the counters for sudden or rapid increments. If the error rate increases significantly, investigate the physical cabling, SFP modules, or upstream switch ports.

Additional Information

Broadcom Compatibility Guide for Cisco VIC