Users may experience vMotion or DRS operations failing with the error message:
"The migration was canceled because the amount of changing memory for the virtual machine was greater than the available network bandwidth."
Key log entries indicate driver-firmware problems include messages similar to:
vmkernel.log:
WARNING: bnxtnet: bnxtnet_check_link:3185: [vmnic4 : 0x452140054000] Unable to receive DCB data from firmware
WARNING: bnxtnet: bnxtnet_dcb_get_setting:1653: [vmnic5 : 0x452140f40000] invalid dcbx_cap 0
Especially in conjunction with firmware trace dump messages, such as:
2024-08-30T20:55:43.337Z In(182) vmkernel: cpu40:58235078)bnxtnet: dev_dump_fw_trace:153: [vmnic5 : 0x452140f40000] Dumping FW Trace:
These entries suggest communication issues between the driver and firmware, which could lead to network performance problems and vMotion failures. This issue can occur even when there appears to be sufficient network bandwidth available.
In certain configurations, this error message can occur due to network driver or firmware issues rather than actual memory changes or bandwidth limitations. These issues may prevent proper network communication between hosts during vMotion.
For general troubleshooting of vMotion failures with this error message, see vMotion fails with the error: The migration was cancelled because the amount of changing memory for the VM was greater than the available network bandwidth
If you've tried the steps in KB 332734 and continue to experience issues, check for signs of network driver or firmware problems: