2025-10-07T07:52:39.009Z kernel - - - [402863.323027] NETDEV WATCHDOG: eth0 (vmxnet3): transmit queue 1 timed out2025-10-07T07:52:39.010Z kernel - - - [402863.323041] WARNING: CPU: 5 PID: 0 at net/sched/sch_generic.c:477 dev_watchdog+0x25b/0x260[....]2025-10-07T07:52:39.085Z kernel - - - [402863.323319] vmxnet3 0000:03:00.0 eth0: tx hang2025-10-07T07:52:39.085Z kernel - - - [402863.323445] vmxnet3 0000:03:00.0 eth0: resetting2025-10-07T07:52:39.557Z kernel - - - [402863.331644] vmxnet3 0000:03:00.0 eth0: Failed to activate dev: error 12025-10-07T07:52:40.198Z In(05) vcpu-5 - VMXNET3 user: Quiesce device 0.2025-10-07T07:52:40.205Z In(05) vcpu-5 - VMXNET3 user: Activate device 0.2025-10-07T07:52:40.206Z In(05) vcpu-5 - VMXNET3 user: failed to activate 'Ethernet0', status: 0xbad00142025-10-07T07:52:40.206Z In(05) vcpu-5 - VMXNET3 user: Activate request failed for device 0.2025-10-07T07:52:31.367Z In(182) vmkernel: cpu39:11080375)NetPort: 1618: enabled port 0x600014c with mac 00:50:xx:xx:xx:xx.[....]2025-10-07T07:52:31.367Z In(182) vmkernel: cpu39:11080375)ENS: Ens_DoDetachPort:6038: Detach and free port 0x600014c resources (interrupt) from ENS handle: 0x450307a99cc0VMware vSphere ESXi
High host memory fragmentation can prevent VM vNICs from allocating contiguous physical memory after a vMotion. This leads to vNIC activation failure and the VM becoming unresponsive post migration.
The issue has been resolved in ESXi version 8.0p06 (build-24859861).
In the below release notes, we can search for - "PR 3480453: Virtual machine vNIC fails to activate after a vSphere vMotion operation":
VMware ESXi 8.0 Update 3g Release Notes