The issue is related to the Guest OS experiencing TX hang when using a UCS chassis and multiple TX queues with RSS enabled.
Kernel logs will show examples like
kernel: vmxnet3 0000:0b:00.0 eth#: tx hangkernel: vmxnet3 0000:0b:00.0 eth#: resettingvar/run/log/vmkernel.log will show VM hang detected
Vmxnet3: 24934: <VMNAME>,<MAC ADDR>, portID(6710##71): Hang detected,numHangQ: 1, enableGen: 0Note: Tx hang can be due to many different reasons during packet processing, and not completing packets in a timely manner.
Validate via command: Edit the vmnic# as per your environment
vsish -e get /net/pNics/vmnic#/propertiesproperties { Driver Name:nenic Driver Version:1.0.##.0 Driver Firmware Version:4.3(3c) System Device Name:vmnic#vsish -e get /net/pNics/vmnic#/txqueues/queueCount8vsish -e get /net/pNics/vmnic#/statstx_queue_[0]_frames_ok: 3750270419tx_queue_[1]_frames_ok: 0tx_queue_[2]_frames_ok: 0tx_queue_[3]_frames_ok: 0tx_queue_[4]_frames_ok: 0tx_queue_[5]_frames_ok: 0tx_queue_[6]_frames_ok: 0tx_queue_[7]_frames_ok: 0rx_rss_queue_[0]_frames_ok: 486774846rx_rss_queue_[1]_frames_ok: 640330374rx_rss_queue_[2]_frames_ok: 830320848rx_rss_queue_[3]_frames_ok: 800527260rx_rss_queue_[4]_frames_ok: 1039549939rx_rss_queue_[5]_frames_ok: 464512348rx_rss_queue_[6]_frames_ok: 1108965446
Note: As seen, the nenic driver is being used, which is configured for 8 Tx queues, but only 1 Tx queue is being used.
VMware vSphere ESXi
A TX hang can occur when the incorrect configuration is applied to the network adapter policy of a UCS adapter.
In this instance, it is recommended to contact the hardware vendor for network adapter policy when it comes to TX and RX queue settings with RSS/netqueue.