oc get nodes command is ran inside the OCP Node, one or multiple nodes shows status as NotReady.NotReady is unreachable over Network.netdbg vswitch instance listnsxdp-cli vswitch instance listvmxnet3 0000:02:01.0 ens33: tx hang
vmxnet3 0000:02:01.0 ens33: resetting
vmxnet3 0000:02:01.0 ens33: intr type 3, mode 0, 9 vectors allocated
vmxnet3 0000:02:01.0 ens33: Failed to activate dev: error 1
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: Quiesce device 0.
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: UPT support is not requested
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - Ethernet0 MAC Address: 00:50:56:##:##:##
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: Ethernet0 RSS fields requested by vmx: f
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: Activate device 0.
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: failed to activate 'Ethernet0': invalid magic value
YYYY-MM-DDTHH:MM:SS.SSSZ In(05) vcpu-14 - VMXNET3 user: Activate request failed for device 0.
txdescriptor ring corruption is observed for the switchport where Virtual Network Adapter of the impacted node is connected. Following dump can be observed in /var/run/log/vmkernel.logYYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488471)Vmxnet3: 2835: Tq: 1 port: 0x6000021 start desc: 451 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488471)Vmxnet3: 2846: Tq: 1 port: 0x6000021 start desc: 455 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488471)Vmxnet3: 2835: Tq: 1 port: 0x6000021 start desc: 459 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488471)Vmxnet3: 2846: Tq: 1 port: 0x6000021 start desc: 463 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu0:2488472)Vmxnet3: 2835: Tq: 2 port: 0x6000021 start desc: 126 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488474)Vmxnet3: 2835: Tq: 4 port: 0x6000021 start desc: 316 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488474)Vmxnet3: 2846: Tq: 4 port: 0x6000021 start desc: 320 hexdump: 0xff00000000 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu0:2488472)Vmxnet3: 2846: Tq: 2 port: 0x6000021 start desc: 130 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu0:2488472)Vmxnet3: 2835: Tq: 2 port: 0x6000021 start desc: 134 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488474)Vmxnet3: 2835: Tq: 4 port: 0x6000021 start desc: 324 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488474)Vmxnet3: 2846: Tq: 4 port: 0x6000021 start desc: 328 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu0:2488472)Vmxnet3: 2846: Tq: 2 port: 0x6000021 start desc: 138 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488475)Vmxnet3: 2835: Tq: 5 port: 0x6000021 start desc: 110 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488475)Vmxnet3: 2846: Tq: 5 port: 0x6000021 start desc: 114 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488475)Vmxnet3: 2835: Tq: 5 port: 0x6000021 start desc: 118 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
YYYY-MM-DDTHH:MM:SS.SSSZ In(182) vmkernel: cpu4:2488475)Vmxnet3: 2846: Tq: 5 port: 0x6000021 start desc: 122 hexdump: 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff 0xffffffffffffffff
VMware vSphere ESXi
Server with AMD CPUs
Red Hat Enterprise Linux Operating System
NOTE: The issue has currently been isolated to Red Hat, and their engineering team is actively working on it.
Reference: VMware guest with large memory hangs
There are two workarounds applicable at the moment which are as follows:
1. Set iommu=pt as the boot parameter.
For any assistance required to configure the above parameter, please reach out to Red Hat customer support.
2. Set the parameter vvtd.enable to false in the vmx configuration file of the impacted Virtual Machine as follows:
vvtd.enable = "FALSE"
For any assistance required to configure the above parameter, please open a Case with Broadcom Support and mention this Knowledge base article.