After the Upgrade of ESXi's to either 6.5 or 6.7 you see alert "The target host does not support the virtual machine's current hardware requirements" and unable to reconnect hosts back to EVC enabled Cluster
book
Article ID: 339981
calendar_today
Updated On:
Products
VMware vSphere ESXi
Issue/Introduction
You had ESXi 6.0 on latest Build which has enabled Intel MDS mitigation
You upgraded to latest 6.5 version which doesn't have Intel MDS mitigation enabled
EVC Cluster is not allowing you to reconnect Hosts back after the upgrade
EVC baseline is SandyBridge
You are able to add IvyBridge ESXi's to the EVC Enabled Cluster
You are unable to reconnect ESXi's with Sandybridge CPU to EVC enabled Cluster with SandyBridge baseline
Also if you have SandyBridge EVC enabled Cluster with Sandybridge and IvyBridge CPU ESXi's, on latest ESXi 6.5 and ESXi 6.7 Build versions you will not be able to use vMotion between ESXi's due to MDCLEAR flag difference between them and EVC will complain
Error stack received in vpxd is similar to below:
<timestamp>.168-03:00 info vpxd[15928] [Originator@6876 sub=Default opID=AddHostWizard-apply-23910-ngc:70003712-35] [VpxLRO] -- ERROR task-29922 -- host-<id>-- vim.HostSystem.reconnect: vim.fault.EVCAdmissionFailedCPUFeaturesForMode: --> Result: --> (vim.fault.EVCAdmissionFailedCPUFeaturesForMode) { --> faultCause = (vmodl.MethodFault) null, --> faultMessage = <unset>, --> productName = <unset>, --> productVersion = <unset>, --> faults = (vmodl.MethodFault) [ --> (vim.fault.FeatureRequirementsNotMet) { --> faultCause = (vmodl.MethodFault) null, --> faultMessage = <unset>, --> featureRequirement = (vim.vm.FeatureRequirement) [ --> (vim.vm.FeatureRequirement) { --> key = "cpuid.MDCLEAR", --> featureName = "cpuid.MDCLEAR", --> value = "Bool:Min:1" --> } --> ], --> vm = <unset>, --> host = 'vim.HostSystem:ha-host' --> msg = "The target host does not support the virtual machine's current hardware requirements." --> } --> ], --> currentEVCModeKey = "" --> msg = "The host's CPU hardware should support the cluster's current Enhanced vMotion Compatibility mode, but some of the necessary CPU features are missing from the host. Check the host's BIOS configuration to ensure that no necessary features are disabled (such as XD, VT, AES, or PCLMULQDQ for Intel, or NX for AMD). For more information, see KB articles 1003212 and 1034926." --> } --> Args: --> --> Arg cnxSpec: --> (vim.host.ConnectSpec) { --> hostName = "<hostname>", --> port = -1, --> sslThumbprint = <unset>, --> userName = "root", --> password = (not shown), --> vmFolder = 'vim.Folder:<id>e:group-v3', --> force = true, --> vimAccountName = <unset>, --> vimAccountPassword = <unset>, --> managementIp = <unset>, --> lockdownMode = <unset>, --> hostGateway = (vim.host.GatewaySpec) null --> } --> Arg reconnectSpec: -->
Some of the perVM EVC test cases failed because of a missing MDCLEAR feature which isn't supported for SandyBbridge and lower hosts. The Intel microcode was not updated for this feature. Issue is the flag MDCLEAR.
Resolution
Patch the hosts to the Build version in which MDCLEAR flag enabled for SandyBridge CPU generation Hosts
This issue has been addressed in below mention vSphere releases 1. vsphere60p08 2. vsphere65p04 3. vsphere67p01
Workaround:
Revert ESXi Upgrade to previous version on ESXi 6.0 using SHIFT + R because Patch for 6.0 is already released