VCF 9.1 Workload Domain deployment fails at 'Validate vMotion Network Connectivity'
Article ID: 442120
Updated On:
Products
Issue/Introduction
- While creating a new workload domain in VMware Cloud Foundation (VCF) 9.1, the workflow fails during the validation phase with the following error:
Message: Validate vMotion Network Connectivity Remediation Message: Verify underlying hosts have vMotion connectivity
- In SDDC manager /
var/log/vmware/vcf/domainmanager/domainmanager.log, the following error is observed:
YYYY-MM-DDTHH:MM:SS.360+0000 ERROR [vcf_dm,0000000000000000,0000] [c.v.e.s.c.h.n.EsxiHostNetworkingUtil,dm-exec-2692] Unable to ping ##.##.##.## via VM Kernel vmk## with MTU 8972 with ip address version IPv4 from ESXi host <FQDN_OF_ESX_HOST> and netStack null:. Cause: null com.vmware.evo.sddc.common.client.vmware.esx.models.LocalizedMethodFault: sendto() failed (Host is down) at java.base/jdk.internal.reflect.DirectConstructorHandleAccessor.newInstance(DirectConstructorHandleAccessor.java:62) at java.base/java.lang.reflect.Constructor.newInstanceWithCaller(Constructor.java:502) at java.base/java.lang.reflect.Constructor.newInstance(Constructor.java:486) YYYY-MM-DDTHH:MM:SS.993+0000 ERROR [vcf_dm,6a106fc######d3f6d4#####3983ae,1290] [c.v.e.s.o.model.error.ErrorFactory,dm-exec-2656] [####] VALIDATE_VMOTION_CONNECTIVITY Validate vMotion Network Connectivity com.vmware.evo.sddc.orchestrator.exceptions.OrchTaskException: Validate vMotion Network Connectivity at com.vmware.vcf.common.fsm.plugins.action.impl.ValidateEsxiHostNetworkConnectivityAction.validateHostsConnectivity(ValidateEsxiHostNetworkConnectivityAction.java:266) at com.vmware.vcf.common.fsm.plugins.action.impl.ValidateEsxiHostNetworkConnectivityAction.execute(ValidateEsxiHostNetworkConnectivityAction.java:175) at com.vmware.vcf.common.fsm.plugins.action.impl.ValidateEsxiHostNetworkConnectivityAction.execute(ValidateEsxiHostNetworkConnectivityAction.java:68)
Caused by: java.lang.IllegalStateException: ESX Host '<FQDN_OF_ESX_HOST>' -> '##.##.##.##' VLAN '##' with MTU 9000 fail to vmkping. Possible no network connection or not reachable, ESX Host '<FQDN_OF_ESX_HOST>'->'##.##.##.##' VLAN '###' with MTU 9000 fail to vmkping.
Possible no network connection or not reach ESX Host'<FQDN_OF_ESX_HOST>' -> '##.##.##.##' VLAN '###' with MTU 9000 fail to vmkping. Possible no network connection or not reachable
Environment
VCF 9.1
Cause
This issue occurs when the preconfigured switch profile selected during the deployment (e.g., "Storage Traffic Separating") automatically assigns physical network adapters (vmnics) that do not match the actual physical cabling or Top-of-Rack (ToR) switch configuration of the ESXi hosts.
For example, VCF may automatically assign vmnic4 and vmnic5 based on the profile, but the physical environment is cabled and configured to use vmnic5 and vmnic6. Because the incorrectly assigned vmnic lacks the necessary physical connections VLAN tags, the automated vmkping tests fail, halting the deployment.
Resolution
To resolve this issue, you must bypass the preconfigured profiles and manually map the correct vmnics using a custom switch configuration.
Restart the Workflow: Cancel the current failed task and restart the Workload Domain creation process in the SDDC Manager UI.
Bypass Preconfigured Profiles: Proceed through the wizard until you reach the Switch Configuration page. Do not select a preconfigured profile (such as "Storage Traffic Separating").
Create Custom Configuration: Select CREATE CUSTOM SWITCH CONFIGURATION.
Map Correct Interfaces: Manually map the correct, physically cabled
vmnics(e.g.,vmnic5andvmnic6) to the required traffic types for the workload domain:Management
vMotion
NSX-Overlay
Proceed with Deployment: Continue to the validation phase. The network checks should now pass using the correctly mapped physical uplinks.
Feedback
