VMware Networking Community
Henrique_Cicuto
Enthusiast
Enthusiast
Jump to solution

NSX-T "Failed to add VTEP"

Good morning,

Trying to do a small vSphere cluster prep in NSX-T for a POC and I've been unable so far.

NSX-T: 2.4.0.0.0.12456646

ESXi:6.7.0 build 14320388

vCenter: 6.7.0.41000 build 14836122

Already checked matrix and those versions are supported.

A 2 host cluster. 2 nics (vmnic0 and vmnic1) are used in a default vDS with management and vMotion.

vmnic1 and vmnic2 are free for n-vds usage.

Transport Zone ESXi TZ:

--> N-VDS name: vSphere NVDs

--> Enhanced Datapath

--> Overlay

Transport Node Profile:

--> TZ: ESXi TZ

--> N-VDS vSphere NVDs

-----> Uplink Profile: nsx-edge-multiple-vteps-uplink-profile (default)

-----> LLDP Profile: LLDP [Send Packet Disabled]

-----> IP Assignment: Use IP Pool

-----> IP Pool: VTEP Addressing

-----> Physical NICs: vmnic2 - uplink-1, vmnic3 - uplink-2

Uplink Profile nsx-edge-multiple-vteps-uplink-profile:

--> Teaming Policy: Load balance source

--> Active Uplinks: uplink-1,uplink-2

--> Using global MTU

IP Pool VTEP Addressing:

--> IP Ranges: 10.0.0.11-10.0.0.14

--> CIDR: 10.0.0.0/24

--> No gateway or DNS server/suffix

Configuring cluster with NSX finishes with a "Partial Success". Hosts are TZ members and N-VDS is created, but the TEPs aren't.

"Partial Success" message shows:

Host configuration: VTEP [vmk10,10.0.0.12] failed to be applied: Failed to add VTEP: {'msg': 'An error occurred during host configuration.', 'fault': 'PlatformConfigFault', 'faultMessage': ['Operation failed, diagnostics report: Unable to complete Sysinfo operation. Please see the VMkernel log file for more details.: Sysinfo error: Not readySee VMkernel log for details.']}; VTEP [vmk10,10.0.0.14] failed to be applied: Failed to add VTEP: {'msg': 'An error occurred during host configuration.', 'fault': 'PlatformConfigFault', 'faultMessage': ['Operation failed, diagnostics report: Unable to complete Sysinfo operation. Please see the VMkernel log file for more details.: Sysinfo error: Not readySee VMkernel log for details.']}; App Init configuration: AppInitConfig skipped. LogicalSwitch full-sync: LogicalSwitch full-sync skipped.

For both hosts (only difference are the IPs used for each one).

Tried checking manager.log and vmkernel.log but couldn't find anything that seemed like errors.

Tried creating a vDS and vmkernels using those exact same addresses manually and works ok.

Anyone has ever seen anything like it and could give some directions where to look further?

Thank you very much

Reply
0 Kudos
1 Solution

Accepted Solutions
Henrique_Cicuto
Enthusiast
Enthusiast
Jump to solution

Not sure why but creating a new TZ using "Standard" instead of "Enhanced Datapath" did the trick.

Not exactly a solution since I have no idea why enhanced caused an issue because no specific logs are generated anywhere.

View solution in original post

Reply
0 Kudos
3 Replies
Henrique_Cicuto
Enthusiast
Enthusiast
Jump to solution

Not sure why but creating a new TZ using "Standard" instead of "Enhanced Datapath" did the trick.

Not exactly a solution since I have no idea why enhanced caused an issue because no specific logs are generated anywhere.

Reply
0 Kudos
lizburke
Contributor
Contributor
Jump to solution

Enhanced Datapath is for NFV workloads, and the NIC needs to be able to support it.  Probably why the install failed.

Reply
0 Kudos
diegodco31
Leadership
Leadership
Jump to solution

As Liz said, Enhanced Datapath is for NFV workloads.

See:

Enhanced Data Path

Diego Oliveira
LinkedIn: http://www.linkedin.com/in/dcodiego
Reply
0 Kudos