New to VIO 6 and looking for either a step by step guide or a HOL. Anything anyone could recommend? The installation and Admin guides dont go into details of how to obtain VM images, or how to properly configure External and Internal Networks and SR-IOV.
I am for example getting errors when creating networks on the GUI but the configuration fail without any explanation. Would be great to see where logs are being generated.
There is no HOL for VIO 6.0 as of yet. I don't know when there is one planned to be made available.
For networking you can check out this guide.
OpenStack&NSX-T - Network Topologies Configuration Guide
For images, this covers importing and the supported image types. Where you get them is up to you.
Hi
I’m currently testing VIO 6 in a poc.
I installed nested vCenter with 3 ESXi und NSX-T using this guide “vGhetto Automated vSphere Lab Deployment for vSphere” https://github.com/lamw/vghetto-vsphere-automated-lab-deployment in the version 6.7 U2 + NSX-T 2.31. In addition I have Software Router for VLAN separation of VIO-Mgmt and VIO-API Subnets and more, Active Directory with DNS, and a Jump VM.
Once NSX-T is configured T0 is created and the necessary DHCP Profile and MetaData Proxy exists the OVF VIO Appliance will be deployed and configured to work with the vCenter and the NSX-T.
It is working so far. I can logon onto VIO, create projects, authorise LDAP users, and create VM in GUI and via OpenStack native CLI. I can connect VIO VM’s over Edge external VLANand so on. I’m investigating currently more automation possibility and storage features.
The VIO 6 with the kubernetes architecture technique behind the scene is a “little bit” complicated. Log access over kubectl commands looks different as vmware usually. The pods architecture not clear to me for now.
Fewer articles in VMware KB for the VIO Version 6 let me guess that not many use the v6 at this time. The articles for the v5 are mostly not adequate for v6. VMUG and VMTN is also not better.
With time we will see if VIO 6 will become popular or not.
Thanks. I finally got my system up and running. Best way I found to troubleshoot what searching through the PODs logs using the failed request ID.
kubectl -n openstack logs <pod> | grep <req_id>
that is cool. In my case I've got real servers so trying an automated install usually takes more time as there are a bunch of prerequisites you need to have in place.