In this post I just want to share with the community our experience with one issue that we faced recently in one of the new NSX-T projects that our team is delivering.
This is a fresh deployment of NSX-T 18.104.22.168 in preparation for a lift and shift migration from NSX-v.
After NSX-T was deployed, two vCenter compute managers were added and then we noticed a violation for the configuration maximums (see below screenshot) which was really weird for us. Are we back to the old NSX-v days and the 1:1 relationship between NSX and vCenter? Well, this should NOT be the case!!
We consulted the GSS team to check what is wrong with the deployment. We thought primarily that it is a license issue forcing some limitations. After checking, the GSS team told us that the NSX-T managers were deployed with the MEDIUM size and it only supports adding one compute manager. Wait, does this mean that NSX-T manager sizing in the new version 22.214.171.124 is not only related to the number of hypervisor nodes (as before) and is also affecting the number of the supported compute managers ? YESSS !!
This information is not published publicly yet as NSX-T 3.2 documentation is still being updated. So for now, just take the below into consideration until things are more clear.
- Medium NSX-T manager supports One compute manager
- Large NSX-T manager supports up to 16 compute managers
Thanks for reading,
I hope this post was informative,