The total number of virtual CPU present or requested in the configuration of the virtual machine has exceeded the limit on that host

There are 2 clusters in a VC, and one of the virtual machines in the cluster cannot Power ON. Prompt:

The total number of virtual CPU present or requested in the configuration of the virtual machine has exceeded the limit of 96 on that host

The total number of virtual CPUs present or requested in virtual machines' configuration has exceeded the limit on the host 96

But directly from the ESXi host can be normal Power ON.

Another cluster can be used normally.

Inexplicably, this error occurs, and the physical resources are fully sufficient. After searching the Internet for a long time, I couldn't find a solution.

During this period, VC is also restarted, and the host is removed and re-added or fruitless.

Later, after consulting in the QQ group, some netizens mentioned that there is a ratio setting of vCPU to pCPU in DRS, but at first I was thinking that I didn't even enable DRS, which doesn't make any sense.

Later, there was no way to enable DRS, and you can see a parameter in the advanced options of DRS:

The MaxVcpusPerClusterPct value is 80. After changing this value to 200, the boot prompt is the same again, but the later number is changed to 200.

Decisively reconfigure DRS to delete this parameter, and the virtual machine will boot normally.

What I don't understand is why this parameter is still valid without DRS.