Deepstack alert cancelled, Deepstack restarted

i got Blueiris installed on ESXI 7.0
Cpu 6700k
8gm RAM
GPU GTX 1660
I installed the CPU version of deep stack yesterday but the alert time was in the high 3 sec - 10 sec range which caused alot of missed recordings.
i decided to install my GPU version today with the 1660. I uninstalled the CPU version of the deepstack.
Installed cuda_10.1.105_418.96_win10
installed cuda_10.1.168_425.25_win10 ( Update 1 )
installed cuda_10.1.243_426.00_win10 ( Update 2 )
copied the contents of cudnn-10.1-windows10-x64-v8.0.5.39 into C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1
then finally installed DeepStack-Installer-GPU-2021.02.1.

When i started it all back up the alerts was cancelling with a 15 sec response and kept stating that deep stack is restarting.
i moved the deepstack to port 8085 incase something was blocking it and its the same.
anyone know what is wrong?

I think when I installed the GPU version I used the below
CUDA Toolkit 10.2 Download, Patch1, and Patch2
CUDA Toolkit 10.2 Download
cuDNN v8.2.4 (September 2nd, 2021), for CUDA 10.2
cuDNN Library for Windows10 (x86)

its working fine now.
i think it was due to when restarting the VM the GPU was in a error 43 state inside device manager. had to reboot the entire rig to get back to normal state.
i also upgrade to the latest 11.x build and is still working.