Hello,
i took a look on other threads, but couldn´t find a solution for my problem.
Overview:
-
Hardware is a DELL PowerEdge R740: 576 GB RAM, 2x Tesla M60, 2x Intel Xeon Gold 6142, 1,8 TB SSD RAID10 storage, 2x 10G Network
-
XenServer 8, actual release, XenServer Premium Edition-Licence, 2 Sockets licenced
-
XenCenter actual release 2024.4.0
-
Downloaded and installed via “XenCenter → Tools - Install Updates” the .ISO of “Complete vGPU 17.4 package for Citrix XenServer 8 including supported guest drivers” (XenServer supplemental pack)
-
both M60 cards GPU Mode 300 → graphics mode
-
deactivated Option for the XenServer-Host in “Properties → GPU” not using the integrated GPU is checked
-
nvidia licence server in the nvidia cloud running with vWS Licences (were running GRID-Servers with GPU passtrough, and those servers are visible in the cloud-GUI of the nvidia licence portal)
Problem:
i can´t choose vGPU porfiles when i setup a Win 11 VM on the XenServer Host. The drop down menu shows the Tesla Cards themselves and no further vGPU profiles are listed. So only GPU passtrough is available and as i said, both Tesla M60 are shown. i assume, that theres no connection to the nvidia licence server at the nvidia licence portal?! If thats right, how do i setup the connection from the xenserver host (or xencenter?) to the nividia licence portal, where our nividia licence server is running? Or am i on the wrong track?
Thanks for your advice and regards!