Hello community,
Since a few weeks we have 5 Dell R740 servers with Tesla NVidia cards and ESXi 6.5 installed on it. We use these servers to deploy VDI Windows 10 computers on it.
We have Dell R730 servers with ESXi 6.0 installed on it and no problems there. (same VCenter but in different Clusters) VDI image is the same.
At random these new servers are crashing.
When the Tesla cards are placed in the machine the error results in a PSOD.
When we remove the Telsa cards, the systems are still failing, but then with a complete system stop and a CPU1 or CPU2 error in the logging most of the time.
So It isn't just the Telsa cards wich is giving the problems.
We have tried the Dell ESXi 6.5, 6.5 cu1 and 6.5 cu2 image, still the same results.
We have updated the Bios version to 1.3.7 and updated the iDrac to version 3.18.18.18, still the same results.
Hardware is replaced by Dell on several servers (CPU, Mainboard, PSU, NIC, SD module) but no results.
It isnt just a hardware problem it seems.
We have send several VMWare logs and Dell system logs to Dell support. They are still investigating but so far no results.
Maybe some of you have an idea or had a similar problem. Please let me know!
Thnx in advance!
Dennis Baaijens
Here are some of the PSOD screens. Het bericht is bewerkt door: Dennis Baaijens