VMware ESXi 3.5 238493 PSOD on AMD Opteron 8356

We have a major incident in our vSphere Environment on 2 ESXi server running ESXi 3.5 Patch 238493. These servers are HP Dl585 G5 with 4x Quad Core AMD Opteron 8356 and 64 GB Memory.

Everytime VM’s getting migrated to this server using VMotion and DRS the server crashed. I have opened a Support Request on VMware site.

Steps to reproduce:

  1. add server (HP Dl585 G5 AMD 8356) to cluster
  2. exit maintenance mode
  3. set another server in this cluster in maintenance mode
  4. VM’s get live migrated to new server
  5. after 3 – 14 VM’s the new server crash’s with PSOD
  6. cancel maintenance mode on other server to save the environment

I checked our Hardware if there is a HW issue but on/in our Boxes everything is working perfectly.

Currently I test some workarounds to get our environment back stable.

  1. Install server with ESXi 3.5 Patch 207095 which is the basic Update 5 without any patches. –> currently testing
  2. Upgrade server to ESXi 4 –> pending results of the first workaround

So my current suggestion is DO NOT PATCH your ESXi installable to 3.5i Patch 238493.

to be continued … 🙁