ESXi 6.5.0 PSOD

[PeterRizk]

Hi,

I built an 8 ESXi 6.5.0 cluster that is managed by vCenter Server 6.5. The 8 ESXi host are identical.

Each host consists of:

- HP Proliant  DL380 Gen 8 (dual Xeon ES-2670 2.6 GHZ, and 385 GB RAM)

- Dual 10 GigE Intel NICs

- Eight 2.5 inch HDDs

I pushed the cluster to its limits by loading 3000 VMs (Windows, Linux, etc) and caused all the hosts in the cluster to max out on CPU and memory. I wanted to test the stability of the cluster under intense workload.

After several hours of this stress testing some of the ESXi host crash with a PSOD such as the one attached to this post. Other hosts continue to run fine without any issues. The other hosts that didn't fail continued to handle the stress workload for several days (and again without any issues).

Any idea on what might be causing this crash and how to resolve it?

Thanks.

[daphnissov]

You are critically behind in ESXi patches for the 6.5 release. A recent patch, released a couple weeks ago, has a large number of fixes, including at least one for 10 GbE adapters. It's recommended you patch up to that revision and see if you experience the same PSOD.

[PeterRizk]

Ok thanks for the response. I'll update to the latest patches and see what happens.

[admin]

Issue could be either hardware or software related.Can you please check patch version and Hp Firmware on the servers . I had same issue after upgrading the server firmware did not experience the same PSOD.

Regards,

Randhir

[PeterRizk]

Hi, thanks for the response. I will update the HP firmware soon and will see if the combination of the firmware update as well as the latest ESXi patches fixes this issue.