I'm getting purple screen and "Spin count exceeded - possible deadlock with PCPU" Vmware ESXi 6.5.0 Releasebuild 4887370 HPz620

[cronos1298]

Hello community

I'm testing this ESXi 6.5.0 and I'm facing a situation with two guest VM linux  after a while I'm getting this error,

any recommendation or comment about it?

[msripada]

Hello cronos1298,,

Have you checked the VMware / HP compatibility for ESXi 6.5. I do not see the HP z620 in the list of compatible hardware.

VMware from HP Certification and Support | HP®

Thanks,

MS

[fomer]

H/W driver  issue.  Please update the server with latest HP firmware and bios update. this should take of the problem

[cronos1298]

It is a server that was running Vmware workstation with windows 7 and Linux,  so I'm trying to migrate to ESXi and check the performance of those VMs, I cannot see the HP Z629 neither,

Thanks for your valuable  help

[generalc]

I am running 6.5 on an ML350 Gen9, which is definitely in the list. Every driver and firmware version was the latest supplied by HP. Still happened.

Just happened again today, but I see that HP just released a new SPP last week. I will try updating again. I hope HP got this worked out.

[eugeneng0209]

Hi generalc

Did the SPP work for you?

I'm having a similar issue but looking through the list of component updates, firmware versions were not updated.

[oldsk00l]

did you solve the problem? I have the same with my HP Server and ESXi 6.5

[priscillarega]

i'm having the same problem with an UCS B200 M4. We eventually isolated the problem to a virtual machine that when was powered on, in about 30 to 60 minutes would cause the PSOD (spin count problem). We moved this virtual machine to all the hosts on the cluster, and they all got de PSOD eventually.

This virtual machine is a firewall, so i'm gessing there's something wrong with the network driver when you have high loads of packets, or something like that because it's highly unlikely one small VM crash the host. Physically we got the support from Cisco and they said everything was fine

I have an open ticket with VMware, they are investigating. When i hear back from them i'll update here.

Just so you guys know, i'm running Vmware-ESXi-6.5a.0-4887370-Custom-Cisco-6.5.0.3 image.

[vigilin]

I have the same issue.

I had restored a volume from a server being decommissioned. Turns out that the vmdk that was restored was at HW ver 4, I'm guessing because the server the backup came from was esxi version 4.  I got this purple screen of death while copying files to this volume.

I had to attach this vmdk as an IDE drive.  I wonder if you guys have a similar scenario.

I ended up converting the drive to a newer hardware version and re-attaching it as a SCSI drive.  I converted a day ago and have been copying to this volume without an issue so far.

[stagnant_pondie]

Same issue here with UCS B440 M2 blades running 3.1(2a) firmware. We just updated to latest Cisco custom ISO (Vmware-ESXi-6.5.0-5969303-Custom-Cisco-6.5.1.1.iso) hoping the issue was resolved, but it is not. Obviously it could differ from Cisco to HP to Dell, but there may be a specific type of driver causing this, like network? Has anyone found a solution with VMware support? Details would be appreciated!