ESXi 5.5 PSOD upon unplugging nework cable

[ycheung]

Encountering this strange issue on a few host. The host will PSOD if I unplug any network cables that is connected to the QLogic 57810 10GB network card. The hardware ESXi is running on is a Dell PowerEdge R720. The ESXi build is the latest (6480324) and I am running the Dell version of ESXi 5.5. I don't think this is a hardware issue because I have more than one host that has this issue. I updated the BIOS and did the service update utility but issue persist. I also did a clean install of ESXi and before I did any configuration, I unplugged a network cable, and it just PSOD.

Driver Info:

         Bus Info: 0000:41:00.0

         Driver: bnx2x

         Firmware Version: FFV10.01.00 bc 7.14.5 phy 1.45

         Version: 2.713.10.v55.4

[rsk007]

It seems either Memory module or motherboard faulty. Collect the logs and contact the Dell Support.

however, try a reboot some time this fixes the issue. But I suggest you to reach the hardware vendor for complete fix.

There is VMware KB article (https://kb.vmware.com/s/article/2149043?language=en_US) for the same PSOD as a known bug this for HP ProLient server.

Have a look:

To work around this issue, re-enable the Intel IOMMU interrupt remapper on the ESXi host:

1. Connect to the ESXi host with an SSH session and root credentials.
2. Run this command:
   
   esxcli system settings kernel set --setting=iovDisableIR -v FALSE
   
   
3. Reboot the ESXi host.
4. Ensure that the iovDisableIR setting is set to FALSE by running this command:
   
   esxcli system settings kernel list -o iovDisableIR
   
   For example:
   
   esxcli system settings kernel list -o iovDisableIR
   
   Name Type Description Configured Runtime Default
   ------------ ---- --------------------------------------- ---------- ------- -------
   iovDisableIR Bool Disable Interrupt Routing in the IOMMU... FALSE FALSE TRUE

[ycheung]

The host is currently in maintenance mode and has been rebooted many times. I have at least 2 host with the same issue, it is unlikely a hardware issue because the effected network card model are the same (QLogic 57810). From QLogic's website they described similar issue but it only effects the 11th Gen of the PowerEdge server. (QLogic Support ). I tried to turn off the C settings in the Bios but on the 12th Gen server, I do not see this setting.