Currently there is no resolution. VMware engineering is aware of this issue.
Please subscribe to this article, using the [SUBSCRIBE] button in the top right corner of this article, to be made aware when a fix has been made available.
Workaround:
To work around this issue, reduce the polling rate for the ipmi module using the following command:
# esxcfg-module --set-options "regMaxPolls=200000 regPollIntrvUs=4" ipmi
Afterwards reboot the host to apply the change.
The default values for the options are:
regMaxPolls=400000
regPollIntrvUs=2
Should you find that the issue persists despite making this change, please reset both options to their default values:
# esxcfg-module --set-options "regMaxPolls=400000 regPollIntrvUs=2" ipmi
Reboot the host again to apply the change, then please file a support request with the VMware Support.