Known Issues

a large number of HW false alarms are outputted on ESXi6.5u3/6.7u3

(:Last updated)

Summary

In ESXi 6.7 Update 3, we have confirmed an issue with a Sensor -1 type hardware alarm on an ESXi host, even though there is no actual problem.
(VMware reports that the same issue will occur with ESXi6.5 Update 3.)

This can cause excessive email alerts to be sent if the vCenter Server system is configured for email notifications for hardware sensor status alarms.

The impact of this is that these emails can cause storage issues in the vCenter Server database and prevent access to the vCenter service if the statistics, events, alarms, and tasks (SEAT) directories exceed the 95% threshold.

Solution

To avoid the issue, update to the following releases or higher. (ESXi host reboot is required)

・ ESXi 6.5 P07 (Build number: 18678235/released October 2021)
VMware ESXi 6.5, Patch Release ESXi650-202110001

・ESXi 6.7 EP 13 (Build number: 15018017/Released November 2019)
VMware ESXi 6.7, Patch Release ESXi670-201911001

Workaround (the way not to patch ESXi hosts)

You can work around the problem by doing 1) or 2) below.

1) Disabling WBEM Services

You can work around the problem by disabling the WBEM service.
*If you disable the WBEM service, you will not be able to check the hardware health status.

SSH into the ESXi host and execute the following command.
esxcli system wbem set --enable false
When an ESXi host is rebooted and if a 3rd party providers are present, the WBEM service will become enabled on reboot.
To disable it again, rerun the above after reboot.

To prevent the WBEM service from being enabled on every reboot, do the following:
chkconfig sfcbd-watchdog off

2) Limit sending alert emails

To avoid sending a large number of alert emails without disabling the WBEM service, perform the following settings to restrict the sending of alert emails.

  1. Log in to the vCenter Server using the vSphere Client.
  2. Select vCenter Server and click Configure.
  3. Click Alarm Definitions.
  4. Select the Alarm Name Host hardware sensor state.
  5. Click Edit > Next.
  6. In the Alarm Rule section, click Add Argument.
  7. Select argument as sensorNumber and select operator as does not start with and provide sensor number as -1 which is causing alarm to trigger the alert.
  8. Click Next and Save.

 

for reference)  VMware KB
Excessive Hardware health alarms being triggered for “Sensor -1 type” on ESXi hosts running vSphere 6.7/6.5 (74607) (vmware.com)