I have multiple VMs running on a host, and all but 2 machines (Both DCs) are reporting as okay. However, these 2 machines are being reported as an error stating that the machine could not be found and are they powered off.
They are not powered off, and are working fine, but PRTG cannot report on them. Any ideas on why?
Article Comments
Hi, thanks for your response. I have confirmed that the UAC is turned off, and recreated the sensor but still the same thing.
Oct, 2015 - Permalink
Hi,
could you please run the sensor manually? The sensor is located at <PRTG Install Directory>\Sensor System and named SCVMMSensor.exe. Pleas run the same with the following call:
SCVMMSensor.exe "-host=<host>" -act=GetHostData "-user=<domain>\<user>" "-pass=<password>" "-id=<hostserver>"
please replace the params in <> accordingly. What results do you get?
Oct, 2015 - Permalink
This is the response from running it manually ...
<prtg>
<error>1</error>
<text>The term 'Get-VMMServer' is not recognized as the name of a cmdlet
, function, script file, or operable program. Check the spelling of the name, or
if a path was included, verify that the path is correct and try again.</text>
</prtg>
This is confusing as other sensors on the same host are working for other VMs?
Oct, 2015 - Permalink
Hello Brent, Kindly confirm: Which of the following sensors are you using?
The test above SCVMMSensor.exe "-host=<host> ..." refers to the SCVMM Virtual Machine Sensor.
The Hyper-V Virtual Machine Sensor can display the following message when the Virtual Machine is powered off or if it was renamed:
| WMI: The VM could not be found. Is it powered off? (code: PE068) -- PerfCounter: No data to return. (Performance Counter error 0x800007D5) |
The workaround in this case is to deploy the sensor for that virtual machine anew. Ideally you shouldn't rename virtual machines when monitoring them using the Hyper-V Virtual Machine Sensor.
Best Regards,
Oct, 2015 - Permalink
Hi,
I have created a new sensor (I am using the "Hyper-V Virtual Machine" sensor, which is the same sensor monitoring all of my other virtual machines running on the same host that are working perfectly) and the same problem occurs.
(When I create the sensor, the "wizard" correctly detects the VM)
Oct, 2015 - Permalink
Hello Brent.
Are you running the latest version of PRTG?
Does the wizard correctly detect the VM but it goes into the previously described error as soon as you add it?
To further troubleshoot this I must ask that you contact us using this form or through a Support Bundle, we'll need logs and other details for analysis. Kindly mention this KB post when contacting us.
Best Regards,
Oct, 2015 - Permalink
If a Hyper-V virtual machine contains a slash in it's name, i.e:
| Windows Server 2012/R2 |
PRTG will not be able to monitor the virtual machine as the slash in it's name affects the WMI query, the result is a (down) sensor with the message:
The sensor could not find the Hyper-V virtual machine. To resolve this issue, check if the virtual machine might be powered off. Also make sure that the virtual machine does not contain a slash or backslash in its name. (code PE068) -- Perfcounter: The wait operation timed out (Performance Counter error 0x102)
The workaround is to rename the virtual-machine and deploy the sensor anew.
Best Regards,
Oct, 2015 - Permalink
Hi,
could you try turnig off UAC on the Guest Operating System of these VMs? Afterwards try creating the sensor again.
Does that work?
Oct, 2015 - Permalink