When adding the CPU Load monitor some of our servers are having an additional CPU added. I've seen some other posts regarding this happening on 64bit servers with a high core count but this is happening on servers with just 2 cores.
Is there a fix for this?
__GENUS : 2 __CLASS : Win32_Processor NumberOfLogicalProcessors : 2
Article Comments
Is there anyway to run a query via the API to display core counts stored in PRTG? Having to manually check over 400 devices to correct this issue is a pain to say the least.
Oct, 2015 - Permalink
Dear vmorrison
This is not directly possible. However you can get a channel list of a sensor. This example uses sensor ID 2123.
/api/table.csv?id=2123&content=channels&columns=name,objid,lastvalue
Replace "table.csv" with "table.xml" to get an XML output instead.
To get a list of all sensor IDs with the tag "wmicpuloadsensor", please use
/api/table.csv?content=sensors&columns=objid,name&count=*&filter_tags=@tag(wmicpuloadsensor)
Oct, 2015 - Permalink
Dear vmorrison
In some cases, WMI appears to return a record an additional CPU for a short amount of time, which lets PRTG create a new channel. This false channel can be manually hidden from graphs and tables. Do do so, please click on its gauge to open the channel options and select the according options.
The channel will still be listed in the channel table on the overview tab, because PRTG does not allow the actual deletion of a sensor channel.
Oct, 2015 - Permalink