[[inputs.win_perf_counters]]
[[inputs.win_perf_counters.object]]
ObjectName = "AD FS"
Counters = ["Token Requests"]
Instances = ["*"]
Measurement = "win_adfs"
#IncludeTotal=true #Set to true to include _Total instance when querying for all (*).
verified perf counter exists and has a value in it
expect it to end up in prometheus
getting error [inputs.win_perf_counters]: Error in plugin: No data to return.
Which Telegraf version?
Version 1.10
At least I believe that is the version I am running. When I look at details on the exe in windows I don鈥檛 see a version
To get the version, simply run telegraf.exe --version
Sorry I run that and I get Telegraf unknown
What is the output when you run typeperf "Token Requests"?
Have you tried repairing your performance counters? Guide is located here.
I had this exact issue. You need to do the instance without the * so like Instances = [""] Then it should work.
i also had the similar issue and tried to repair the performance counter but it doesn't fixed.
@danielarkins Does the suggestion by @takenalias above solve this issue for you?
This bug is so strange. I went back to basics and reset my perf counters:
C:\Windows\System32>lodctr /r
Info: Successfully rebuilt performance counter setting from system backup store
And then tested each individually, for CPU, Mem, System I get:
2020-07-11T19:24:20Z E! [inputs.win_perf_counters] Error in plugin: No data to return.
For LogicalDisk and NICs I get data. Tried the latest nightly and also all 4 versions previously. No go. Instances=[""] also doesn't work. Weird.
just had this, massive nightmare.
I followed Biliskn3r's idea about the rebuild then I deleted the conf file, got it rebuilt (mine are rolled automatically) and then are restart of the service.
On some of the boxes I also did a uninstall as well before deleting the conf file. Weird.
Most helpful comment
I had this exact issue. You need to do the instance without the * so like Instances = [""] Then it should work.