Telegraf: Collect Hyper-V CPU Wait Time per Dispatch (CPU Ready): Unterschied zwischen den Versionen

Aus Wiki-WebPerfect
Wechseln zu: Navigation, Suche
(Die Seite wurde neu angelegt: „CPU Wait Time per Dispatch asdasd Kategorie:TIG-Stack Kategorie:Hyper-V / VMM“)
 
 
(8 dazwischenliegende Versionen des gleichen Benutzers werden nicht angezeigt)
Zeile 1: Zeile 1:
−
[[CPU Wait Time per Dispatch]] asdasd
+
More informations about the ''CPU Wait Time per Dispatch (CPU Ready)'' here: [[Hyper-V: Performance (Counters)]]
  
 +
To calculate the CPU Wait Time per Dispatch for each VM you need following Telegraf Plugins:
 +
*'''Input:''' win_perf_counters
 +
*'''Processor:''' regex
 +
*'''Aggregator:''' basicstats
 +
 +
 +
== Input: win_perf_counters ==
 +
'''''Collects the data for each virtual processor''' about CPU Wait Time per Dispatch from Perfmon.''
 +
<pre>
 +
[[inputs.win_perf_counters.object]]
 +
  ObjectName = "Hyper-V Hypervisor Virtual Processor"
 +
  Instances = ["*"]
 +
  Counters = [
 +
    "CPU Wait Time per Dispatch",
 +
  ]
 +
  Measurement = "hyperv_virtual_processor"
 +
</pre>
 +
 +
== Processor: regex ==
 +
'''''Extract the VM-Name''' from the Perfmon Counter Instance and create a new Tag "VMName" with it.''
 +
<pre>
 +
[[processors.regex]]
 +
  namepass = ["hyperv_virtual_processor"]
 +
  [[processors.regex.tags]]
 +
    # Extract the VM-Name from the Perfmon Counter Instance and create a new Tag "VMName" with it
 +
    key = "instance"
 +
    pattern = '^(?P<VMName>.+)\:.+$'
 +
    replacement = "${VMName}"
 +
    result_key = "VMName"
 +
</pre>
 +
 +
== Aggregator: basicstats ==
 +
'''''Calculates the sum''' of all CPU Wait Time per Dispatch '''foreach VM'''.''
 +
<pre>
 +
# Aggregate the sum of the virtual processors wait time per VM.
 +
[[aggregators.basicstats]]
 +
  namepass = ["hyperv_virtual_processor"]
 +
  fieldpass = ["CPU_Wait_Time_per_Dispatch"]
 +
  taginclude = ["VMName", "host"]
 +
 
 +
  ## The period on which to flush & clear the aggregator.
 +
  period = "30s"
 +
 +
  ## If true, the original metric will be dropped by the
 +
  ## aggregator and will not get sent to the output plugins.
 +
  drop_original = true
 +
 +
  ## Configures which basic stats to push as fields
 +
  stats = ["sum"]
 +
</pre>
  
  

Aktuelle Version vom 5. Januar 2022, 17:17 Uhr

More informations about the CPU Wait Time per Dispatch (CPU Ready) here: Hyper-V: Performance (Counters)

To calculate the CPU Wait Time per Dispatch for each VM you need following Telegraf Plugins:

  • Input: win_perf_counters
  • Processor: regex
  • Aggregator: basicstats


Input: win_perf_counters

Collects the data for each virtual processor about CPU Wait Time per Dispatch from Perfmon. 

[[inputs.win_perf_counters.object]]
  ObjectName = "Hyper-V Hypervisor Virtual Processor"
  Instances = ["*"]
  Counters = [
    "CPU Wait Time per Dispatch",
  ]
  Measurement = "hyperv_virtual_processor"

Processor: regex

Extract the VM-Name from the Perfmon Counter Instance and create a new Tag "VMName" with it. 

[[processors.regex]]
  namepass = ["hyperv_virtual_processor"]
  [[processors.regex.tags]]
    # Extract the VM-Name from the Perfmon Counter Instance and create a new Tag "VMName" with it
    key = "instance"
    pattern = '^(?P<VMName>.+)\:.+$'
    replacement = "${VMName}"
    result_key = "VMName"

Aggregator: basicstats

Calculates the sum of all CPU Wait Time per Dispatch foreach VM. 

# Aggregate the sum of the virtual processors wait time per VM.
[[aggregators.basicstats]]
  namepass = ["hyperv_virtual_processor"]
  fieldpass = ["CPU_Wait_Time_per_Dispatch"]
  taginclude = ["VMName", "host"]
  
  ## The period on which to flush & clear the aggregator.
  period = "30s"

  ## If true, the original metric will be dropped by the
  ## aggregator and will not get sent to the output plugins.
  drop_original = true

  ## Configures which basic stats to push as fields
  stats = ["sum"]
Von „http://wiki.webperfect.ch/index.php?title=Telegraf:_Collect_Hyper-V_CPU_Wait_Time_per_Dispatch_(CPU_Ready)&oldid=3160“