Here's an example of the first few lines of "Probe State (Scheduler Debug Data).txt"
Requests (Timings are Average/Last/Delay/Work/Kernel/User/Count):
* Host: 127.0.0.1
1 2014.01.14 18:17:15:015 Open: 1 Interval: 60 Request: 3735730487 Id: 2965 Timing: 11452/34666/0/9041/2/1/32 WMINetwork on 127.0.0.1 Mem: 1947 Action=12
What do the following values measure and what would cause them to increase. Specifically looking for if each number represents a slowness in PRTG, outside PRTG, or both. Which numbers are CPU time and which are real time. Would also like more details on Count, since that doesn't appear to be timing information.
Average
Last
Delay
Work
Kernal
User
Count
Article Comments
The Probe Debug log can be translated accordingly:
13 2013.08.28 09:53:27:997 Open: 0 Interval: 30 Request: 3824878 Id: 3881 Timing: 2/2/0/0/0/0/2507 Ping on 10.0.0.1 Mem: 721 Action=21
13 = This was the 13th Sensor put on this device
2013.08.28 09:53:27:997 = Last time the sensor was run
Open: 0 = Is a request currently running or not
Interval: 30 = Scanning interval in Seconds
Request: 3824878 = This is an interal ID for the request
Id: 3881 = Sensor ID
Timing:
2= Average response time (Average)
2= Last Response time (Last)
0 = What the percentage delay there is for the sensor (Again how long of a delay there is with this sensor's answer)(Delay)
0 = Total average work time for the sensor (Work)
0 = Total average kernel time (Kernel)
0 = Total average time in user space (User)
2507 = How many times this sensor has been scanned (Count)
Ping on 10.0.0.1 = Name of the sensor
Mem: 721 = Memory usage of the request data
Action=21 = Where this sensor is in the schedule of the scheduler
These numbers can change or alter due to both internal and external influences. If there are more and more sensors being created on the server or the sensors you have are answering slower, these numbers can go up. The real time is the average or last and the Resource usage time is best represented by the Kernel time.
Jan, 2014 - Permalink
The Probe Debug log can be translated accordingly:
13 = This was the 13th Sensor put on this device
2013.08.28 09:53:27:997 = Last time the sensor was run
Open: 0 = Is a request currently running or not
Interval: 30 = Scanning interval in Seconds
Request: 3824878 = This is an interal ID for the request
Id: 3881 = Sensor ID
Timing:
2= Average response time (Average)
2= Last Response time (Last)
0 = What the percentage delay there is for the sensor (Again how long of a delay there is with this sensor's answer)(Delay)
0 = Total average work time for the sensor (Work)
0 = Total average kernel time (Kernel)
0 = Total average time in user space (User)
2507 = How many times this sensor has been scanned (Count)
Ping on 10.0.0.1 = Name of the sensor
Mem: 721 = Memory usage of the request data
Action=21 = Where this sensor is in the schedule of the scheduler
These numbers can change or alter due to both internal and external influences. If there are more and more sensors being created on the server or the sensors you have are answering slower, these numbers can go up. The real time is the average or last and the Resource usage time is best represented by the Kernel time.
Jan, 2014 - Permalink