Understanding System Counter
VirtualMetric provides complete visibility with its pre-configured more than 650 performance counters.
Select type of server and module.
VirtualMetric offers 4 different Statistics Levels by default:
· Level 1: Contains critical counters for standard performance monitoring. *Recommended*
· Level 2: Contains recommended counters for detailed performance monitoring.
Level 2 includes all the Level 1 counters.
· Level 3: Contains deep dive counters for deep dive performance monitoring.
Level 3 includes all the Level 2 and Level 1 counters.
· Level 4: Contains debug counters for debug level performance monitoring.
Level 4 includes all the Level 3, Level 2 and Level 1 counters.
We suggest you to enable Level 4 and Level 3 counters when you only need to do deep dive troubleshooting.
Notification Options
In the System Counter settings, you can enable or disable all alerting systems including Email, SMS and Heatmap. By modifying thresholds of the counters, you can set their threshold limits and enable them on heatmap or activate notifications
If you enable thresholds for a system counter, you should also set threshold limits to activate. You can also choose to receive notifications or see their health status on Heatmap
Thresholds
If thresholds option is enabled, VirtualMetric uses notification thresholds to determine counter health status which are healthy, monitor, warning or critical.
VirtualMetric offers 4 different Thresholds for alerting:
Healthy: That means counter is in recommended thresholds and considered as healthy
Monitor: That means counter is not in recommended thresholds but also it is not in danger zone. You should monitor activity of this counter because soon it may be in warning thresholds.
Warning: That means counter is in warning thresholds and considered as not healthy. You should troubleshoot issue before counter gets into critical level.
Critical: That means counter is in critical thresholds and considered as not healthy. Counter is in critical state so you must troubleshoot issue fast before your system becomes unresponsive
VirtualMetric comes with some pre-defined threshold limits based on Microsoft and VMware recommendations.
Threshold Exercises
We can use different scenarios to cover different threshold configurations.
Exercise 1: CPU counter, 0%-60% is healthy, 61%-85% is monitor, 86%-95% is warning, %96-%100 is critical.
Healthy Min: 0%
Healthy Max: 60%
Monitor Min: 61%
Monitor Max: 85%
Warning Min: 86%
Warning Max: 95%
Critical Min: 96%
Critical Max: 100%
Exercise 2: CPU counter, %96-%100 is critical, other values are healthy.
Healthy Min:
Healthy Max:
Monitor Min:
Monitor Max:
Warning Min:
Warning Max:
Critical Min: 96
Critical Max: 100
If you don't type % for unit, VirtualMetric will add automatically.
Exercise 3: Disk counter, more than 5000 IOPS is critical, other values are healthy.
Healthy Min:
Healthy Max:
Monitor Min:
Monitor Max:
Warning Min:
Warning Max:
Critical Min: 5001
Critical Max:
Exercise 4: Disk counter, 0-500 IOPS is healthy, 501-1000 IOPS is monitor, more than 1000 IOPS is warning.
Healthy Min: 0
Healthy Max: 500
Monitor Min: 501
Monitor Max: 1000
Warning Min: 1001
Warning Max:
Critical Min:
Critical Max:
Using unit is not necessary for units like %, IOPS, packets etc.
Exercise 5: Paging counter, 0-300 MB is healthy, 300-500 MB is monitor, 500 MB-1 GB is warning, more than 1 GB is critical.
Healthy Min: 0
Healthy Max: 300 MB
Monitor Min: 300 MB
Monitor Max: 500 MB
Warning Min: 500 MB
Warning Max: 1 GB
Critical Min: 1 GB
Critical Max:
You can use different units like MB, GB in same threshold configuration.
Last updated