add cgroups v2 PSI metrics to Instana

With the cgroupv2 starting to emerge maybe is a good idea to start collecting and using the PSI metrics available in machines/containers that enabled cgroupsv2. They are valid for Servers and containers.

You can find details here about the PSI metrics:

https://www.kernel.org/doc/html/latest/accounting/psi.html#psi

https://adil.medium.com/how-to-monitor-server-via-psi-pressure-stall-information-and-cgroupv2-2d944a9e732e

Looking at Load Average (existing metric) doesn't give you the full perception of what is happening in your systems. Also Load average gives you only a 1min average value which in the container world is too long. Also, you need to relate Load average with other metrics to understand if you have problems.

With PSI (Pressure Stall Information), it identifies and quantifies the disruptions caused by such resource crunches and the time impact it has on complex workloads or even entire systems.

It creates 3 different files:

/proc/pressure/cpu ,

/proc/pressure/memory ,

/proc/pressure/io

Inside those files you have the pressure metrics as 10, 60, 300 sec average.

Example for CPU:

root:~# cat /proc/pressure/cpu
some avg10=0.03 avg60=0.07 avg300=0.06 total=5376072182

Avg10: How long have the processes stalled for the last 10 seconds
Avg60: How long have the processes stalled for the last 60 seconds
Avg300: How long have the processes stalled for the last 300 seconds
Total: How long have the processes stalled since the server booted

If a process was starved of the CPU for 5 seconds in the last 10 seconds, the Avg10 column will be 50, which means 50% of the last 10 seconds.