Task 1

There're two solutions for the task: The first one is python based and the second solution are just bunch of bash commands.

Just run the script_1.0.sh or script_2.0.sh :)

Task 2

1. Below are the metrics I consider as interesting to monitor

a) Metrics on CPU

Temperature
Used part
Load

b) Metrics on memory

Mem. usage
Mem. used

Disk space usage

c) Input and Output stats

Read/Write stats

d) Network

Traffic info

2. Custom bash script could be written for each metrics value for special cases, if necessary, but in a usual case I would monitor them with Zabbix and set relevant custom notification alerts for various overloads and limit reaches.

Challenges of monitoring

1. Deciding which metrics are really necessary and which are not important for the case.

2. Proper choice and setting of the monitoring system, so it won't influence server's performance.

3. Making sure alerts set are precise and couldn't cause false cautions.