Answer the question
In order to leave comments, you need to log in
What is the best Zabbix process monitoring implementation?
I would like to know the opinion of the participants on the topic: how best to organize monitoring of a certain process on the server, specifically and in principle, how much any process loads the processor? Preferably as a percentage. That is, I have some kind of abstract process "process.exe" and I need to find out how much processor resources it eats.
Answer the question
In order to leave comments, you need to log in
Necroposting, of course, but I found this post when I was looking for a solution to this issue.
For those who stumble in the future - solution:
Zabbix for Windows has Zabbix agent-checks:
" proc_info " - to get information about the process.
" proc.num " - to count the number of processes on the server.
It is described in detail in the manual:
proc_info
proc.num
For most unix systems, proc.mem is used instead of proc_info .
To find out if a particular data element is supported on your system, check the table in the manual .
In the case of the author, the item key should look like this:
This item displays the amount of memory consumed in KB.
Uncomfortable? We put in the settings of the data element:
Units (Unit of measure) - B
Use custom multiplier (Custom multiplier) - 1024
There is also the possibility of monitoring processes through perf_counter, for example:
perf_counter["\Процесс(explorer)\Счетчик потоков", 300]
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question