问题描述:

I have the following default chart in zabbix, but I have no idea how to interprete these values. Can anyone explain?

网友答案:

An OS is a very busy thing, particularly so when you have it doing something (and even when you aren't). And when we are looking at an active enterprise environment, something is always going on. (From Wikipedia: zabbix "is designed to monitor and track the status of various network services, servers, and other network hardware.")

Most of this activity is "bursty", meaning processes are typically quiescent with short periods of intense activity. This is certainly true of any type of network-based activity (e.g. processing PHP requests), but also applies to OS maintenance (e.g. file system maintenance, page reclamation, disk I/O requests). I won’t even get into modern power saving technologies.

If you take a situation where you have a lot of such bursty processes, you get a very irregular and spiky CPU usage plot.

PS As “500 – Internal Server Error” says (love that handle!), the high number of context switches are going to make the situation even worse.

PPS The physics nerd in me just has to mention that this is a very common phenomenon in situations where you have a somewhat large number of bursty events (say particle collisions or atomic decay). Once you get into an extremely large number of such events (think Avogadro’s Number), things smooth out.

相关阅读:
Top