[wellylug] How to monitor CPU/Memory/Disk usage

[Pete Black]

simply observe the values in the /proc filesystem

e.g. /proc/stats, /proc/loadavg etc. etc.

You could also set up snmp and script a simple snmp query using the 
command line snmp client - which is also how how cricket/mrtg etc 
usually expect to gather their stats.

Nagios is pretty nice but it is nice to know how to find the 'raw' 
information as well.

-Pete

> Nagios is your friend :)
>
> You can be alerted when services such as disk / memory or whatever 
> gets to a critical or warning state
>
> It's also built as a debian package
>
> Jamie Dobbs wrote:
>
>> I'm a little worried that my 'mail server' PC is getting overloaded 
>> and I
>> am losing emails etc., or that I am having delays in outgoing mail.
>> How can I set up some sort of monitoring to look at CPU, memory and disk
>> usage - preferably in a graphical format.
>> I seem to recall that cricket and/or mrtg might be able to do this but
>> searching around has not shown me any way to monitor a local PC's 
>> stats -
>> can anyone point me in the right direction?
>> I'd like to be able to do this on Debian and Gentoo as there are a 
>> couple
>> of other PCs I'd like to be able to monitor as well.
>>
>>
>
>