|
|||
|
You could always download the source, mirrors located from here: http://mrtg.hdl.com/mrtg.html
There's a good tutorial here for configuring mrtg : http://www.siliconvalleyccie.com/linux-hn/mrtg.htm |
|
|||
|
On Centos, you should find the following file
/etc/yum.conf Code:
[main] exclude=mrtg httpd kernel kernel-smp mysql php mod_* perl You can add packages to that file to be excluded, include/exclude as many as you need to , those listed above would be just an example of some that you don't really want updated as a rule automatically by yum updates. (which cpanel runs each night if you have that option for security updates set to on in whm). Hope this helps. |
|
|||
|
Has anyone got mrtg running where the server load is not measured by snmp but instead reports on the load averages that 'top' would report?
I ask because the two are not often (ever?) the same. I've had a few occasions recently where a server has ground to a halt because of a runaway process and 'top' was indeed reporting load averages as being very high. But at the time I was looking a nice mrtg graph of cpu load that uses snmp for it's server load info, and everything looked normal. In other words, it's not much point me having a nice mrtg graph of server load ('cpu utilization') if it's not giving me useful information.
__________________
|
![]() |
| Thread Tools | |
| Display Modes | |
|
|