[CentOS] monitoring a workstation
jdmls at yahoo.com
Wed Feb 3 10:43:42 UTC 2010
From: Christoph Maser <cmr at financial.com>
> Nagios does not do snmp at all (some plugins do), also metrics (aka
> performance data) is just the weakest spot in nagios. To get cpu
> utilization per user you propably need psacct.
At the same time, it is "easy" to create a plugin with any data you want (by using psacct by example).
Only thing is that the data length is limited...
By example, the plugin just has to return on stdout something like:
Example for a memory plugin, just output:
MEMORY OK - RAM Used: 23% (728MB/3041MB), SWAP Used: 0% (0MB/2932MB)|RAM=728MB;2736;2888;0;3041 SWAP=0MB;2902;2932;0;2932
But again, the part after the '|' is limited in length (I think it used to be 229 chars max from old tests)...
So if you have a lot of users data, you won't be able to put them all in this line.
Unless you create like 1 plugin per user or some trick like that...
And to graph, you can use the PnP nagios plugin.
But all this is only worth it if you need to monitor many servers...
More information about the CentOS