JP Vossen on 2 Sep 2009 12:54:57 -0700

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: [PLUG] server monitoring- set up threshold alerts

On Wednesday, September 2, 2009, Ron Kaye Jr <> wrote:
 > need to monitor server activity and send ALERTS for exceeded
 > thresholds. have 2 rhel5, 2 ubuntu for now
 > 1) any suggestions on what to monitor (? ie?? >80% processor,
 >    80%disk usage, failed logins etc)
 > 2) any suggestions on the best tool ( i installed webmin which
 >    appears, at first glance, to be pretty good)
 > 3) how about network activity?

Monit =+ 1;

I've been happy with Monit (in the Debian/Ubuntu repos, not sure about 
CentOS/RHEL), with some caveats.

It's a lot easier to set up than Cacti, Nagios, Munin, etc., but it is 
less capable and is per-host only.  4 servers is probably starting to 
approach the limit for per-host distributed verses centralized, in terms 
of admin hassle.  So it's quick and easy to deploy on a few hosts, but 
will not scale up well.

There are useful config examples:

I don't think it does network stats, but you can have it check if all 
sorts of things (ICMP, ssh, HTTP, DNS, other) are available on host B 
from host A.  So I run a big config on my main server that checks local 
resources and makes sure it can "see" a bunch of other hosts, both local 
and over the 'Net.  Some other hosts run their own local instance, since 
Monit can only check LOCALLY for things like disk or CPU utilization 
(unlike the centralized, agent-based aps like Nagios or Cacti).

JP Vossen, CISSP            |:::======|
My Account, My Opinions     |=========|
"Microsoft Tax" = the additional hardware & yearly fees for the add-on
software required to protect Windows from its own poorly designed and
implemented self, while the overhead incidentally flattens Moore's Law.
Philadelphia Linux Users Group         --
Announcements -
General Discussion  --