When using this facility, I've found it will show an alert for a service
that is not being monitored. For example:
host.cfg entry:
10.1.2.3 whatever.au # http smtp NAME:"Sample Entry"
alert.cfg entry:
HOST=whatever.au
MAIL user-b6fdd911a904@xymon.invalid COLOR=RED
So, if I do this test:
xymoncmd xymond_alert --test vho64pl09.orix.com.au blah
it will return an alert.
I understand that the alert.cfg just says to alarm on any RED condition,
but it would be useful for this test to actually check that the service is
in fact one being tested for. This already tests that the host has a
hosts.cfg entry (very useful when working from a separate host list), it
would be nice to extend this to the actual test as well - a single cross
check on the whole chain of config files to ensure alerts are working as
expected.
thanks, Phil
Phil:
If you want to alert on a particular service, add "SERVICE=XXX,YYY" to the
entries in your alerts.cfg, a la:
MAIL user-9ceda1105de1@xymon.invalid,user-4d0103ceb099@xymon.invalid COLOR=red SERVICE=conn,cpu,disk
Which would send an email to the aforementioned email addresses, on a red
status, for services/facilities conn, cpu and disk.
This option, and more, are available in the MAN pages for the alerts.cfg.
--
Mike Burger
http://www.bubbanfriends.org
"It's always suicide-mission this, save-the-planet that. No one ever just
stops by to say 'hi' anymore." --Colonel Jack O'Neill, SG1