1) What is the behavior of upalerts with respect to the "exit=value"
parameter? I don't seem to be getting upalerts when I use this parameter;
should I be getting an upalert when the monitor either returns success or
its exit value is no longer in the threshold, or just when it return
success?
2) Is it possible to call an alert script without mon registering it as an
alert? I know this sounds like a bizarre idea, but I am trying to use the
"exit=" parameter to the alert script to limit when an email alert is sent
out, while still logging all failures. The problem is, I would like the
watch to come up as a "yellow alert" in mon.cgi when the monitor fails but
is below the threshold, and have it go to "red alert" (and send email) when
the exit value of the monitor exceeds a threshold. The problem is I would
also like to use file.alert to log all failures do different logfiles per
hostgroup, but when file.alert is called mon sees an alert as having been
sent, and mon.cgi ends up in the "red alert" state.
This snippet of the config file might make it easier to see what I mean:
service diskmonitor
description monitor disk space on Unix Production servers
interval 15m
monitor netsnmp-freespace.monitor -T90 _SNMP_OPTS_
depend SELF:ping
period _ANYTIME_
numalerts 1
alert exit=92-100 mail.alert _UNIX_ADMIN_
upalert exit=92-100 mail.alert -m _UNIX_ADMIN_
alert file.alert _UNIX_PRODUCTION_ALERT_LOG_
upalert file.alert _UNIX_PRODUCTION_ALERT_LOG_
Thanks,
Dan
--
[EMAIL PROTECTED]
[EMAIL PROTECTED]
http://world.std.com/~durist
