Hello, > We belive that this is due to a individual machine being down for a > check (and then back up again) and then a seperate machine being down > the next check (and so on). This is confusing mon and it thinks that it > failing.
This is confusing you not mon. If a host fails the group fails. If you don't want that consider one group per host. > *2* > machine flapping. > my plan on this was to somehow check if an email was sent in the > alerting period already for that machine and not send it. See alertevery and alertafter > also.. > what do people think of a new type of group/alerting mechanism.. > a 'load-balanced' group where you can get alerts when X% of machines > in the group are not responding You just have to modify the monitors, doing X% tests. If your service is well load-balanced then a host down must not be seen by mon since it checks the service and not the host. > (does anyone know of a GPL/BSD package out there which works like mon, > and does the above already?) Mon itself. -- Au revoir, 33 (0) 2 99 78 62 49 Gilles Lamiral. France, L'Hermitage (35590) 33 (0) 6 20 79 76 06 http://www.sri.ucl.ac.be/SRI/frfc/rfc1855.fr.html _______________________________________________ mon mailing list [EMAIL PROTECTED] http://linux.kernel.org/mailman/listinfo/mon
