It's my understanding that Zenoss's parent/child dependencies are all Layer 3 based, via an ICMP test. These dependencies are can either be auto discovered (routing table discovery) or manually entered.
Although it's not as quite a easy to manually manipulate as Nagios' dependencies it can be done, but the entire concept of layer 3 ICMP based testing of dependencies become decreasingly useful as network complexity increases. By relying on this type of dependency model how do you distinguish between a failure of the management engine as opposed to actual device failure? Granted failure of the management plane on a device is important, but many devices will continue to route/switch/etc.. even after the management has gone to lunch. In a situation where you need reliable outage metrics it would be wrong to assume that a management outage equals an outage to all connected devices. -------------------- m2f -------------------- Read this topic online here: http://community.zenoss.com/forums/viewtopic.php?p=20953#20953 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
