Automatic items per host

[cpicton]

Hi

I have a few hosts (about 100) I am migrating from nagios to zabbix.

I have a custom nagios script installed on the hosts which checks local resources and submits the (passive) checks to the nagios server automatically.

This includes things like free space per partition or disk statistics per drive, or bandwidth per network interface.

These stats, and the device or partition they apply to are all dynamically determined and submitted.

I am looking through the zabbix frontend, and only see a way to statically define partitions/drives/interfaces to monitor.

As my servers are all heterogeneous, is there a way to get zabbix to automatically determine lists of items to monitor on the servers, as above.

[cpicton]

I suppose that another way to do this would be to allow multiple results per check. Ie, create an item that is free space on all drives listed in /etc/fstab (or /proc/partitions)

The check would return one entry per partition.

Does this fit into zabbix's design at all

[Alexei]

Well, while autodiscovery is not supported currently, I would suggest creating set of templates which would include monitoring of all possible resources. Items related to nonexistent resources will become unsupported and you won't see them.