Monitoring NFS share resulting in far too many alerts - Filtering options?

[tiarnaigh]

Hi all.
I have a few NFS shares from a NetApp device mounted across several nodes in my infrastructure. Originally, when I added these nodes, the NFS shares weren't discovered at all, but I eventually found out how to enable these via the "Mounted filesystem discovery" rule (I had to update the settings for "File systems for discovery" under "Administration/General/Regular expressions" from this:

^(btrfs|ext2|ext3|ext4|jfs|reiser|xfs|ffs|ufs|jfs| jfs2|vxfs|hfs|ntfs|fat32|zfs)$

to:

^(btrfs|ext2|ext3|ext4|jfs|reiser|xfs|ffs|ufs|jfs| jfs2|vxfs|hfs|ntfs|fat32|zfs|nfs|nfs3|nfs4)$

As such, the NFS shares are now being discovered on all of my nodes, and are reporting as expected.

The problem however is that I don't want to necessarily monitor this same volume from all nodes at the same time, or more to the point, I don't wish to receive the same alarm from all nodes each time the capacity on these volumes breaches the trigger (80%).

Ideally, I'd like to monitor the NFS share only from one "admin" type host (i.e. the Zabbix server itself say) and ignore on all the rest. As such, if the volume reaches the 80-90% mark, I get one Problem alert only, as opposed to multiple.

Any idea on how can do this?

Thanks,
M

[ingus.vilnis]

Hi,

Solution here would be to use multiple regular expressions. You can use the standard one on regular hosts but on some specific hosts you can use extended regexp. That can be modified per each host if needed.

If you are not happy with the thresholds then you can try to use macros on selected hosts of clone the LLD rules themselves and use different threshold settings there.

Best Regards,
Ingus

[tiarnaigh]

Hey Ignus.
I'm not very clear on what you mean here. The Regular Expression I altered seems to be across the board (i.e. for all the file discovery monitors and has since been picked up by all hosts). How would I go about definining a specific one to be used for 1 host only?

Thanks,
M