check_disk returns 0MB in use for zfs filesystem

[Christopher Johnson]

In the performance data returned by check_disk the value for MB in use is always zero.
$ /usr/lib/nagios/plugins/check_disk -w '10%' -c '5%' -e -p /tank -p /
DISK OK| /tank=0MB;3005801;3172790;0;3339779 /=7522MB;84365;89052;0;93739

Tank is a ZFS mountpoint/filesystem.
$ df /tank /
Filesystem 1K-blocks Used Available Use% Mounted on
tank 3419934254 32 3419934222 1% /tank
/dev/sda1 95989516 7703148 83403632 9% /

Results from check_disk are the same when run as root

ProblemType: Bug
DistroRelease: Ubuntu 13.10
Package: nagios-plugins-basic 1.4.16-2ubuntu2
ProcVersionSignature: Ubuntu 3.8.0-25.37-generic 3.8.13
Uname: Linux 3.8.0-25-generic x86_64
NonfreeKernelModules: nvidia
ApportVersion: 2.12.5-0ubuntu2.1
Architecture: amd64
Date: Thu Oct 31 09:51:01 2013
EcryptfsInUse: Yes
InstallationDate: Installed on 2013-05-29 (154 days ago)
InstallationMedia: Ubuntu 13.04 "Raring Ringtail" - Release amd64 (20130424)
MarkForUpload: True
SourcePackage: nagios-plugins
UpgradeStatus: Upgraded to saucy on 2013-10-17 (13 days ago)

[Jan Wagner (waja)]

I guess the 32k from 'df' is something like 0%? I can't see anything wrong from your output for the first view!

[Christopher Johnson (chris-j-s)]

How do I cancel this stupidity on my part? :-(
tank 3419934252 32 3419934220 1% /tank
tank/Private 3812933422 392999202 3419934220 11% /tank/Private
tank/netdisk 3434754022 14819802 3419934220 1% /tank/netdisk

It seems that ZFS reports that the pool has only 32K used which is what check_disk picks up. In looking at the df output I missed because I saw that the mountpoints created under the pool DO report amount used.

Please close this bug report. Luser error.

[Jan Wagner (waja)]

you maybe want to recreate that in the corresponsing zfs/kernel package.