zpool iostat -y <pool> returns the current stats, this is not suitable as zabbix queries this value every 60 seconds.
zpool iostat -y <pool> 1 1 returns the average stats in the last second. This is better but also not a 100% accurate.
zpool iostat <pool> returns the average stats since the last boot. Multiplying this with the uptime in seconds it should be correct. Me and some other people believed this to be overall, not average: https://github.com/openzfs/zfs/issues/12866
The template should use zpool iostat <pool> * uptime using a Calculated Item.
zpool iostat -y <pool>
returns the current stats, this is not suitable as zabbix queries this value every 60 seconds.zpool iostat -y <pool> 1 1
returns the average stats in the last second. This is better but also not a 100% accurate.zpool iostat <pool>
returns the average stats since the last boot. Multiplying this with the uptime in seconds it should be correct. Me and some other people believed this to be overall, not average: https://github.com/openzfs/zfs/issues/12866The template should use
zpool iostat <pool> * uptime
using a Calculated Item.