Telegraf on FreeNAS

start-telegraf.sh

#!/bin/bash

# full path to the telegraf executable in the jail it was installed in
TELEGRAF_BIN='/mnt/tank0/jails/telegraf-test/usr/local/bin/telegraf'
# dir for storing the pid, config, and log files
BASE_DIR='/mnt/tank0/home/aheadley'

if [ -f $BASE_DIR/telegraf.pid ] && kill -0 $(cat $BASE_DIR/telegraf.pid) &>/dev/null; then
    # don't start it again if it's already running
    echo "Telegraf running ($(cat $BASE_DIR/telegraf.pid))..."
else
    # start it
    $TELEGRAF_BIN --config $BASE_DIR/telegraf.conf &
    echo "$!" > $BASE_DIR/telegraf.pid
fi

telegraf.conf

[global_tags]
  role = "freenas"
  fqdn = "<YOUR-FREENAS-FQDN>"

[agent]
  hostname = "<YOUR-FREENAS-HOSTNAME>"
  omit_hostname = false
  interval = "10s"
  round_interval = true
  metric_buffer_limit = 1000
  flush_buffer_when_full = true
  collection_jitter = "0s"
  flush_interval = "10s"
  flush_jitter = "0s"
  debug = false
  quiet = false
  # see BASE_DIR in start-telegraf.sh
  logfile = "/mnt/tank0/home/aheadley/telegraf.log"

#
# OUTPUTS:
#
[[outputs.influxdb]]
  database = "<INFLUXDB-DATABASE-NAME>"
  username = "<INFLUXDB-USERNAME>"
  password = "<INFLUXDB-PASSWORD>"
  urls = ["http://INFLUXDB-HOST:8086"]

#
# INPUTS:
#
[[inputs.cpu]]
  percpu = true
  totalcpu = true
#[[inputs.disk]]
[[inputs.internal]]
[[inputs.io]]
[[inputs.mem]]
[[inputs.net]]
[[inputs.swap]]
[[inputs.system]]
[[inputs.diskio]]
[[inputs.netstat]]
[[inputs.zfs]]
  poolMetrics = true
  kstatMetrics = ["arcstats", "zfetchstats", "vdev_cache_stats", "zcompstats", "zio_trim"]

How is telegraf able to get information about your pool, system, etc via the jail? for example, the new smarmon/smartctl plugin needs to be able to talk to the disks...AFAIK a jail won't allow this sort of low level access. Did you find a way around this?