Librato output : disk input: Couldn't build gauge: unable to extract value from Fields, undeterminable type

[W13ts38]

Relevant telegraf.conf:

[[outputs.librato]]
api_user = "[email protected]" # required.
api_token = "myapicode" # required.
template = "host"

[[inputs.disk]]
mount_points = ["/"]
ignore_fs = ["tmpfs", "devtmpfs", "devfs"]

System info:

core@docker ~/telegraf $ cat /etc/os-release
NAME="Container Linux by CoreOS"
ID=coreos
VERSION=1688.5.3
VERSION_ID=1688.5.3
BUILD_ID=2018-04-03-0547
PRETTY_NAME="Container Linux by CoreOS 1688.5.3 (Rhyolite)"
ANSI_COLOR="38;5;75"
HOME_URL="https://coreos.com/"
BUG_REPORT_URL="https://issues.coreos.com"
COREOS_BOARD="amd64-usr"

2018-05-15T12:48:57Z I! Starting Telegraf v1.6.2

Steps to reproduce:

using docker-compose file below with the telegraf.conf adjustments above:
version: '2'
services:
telegraf:
image: telegraf
restart: always
environment:
HOST_PROC: /hostfs/proc
HOST_MOUNT_PREFIX: /hostfs
hostname: localhost
network_mode: host
volumes:

• /:/hostfs:ro
• /data/docker/telegraf/telegraf.conf:/etc/telegraf/telegraf.conf:ro

Expected behavior:

disk metrics of the mount / being sent to librato

Actual behavior:

receive an error :
2018-05-15T12:48:00Z D! Output [librato] buffer fullness: 1 / 10000 metrics.
2018-05-15T12:48:00Z I! unable to build Gauge for disk, skipping
2018-05-15T12:48:00Z D! Couldn't build gauge: unable to extract value from Fields, undeterminable type 6520340480

2018-05-15T12:48:00Z D! Output [librato] wrote batch of 1 metrics in 94.728µs

Additional info:

when i disable the librato output and enable file output everything works as expected. so that suggests the problem is in the librato output plugin

[Include gist of relevant config, logs, etc.]

[danielnelson]

Sorry about that, the fix will be in 1.6.3 early next week. You can also try the nightly builds, just give it until tomorrow to be incorporated.