I’d like to expand HomeAssistant with better graphs, by using Grafana (in combination with InfluxDB). My Home Assistant is running as a VM on my Unraid server. I’m used to running everything as its own (micro) service, everything in its own docker. For example Tautulli, Netdata, Plex, are all running as a docker container. However, if I lookup tutorials on the internet, I see everyone adding InfluxDB and Grafana directly onto the HAOS vm/machine. This got me thinking whether running InfluxDB and Grafana outside of HAOS would be a good choice or not.
Pro’s of running them on the host;
- More resources available (all cpu cores, all memory, faster IO (as it can directly talk to the disk))
- Data remains available even if my VM would for some reason crash and fail to start again
- Newer Grafana/InfluxDB versions probably faster available
- VM disk can remain small, even if I’d retain data in InfluxDB forever
Pro’s of running them on the home assistant vm;
- Everything related to home assistant in 1 package (vm)
- If an update to home assistant is available, I could snapshot the whole VM, and in case of any problems restore the snapshot in one go
I’m new to Home Assistant, am I overseeing some things that would make running these services as dockers outside of HAOS a bad idea?
I run both of them outside of Home Assistamt OS just fine, no issues.
Burningstone, do you use the HA integration? I cannot figure out how to get the current HA-supervised instance to talk to my external influxdb server (it is on the same LAN). How are you doing it?
Thanks, that did it for me too.
Sorry to highjack this thread, but it seems to be the closest I can get, though following some of the instructions didn’t help.
NEWBIE ALERT.
I have installed influxdb 2_2.7.4 on an old Windows Surface 3 running Linux Mint. I have also installed mariadb on the same machine and this is running very well.
I have installed grafana on the HA x64 box running HAOS using the HA add-on capability.
I cannot get grafana to talk to influxdb.
I can login to influxdb from both the machine running influxdb and another machine on the same network.
I know that the HA influxdb configuration (below) is correct because I can inspect the influxdb data from a remote machine successfully by logging in as above.
influxdb:
api_version: 2
ssl: false
host: 192.168.xx.yy
port: 8086
token: !secret influxdb_api_token
organization: !secret influxdb_organizationid
bucket: HA
precision: s
max_retries: 3
tags:
source: HA
tags_attributes:
- friendly_name
default_measurement: units
exclude:
entities:
- zone.home
domains:
- persistent_notification
- person
include:
domains:
- sensor
Within grafana I am using the following configuration. I have checked the redacted urls several times, and they are correct.
But on clicking the Save & test button I get the following error:
InfluxDB returned error: error reading influxDB
This is the configuration of grafana in HA:
I know I am doing something incorrectly, but I am neither skilled or experienced enough to know what I need to change. So all help is greatly appreciated.
@Burningstone & @txNgineer sorry for prodding you directly, but I am desperate.
