netdata

By default, netdata will pull from a UID for the container itself to display in the list of netdata servers you have, so you would see something like '0f2342dac'. To define this and make it more readable/recognizable for you (In case you have multiple netdata servers):

  1. Stop the netdata container.
  2. Edit or Create this file: ~/.docker/compose/docker-compose.override.yml and change newnetdataname to friendlynamefornetdata.
  3. Once this is done, re-run sudo ds -c

For Reverse Proxy configuration, we'll use this template from guys who already thought of this at organizrTools.

Template from OrganizrTools

Example:

location = /netdata {
    return 301 /netdata/;
}

location ~ /netdata/(?<ndpath>.*) {
    include /config/nginx/proxy.conf;
    resolver 127.0.0.11 valid=30s;
    proxy_redirect off;
    proxy_set_header Host $host;
    proxy_set_header X-Forwarded-Host $host;
    proxy_set_header X-Forwarded-Server $host;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_pass_request_headers on;
    proxy_set_header Connection "keep-alive";
    proxy_store off;
    proxy_pass http://netdata:19999/$ndpath$is_args$args;
    proxy_headers_hash_max_size 512;
    proxy_headers_hash_bucket_size 64;
    gzip on;
    gzip_proxied any;
    gzip_types *;
}

Notifications

Add health_alarm_notify.conf to your netdata config directory. Populate the notification service(s) you want with login, tokens or similar that is applicable. Instructions found in health_alarm_notify.conf.

Create health.d directory in netdata config directory. Add conf files from health.d for which modules you want alarms. Also note that one can remove specific alarms by commenting them in .conf files.

Get CPU temp from raspberry pi

Netdata will not pick up cpu temp per default for raspberry pi. To activate chart for pi cpu temp add a file with name charts.d.conf in netdata config directory and add the following line. sensors=force

Get data for home assistant

To identify the correct data group and element to input in netdata home assistant component use http://yournetdataip:19999/api/v1/allmetrics?format=json

Monitor services with netdata

Create python.d directory in netdata config directory. Add httpcheck.conf to your python.d directory. Edit according to instructions in file, suggestion is to add after last line in conf file. See example below.

# This plugin is intended for simple cases. Currently, the accuracy of the response time is low and should be used as reference only.

Hydra:
    url: 'http://192.168.86.60:5076/nzbhydra/'
    timeout: 1
    redirect: no
    status_accepted:
        - 200
    regex: '.*hydra.*'

Ombi:
    url: 'http://192.168.86.60:3579/ombi/landingpage'
    timeout: 1
    redirect: yes
    status_accepted:
        - 200
    regex: '.*ombi.*'

You will now get charts in netdata for ombi and hydra. Please add your ip and ports accordingly.

To get alarms add httpcheck.conf to your health.d directory. Don't forget to comment the unwanted alarms. Slow response alarm can be quite annoying.

Netdata badges

Coming soon.