River Level 🏞️

Docker based services that serves prometheus metrics of Environment Agency river level & rainfall API data using the prometheus python client.

Available in AMD, and ARM builds.

Why?

Theres already a tool to scrape json api data and load it into prometheus JSON Exporter but I couldn't get it to work. Because of this i thought it would be a good beginner's challenge to write a python program to do the same and then containerise it so i could run it on my home server.

Deploy

Running as a docker image

Docker Compose

1. cd to the project folder
2. Copy docker-compose-example.yml to docker-compose.yml
3. Update the five environmental variables (water level station, water level measure, rainfall station, rainfall measure, metrics port) in the compose file
4. Run docker-compose up -d to start the container
5. go to ipofmachine:8897 to check the prometheus guages are being published

Docker

In the terminal run:

docker run -d -i \
--name riverleveltest \
-p 8897:8897 \
-e RIVER_MEASURE_API=https://environment.data.gov.uk/flood-monitoring/id/measures/examplemeasure.json \
-e RIVER_STATION_API=https://environment.data.gov.uk/flood-monitoring/id/stations/examplestation.json \
-e RAIN_MEASURE_API=https://environment.data.gov.uk/flood-monitoring/id/measures/examplemeasure.json \
-e RAIN_STATION_API=https://environment.data.gov.uk/flood-monitoring/id/stations/examplestation \
-e METRICS_PORT=8897 \
magammon/riverlevel:latest

Ensure that you update the station, measure and metrics port environmental variables.

Running the python program

riverlevel.py will run outside of the docker container:

1. Use pip to install the dependencies in requirements.txt
2. Update the four variables (water level station, water level measure, rainfall station, rainfall measure) in riverlevel.py
3. Run python riverlevel.py

Finding a station and measure URL

• the documentation for the river-level api is available here and for the rainfal api here.
• the simplest way to find stations and measures is to use the built in location function of the API.
• To find stations:
  • https://environment.data.gov.uk/flood-monitoring/id/stations.html?parameter=rainfall&lat=51.48&long=-2.77&dist=10 provides a list of rainfall stations within 10 km of portishead. Update the lat and long to your desired location to find local stations
  • https://environment.data.gov.uk/flood-monitoring/id/stations.html?parameter=level&lat=51.48&long=-2.77&dist=10 provides a list of water level stations within 10 km of portishead. Update the lat and long to your desired location to find local stations
  • Click on the link of the station you would like to use. Take the URL and update .html to .json to get the URL of the station to add to your docker-compose.yml file
• To find measures:
  • on your station page e.g. https://environment.data.gov.uk/flood-monitoring/id/stations/E72639.html the measures available at that station are listed below.
  • Click on the link of the measure you would like to use. Take the URL and update .html to .json to get the URL of the station to add to your docker-compose.yml file

Setting your prometheus.yml file

To get prometheus to scrape the published gauges you need to update your prometheus.yml file i.e.

- job_name: 'json_export_river'
  scrape_interval: 60s
  honor_labels: true
  metrics_path: /probe
  static_configs:
    - targets:
      - 192.168.1.999:8897 # set to IP of the machine running the container. 8897 is the default port.

To Do

• if the reading is 0m then skip updating gauge