We will not touch our application code in this step and focus instead on the underlying database that we will use to persist the data. In the next step, we will see how to integrate it on our application.
To persist the data, we will use InfluxDB, a great open source Time Serie Database.
If you want to know more about InfluxDB, I hightly recommend you have a look at the documentation.
- Go to the Docker Hub page of the official Influxdb image
A lot of information are provided concerning the usage of this image, its configuration, the port that needs to be exported, ...
- Use the following command to generate the default configuration file
docker container run --rm influxdb influxd config > influxdb.conf
- Modify the admin part of the configuration file so that the administration interface is enabled
[admin]
enabled = true
bind-address = ":8083"
https-enabled = false
https-certificate = "/etc/ssl/influxdb.pem"
- Create a volume named influxdata
For this, we use the Docker's volume CLI with the default driver
docker volume create influxdata
- Run a container using the named volume and the modified configuration file
To run a InfluxDB container, several option need to be provided to the docker container run command to
- specify the port to publish on the host (8086 is the API port, 8083 is the port used to save the administration web interface)
- bind mount the configuration file from the local folder into the container's /etc/influxdb folder
- specify the volume to use to persist the data
- specify a name that will make it handy to get container's information (more on that soon)
docker container run -p 8083:8083 -p 8086:8086 \
-v $PWD/influxdb.conf:/etc/influxdb/influxdb.conf:ro \
-v influxdata:/var/lib/influxdb \
--name influx \
influxdb -config /etc/influxdb/influxdb.conf
For the next step, we will need to get the IP address of the Influx container, let's get it with the following command
$ docker container inspect -f '{{ .NetworkSettings.IPAddress }}' influx
172.17.0.2
Note: you might not get the same one in your machine
Influxdb administration interface should be available on (http://localhost:8083)
We will start be creating a database named iot using the following HTTP request targeting Influxdb's API
curl -i -XPOST http://localhost:8086/query --data-urlencode "q=CREATE DATABASE iot"
The database creation can easily be done using the administration web interface but it's interesting to see that under the hoods everything happends through HTTP requests.
Now that the database is ready, let's create a test entry using InfluxDB's HTTP API.
curl -i -XPOST 'http://localhost:8086/write?db=iot' --data-binary 'data,type=temp,sensor_id=123 value=34 1483481572000000000'
This HTTP Post request creates a point, in the data measurement. This point has the fields sensor_id and value with the respective values of 123 and 34. type is a tag with the value temp. measurement, point, field, tag, ... are all InfluxDB terminology that is cover in the InfluxDB key concepts.
Basically, in a first approximation, we can see a measurment as a regular SQL table, a point as a record of this table. A point is a list of fields and their value and a list of tags and their value, tags and fields are different in the sense that tags are indexed when fields are not.
- Verify using the administration interface
Issuing the following SQL command within the administration interface enables to retrive the point created above.
