bjornoleh / ns-exporter

Nightscout exporter to InfluxDB

ns-exporter

Nightscout exporter to InfluxDB

usage variants:

1. inline

go build
./ns-exporter

2. docker

docker build -t ns-exporter .
docker run -d ns-exporter:latest

arguments:

mongo-uri       - MongoDb uri to download from
mongo-db        - MongoDb database name
ns-uri          - Nightscout server url to download from
ns-token        - Nigthscout server API Authorization Token
limit           - number of records to read per request
skip            - number of records to skip from beginning (`skip=100 take=50` means 'take records from 101 to 150')
influx-uri      - InfluxDb uri to download from
influx-token    - InfluxDb access token
influx-org      - (optional, default = 'ns') InfluxDb organization to use
influx-bucket   - (optional, default = 'ns') InfluxDb bucket to use
influx-user-tag - (optional, default = 'unknown') InfluxDb 'user' tag value to be added to every record - to be able to store multiple users data in single bucket

arguments also can be provided via env with NS_EXPORTER_ prefix:

NS_EXPORTER_MONGO_URI=
NS_EXPORTER_MONGO_DB=
NS_EXPORTER_NS_URI=
NS_EXPORTER_NS_TOKEN=
NS_EXPORTER_LIMIT=
NS_EXPORTER_SKIP=
NS_EXPORTER_INFLUX_URI=
NS_EXPORTER_INFLUX_TOKEN=
NS_EXPORTER_INFLUX_ORG=
NS_EXPORTER_INFLUX_BUCKET=
NS_EXPORTER_INFLUX_USER_TAG=

So you can choose the data source: direct MongoDB or Nightscout REST API. Supplying required set of parameters will trigger related consumer. You can even supply both and get from both sources :)

For NS API access you need provide security token. For security reason it is better to go to 'Admin tools' and create special token for NS-Exporter only instead of using admin security key. Since exporter only requires read access, creating role with two permissions will be enough:

• api:treatments:read
• api:devicestatus:read

Presentation

I'm using Grafana dashboard for viewing data. To setup grafana with InfluxDB you need to follow InfluxDB's instructions. The sample dashboard can be imported from grafana.json. It uses both InfluxQL and Flux datasources for different panels. Some can be omitted, some can be reworker based on other InfluxDB datasource query type. Anyway they're provided as samples, for educational purpose :)