Status Cake Exporter

Status Cake Exporter is a Prometheus exporter for StatusCake.

Metrics are exposed on port 8000 when using the provided examples/manifest.yml](examples/manifest.yml) in Kubernetes, e.g.

http://status-cake-exporter.default.svc:8000

Requirements

• Python 3.7 (not tested with anything below this)
• Python dependencies from requirements.txt
• Docker
• Kubernetes (optional)
• Helm 3 (optional)

Usage

Setting	Required	Default
USERNAME	Yes	Null
API_KEY	Yes	Null
TAGS	No	Null
LOG_LEVEL	No	info
PORT	No	8000

Docker

The following will expose the exporter at localhost:8000:

export USERNAME=statuscakeuser
export API_KEY=xxxxxxxx
docker run -d -p 8000:8000 --env USERNAME --env API_KEY chelnak/status-cake-exporter:latest

Kubernetes

To get up and running quickly, use examples/manifest.yml as an example. You will need to create a secret named status-cake-api-token containing your USERNAME and API_KEY first.

Otherwise, you can use the Helm Chart provided in chart/status-cake-exporter.

Terminal

usage: app.py [-h] [--username USERNAME] [--api-key API_KEY]
              [--tests.tags TAGS] [--logging.level {debug,info,warn,error}] [--port PORT]

If an arg is specified in more than one place, then commandline values
override environment variables which override defaults.

optional arguments:
  -h, --help            show this help message and exit
  --username USERNAME   Username for the account [env var: USERNAME]
  --api-key API_KEY     API key for the account [env var: API_KEY]
  --tests.tags TAGS     A comma separated list of tags used to filter tests returned from the api [env var: TAGS]
  --logging.level       {debug,info,warn,error} Set a log level for the application [env var: LOG_LEVEL]
  --port                The TCP port to start the web server on [env var: PORT]

Metrics

Name	Type	Description
status_cake_test_info	Gauge	A basic listing of the tests under the current account.
status_cake_test_uptime_percent	Gauge	Tests and their uptime percentage

Prometheus

Prometheus config needs to be updated in order to see the new exported. Use the following scrape config as an example:

scrape_configs:
    - job_name: status-cake-exporter
    honor_timestamps: true
    scrape_interval: 10m
    scrape_timeout: 1m
    metrics_path: /
    scheme: http
    static_configs:
    - targets:
        - status-cake-exporter.default.svc:8000

Grafana

Data collected by Prometheus can be easily surfaced in Grafana.

Using the Statusmap panel by flant you can create a basic status visualization based on uptime percentage:

PromQL

status_cake_test_info * on(test_id) group_right(test_name) status_cake_test_uptime_percent

Development

This repository uses Tilt for rapid development on Kubernetes.

To use this, run:

cd chart/status-cake-exporter
tilt up

Tilt will reload your environment when it detects changes to your code.

Note: You will need to provide valid credentials for StatusCake in your Tiltfile for this to work.