synse-prometheus provides a Prometheus exporter for the metrics being provided by your data center and IT equipment. You can use this, in conjunction with Prometheus to monitor these metrics and alert on them.
Kick the tires
To keep everything tied together, this starts a prometheus server up for you. Send your browser to
http://localhost:9090 and play around.
To stop it, you can run:
Note: the default configuration uses the embedded synse-server and ipmi emulator. If you'd like to use a different backend, check out
docker run -it --rm vaporio/synse-prometheus python runserver.py --help
- To specify which synse-graphql instance to collect data from, set the SYNSE_GRAPHQL environment variable to the url of the graphql server.
- Configure your prometheus server. For more detailed documentation, see the [prometheus docs]. An example config:
global: scrape_interval: 15s scrape_configs: - job_name: 'synse' # Override the global default scrape_interval: 300s static_configs: - targets: ['synse-prometheus:9243']
Run the server
make build dev python runserver.py
- From outside the container (or inside it), you can run
Run the tests (as part of development)
make build dev tox
See nosetests for some more examples. Adding
@attr('now') to the top of a function is a really convenient way to just run a single test.