healthy

healthy is a service that via HTTP REST API exposes current status of configured health checks.

The main use case for this service is to have single endpoint for checking current status of an entire infrastructure. For example, in CI, after deploying or modifying your infrastructure the test can easily check whether the infrastructure is fully operational before executing its scenario. The same in product, having a single check is very convenient for the Ops team to quickly verify product status.

configuration

Configuration can be either JSON or YAML or actually anything that is supported by the spf13/viper library. Configuration options can also be set via environment variables with the names of the variables prefixed with HEALTHY_, e.g.:

HEALTHY_SERVER_LISTEN_ON=localhost:8888 ./healthy --config ./healthy.yml 

There are two main configuration sections: server and checks, for details please see here.

example

server:
listen_on: 127.0.0.1:8199
checks:
google:
type: http
url: http://google.com
rabbit:
type: dial
address: 127.0.0.1:5672
echo:
command: echo "hello world!"

Other examples can be found in examples/.

endpoints

Currently, there are two endpoints available:

• getting status of selected checks

  curl -f http://localhost:8199/v1/check/google
  curl -f http://localhost:8199/v1/check/google,rabbit

• getting status of all configured checks

  curl -f http://localhost:8199/v1/status

In both cases, success is determined by response with status code 200. Other status codes mean the check has failed.

docker images

healthy is also released as docker image available for example as zkostrzewa/healthy:0.1.0.

Until #20 is fixed docker image requires mounting custom configuration file to /etc/healthy/healthy.yml or /healthy.yml.