This project provides an HTTP server that performs health checks to the specified services and exposes HTTP endpoints to get status of configured health checks. Main use-cases of this project are:
- Exposing HTTP health check endpoints for services that don't have a native HTTP health check API (i.e. databases or containers).
- Exposing HTTP health check endpoints from the private network to the public (i.e. you are using external monitoring tool/service that doesn't have access to the private network).
- Supports all health checks from hellofresh/health-go library except gRPC (will be enabled later), and additionally:
- DNS hostname resolve check (A, CNAME, PTR, TXT records)
- ICMP ping
- Pull and run
ghcr.io/buzzer13/healthpose:<tag>image (tag can be eitherlatest, or a specific version likev1.2.0).- Command:
docker run --volume="healthpose-config:/config" --name=healthpose -it "ghcr.io/buzzer13/healthpose:latest" - To use ping health check with runtimes other than Docker engine - you may need to apply
net.ipv4.ping_group_range=0 2147483647sysctl variable to the container.
- Command:
- Update an example configuration file that was created in the volume.
Config file can be placed at the following paths:
/etc/healthpose/healthpose.yaml/config/healthpose.yamlhealthpose.yaml- At the path, specified by the
CONFIG_FILEenvironment variable.
Here is the general configuration file structure:
# Internal HTTP server configuration dictionary.
http:
# Address and port HTTP server should listen at
listen: :8080
# Health check configuration dictionary.
# Key is the service name, that will be used in the heath check URL.
# If you define `test` service below - then `GET /test` endpoint will be created
# in the internal HTTP server.
services:
test:
# Required. Name of the service.
name: example
# Required. Version of the service.
version: v1.0
# Required. List of service health checks, that should be performed.
checks:
- # Required. Health check name. It will be shown in the endpoint response if check fails.
name: cassandra
# Required. Health check interval.
timeout: 60
# Optional. If check is marked as optional - it won't fail health check of the whole service.
optional: true
# Required. One of the available health check configuration dictionaries.
# Examples can be found here: https://github.com/buzzer13/healthpose/blob/main/misc/config/healthpose.yaml
dns:
address: example.com
server: 8.8.8.8:53
type: a
request_timeout: 5
fallback_delay: 0.3You can check out example config here.