k8s-ingress-health-bot

k8s-ingress-health-bot

A K8s Ingress Health Bot is a lightweight application to check the health of qualified ingress endpoints for a given kubernetes namespace.

How it works

The Bot will start a web server in the background.

Users can use one of the http paths to retrieve the health of qualified ingress endpoints in real time or through the latest populated prometheus metrics.

/health

Return status 200 if the Bot is running.

/metrics

Expose the prometheus metrics for all qualified ingress endpoints to be monitored. Periodically (default every minute), the bot will check the health of each qualify ingress endpoint within a given k8s namespace and populate the health metrics for each endpoint.

Prometheus Metrics

  endpoint_health{namespace="example",status_code="200",url="https://aaron.example.com/-/healthy"} 1

  cert_validity_days{namespace="example",url="https://aaron.example.com/-/healthy"} 88

/namespace-health

Return the real time health of the all qualified ingress endpoints within a given k8s namespace.

All endpoints are up.

There is no ingress endpoint to monitor.

Internal error, unable to retrieve ingress endpoints.

One or more endpoints are down.

Build

go mod vendor
go install ./cmd/...

Usage (Flags)

usage: k8s-ingress-health-bot [<flags>]

Flags:

  -appName string
        Application name of this bot. (default "k8s-ingress-health-bot")

  -excludedIngressNames string
        A comma delimited ingress list that the bot won't monitor.

  -httpClientTimeOut duration
        Maximum time that the HTTP client will wait when pinging an endpoint. (default 15s)

  -ingressHealthCheckUrlsAnnotationKey string
        Ingress annotation key that the bot will look for. (default "ingress.endpoint.healthcheck.urls")

  -kubeconfig string
        Path to a kubeconfig. Only required if the application is running outside of K8s cluster.

  -namespace string
        K8s namespace that this bot will watch. (default "default")

  -port int
        Web server Port. (default 8088)

  -prometheusMetricsPopulateInterval duration
        Time between populate Prometheus metrics for a monitored endpoints. (default 1m0s)

  -prometheusMetricsResetPeriod duration
        Controls how often Prometheus health metrics cache will be cleared. (default 10m0s)

  -retryAttempts int
        Number of attempts before declaring the endpoint is not accessible. (default 3)

  -retryInterval duration
        Time to wait before pinging an endpoint in case of error. (default 3s)