I have kubernetes cluster with multiple services, each service has an actuator and an health-check (readiness+liveness probes) I tried to create a Prometheus job for one of the services to scrape the response (200 response indicating service is good), But from my understanding, Prometheus is storing the data as a "time series" key-value so I'm not sure I am doing it right. This is the actuator response:
{
"status" : "UP",
"redisson" : {
"status" : "UP"
},
"jedis" : {
"status" : "UP",
}
}
And this is my job:
- job_name: 'service'
scheme: https
metrics_path: /health
kubernetes_sd_configs:
- role: service
namespaces:
names:
- default
relabel_configs:
- source_labels: [__meta_kubernetes_service_name]
action: keep
regex: service
- target_label: __address__
replacement: service.iot.test.service.co.il:8090
- source_labels: [__param_target]
target_label: instance
- action: labelmap
regex: __meta_kubernetes_service_label_(.+)
- source_labels: [__meta_kubernetes_namespace]
target_label: default
- source_labels: [__meta_kubernetes_service_name]
target_label: service
Is there a way to scrape the json response that holds the "status"? And the "INVALID" is not a valid start token is caused by a bad scraping?