Hi! I have a Kubernetes/Prometeus newbie question.
I have a Kubernetes application that I want to scale up via Horizontal Pod Autoscaler based on custom metrics reflecting the length of the app's work queue backlog in the app's internal database. I can expose the metrics on a pod port in Prometheus format, so there is more than one scrape endpoint for fail-over purposes. However I do not want Prometheus to call the scrape endpoint on every pod's instance, only one instance at a time and failover to another instance only if necessary. All instance endpoints report the same data, and calling them in parallel would serve no useful purpose and only increase a workload on storage queries behind the app runtime. Right now the application is packaged as a Kubernetes Deployment. Do I need to convert it for the above as a Service, or can Prometeus understand intrinsically or be instructed in the configuration that only one instance is to be called at a time? Also, can someone please provide an example of the snippet I need to insert into https://github.com/prometheus/prometheus/blob/master/documentation/examples/prometheus-kubernetes.yml so Prometheus picks up the endpoint? Assuming, for example, it is exposed on pod port 3333, path /metrics, service/pod named ABC or with label appname=LABC? Or do I simply tag the pod/service/deployment with some tag and Prometeus starts to collect on it? If yes, how do I pack the endpoint port+protocol+path into the tag? Thanks for the advice. -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/cabf6ba3-79c8-4f5e-865d-1fc51ba52c72%40googlegroups.com.
