You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am trying to make Prometheus scrape the APISIX endpoint without success.
I don't want to use the ServiceMonitor because it's not managed directly in the cluster, so basically I just need to add the annotations (done via apisix.podAnnotations) and then expose the port, which is the blocking part.
It looks like there is no way to expose more ports from the pods ?
I tried the gateway.http.additionalContainerPorts but then it get in conflicts as it's also setting up APISIX to listen on this port.
I'm quite surprised to use ServiceMonitor that is not really standard nor useful for this kind of case, as all what it is setting up can be setup using pod annotations and would work even without dependencies installed (here I can't use the ServiceMonitor because the CRD is not supported)
Should we add a simple gateway.prometheus.enabled variable that set this up ? Also an gateway.additionalContainerPorts that only add ports to the pod would make sense.
Thank you.
The text was updated successfully, but these errors were encountered:
I am not K8s cluster admin and have limited permission.
The helm chart should be able to add ingress to metrics endpoint and then we can use the standard http way to scrape.
Currently I manually added the ingress, but it should better be managed together with the values yaml.
Hello.
I am trying to make Prometheus scrape the APISIX endpoint without success.
I don't want to use the
ServiceMonitor
because it's not managed directly in the cluster, so basically I just need to add the annotations (done viaapisix.podAnnotations
) and then expose the port, which is the blocking part.It looks like there is no way to expose more ports from the pods ?
I tried the
gateway.http.additionalContainerPorts
but then it get in conflicts as it's also setting up APISIX to listen on this port.I'm quite surprised to use
ServiceMonitor
that is not really standard nor useful for this kind of case, as all what it is setting up can be setup using pod annotations and would work even without dependencies installed (here I can't use the ServiceMonitor because the CRD is not supported)Should we add a simple
gateway.prometheus.enabled
variable that set this up ? Also angateway.additionalContainerPorts
that only add ports to the pod would make sense.Thank you.
The text was updated successfully, but these errors were encountered: