search

redhat-cop / global-load-balancer-operator

A global load balancer operator for OpenShift
Apache License 2.0
53 stars 16 forks source link

ServiceMonitor rejected on OCP 4.14 #45

Open vrutkovs opened 1 year ago

vrutkovs commented 1 year ago

Installed global-load-balancer on OCP 4.14 and it appears user level monitoring is rejecting the servicemonitor for global-load-balancer. PrometheusOperatorRejectedResources alert appears and user level monitor has:

2023-09-26T08:29:50.229341706Z level=warn ts=2023-09-26T08:29:50.229243216Z caller=resource_selector.go:174 component=prometheusoperator msg="skipping servicemonitor" error="it accesses file system via bearer token file which Prometheus specification prohibits" servicemonitor=global-load-balancer-operator/global-load-balancer-operator-controller-manager-metrics-monitor namespace=openshift-user-workload-monitoring prometheus=user-workload

Not sure if its user monitoring issue or this operator's ServiceMonitor needs additional changes

sabre1041 commented 1 year ago

@vrutkovs can you share which version of 4.14 you are seeing this on. I deployed rc2 and did not experience this situation, but did observe that a service account token was present in the pod

vrutkovs commented 1 year ago

It was rc1 and rc2 on GCP, most likely the trigger was user-level monitoring. I'll play with this a few more times to see what's causing it