-
Bug
-
Resolution: Won't Do
-
Critical
-
None
-
4.13.z, 4.14.z
-
None
Description of problem:
HA Proxy router should have ROUTER_METRICS_HAPROXY_SERVER_THRESHOLD variables configurable.This feature was available in ocp 3.11.
Version-Release number of selected component (if applicable):
How reproducible:
Verify that haproxy_exporter_server_threshold{type="current"} exceed the 500 limit:
Steps to Reproduce:
$ oc exec -c prometheus prometheus-k8s-0 -n openshift-monitoring -- sh -c 'curl -sk -H "Authorization: Bearer $(cat /var/run/secrets/kubernetes.io/serviceaccount/token)" https://<router pod IP>:1936/metrics' |grep haproxy_exporter_server_threshold # HELP haproxy_exporter_server_threshold Number of servers tracked and the current threshold value. # TYPE haproxy_exporter_server_threshold gauge haproxy_exporter_server_threshold{type="current"} 2000 haproxy_exporter_server_threshold{type="limit"} 500
Actual results:
This variable is a measure of the max number of servers allowed to be measured when reporting metrics (default 500).
Expected results:
This feature was available in OCP 3.11 and the default setting for this parameter in OCP 4 is insufficient for the metrics gathering with prometheus
Additional info:
- duplicates
-
RFE-2195 Ingress router tuning
- Accepted