-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
4.15.z, 4.16.z
-
None
Description of problem:
The prometheus operator fails to reconcile when proxy settings like no_proxy are set in the Alertmanager configuration secret.
Version-Release number of selected component (if applicable):
4.15.z and later
How reproducible:
Always when AlertmanagerConfig is enabled
Steps to Reproduce:
1. Enable UWM with AlertmanagerConfig enableUserWorkload: true alertmanagerMain: enableUserAlertmanagerConfig: true 2. Edit the "alertmanager.yaml" key in the alertmanager-main secret (see attached configuration file) 3. Wait for a couple of minutes.
Actual results:
Monitoring ClusterOperator goes Degraded=True.
Expected results:
No error
Additional info:
The Prometheus operator logs show that it doesn't understand the proxy_from_environment field. The newer proxy fields are supported since Alertmanager v0.26.0 which is equivalent to OCP 4.15 and above.
- blocks
-
OCPBUGS-38398 Cannot use new proxy settings in Alertmanager configuration
- Closed
- is cloned by
-
OCPBUGS-38398 Cannot use new proxy settings in Alertmanager configuration
- Closed
- is depended on by
-
MON-3962 Modify the default Alertmanager configuration secret to use proxy_from_environment
- Closed
- links to
-
RHEA-2024:6122 OpenShift Container Platform 4.18.z bug fix update