-
Bug
-
Resolution: Done-Errata
-
Blocker
-
None
-
None
-
None
-
None
-
1
-
False
-
None
-
False
-
-
-
Tracing Sprint # 256
Version of components:
opentelemetry-operator.v0.102.0-2
OpenShift version 4.16.1
Description of the problem:
When a OpenTelemetry Collector instance is created on OpenShift version 4.16, we see a perpetual creation of secrets in the namespace. Around 15-20 secrets are created per second. We need to get this fixed as soon as possible as it can overwhelm the API server causing the cluster to be instable.
cluster-collector-collector-dockercfg-wz452 kubernetes.io/dockercfg 1 70s cluster-collector-collector-dockercfg-wzdqm kubernetes.io/dockercfg 1 98s cluster-collector-collector-dockercfg-x2vjr kubernetes.io/dockercfg 1 66s cluster-collector-collector-dockercfg-x67gl kubernetes.io/dockercfg 1 25s cluster-collector-collector-dockercfg-x79rb kubernetes.io/dockercfg 1 43s cluster-collector-collector-dockercfg-x8jxh kubernetes.io/dockercfg 1 47s cluster-collector-collector-dockercfg-xcb46 kubernetes.io/dockercfg 1 77s cluster-collector-collector-dockercfg-xjswd kubernetes.io/dockercfg 1 105s cluster-collector-collector-dockercfg-xp77b kubernetes.io/dockercfg 1 50s cluster-collector-collector-dockercfg-xqlgh kubernetes.io/dockercfg 1 45s cluster-collector-collector-dockercfg-xqlsg kubernetes.io/dockercfg 1 74s i% oc get secrets| wc -l 852 % oc get secrets| wc -l 864 % oc get secrets| wc -l 882 % oc get secrets| wc -l 892
Steps to reproduce the issue:
- Deploy a OpenShift 4.16 cluster.
- Create a OpenTelemetry collector instance.
- Check the OpenTelemetry Collector instance namespace for the secrets.
- Notice that the secrets are continuously being generated.
Additional details:
The issue occurs on OpenShift version 4.16 only.
- relates to
-
OCPBUGS-36833 4.16 "Bad" reconciliation loops can cause unbounded dockercfg secret creation
- Closed
- links to
-
RHBA-2024:135619 Red Hat OpenShift distributed tracing 3.2.0 operator/operand containers