proxy not configurable from CR

[sanbornick]

Expected Behavior

Following the docs should have let me connect to update through a proxy. I used

- name: CURL_OPTIONS
  value: '-L -x http://<proxyhostname>:8080'

with proxyhostname actually replaced.

Actual Behavior

The proxy configuration added to the CR didn't seem to make it to the running pods. deleting pods caused a new pod that still didn't have the configuration.

Steps to Reproduce the Problem

1. follow the docs

Additional Info

• Kubernetes version:

  Output of kubectl version:

kubectl version
Client Version: version.Info{Major:"1", Minor:"19", GitVersion:"v1.19.2", GitCommit:"f5743093fd1c663cb0cbc89748f730662345d44d", GitTreeState:"clean", BuildDate:"2020-09-16T21:51:21Z", GoVersion:"go1.15.2", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"18+", GitVersion:"v1.18.3+b0068a8", GitCommit:"b0068a8", GitTreeState:"clean", BuildDate:"2020-09-11T20:03:06Z", GoVersion:"go1.13.4", Compiler:"gc", Platform:"linux/amd64"}

• Jenkins Operator version:

$ kubectl describe  csv jenkins-operator.0.7.0
...
      Deployments:
        Name:  jenkins-operator
        Spec:
          Replicas:  1
          Selector:
            Match Labels:
              Name:  jenkins-operator
          Strategy:
          Template:
            Metadata:
              Creation Timestamp:  <nil>
              Labels:
                Name:  jenkins-operator
            Spec:
              Containers:
                Command:
                  /manager
                Env:
                  Name:  WATCH_NAMESPACE
                  Value From:
                    Field Ref:
                      Field Path:  metadata.annotations['olm.targetNamespaces']
                  Name:            POD_NAME
                  Value From:
                    Field Ref:
                      Field Path:   metadata.name
                  Name:             OPERATOR_NAME
                  Value:            jenkins-operator
                  Name:             DEFAULT_JENKINS_IMAGE
                  Value:            quay.io/redhat-developer/origin-jenkins@sha256:a8e94991bd42290c51310ad1d34554f1d1604813a9f15f8acf96e18f71ff65c7
                  Name:             JENKINS_SIDECAR_IMAGE
                  Value:            quay.io/redhat-developer/jenkins-kubernetes-sidecar@sha256:590f088f15cc0c010d1730d68780bba3cee692b9c07d9322aa758f733ba08977
                  Name:             JENKINS_IMAGE_BUILDER_NAME
                  Value:            quay.io/redhat-developer/openshift-jenkins-image-builder@sha256:473d6dfb011c69f32192e668d86a47c0235791e7e857c870ad70c5e86ec07e8c
                  Name:             JENKINS_BACKUP_IMAGE
                  Value:            registry.redhat.io/ubi8/ubi-minimal@sha256:c6473956faa4985a41776701993718eb27ab933078bf579768e8ca7d967459ee
                Image:              quay.io/redhat-developer/openshift-jenkins-operator@sha256:8917d4159b0ab2a69d0465dff76cbbe75525034e43c9d60d4b5ea519eeb59191
                Image Pull Policy:  IfNotPresent
                Name:               jenkins-operator
                Resources:
              Service Account Name:  jenkins-operator
...
Version:  0.7.0

Configuring the proxy in the UI at /pluginManager/advanced allowed it to work. It was empty even after env vars were set. No auth was in play in this context. I initially corrected the CR after creation. I also created a new CR with the value set as intended to rule out the operator not reconciling it. This is as far as I went.