Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-33467

[4.14] vsphere-problem-detector-operator does not respect cluster wide proxy

XMLWordPrintable

    • ?
    • Important
    • Yes
    • False
    • Hide

      None

      Show
      None
    • Hide
      *Cause*: vsphere-problem-detector did not have HTTP(S) proxy configured.
      *Consequence*: vsphere-problem-detector could not connect to vCenter. All checks failed and alerted about invalid cluster configuration.
      *Fix*: What was done to fix the bug. vsphere-problem-detector now uses the same HTTP(S) proxy as the rest of the cluser.
      *Result*: vsphere-problem-detector checks succeed.
      Show
      *Cause*: vsphere-problem-detector did not have HTTP(S) proxy configured. *Consequence*: vsphere-problem-detector could not connect to vCenter. All checks failed and alerted about invalid cluster configuration. *Fix*: What was done to fix the bug. vsphere-problem-detector now uses the same HTTP(S) proxy as the rest of the cluser. *Result*: vsphere-problem-detector checks succeed.
    • Bug Fix
    • In Progress

      This is a clone of issue OCPBUGS-28879. The following is the description of the original issue:
      —
      Description of problem:

      In a recently installed cluster running 4.13.29, after configuring the cluster-wide-proxy, the "vsphere-problem-detector" is not taking the proxy configuration.
As the pod cannot reach vSphere  it's failing to run checks:
2024-02-01T09:28:00.150332407Z E0201 09:28:00.150292       1 operator.go:199] failed to run checks: failed to connect to vsphere.local: Post "https://vsphere.local/sdk": dial tcp 172.16.1.3:443: i/o timeout  

The pod doesn't get the cluster proxy settings as expected:
   - name: HTTPS_PROXY
     value: http://proxy.local:3128
   - name: HTTP_PROXY
     value: http://proxy.local:3128

Other storage related pods get the configuration expected as above.

This causes the vsphere-problem-detector to fail connections to vSphere, hence failing the health checks.

       

      Version-Release number of selected component (if applicable):

        4.13.29

      How reproducible:

         Always

      Steps to Reproduce:

          1.Configure cluster-wide proxy in the environment. 
    2. Wait for the change
    3. Check the pod configuration

      Actual results:

          vSphere health checks failing

      Expected results:

          vSphere health checks working through the cluster proxy

      Additional info:

              rhn-engineering-jsafrane Jan Safranek
              openshift-crt-jira-prow OpenShift Prow Bot
              Wei Duan Wei Duan
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated:
                Resolved: