Bug 2057696 - Removing console still blocks OCP install from completing
Summary: Removing console still blocks OCP install from completing
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Management Console
Version: 4.9
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
: 4.11.0
Assignee: Jakub Hadvig
QA Contact: Yadan Pei
URL:
Whiteboard:
Depends On:
Blocks: 2076453
TreeView+ depends on / blocked
 
Reported: 2022-02-23 21:02 UTC by David Vossel
Modified: 2022-08-10 10:51 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: No Doc Update
Doc Text:
Clone Of:
Environment:
Last Closed: 2022-08-10 10:50:55 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
cluster operator yaml (41.41 KB, text/plain)
2022-02-23 21:02 UTC, David Vossel
no flags Details


Links
System ID Private Priority Status Summary Last Updated
Github openshift console-operator pull 642 0 None open Bug 2057696: Console operator should not block installation/upgrade process when set to Removed state 2022-03-06 20:57:04 UTC
Red Hat Product Errata RHSA-2022:5069 0 None None None 2022-08-10 10:51:12 UTC

Description David Vossel 2022-02-23 21:02:14 UTC
Created attachment 1863080 [details]
cluster operator yaml

Description of problem:

A removed management console still blocks an OCP install from completing.

We're working on a hypershift platform that uses CNV to host OCP worker nodes. While we plan on supporting the management console in the future, we're at a phase now where it will take some time before we can get connectivity to the console working properly.

In the meantime, we have attempted to remove the management console using the steps here, https://access.redhat.com/solutions/3808951, where we set  "Management State:  Removed" on the console operator resource. That successfully removes the management console, however the OCP install is still complains that we're waiting on the management console to become available.

Version-Release number of selected component (if applicable):
4.9.21

How reproducible:
100%

Steps to Reproduce:
1. Set "ManagementState: Removed"  on console operator while management console connectivity is failing.

Actual results:

OCP install still won't move to completed due to waiting on console to become available

Expected results:

OCP install should not wait on a management console to become available that is removed.

Comment 2 Jakub Hadvig 2022-02-25 13:19:25 UTC
Hey David,
thanks for reporting the issue. Unfortunately console is a core component, thats why you wont be able to finish cluster installation without it, or at least Im not familiar with any change...
Please open an RFE since this is architectural change.

Comment 3 Samuel Padgett 2022-02-25 17:37:47 UTC
Jakub, this is setting `managementState: Removed` in the console operator config, which is already supported. It should not block upgrade. The operator is reporting a degraded state after console is successfully removed.

Comment 4 Jakub Hadvig 2022-02-28 07:13:20 UTC
So Im a bit confused right now,I thought this is not about the upgrade but rather the installation process, where the use case is to remove the console, due to which the installation will not get finished.
David, which one is it?

Comment 5 David Vossel 2022-03-01 14:07:10 UTC
>  due to which the installation will not get finished.
> David, which one is it?

install

Comment 6 Jakub Hadvig 2022-03-03 15:21:26 UTC
Reassigning to Installer component since this is not issue of the Console.

Comment 8 Jakub Hadvig 2022-03-06 20:56:39 UTC
So I was in an impression that the installation is blocked by the fact that OCP can not be installed without the console, since its the core component.

Comment 11 Yadan Pei 2022-04-19 01:54:42 UTC
1. spin up a 4.11 nightly cluster 4.11.0-0.nightly-2022-04-15-025856
2. set console.operator ManagementState to Removed
$ oc get console.operator cluster -o json | jq .spec.managementState
"Removed"
3. check co/console status, it shows 'Degraded: False' and 'Available: True'
$ oc get co console -o json | jq .status.conditions
[
  {
    "lastTransitionTime": "2022-04-18T07:12:44Z",
    "message": "All is well",
    "reason": "AsExpected",
    "status": "False",
    "type": "Degraded"
  },
  {
    "lastTransitionTime": "2022-04-18T07:15:49Z",
    "message": "All is well",
    "reason": "AsExpected",
    "status": "False",
    "type": "Progressing"
  },
  {
    "lastTransitionTime": "2022-04-18T07:15:45Z",
    "message": "All is well",
    "reason": "AsExpected",
    "status": "True",
    "type": "Available"
  },
  {
    "lastTransitionTime": "2022-04-18T07:12:45Z",
    "message": "All is well",
    "reason": "AsExpected",
    "status": "True",
    "type": "Upgradeable"
  }
]
4. trigger an upgrade from 4.11.0-0.nightly-2022-04-15-025856 to 4.11.0-0.nightly-2022-04-15-153812, the upgrade succeeded, console removed state doesn't block upgrade

Comment 14 errata-xmlrpc 2022-08-10 10:50:55 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (Important: OpenShift Container Platform 4.11.0 bug fix and security update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2022:5069


Note You need to log in before you can comment on or make changes to this bug.