Bug 1895372

Summary: Web console going blank after selecting any operator to install from OperatorHub
Product: OpenShift Container Platform Reporter: Yash Chouksey <ychoukse>
Component: Management ConsoleAssignee: Samuel Padgett <spadgett>
Status: CLOSED ERRATA QA Contact: Yanping Zhang <yanpzhan>
Severity: high Docs Contact:
Priority: high    
Version: 4.5CC: aos-bugs, ivn13452, jokerman, rhamilto, spadgett, yapei
Target Milestone: ---   
Target Release: 4.7.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Previously, an OperatorGroup with a missing `status` block in any namespace could cause a runtime error in the web console when installing a new operator from OperatorHub. The problem has been fixed.
 Story Points: ---
Clone Of: Environment:
Last Closed: 2021-02-24 15:31:26 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1896600    
Attachments:
Description Flags
Console pod logs none

Description Yash Chouksey 2020-11-06 14:17:15 UTC 
Description of problem:
Operator hub is showing blank screen after selecting any operator to install. 
Gif: https://attachment-viewer.cee.redhat.com/#/viewer/02769012?uuid=ed1dbd5f-85e6-498a-b583-24c820efbcba 

Issue occurs with all the operators, irrespective of their provider type.
The cluster has all packagemanifests, opsrc and catsrc present. 


Version-Release number of selected component (if applicable):
Baremetal, connected installation. Cluster version is at 4.5.16 (issue occurs on similar cluster at version 4.5.14)

How reproducible:
Always

Steps to Reproduce:
1. Login the Web console as the cluster-admin user(or any user).
OCP web-console -> Operator Hub -> Select any operator (Click install) ->  Blank white screen


Actual results:
The page jumps in the blank status, and cannot recover after clicking the back button. URL redirection doesn't work, a new window needs to be opened for console.

Expected results:
I know something is wrong with the operator hub(maybe), but we should pop-up errors instead of the blank page.


Additional info:
Nothing suspicious in marketplace pods, or any other related pods, all are in running state.
However, the openshift-console shows logs:

2020-11-05T16:10:02Z server: authentication failed: http: named cookie not present
2020-11-05T16:10:04Z server: authentication failed: http: named cookie not present

Restarting console pods didn't help.
Comment 2 Robb Hamilton 2020-11-09 16:28:40 UTC 
Jakub Hadvig tried and failed to reproduce on a fresh 4.5.16 cluster.  I tried and failed to reproduce using Firefox 82.0.3 for macOS 10.15.7, namespace 'gitlab-runner', and operator 'gitlab' on a fresh 4.5.14 cluster. Are there any browser console errors present when the error occurs?
Comment 4 Yash Chouksey 2020-11-09 17:44:24 UTC 
Created attachment 1727892 [details]
Console pod logs
Comment 5 Robb Hamilton 2020-11-09 19:23:46 UTC 
> However, we didn't see any errors on the developer console while replicating the issue on Cu's env. 

Did you mean the web console (e.g., https://developer.mozilla.org/en-US/docs/Tools/Web_Console)?

Typically when the entire browser window goes white (empty), that means the entire OpenShift web console app has crashed, and I would expect to see one or more errors in the web console.
Comment 7 Samuel Padgett 2020-11-10 15:12:42 UTC 
We've been able to reproduce. This is happening because one of the OperatorGroup resources doesn't have a `status` stanza, which causes a runtime error in console.

Uncaught TypeError: og.status is undefined
Comment 9 Samuel Padgett 2020-11-10 15:40:17 UTC 
We're working on a fix, but you should be able to workaround this problem in the meantime by adding editing any OperatorGroup without a `status` and adding `status: {}` to the YAML.
Comment 10 Samuel Padgett 2020-11-10 15:43:56 UTC 
(In reply to Samuel Padgett from comment #9)
> We're working on a fix, but you should be able to workaround this problem in
> the meantime by adding editing any OperatorGroup without a `status` and
> adding `status: {}` to the YAML.

Actually it probably has to be

status:
  namespaces: []
Comment 14 Yanping Zhang 2020-11-11 11:37:51 UTC 
Checked on ocp 4.7 cluster with payload 4.7.0-0.nightly-2020-11-11-055454, create operatorgroup with file in attachment. Then go to operatorhub page, choose an operator, click "Install", the page shows up correctly, it's not blank now.
The bug is fixed.
Comment 21 errata-xmlrpc 2021-02-24 15:31:26 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 (Moderate: OpenShift Container Platform 4.7.0 security, bug fix, and enhancement 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-2020:5633
Comment 22 Allenwood 2022-10-22 06:41:15 UTC Comment hidden (spam)