Bug 2090457

Summary:	openshift-debug-node- namespaces do not get deleted for debug terminals launched via console
Product:	OpenShift Container Platform	Reporter:	roarora
Component:	Management Console	Assignee:	Jakub Hadvig <jhadvig>
Status:	CLOSED DEFERRED	QA Contact:	Xiyun Zhao <xiyuzhao>
Severity:	low	Docs Contact:
Priority:	unspecified
Version:	4.7	CC:	jhadvig, talessio, yapei
Target Milestone:	---
Target Release:	4.12.z
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2023-03-09 01:20:06 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:

Description roarora 2022-05-25 19:34:34 UTC

Description of problem:

For node debug terminals opened via web console, pods and the corresponding projects do not get deleted 

# oc get pods -A |grep debug
openshift-debug-node-wrcww                         ip-10-130-82-35.us-west-2.compute.internal-debug                       0/1     Comple

Version-Release number of selected component (if applicable):
4.7.40

How reproducible:
For affected customer, exact sequence of reproducing is not known as multiple teams use the cluster. I can still reproduce by opening debug terminal for the node and then closing the browser tab without navigating to any other page even on 4.10. However exact reproducing steps for customer is unknown and can be something other than this too


Actual results:
pod/project do not get deleted

Expected results:
pod/project should be deleted

Additional info:
Similar problem was reported and fixed in https://bugzilla.redhat.com/show_bug.cgi?id=1947430  , as per the Pull request, seems changes were done to oc cli. What about web console? The errata for that bug was released for 4.7.8

Comment 2 Jakub Hadvig 2022-05-26 08:43:50 UTC

Thanks for reporting the issue.
Given its severity of `low` and the fact that 4.7 is in maintenance support, I will ask you if you could reproduce this issue on supported clusters - 4.9.+
Otherwise I am closing this bug as "During the Maintenance Support phase, qualified Critical and Important Security Advisories (RHSAs) and Urgent and Selected High Priority Bug Fix Advisories (RHBAs) may be released as they become available. Other Bug Fix (and Enhancement (RHEA) Advisories may be released at Red Hat’s discretion, but should not be expected." per https://access.redhat.com/support/policy/updates/openshift.

Comment 3 roarora 2022-05-26 17:49:40 UTC

Reproducible on a 4.10 : 

# oc get pods -n openshift-debug-node-z8fxv
NAME                                               READY   STATUS      RESTARTS   AGE
ip-10-0-140-89.ap-south-1.compute.internal-debug   0/1     Completed   0          22h

For reproducing above, a terminal session was launched from the console and then the browser tab was closed without navigating anywhere else.

Comment 7 Xiyun Zhao 2022-06-20 13:17:54 UTC

Hi zherman could you help check if the below reproduce/verification step is workable in your environment?

I failed to verify this bug in payload 4.11.0-0.nightly-2022-06-15-222801 (attached snapshot for reference)

Verification Step:
1. Create 2-5 Pods with error 'Crashloopback-off' error (sample as below)
kind: Pod
apiVersion: v1
metadata:
  name: crash-pod 
  labels:      
   app: test
   deploymentconfig: nodejs-ex-git 
spec:   
  containers:     
   - name: crash-app      
     image: quay.io/openshifttest/crashpod   
  restartPolicy: Always
2. Do not close the Pod page, Open the debug container on another Tab of the Browser, and wait for a while
3. Go back to the Workloads - Pods page, close the tab
4. Verify if the debug container is being deleted

Result:
4. The debug container is still can be found on the Pod list page with the Running state

Comment 10 Xiyun Zhao 2022-07-21 02:01:01 UTC

Hi @gvudinh
The mentioned issue on comment7 is not entirely fixed, the issue still can be reproduced on payload 4.12.0-0.nightly-2022-07-20-030220

Comment 15 Shiftzilla 2023-03-09 01:20:06 UTC

OpenShift has moved to Jira for its defect tracking! This bug can now be found in the OCPBUGS project in Jira.

https://issues.redhat.com/browse/OCPBUGS-9286