Bug 2041921 - Documented uninstall procedure leaves many leftovers
Summary: Documented uninstall procedure leaves many leftovers
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Advanced Cluster Management for Kubernetes
Classification: Red Hat
Component: Installer
Version: rhacm-2.4
Hardware: Unspecified
OS: Unspecified
unspecified
medium
Target Milestone: ---
: rhacm-2.5
Assignee: Jakob
QA Contact: txue
Christopher Dawson
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-01-18 14:21 UTC by Nicolas Comete
Modified: 2022-06-09 02:09 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2022-06-09 02:08:48 UTC
Target Upstream Version:
Embargoed:
bot-tracker-sync: rhacm-2.5+


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Github stolostron backlog issues 19204 0 None None None 2022-01-18 15:55:40 UTC
Red Hat Product Errata RHSA-2022:4956 0 None None None 2022-06-09 02:09:16 UTC

Description Nicolas Comete 2022-01-18 14:21:54 UTC
Description of the problem:
After having uninstalled RHACM as per documentation, there are many leftovers in the cluster.

Release version: 2.4

OCP version: 4.7

Browser Info:

Steps to reproduce:
1. Install RHACM as per doc
2. Remove RHACM with command line as per https://access.redhat.com/documentation/en-us/red_hat_advanced_cluster_management_for_kubernetes/2.4/html/install/installing#uninstalling (usinng either Console or command-line instructions)

Actual results:
All RHACM-related components have been removed

Expected results:
Many leftovers are found:
- CRDs, e.g:
channels.apps.open-cluster-management.io                               2022-01-14T08:00:19Z
clustermanagers.operator.open-cluster-management.io                    2022-01-14T08:00:27Z
deployables.apps.open-cluster-management.io                            2022-01-14T08:00:30Z
gitopsclusters.apps.open-cluster-management.io                         2022-01-14T08:00:25Z
helmreleases.apps.open-cluster-management.io                           2022-01-14T08:00:24Z
multiclusterhubs.operator.open-cluster-management.io                   2022-01-14T08:00:23Z
multiclusterobservabilities.observability.open-cluster-management.io   2022-01-14T08:00:27Z
observabilityaddons.observability.open-cluster-management.io           2022-01-14T08:00:22Z
placementrules.apps.open-cluster-management.io                         2022-01-14T08:00:22Z
submarinerconfigs.submarineraddon.open-cluster-management.io           2022-01-14T08:00:27Z
subscriptions.apps.open-cluster-management.io                          2022-01-14T08:00:21Z
- ClusterRoles
- ClusterRoleBindings


Additional info:
Also used the shell script in Troubleshooting section, still not sufficient.

Comment 1 Jakob 2022-01-20 17:20:02 UTC
The operator lifecycle manager that deploys ACM intentionally preserves CRDs after uninstall to prevent data loss risk. See https://github.com/operator-framework/operator-lifecycle-manager/issues/1326. You can delete the CRDs after, but we don't attempt to clean them up at the risk of deleting existing instances of those resources.

Comment 2 Dimon 2022-01-20 17:38:44 UTC
There are much more left-overs, e.g. cluster roles and cluster rolebindings. But more hurting are the gatekeeper constrainttemplates and constraints, which are being processed by gatekeeper after RHACM uninstall.

Comment 3 kurwang 2022-05-02 19:37:09 UTC
Moving to Modified, talked to @cwall about there being new leaks after testing with the old leaks cleared up.

Comment 5 bot-tracker-sync 2022-05-11 05:11:45 UTC
G2Bsync 1123171357 comment 
 nelsonjean Wed, 11 May 2022 04:26:36 UTC 
 G2Bsync

@kurwang , are we planning to continue fixing this in 2.5 or is there a workaround that can be documented?

Comment 6 bot-tracker-sync 2022-05-11 05:11:46 UTC
G2Bsync 1123171415 comment 
 nelsonjean Wed, 11 May 2022 04:26:42 UTC 
 G2Bsync

@kurwang , are we planning to continue fixing this in 2.5 or is there a workaround that can be documented?

Comment 7 bot-tracker-sync 2022-05-11 20:08:27 UTC
G2Bsync 1124237509 comment 
 cameronmwall Wed, 11 May 2022 20:04:29 UTC 
 G2Bsync

@nelsonjean 

We have made all the fixes we intend to for 2.5. The majority of leaks were cleaned up and any left over should be able to be worked around using the [documented cleanup script](https://github.com/stolostron/rhacm-docs/blob/2.4_stage/install/uninstall.adoc).

Persisting leaks and new ones that have sprung up over the development cycle will be handled in 2.6 as part of [this epic](https://app.zenhub.com/workspaces/engineering-backlog-do-not-delete-604fab62d4b98d00150a2854/issues/stolostron/backlog/22481)

I believe this issue can be closed

Comment 8 kurwang 2022-05-12 12:53:56 UTC
Discussed and will be closed, QE has verified the bug.

Comment 11 errata-xmlrpc 2022-06-09 02:08:48 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: Red Hat Advanced Cluster Management 2.5 security updates, images, and bug fixes), 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-2022:4956


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