Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1966488

Summary:	[master] Assisted Service operator's controllers are starting before the base service is ready
Product:	OpenShift Container Platform	Reporter:	Osher De Paz <odepaz>
Component:	assisted-installer	Assignee:	Rom Freiman <rfreiman>
assisted-installer sub component:	Deployment Operator	QA Contact:	Omri Hochman <ohochman>
Status:	CLOSED CURRENTRELEASE	Docs Contact:
Severity:	high
Priority:	urgent	CC:	aos-bugs, mfilanov, yobshans
Version:	4.8	Keywords:	Triaged
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:	AI-Team-Hive
Fixed In Version:	OCP-Metal-v1.0.22.1	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:
Clones:	1968455 1969410 (view as bug list)		Environment:
Last Closed:	2022-08-28 08:45:59 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:	1968455, 1969410, 1971300

Description Osher De Paz 2021-06-01 09:23:47 UTC

Description of problem:
The process of starting controllers in the assisted-service operator doesn't make into account the current state of backend REST API service.
This makes it start before we've uploaded rhcos base image, or before we've validated the pull-secret for auxiliary images (controller, agent, installer).

How reproducible:
REST API preparation might be slow or fast, and we might make use of the operator in different timings.
But pull-secret validation for example is something that fails consistently, and might be a good candidate to replicate this bug.


Steps to Reproduce:
1. Start installation of the operator, using images not from ordinary registries (not from quay.io, redhat.registry.io, etc.)
2. be aware of when the assisted-service pod actually becomes ready
3. apply cluster deployment with all the objects
4. look on the agentclusterinstall conditions. They will not be resolved yet

Actual results:
kube-api interface is ready, REST API is not.

Expected results:
kube-api interface not being ready until REST API is

Additional info:
This failure as an example
https://prow.ci.openshift.org/view/gs/origin-ci-test/pr-logs/pull/openshift_assisted-service/1811/pull-ci-openshift-assisted-service-master-e2e-metal-assisted-operator-ztp/1399582336479137792

Comment 2 Michael Filanov 2021-06-01 09:28:41 UTC

In main, we have `ApiEnabler`, that enable the api after we finished uploading rhos images, suggesting to add another function to it, IsEnabled and start the operators only after the api is ready.

Comment 3 Raz Regev 2021-06-09 09:56:19 UTC

controller will now wait for REST api to be ready
merged to master, commit 7cb0e2eb5a30fee6a221f22ccba0a082492e578e