Bug 2439690

Summary:	Changes/Build FCOS on Fedora Konflux
Product:	[Fedora] Fedora	Reporter:	Allison King <alking>
Component:	Changes Tracking	Assignee:	Joel Capitao <jcapitao>
Status:	ASSIGNED ---	QA Contact:
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	rawhide	CC:	alking, duffy, dustymabe, jcapitao, joel.capitao93, zbyszek
Target Milestone:	---	Flags:	duffy: needinfo? (dustymabe) duffy: needinfo? (joel.capitao93) fedora-admin-xmlrpc: mirror+
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	---
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:		Type:	---
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:	2387673

Description Allison King 2026-02-13 14:45:41 UTC

This is a tracking bug for Change: Changes/Build FCOS on Fedora Konflux
For more details, see: https://fedoraproject.org/wiki/Changes/Build_FCOS_on_Fedora_Konflux

We want to build Fedora CoreOS updates payloads in Konflux, instead of Jenkins.

If you encounter a bug related to this Change, please do not comment here. Instead create a new bug and set it to block this bug.

Comment 1 Zbigniew Jędrzejewski-Szmek 2026-02-24 18:12:05 UTC

What is the status here? Will this be ready for F44?

Comment 2 Dusty Mabe 2026-02-24 18:25:41 UTC

This is currently not implemented, but I'd prefer not to defer just yet. The blocker here was getting the Fedora konflux cluster re-deployed https://issues.redhat.com/browse/KFLUXINFRA-2796

The good news is that we can easily switch on and off the konflux building in the CoreOS pipelines AND this change itself is transparent to the user (i.e. the user observes no difference with or without konflux). We have successfully built `rawhide` using konflux before and are currently waiting for the re-deployed cluster to switch it back on there and also for 44/branched.

Comment 3 Máirín Duffy 2026-03-10 18:16:36 UTC

@dustymabe This came up in this week's FESCo - any updates on this? It seems the cluster is up now, has auth and dns.

Comment 4 Joel Capitao 2026-03-11 17:14:08 UTC

Thank you for your patience here, and sorry for late answer. 
The cluster is up but it's still not live for the community c.f https://issues.redhat.com/browse/KFLUXINFRA-2789 and https://issues.redhat.com/browse/KFLUXINFRA-2965 
I believe now that the cluster won't be live before F44 GA, so I guess we can switch on the Konflux building again. We won't have logs until the new cluster is fully installed, but at least this would prevent us to defer to F45, whereas all the preliminary work is done i.e: just one knob to switch on.

Comment 5 Máirín Duffy 2026-03-11 17:44:04 UTC

@joel.capitao93 No problem! So the cluster is up, but the community can't access, and it looks like from the issue in Jira you pointed at that it's not anticiapted all setup tasks will be complete before 3/31. The go-no-go call for F44 final is 04/02. But you believe this is something we could switch on around the time of GA of F44 and it's one switch to turn on, and isn't dependent upon anything in the release?

Comment 6 Joel Capitao 2026-03-12 10:24:54 UTC

> But you believe this is something we could switch on around the time of GA of F44 and it's one switch to turn on, and isn't dependent upon anything in the release

Yes, we'll switch on again end of March even if the new cluster is not fully ready. The hybrid workflow Konflux/Jenkins works well, but we had to switch off Konflux building, and hand it off to Jenkins because of an issue with log persistence c.f https://issues.redhat.com/browse/KONFLUX-11323.
So currently, the Continuous Delivery is done with Jenkins only, and CI is done with both Jenkins and Konflux.
Also, there is no dependency upon anything, so it's safe from project management perspective.