Re: Konflux: What is the right time?

[Miroslav Vadkerti <mvadkert@xxxxxxxxxx>]

Hi

On Tue, Mar 11, 2025 at 10:53 AM Jakub Čajka <jcajka@xxxxxxxxx> wrote:

Mar 10, 2025, 23:19 by walters@xxxxxxxxxx:

>
>
> On Sat, Mar 8, 2025, at 3:36 PM, Richard W.M. Jones wrote:
>
>> This is an interesting point.  Is there a Kubernetes story on (eg)
>> POWER, s390x or RISC-V?
>>
>
> Yes. My employer for example has a Kubernetes downstream product called OpenShift that runs on ppc64le and s390x. AFAIK most of the upstream container images include those architectures as well.
>
Will be Fedora provided with additional resources(HW, sysadmins and developers)? AFAIK atm we don't have k8/Openshift Clusters available for these or any upcoming architectures like RISC-V.

IMHO having a working k8/Openshift cluster significantly(IMNHO and unreasonably) shift bar for architecture inclusion in to distribution. Overall I would also expect increased costs for building and maintaining the infrastructure for Fedora.
 Could be/Will be konflux made to run in "single/simple node/executor/..."(I guess having  remote access to podman/OCI runtime socket and/or ssh access on host) configuration? I would see use for it for development and bootstrapping.

We bootstrap Konflux in our integration testing for Testing Farm on kind running on a single VM:

https://github.com/konflux-ci/konflux-ci?tab=readme-ov-file#trying-out-konflux

https://gitlab.com/testing-farm/integrations-konflux/-/blob/main/tests/integration-test/test.sh?ref_type=heads

Note that this is now working on container builds, rpms will be added once ready to be consumed.

Best regards,

/M

 

>> It possibly doesn't matter depending on the
>> total architecture -- eg. with Koji we have an x86-based hub and
>> riscv64 builders,
>>
>
> But also yes to this - a key principle I think we want to take here going forward is that most "baseline" operations should be reproducible by executing a container (or pod) on a standalone system (such as a workstation) - which is exactly what podman (for example) can do.
>
I would expect that whole konflux should be able to mostly run in this "degraded"(just a bunch of containers/pods) state for development and bootstrapping.
JC


> --
> _______________________________________________
> devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
> To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
> Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
> Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
>

--
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue

--

🧑‍🔧 Miroslav Vadkerti :: Senior Principal QE :: Testing Farm / Linux QE

🎩 Red Hat Slack :: mvadkert :: #testing-farm #tmt
📱 Fedora Matrix :: mvadkert :: #Fedora CI

🌍️ Remote Czech Republic :: Red Hat Czech s.r.o

-- 
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue