On Thu, Jan 9, 2025 at 3:53 PM Michel Lind <michel@xxxxxxxxxxxxxxx> wrote:
On Wed, 2025-01-08 at 20:43 -0800, Brendan Conoboy wrote:
> Over the last few months on chat.fp.o and at conferences I have
> started joining into conversations- listening, asking questions,
> making assertions, telling bad jokes. All this activity has led to a
> recurring question by many people in the community, “uhh, who are
> you?”
>
>
...
> But this isn’t just an introduction, this is a reintroduction. My
> first foray into the Fedora community, beyond just running Fedora,
> was in 2012 when my team at Red Hat started to explore the potential
> for ARM servers. We, with many of you, bootstrapped armv7hl, then
> aarch64. We started with a Seneca College partnership, we gave away
> hardware, we made ugly uboot hacks, ported hundreds of packages,
> helped establish the Linaro Enterprise Group to do more ports &
> standardization, and solved seemingly thousands of circular
> dependencies- and hired a bunch of great contributors along the way,
> folks like Marcin Juszkiewicz and Paul Whalen. Several of us took
> what we learned from the experience and replicated the work inside
> Red Hat with RHEL 7, first for ppc64le, then aarch64- two
> architectures that are mainstream in RHEL today, but scarcely existed
> a decade ago. For me that evolved into driving general RHEL
> development as an individual, then management, then last year I had
> the opportunity to come back to support Red Hat’s Linux communities
> in general and Fedora in particular.
>
Ahh, the perfect person to deal with our s390x capacity crunch ;)
Yes! Actually, I think Kevin and IT just applied a bandaid that has made a significant improvement in wait times. That's the thing about utilization, it's in the domain of queue theory where linear increases in utilization lead to exponential wait increases in wait time. Currently Fedora is using part of a z16 in one of our data centers and there isn't another one available for our use. If we want to increase capacity further our next easiest avenue is to look for space on a z15 or make use of IBM's Cloud where space is available. We've generally tried to avoid using clouds for builders since koji's architecture isn't well suited to it, but I'm hopeful as we find ways to take advantage of konflux we can make use of external clouds for demand bursts. I'm old so I mostly think of builders as the thing people care about, but if other workloads are being held back that are amenable to cloud semantics or where we don't currently need z16 we can explore that.
> I think it’s fair to say that most of the RHEL management team has,
> as a whole, tended to be hands-off in Fedora’s affairs. There are
> some notable exceptions of course, but on average, most people
> working as managers are less involved in Fedora. There’s no written
> rule about this, so for my part I’ve inferred that most managers
> don’t think they have a lot of value to contribute compared to the
> people who they manage who are engaged in Fedora. Even in writing
> this blog I feel apprehensive: do I have anything to say people want
> to hear? Maybe just one thing…
>
Thank you for engaging! I agree with your analysis, and happen to think
that more engagement from management - done properly - is a good thing.
After all certain issues require investment to fix, and Fedora depends
on Red Hat for its infrastructure...
Looking forward to your next posts, and future collaborations,
Cheers!
Brendan Conoboy / Community Linux Engineering / Red Hat, Inc.
-- _______________________________________________ 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
