Re: Why so long for EPEL-8?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Jul 22, 2021 at 1:56 AM Nico Kadel-Garcia <nkadel@xxxxxxxxx> wrote:
>
> On Mon, Jul 19, 2021 at 7:04 AM Fabio Valentini <decathorpe@xxxxxxxxx> wrote:
> >
> > On Mon, Jul 19, 2021 at 12:06 PM Jiri Vanek <jvanek@xxxxxxxxxx> wrote:
> > >
> > > On 7/17/21 11:15 PM, Ron Olson wrote:
> > > > Hey all-
> > > >
> > > > I’m curious as to why submitting a new build to Bodhi takes a week to be pushed to stable, but two weeks for EPEL-8. Is the presumption that it’s just that much more time for folks to test and verify?
> > >
> > > lack of reviewers, so longer in buildroot is kinda test...
> >
> > Surely not? The user base of EPEL is, by some measures, considerably
> > bigger than the user base of Fedora.
> > Those users might not participate in testing as much, because they
> > rely on the stability of their systems more, but this is certainly not
> > a *lack of reviewers*.
>
> There are odd trade-offs. Many EPEL users do so as a matter of course,
> treating EPEL as an adjunct to RHEL, and would find RHEL or CentOS
> quite useless without EPEL. The availability of python36 for RHEL 7,
> for example, was an excellent step towards python3 for RHEL 7 and
> CentOS 7 users, and I'd have had to consider migrating from RHEL
> without the stable python36 tools from EPEL, which avoided having to
> build up a python suite myself. I do wish our colleagues at RHEL, or
> that Fedora itself, had kept the "python34", "python36", etc.
> numbering from EPEL.  It would have eased demands I'm seeing for
> people to build and ihstalll and manage python in their home
> directories with "linuxbrew", which does not work, or "pyenv", which
> is vulnerable to dependency skew adventures.
>
> I consider it a crying shame that the numbered versions, like python36
> and python38, were abandoned. Unfortunately, it can't always be that
> stable because it's driven by things people want and are willing to do
> the work to put in. Don't get me started on the chromium and ansible
> regressions I've encountered in the last few years. tools like those
> which are used as part of automated testing suites are very vulnerable
> to quite small changes in their features or API's leading to quite
> adventuresome breakdowns.

What are you talking about? The Python interpreter *is* versioned in
Fedora now. We use python3.Y for the package name of the interpreter,
and only the *default* interpreter package produces the python3
package. It *started* in Fedora and made its way to RHEL.

What we're *not* doing yet is having the Python modules built
versioned against multiple versions at once, because that requires
more adjustments to how we build Python modules. Honestly, I'd *like*
to do it, but I'd like to do it without costing my sanity when
debugging Python packaging spec files.



-- 
真実はいつも一つ!/ Always, there's only one truth!
_______________________________________________
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 on the list, report it: https://pagure.io/fedora-infrastructure




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Users]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]

  Powered by Linux