On Fri, 2023-08-25 at 20:32 -0400, Solomon Peachy via devel wrote:
> On Wed, Aug 23, 2023 at 08:22:42PM +0200, Miroslav Suchý wrote:
> > Do you want to make Fedora 39 better? Please spend 1 minute of your time 
> > and try to run:
> 
> I ran this on a bunch of my fleet, lots of problems, and they all seem 
> to be related to the Python 3.12 bump.

It's a lot of output, but not *so* many problems when you boil it down.

A lot of it is fail2ban, which is not working with Python 3.12 because
it uses asynchat , which has been removed in Python 3.12:

https://docs.python.org/3.12/whatsnew/3.12.html#removed

Upstream is aware of this and working on it, but it's not done yet:
https://github.com/fail2ban/fail2ban/issues/3487

so, there's not much we can do about this, unless we want to write the
port for them. (Not me! I don't volunteer!)

I suppose we could just vendor asynchat and asyncore back into
fail2ban, but that feels pretty ugly. Maybe worth doing in extremis if
upstream can't get it ported before F39 Final, since I think fail2ban
is quite popular?

openocd seems to not be building ATM:
https://bugzilla.redhat.com/show_bug.cgi?id=2226064
I think this is likely due to libgpiod getting a major bump (from 1.6
to 2.0). There's an upstream ticket on this:
https://sourceforge.net/p/openocd/tickets/306/
but last update from upstream was in 2021. pbrobinson pinged it in
March with no response. Again not much we can do here unless we want to
do the porting work for upstream.

PyDrive was retired nine months ago:
https://src.fedoraproject.org/rpms/PyDrive/c/17c7af2339900db6584b15b89c0be88a2853cfd5?branch=rawhide
it probably needs to be obsoleted as part of
https://bugzilla.redhat.com/show_bug.cgi?id=2233409 . I've added a
comment there.

python-pathtools is FTBFS:
https://bugzilla.redhat.com/show_bug.cgi?id=2226277
because it uses the imp module that was removed in Python 3.12. This is
a pretty common case, we've fixed several packages for this, but not
this one yet. It's using it for something pretty stupid, so I've done
an equally stupid patch to fix it:
https://bodhi.fedoraproject.org/updates/FEDORA-2023-9235b56665

There's some discussion about async-generator here:
https://bugzilla.redhat.com/show_bug.cgi?id=2220121
seems it may get retired (which would mean we'd need to obsolete it).

matrix-synapse has a bunch of test suite failures, it looks like:
https://kojipkgs.fedoraproject.org//work/tasks/8799/104998799/build.log
upstream seem to be working on this:
https://github.com/matrix-org/synapse/pull/16099
so I guess the next version or so might succeed, or the maintainer
could bump to 1.91.0rc1 and backport that PR if they're really keen...

I *think* that's everything in your list.
-- 
Adam Williamson (he/him/his)
Fedora QA
Fedora Chat: @adamwill:fedora.im | Mastodon: @ad...@fosstodon.org
https://www.happyassassin.net



_______________________________________________
devel mailing list -- devel@lists.fedoraproject.org
To unsubscribe send an email to devel-le...@lists.fedoraproject.org
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@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Reply via email to