Pulled from the koji builds. Works for rootless and rootful containers.

Note that the previously built subpackage podman-quadlet has now been integrated into the podman binary package in this release. You will need to remove podman-quadlet manually if you have one.

@mhjacks doesn't sound quite right. I have added an obsoletes on the older podman-quadlet so it should get automatically removed. Did you actually need to remove the older podman-quadlet ? Might be a packaging error on my part in that case.

I definitely had to remove podman-quadlet manually. I'm looking at the specfile and I see the obsoletes line, and it appears to be correct.

Updating Subscription Management repositories. Transaction ID : 695 Begin time : Thu 09 Feb 2023 08:38:05 AM CST Begin rpmdb : 1e4cd75cbd9ba87adccd9919af8123226f00136441e8e4a2d4cf099eb7ca35b1 End time : Thu 09 Feb 2023 08:38:06 AM CST (1 seconds) End rpmdb : db15e28b7e950e8269ca18d663af7c8c5469298e649d82ee1773ab95bef9c0da User : Martin Jackson <mjackson> Return-Code : Success Releasever : 37 Command Line : erase podman-quadlet Comment : Packages Altered: Removed podman-quadlet-5:4.4.0-1.fc37.x86_64 @@System

(From the manual removal) vs:

Obsoletes: %{name}-quadlet <= 5:4.4.0-1

in the current specfile. So I'm not sure why I would have had to do that...

Seems the current spec still has a %package quadlet:

%package quadlet
Summary: Easily create systemd services using %{name}
Requires: %{name} = %{epoch}:%{version}-%{release}
Conflicts: quadlet

I don't know if that explains the behavior, but I think the conflicts belongs in the main package now, at least

Works great! LGTM! =)

@mhjacks ugh that was my bad. I'll check if the test failures are ok to be waived. If not, If I can't waive them yet, I'll do a new build while i'm at it.