[libvirt PATCH] spec: libvirt-daemon: Add optional dependency on *-client

[Jim Fehlig]

On 11/7/22 04:32, Jiri Denemark wrote:
> On Fri, Nov 04, 2022 at 10:09:31 -0600, Jim Fehlig wrote:
>> On 11/4/22 09:22, Martin Kletzander wrote:
>>> On Thu, Nov 03, 2022 at 05:24:18PM +0100, Jiri Denemark wrote:
>>>> The libvirt-daemon subpackage contains libvirt-guests.sh script (used by
>>>> libvirt-guests service), which requires virsh to actually work. But
>>>> since dynamic libraries were separated from libvirt-client to
>>>> libvirt-libs more than 6 years ago, libvirt-daemon no longer requires
>>>> virsh to be installed. So unless libvirt-client is explicitly installed
>>>> (either manually or by installing the libvirt meta package),
>>>> libvirt-guests will not work.
>>>>
>>>> Just adding libvirt-client as a dependency of libvirt-daemon would go
>>>> against the original idea behind splitting libvirt-client: users may not
>>>> want to install or use any client binaries on the host where the daemon
>>>> runs (either they just use various language bindings or access the
>>>> daemon remotely). To solve this we could possibly turn libvirt-daemon
>>>> into an empty package and separate the daemons and libvirt-guests into
>>>> subpackages to make sure we support both use cases, but marking
>>>> libvirt-client as Recommended for libvirt-daemon does the same job in a
>>>> much simpler way.
>>>>
>>>
>>> Or you could just move the libvirt-guests files to libvirt-client
>>> package since they couldn't work without it anyway.
>>
>> This actually seems like a better approach, especially in the context of modular
>> daemons.
> 
> I think installing system services as part of libvirt-client is even
> stranger than requiring libvirt-client as a dependency of the daemon
> package.

Nod. Neither is a good solution.

> It would make more sense to split the daemon package, separate the
> monolithic daemon, proxy, and libvirt-guests so that one can only
> install the parts that are actually needed when using modular daemons,

I've taken an initial stab at that

https://listman.redhat.com/archives/libvir-list/2022-November/235924.html

> but I think it's quite an overkill for several reasons. First, the
> monolithic daemon is supposed to completely disappear at some point.

In the meantime they could coexist a bit nicer with something like the above.

> And the ancient libvirt-guests service should really be replaced by a
> solution implemented by the daemons themselves. And not only because the
> libvirt-guests service may cause a modular daemon to be started when a
> host is being shut down in case no domain is running.

Agreed, but IMO this is a separate problem to solve.

Regards,
Jim