Closed mbialoru closed 2 years ago
Maybe we could drop the audio device altogether?
I never really ended up using it, and choosing any backend is going to be a rabbit-hole, as that is normally distribution/desktop-environment dependent.
Also, I'm still on Ubuntu 20.04 / QEMU 4.2, where -sounddev
is still supported.
Yeah, I've been using this repository for creating boxes for disposable windows development VMs on libvirt provider. My first solution was also to get rid of the sound device, it simply isn't a priority in my use case.
However when box builds first failed I couldn't find a good hit on google, so after a little bit of troubleshooting I found QEMU documentation mentioning removal of sound device parameter in this form and figured out how to get it going again.
Maybe when someone comes across the same problem later this post here will be indexed by google and provide some explanation with possible solution.
Great repository by the way, helped me out a lot! 😄
I've removed the audio device in https://github.com/rgl/windows-vagrant/commit/4d6d382e2e619e718bb980473a9d5563aee4fe7d.
So I'm going to close this PR and the related issue.
Updated packer files with new qemu parameter for audio device. Besides of
none
, alternatively instead of pulseaudio (pa), open sound system (oss) is available. Build now finishing successfully on qemu 7.1.0-5 (arch linux). Old version with-soundhw
was removed, more here: https://www.qemu.org/docs/master/about/removed-features.html?highlight=soundhw