jstanek pinged me today with some weird errors :) so I got tempted to ask:
Is anyone else experiencing some Vagrant-related issues?
Most probable cause:
- there was a Change, in which `qemu://session` (user session) is used by default. That means root privilleges are not required anymore for vagrant+libvirt
- some network settings are currently limited (f.e. IP change)
- the location of the VM images, even the base one, was moved to home directory (`~/.local/share/libvirt/images/` instead of `/var/lib/libvirt/images`)
- flipping the option back to `qemu://system` disrupts halted machines (no images-moving magic is there)
If you wan't just the IP of the machine, you can simply `vagrant ssh-config`.
If you really need to use some custom networking setup (or need to access your existing box), you can change the libvirt provider's option `qemu_use_session`.
- To change it for one guest, you need to add to your Vagrantfile (where it fits):
config.vm.provider :libvirt do |libvirt|
libvirt.qemu_use_session = false
- To change it for all user's instances, you can add the setting above to `~/.vagrant.d/Vagrantfile` (or create the file). Then you can set it back to `true` on per-guest basis as needed.
- prior to flipping the setting, `destroy` the instance(s) it affects (`vagrant global-status`)
- to remove base image from libvirt storage, use always:
`sudo virsh vol-list --pool default`
to list images and
`sudo virsh vol-delete --pool default fedora-VAGRANTSLASH-29-cloud-base_vagrant_box_image_29.20181024.1.img`
or similar to remove them (without `sudo` for `qemu://session`).
You can read more on the change here:
If you encounter any issues, please ping me (pvalena on #fedora-devel) or create a bug (http://bugzilla.redhat.com/).
We (the Anaconda installer team) want to solve multiple problems by one
solution and we want
In short we are proposing to use custom repo files when configuring
Anaconda for image creation instead of adding even more complexity to
the kickstart repo command. For more details see the link below.
Purpose of this mail is to give everyone interested in the topic a
place to discuss our proposal to find out if we didn't miss something
or better case, if you agree with it :). It could be that this is not
applicable for Lorax / Pungi / live-cd tool in that case please write a
response to the document.
You can read details here:
Please, all the relevant comments should stay in the document if
possible. If that solution won't work because of higher traffic then we
will find a better solution but the point is to have everything on one
Have a great day everyone,
FYI - I'm building cmake 3.15.3 in rawhide now. This is an update from
3.14.5. cmake version updates have been known to trigger build issues
in the past, so I figured it was worth a heads up.
Manager of NWRA Technical Systems 720-772-5637
NWRA, Boulder/CoRA Office FAX: 303-415-9702
3380 Mitchell Lane orion(a)nwra.com
Boulder, CO 80301 https://www.nwra.com/
Do anyone know how to subscribe to Bugzilla bugs for a particular
I'm trying to watch glib-networking bugs. I'm already watching issues
and PRs at https://src.fedoraproject.org/rpms/glib-networking, but that
only watches for Pagure issues, not Bugzilla.
I'm already watching webkit2gtk3 package issues on Bugzilla and this
works exactly as I would expect, but I think I configured that long ago
with pkgdb back when pkgdb still existed.
This seems pretty basic, so it shouldn't be hard. Any ideas?