Fedora-Cloud-Atomic-Vagrant-23-20151119.x86_64.vagrant-libvirt.box
I have reproduced same error on my local computer with
sudo python3 -m unittest tunirtests.cloudtests -v sudo systemctl stop crond.service sudo systemctl disable crond.service sudo reboot
I failed to ssh back to system after rebooting.
Hey aalam,
I think I know the problem here. I believe we have a problem because the network device is not coming up as eth0 on atomic hosts. It actually comes up often with a random name.. so on a subsequent reboot you could get ens2,ens3 etc.. Which means you'll get a different IP from dhcp, which means vagrant loses track of your ip and you can no longer get in to the host.
We have /etc/udev/rules.d/80-net-setup-link.rules symlinked to /dev/null though so it's not clear why this is happening.
-bash-4.3# ls -l /etc/udev/rules.d/80-net-setup-link.rules lrwxrwxrwx. 1 root root 9 Oct 30 02:06 /etc/udev/rules.d/80-net-setup-link.rules -> /dev/null
This page explains that having that symlink in place should be sufficient http://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/ . I guess for now we could add the other disabler which is to have net.ifnames=0 on the kernel command line.
thanks. That explains that why got different IP every-time after reboot.
I sent a patch to the list about this:
https://lists.fedoraproject.org/archives/list/cloud%40lists.fedoraproject.org/message/N5LZN2LN2ILQJZQFTNFKMMVNJXWH7DFG/
Fixed in https://git.fedorahosted.org/cgit/spin-kickstarts.git/commit/?h=f23&id=2878aa006072ac7023765c0386cc90ae63389be4
Should be in the next 2 week atomic release.
Login to comment on this ticket.