Description of problem: 20140219 cloud builds fail to boot. Error reads: ---[ end Kernel panic - not syncing: No working init found. Try passing init= option to kernel. See Linux Documentation/init.txt for guidance. Version-Release number of selected component (if applicable): https://kojipkgs.fedoraproject.org//work/tasks/4826/8994826/Fedora-Cloud-Base-20150219-22.x86_64.qcow2 https://kojipkgs.fedoraproject.org//work/tasks/4827/8994827/Fedora-Cloud-Base-20150219-22.i386.qcow2 How reproducible: Always Steps to Reproduce: 1. Boot cloud image 2. 3. Actual results: Doesn't boot Expected results: Instance boots Additional info: Tested locally with testCloud as well as OpenStack
Proposed as a Blocker for 22-alpha by Fedora user roshi using the blocker tracking app because: This violates the "Release blocking images must boot" Alpha criterion: All release-blocking images must boot in their supported configurations.
The kernel panics if it can't find a userspace init process to start. This is normally an indication that either the initramfs didn't get loaded or the image compose is broken. This isn't a kernel problem.
Discussed at today's blocker review meeting [1]. AcceptedBlocker Alpha - This violates the "Release blocking images must boot" Alpha criterion: All release-blocking images must boot in their supported configurations. http://meetbot.fedoraproject.org/fedora-blocker-review/2015-02-23/
I can confirm this bug. I am trying with Fedora 22 Alpha TC8 image. It can be downloaded from http://dl.fedoraproject.org/pub/alt/stage/22_Alpha_TC8/Cloud-Images/x86_64/Images/Fedora-Cloud-Base-22_Alpha_TC8-20150302.x86_64.qcow2 The host has 3.18.6-200.fc21.x86_64 kernel.
Updated the host system with 3.18.7-200.fc21.x86_64 and after a reboot, the same TC8 image worked well. Strange!!!
Things seem to be resolved for the local boots with TC8 - but waiting to close this until we can determine what's causing images to not work on Openstack.
Images boot fine locally and in EC2. The openstack issue I was seeing seems to be caused by the host infra cloud runs. Our images are v3 qcow2 and qemu on el6 doesn't support v2 qcow images (which is what fed-cloud02 is running). Closing.