I think some things might be good to write down here. * the reason for checking of /dev/sda1 and /dev/sda (and xvda1 and xvda) is that on EC2 these devices do not both exist. Ie, the root device is actually a block device (not a partition) but is named xvda1 or svda1 * the purpose of this code is to tell grub where to install (otherwise, on upgrade of grub, it would complain that the last time it ran it installed to a device that is now not present, and bother the user). * if this fails, the result is not failure to boot, but rather a prompt on grub upgrade.
First issue: * I think you're right that xvda and xvda1 should be added there. Second issue: * I think that this doesn't matter. grub will just check/insist that the devices it thinks should be installed to are there. More testing is needed here, the grub code is complex, and I'm weary of changing this because i haven't actually seen or reproduced an issue, and my mind is fuzzy. for reference: bug 623609 bug 752361 -- You received this bug notification because you are a member of Ubuntu Server Team, which is subscribed to cloud-init in Ubuntu. https://bugs.launchpad.net/bugs/961240 Title: cloud-init does not run grub on PV Xen and KVM has issues To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/961240/+subscriptions -- Ubuntu-server-bugs mailing list Ubuntu-server-bugs@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-server-bugs
