On Sun, 2014-05-25 at 02:01 +0200, Nils Steinger wrote: > On Sat, May 24, 2014 at 05:59:18PM +0200, Bastian Blank wrote: > > On Tue, May 20, 2014 at 11:31:06PM +0200, debian-b...@voidptr.de wrote: > > Please get a name. > Sorry, I forgot to configure that properly. > > > > On-screen output during a failed boot attempt: > > > ------8<------ > > > Decompressing Linux... Parsing ELF... done. > > > Booting the kernel. > > > Loading, please wait...00:03.0 C900 PCI2.10 PnP PHH+3FFC9E10+3FFB9E10 C900 > > > modprobe: can’t load module btrfs (kernel/fs/btrfs/btrfs.ko): unknown > > > symbol in > > > module, or unknown parameter > > > > So you broke the module somehow. > I doubt _I_ broke anything, since I installed the pre-compiled package > from the official repository and rebooted without modifying anything > else. > > > Please use "dmesg" > dmesg's output is several hundred lines long, which isn't exactly easy > to read when busybox's 'more' seems to be an alias for 'cat'… > I couldn't find any errors in the last screenful of output, the kernel > has since panicked and the KVM's web-based control interface has locked > up, so I can't examine the error any further at the moment. I will get > back to you once that's fixed.
There should be a message in the kernel log telling you exactly what was not found in btrfs. You could run: 'dmesg | grep btrfs'. > I also migrated another KVM system from an XFS root filesystem to btrfs > and installed linux-image-3.14-0.bpo.1-amd64 — and it still boots. > The only difference to the two non-booting systems seems to be that the > working system has a separate (ext3-formatted) /boot partition. This should matter only to the boot loader. The initramfs does not mount the /boot partition. Ben. -- Ben Hutchings Teamwork is essential - it allows you to blame someone else.
signature.asc
Description: This is a digitally signed message part
