Re: [Xen-users] XENBUS: Timeout connecting to device: device/vbd/51712 (local state 3, remote state 1)

[Stephen Kent <smkent@xxxxxxxxxx>]

It turns out /dev/xvda is the wrong type of disk specifier, just using "xvda" is correct instead.

However, I'm still having issues attaching my virtual disks. If I manually run losetup /dev/loop1 /home/data/xen/megatron/root.img, and then attach phy:/dev/loop1 in my dom0, /dev/xvda appears and I can mount it. But, if I try to boot my domU with phy:/dev/loop1 instead of the file path, it just enters a reboot loop. On my dom0, when I try to xl block-attach the file directly (without a loop device), the command appears to succeed and xl block-list shows the device, but /dev/xvda is not created and nothing about xvda shows up in dmesg.

It seems like the file backend driver in my dom0 is broken. Does this make sense to anyone?

On Mon, Feb 4, 2013 at 7:34 PM, Stephen Kent <smkent@xxxxxxxxxx> wrote:

I can't seem to attach the block device in domain 0 either:

# xl block-attach 0 file:/home/data/xen/megatron/root.img /dev/xvda

libxl: error: libxl.c:1752:libxl__device_from_disk: Invalid or unsupported virtual disk identifier /dev/xvda

libxl: error: libxl.c:1844:device_disk_add: Invalid or unsupported virtual disk identifier /dev/xvda

libxl: error: libxl.c:1720:device_addrm_aocomplete: unable to add device

libxl_device_disk_add failed.

I still feel like I'm missing a kernel config option or something, but I have no idea where to look. The Xen block backend and frontend drivers are both included, seems strange that I would be missing support for a simple loopback file.

On Mon, Feb 4, 2013 at 5:37 PM, Stephen Kent <smkent@xxxxxxxxxx> wrote:

I checked, PCI is enabled in both my dom0 and domU kernels. I've enabled all the Xen drivers I can find (including the block device frontend and backend drivers).

On Mon, Feb 4, 2013 at 11:45 AM, Rushikesh Jadhav <2rushikeshj@xxxxxxxxx> wrote:

On Tue, Feb 5, 2013 at 12:00 AM, Stephen Kent <smkent@xxxxxxxxxx> wrote:

I just upgraded my computer from an old kernel (2.6.34-xen) to 3.6.11, and upgraded Xen to 4.2.0. I can't seem to get my domU to boot, as it can't find the root filesystem.

I've compiled a domU kernel (also 3.6.11) with xen-blkfront built-in and set the domU config file with this disk configuration:

disk = [ 'file:/home/data/xen/megatron/root.img,xvda,w' ]

(The disk is just a file with an ext3 filesystem, I can mount it on loopback in dom0)

After I xl start my domU, xl block-list in dom0 shows the device connected to the domU but the domU kernel (with root=/dev/xvda) can't find the root filesystem:

[...]

XENBUS: Timeout connecting to device: device/vbd/51712 (local state 3, remote state 1)

console [netcon0] enabled

netconsole: network logging started

ALSA device list:

 No soundcards found.

md: Waiting for all devices to be available before autodetect

md: If you don't use raid, use raid=noautodetect

md: Autodetecting RAID arrays.

md: Scanned 0 and added 0 devices.

md: autorun ...

md: ... autorun DONE.

VFS: Cannot open root device "xvda" or unknown-block(0,0): error -6

Please append a correct "root=" boot option; here are the available partitions:

Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)

Pid: 1, comm: swapper/0 Not tainted 3.6.11-gentoo #8

Call Trace:

Â[<ffffffff815c59ff>] ? panic+0xc5/0x1cc

Â[<ffffffff815c5bd9>] ? printk+0x4c/0x51

Â[<ffffffff819990cd>] ? mount_block_root+0x1d4/0x1f2

Â[<ffffffff8108ebaa>] ? lg_local_lock+0x11/0x14

Â[<ffffffff81002930>] ? kvm_io_bus_sort_cmp+0x1c/0x27

Â[<ffffffff8110fcf8>] ? sys_mknodat+0x12c/0x140

Â[<ffffffff8199920f>] ? mount_root+0x124/0x12b

Â[<ffffffff81999d8f>] ? initrd_load+0x303/0x309

Â[<ffffffff8199937e>] ? prepare_namespace+0x168/0x19e

Â[<ffffffff81998702>] ? kernel_init+0x17f/0x191

Â[<ffffffff8199852b>] ? parse_early_options+0x2a/0x2a

Â[<ffffffff815cf644>] ? kernel_thread_helper+0x4/0x10

Â[<ffffffff815c8338>] ? retint_restore_args+0x5/0x6

Â[<ffffffff815cf640>] ? gs_change+0x13/0x13

Check if you have PCI enabled in domU kernel.
Select the PCI options while you do # make menuconfig

Â

I don't see any earlier kernel messages about the Xen block frontend driver loading. Am I missing a kernel option or something? I'm not sure what else to try.

Thanks!

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users