[Qemu-devel] Possible disk corruption with virtIO?

[Prateek Sharma]

Hello all,
   I am writing to ask whether there is any possibility of silent data
corruption with virtIO/qemu.
My files inside the guest are getting zeroed out. Specifically, bytes
512 to 4096 on some files are being zeroed out. (some 60K files out of
 5Million are showing this so far).  The virtual disk is a file on
ext4 (size of file is 1TB), and i am using virtio, aio-threads=native,
and cache=none. The disk having the virtual disk is on mdadm
raid1---so i am guessing possibility of disk corruption is low

I am sure that there is no process inside the guest writing to the
files (verified with inotify). Is it possible that qemu/virtio can
cause this kind of carnage?
The files getting corrupted are 'random', and there is no relation
between file creation time and corruptio -- the corruption happens
silently at arbitrary time.

I am using ubuntu-10.04.3 with the default qemu on 2.6.32 , with the
guests having the same configuration. I have upgraded qemu(to 1.0) and
the kernel(3.2.2) now, and am trying to reproduce the problem.

Any advice/help is appreciated (I am running a production mail-server
on the virtual machine)

Thanks,
---Prateek