[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] resizing filesystem in a VM

  • To: Xen User-List <xen-users@xxxxxxxxxxxxxxxxxxx>
  • From: "Fajar A. Nugraha" <list@xxxxxxxxx>
  • Date: Thu, 11 Aug 2011 16:53:48 +0700
  • Delivery-date: Thu, 11 Aug 2011 02:55:00 -0700
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

On Thu, Aug 11, 2011 at 4:40 PM, Alexandre Chapellon
<a.chapellon@xxxxxxxxx> wrote:
> Hello,
> I am using Xen 4.0.1 and Xen 3.2.1 dom0 with Debian Lenny and Squeeze dom0s.
> I have a bunch of virtual machines running , all using LVM as backend storage.
> Mostly, an LVM logical volume is mapped to a partition in the VM (eg: xvda1)
> I have tried to hot resize the filesystem in the VM without umounting it but 
> it just do not work... Here is what I did:
>  - lvextend on the dom0,
>  - resize2fs on the VM
> Doing so, the kernel of the virtual machine seems not to see the space 
> available for the partition has changed and resize2fs returns: "xvda already 
> has xxxxxxxxx blocks, Nothing to do"
> This last until I hot remove the disk from the VM using xm block-detach and 
> re-attach it using xm block attach.
> After re-attaching  the disk I can resize2fs as expected.

That is expected behaviour.

> Is there anything I can do to avoid detaching the block device from the VM to 
> let the kernel know the partition has a new size?
> What are the bests practices in Xen to achieve hot resize of filesystem in 
> the VM?

Newer version of xen and domU kernel allows domU to detect block
device change without having to detach/reboot. Try using domU kernel
form Jeremy's git tree.


Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.