[ovirt-users] VM/Template copy issue

[Maor Lipchuk]

On Sat, Jun 3, 2017 at 6:20 AM, Bryan Sockel <Bryan.Sockel at altn.com> wrote:
> This happening to a number  of vm's. All vm's are running and can be stopped
> and re started.  We can read and write data within the vm.
>
> All vms are currently running on a single node gluster file system.  I am
> attempting to migrate to a replica 3 gluster file system when i exprience
> these issues.  The problem always seems to happen when finalizing the move
> or copy.
>
> If it makes a difference the gluster storage we are coping to and from are
> dedicated storage servers.
>
>
> -------- Original message --------
> From: Maor Lipchuk <mlipchuk at redhat.com>
> Date: 6/2/17 5:29 PM (GMT-06:00)
> To: Bryan Sockel <Bryan.Sockel at altn.com>
> Cc: users at ovirt.org
> Subject: Re: [ovirt-users] VM/Template copy issue
>
> ________________________________
> From : Maor Lipchuk [mlipchuk at redhat.com]
> To : Bryan Sockel [Bryan.Sockel at altn.com]
> Cc : users at ovirt.org [users at ovirt.org]
> Date : Friday, June 2 2017 17:27:32
>
> Hi Bryan,
>
> It seems like there was an error from qemu-img while reading sector
> 143654878 .
>  the Image copy (conversion) failed with low level qemu-img failure:
>
> CopyImageError: low level Image copy failed:
> ("cmd=['/usr/bin/taskset', '--cpu-list', '0-31', '/usr/bin/nice',
> '-n', '19', '/usr/bin/ionice', '-c', '3', '/usr/bin/qemu-img',
> 'convert', '-p', '-t', 'none', '-T', 'none', '-f', 'raw',
> u'/rhev/data-center/d776b537-16f2-4543-bd96-9b4cba69e247/e371d380-7194-4950-b901-5f2aed5dfb35/images/9959c6e4-1fb7-455b-ad5e-8b9e2324a0ab/3f4e183b-7957-4bee-9153-9d967491f882',
> '-O', 'raw',
> u'/rhev/data-center/mnt/glusterSD/vs-host-colo-1-gluster.altn.int:_desktop-vdi1/f927ceb8-91d2-41bd-ba42-dc795395b6d0/images/9959c6e4-1fb7-455b-ad5e-8b9e2324a0ab/3f4e183b-7957-4bee-9153-9d967491f882'],
> ecode=1, stdout=, stderr=qemu-img: error while reading sector
> 143654878: No data available\n, message=None",)
>
> Can you verify those disks are indeed valid? Can you IO to them while
> attaching them to a running VM?
>
> On Tue, May 30, 2017 at 9:10 PM, Bryan Sockel  wrote:
>>
>> Hi,
>>
>> I am trying to rebuild my ovirt environment after having to juggle some
>> hardware around.  I am moving from hosted engine environment into a engine
>> install on a dedicated server.  I have two data centers in my setup and
>> each
>> DC has a non-routable vlan dedicated to storage.
>>
>> As i rebuild my setup i am trying to clean up my storage configuration.  I
>> am attempting to copy vm's and templates to a dedicated gluster setup.
>> However  each time i attempt to copy a template or move a vm, the
>> operation
>> fails.  The failure always happens when it is finalizing the copy.
>>
>> The operation does not happen with all vm's, but seems to happen mostly
>> with
>> vm's created from with in Ovirt, and not imported from vmware.
>>
>>
>> I have attached the logs from where i was trying to copy to templates to a
>> new Gluster Filesystem
>>
>> Thanks
>> Bryan
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>


I'm not sure if this is related to the hardware that was juggled, but
it seems like the volume has a bad sector and qemu-img reports it.
Do you have this volume in another storage domain maybe before the
hardware change, so we can eliminate the issue of hardware change.
You can also open a bug so it will be easier to track and investigate
it:  https://bugzilla.redhat.com/enter_bug.cgi?product=vdsm
please also attach the engine and vdsm logs.

Regards,
Maor