On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro S nchez wrote:
> Hello.
> I have a cluster with two nodes and I have to update the lvm2 package.
> Should I remove the node from the cluster before updating the package
> or that is not necesary?

> On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>> Hello.
>> I have a cluster with two nodes and I have to update the lvm2 package.
>> Should I remove the node from the cluster before updating the package
>> or that is not necesary?

> If there will be no "traffic" on the cluster at that moment, it's not
> needed to do anything. Otherwise, it's best to drain the queue before
> upgrading, but removing the node from the cluster is not needed.

> iustin

> 2012/9/25 Iustin Pop <ius...@google.com>:
>> On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>>> Hello.
>>> I have a cluster with two nodes and I have to update the lvm2 package.
>>> Should I remove the node from the cluster before updating the package
>>> or that is not necesary?

>> If there will be no "traffic" on the cluster at that moment, it's not
>> needed to do anything. Otherwise, it's best to drain the queue before
>> upgrading, but removing the node from the cluster is not needed.

>> iustin

> Hello, I have updated the package and everything went just fine.
> I did the update because I wanted to fix an error with the gnt-backup
> of some of the instances I had.
> The problem I have now is that I get a diferent error when I do this:
> gnt-backup export -n node instance

> I get the following errors with all the instances I try to backup.
> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
> xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
> xen-03.grupogaviglio.zp: Cannot find block device
> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
> not visible, size=2048m)>
> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
> xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
> xen-03.grupogaviglio.zp: Cannot find block device
> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
> not visible, size=4096m)>
> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
> Error while executing backend function: [Errno 2] No such file or
> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
> Failure: command execution error:
> Export failed, errors in export finalization, disk export: disk(s) 0, 1

> I think it is because i the logical volumens id had change.
> Any ideas on this?
> Thanks a lot.
> Mauro.

On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro S�nchez wrote:
> 2012/9/25 Iustin Pop <ius...@google.com>:
> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro S�nchez wrote:
> >> Hello.
> >> I have a cluster with two nodes and I have to update the lvm2 package.
> >> Should I remove the node from the cluster before updating the package
> >> or that is not necesary?

> > If there will be no "traffic" on the cluster at that moment, it's not
> > needed to do anything. Otherwise, it's best to drain the queue before
> > upgrading, but removing the node from the cluster is not needed.

> > iustin

> Hello, I have updated the package and everything went just fine.
> I did the update because I wanted to fix an error with the gnt-backup
> of some of the instances I had.
> The problem I have now is that I get a diferent error when I do this:
> gnt-backup export -n node instance

> I get the following errors with all the instances I try to backup.
> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
> xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
> xen-03.grupogaviglio.zp: Cannot find block device
> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
> not visible, size=2048m)>
> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
> xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
> xen-03.grupogaviglio.zp: Cannot find block device
> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
> not visible, size=4096m)>
> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
> Error while executing backend function: [Errno 2] No such file or
> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
> Failure: command execution error:
> Export failed, errors in export finalization, disk export: disk(s) 0, 1

> I think it is because i the logical volumens id had change.
> Any ideas on this?

> On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro Sánchez wrote:
>> 2012/9/25 Iustin Pop <ius...@google.com>:
>> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>> >> Hello.
>> >> I have a cluster with two nodes and I have to update the lvm2 package.
>> >> Should I remove the node from the cluster before updating the package
>> >> or that is not necesary?

>> > If there will be no "traffic" on the cluster at that moment, it's not
>> > needed to do anything. Otherwise, it's best to drain the queue before
>> > upgrading, but removing the node from the cluster is not needed.

>> > iustin

>> Hello, I have updated the package and everything went just fine.
>> I did the update because I wanted to fix an error with the gnt-backup
>> of some of the instances I had.
>> The problem I have now is that I get a diferent error when I do this:
>> gnt-backup export -n node instance

>> I get the following errors with all the instances I try to backup.
>> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
>> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
>> xen-03.grupogaviglio.zp
>> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
>> xen-03.grupogaviglio.zp: Cannot find block device
>> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
>> not visible, size=2048m)>
>> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
>> xen-03.grupogaviglio.zp
>> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
>> xen-03.grupogaviglio.zp: Cannot find block device
>> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
>> not visible, size=4096m)>
>> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
>> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
>> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
>> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
>> Error while executing backend function: [Errno 2] No such file or
>> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
>> Failure: command execution error:
>> Export failed, errors in export finalization, disk export: disk(s) 0, 1

>> I think it is because i the logical volumens id had change.
>> Any ideas on this?

> Huh, why did the ID change?

> iustin

On Wed, Oct 03, 2012 at 07:45:44PM -0300, Mauro S nchez wrote:
> 2012/10/3 Iustin Pop <ius...@google.com>:
> > On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro S nchez wrote:
> >> 2012/9/25 Iustin Pop <ius...@google.com>:
> >> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro S nchez wrote:
> >> >> Hello.
> >> >> I have a cluster with two nodes and I have to update the lvm2 package.
> >> >> Should I remove the node from the cluster before updating the package
> >> >> or that is not necesary?

> >> > If there will be no "traffic" on the cluster at that moment, it's not
> >> > needed to do anything. Otherwise, it's best to drain the queue before
> >> > upgrading, but removing the node from the cluster is not needed.

> >> > iustin

> >> Hello, I have updated the package and everything went just fine.
> >> I did the update because I wanted to fix an error with the gnt-backup
> >> of some of the instances I had.
> >> The problem I have now is that I get a diferent error when I do this:
> >> gnt-backup export -n node instance

> >> I get the following errors with all the instances I try to backup.
> >> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
> >> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
> >> xen-03.grupogaviglio.zp
> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
> >> xen-03.grupogaviglio.zp: Cannot find block device
> >> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
> >> not visible, size=2048m)>
> >> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
> >> xen-03.grupogaviglio.zp
> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
> >> xen-03.grupogaviglio.zp: Cannot find block device
> >> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
> >> not visible, size=4096m)>
> >> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
> >> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
> >> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
> >> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
> >> Error while executing backend function: [Errno 2] No such file or
> >> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
> >> Failure: command execution error:
> >> Export failed, errors in export finalization, disk export: disk(s) 0, 1

> >> I think it is because i the logical volumens id had change.
> >> Any ideas on this?

> > Huh, why did the ID change?

> > iustin

> Sorry, my mistake. The id didn't change. If I do a ls /dev/xenvg the
> volumes are there and they have the same name.
> I don't know why the gnt-backup and gnt-cluster verify complains about
> they don't exist.

> On Wed, Oct 03, 2012 at 07:45:44PM -0300, Mauro Sánchez wrote:
>> 2012/10/3 Iustin Pop <ius...@google.com>:
>> > On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro Sánchez wrote:
>> >> 2012/9/25 Iustin Pop <ius...@google.com>:
>> >> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>> >> >> Hello.
>> >> >> I have a cluster with two nodes and I have to update the lvm2 package.
>> >> >> Should I remove the node from the cluster before updating the package
>> >> >> or that is not necesary?

>> >> > If there will be no "traffic" on the cluster at that moment, it's not
>> >> > needed to do anything. Otherwise, it's best to drain the queue before
>> >> > upgrading, but removing the node from the cluster is not needed.

>> >> > iustin

>> >> Hello, I have updated the package and everything went just fine.
>> >> I did the update because I wanted to fix an error with the gnt-backup
>> >> of some of the instances I had.
>> >> The problem I have now is that I get a diferent error when I do this:
>> >> gnt-backup export -n node instance

>> >> I get the following errors with all the instances I try to backup.
>> >> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
>> >> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
>> >> xen-03.grupogaviglio.zp
>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
>> >> xen-03.grupogaviglio.zp: Cannot find block device
>> >> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
>> >> not visible, size=2048m)>
>> >> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
>> >> xen-03.grupogaviglio.zp
>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
>> >> xen-03.grupogaviglio.zp: Cannot find block device
>> >> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
>> >> not visible, size=4096m)>
>> >> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
>> >> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
>> >> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
>> >> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
>> >> Error while executing backend function: [Errno 2] No such file or
>> >> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
>> >> Failure: command execution error:
>> >> Export failed, errors in export finalization, disk export: disk(s) 0, 1

>> >> I think it is because i the logical volumens id had change.
>> >> Any ideas on this?

>> > Huh, why did the ID change?

>> > iustin

>> Sorry, my mistake. The id didn't change. If I do a ls /dev/xenvg the
>> volumes are there and they have the same name.
>> I don't know why the gnt-backup and gnt-cluster verify complains about
>> they don't exist.

> Are you by any chance running an old ganeti version which don't support
> newer LVM versions? Check the list arhives, there were a number of
> discussions; basically anything older than 2.6.0 _could_ have problems.

> regards,
> iustin

> 2012/10/3 Iustin Pop <ius...@google.com>:
>> On Wed, Oct 03, 2012 at 07:45:44PM -0300, Mauro Sánchez wrote:
>>> 2012/10/3 Iustin Pop <ius...@google.com>:
>>> > On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro Sánchez wrote:
>>> >> 2012/9/25 Iustin Pop <ius...@google.com>:
>>> >> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>>> >> >> Hello.
>>> >> >> I have a cluster with two nodes and I have to update the lvm2 package.
>>> >> >> Should I remove the node from the cluster before updating the package
>>> >> >> or that is not necesary?

>>> >> > If there will be no "traffic" on the cluster at that moment, it's not
>>> >> > needed to do anything. Otherwise, it's best to drain the queue before
>>> >> > upgrading, but removing the node from the cluster is not needed.

>>> >> > iustin

>>> >> Hello, I have updated the package and everything went just fine.
>>> >> I did the update because I wanted to fix an error with the gnt-backup
>>> >> of some of the instances I had.
>>> >> The problem I have now is that I get a diferent error when I do this:
>>> >> gnt-backup export -n node instance

>>> >> I get the following errors with all the instances I try to backup.
>>> >> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
>>> >> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
>>> >> xen-03.grupogaviglio.zp
>>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
>>> >> xen-03.grupogaviglio.zp: Cannot find block device
>>> >> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
>>> >> not visible, size=2048m)>
>>> >> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
>>> >> xen-03.grupogaviglio.zp
>>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
>>> >> xen-03.grupogaviglio.zp: Cannot find block device
>>> >> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
>>> >> not visible, size=4096m)>
>>> >> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
>>> >> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
>>> >> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
>>> >> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
>>> >> Error while executing backend function: [Errno 2] No such file or
>>> >> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
>>> >> Failure: command execution error:
>>> >> Export failed, errors in export finalization, disk export: disk(s) 0, 1

>>> >> I think it is because i the logical volumens id had change.
>>> >> Any ideas on this?

>>> > Huh, why did the ID change?

>>> > iustin

>>> Sorry, my mistake. The id didn't change. If I do a ls /dev/xenvg the
>>> volumes are there and they have the same name.
>>> I don't know why the gnt-backup and gnt-cluster verify complains about
>>> they don't exist.

>> Are you by any chance running an old ganeti version which don't support
>> newer LVM versions? Check the list arhives, there were a number of
>> discussions; basically anything older than 2.6.0 _could_ have problems.

>> regards,
>> iustin

> Ah, that's the problem. I am running Ganti 2.4.5
> I'm going to update it as soon as posible.
> Thanks a lot Iustin.
> Mauro.

On Wed, Oct 03, 2012 at 08:26:59PM -0300, Mauro S nchez wrote:
> 2012/10/3 Mauro S nchez <maur...@gmail.com>:
> > 2012/10/3 Iustin Pop <ius...@google.com>:
> >> On Wed, Oct 03, 2012 at 07:45:44PM -0300, Mauro S nchez wrote:
> >>> 2012/10/3 Iustin Pop <ius...@google.com>:
> >>> > On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro S nchez wrote:
> >>> >> 2012/9/25 Iustin Pop <ius...@google.com>:
> >>> >> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro S nchez wrote:
> >>> >> >> Hello.
> >>> >> >> I have a cluster with two nodes and I have to update the lvm2 package.
> >>> >> >> Should I remove the node from the cluster before updating the package
> >>> >> >> or that is not necesary?

> >>> >> > If there will be no "traffic" on the cluster at that moment, it's not
> >>> >> > needed to do anything. Otherwise, it's best to drain the queue before
> >>> >> > upgrading, but removing the node from the cluster is not needed.

> >>> >> > iustin

> >>> >> Hello, I have updated the package and everything went just fine.
> >>> >> I did the update because I wanted to fix an error with the gnt-backup
> >>> >> of some of the instances I had.
> >>> >> The problem I have now is that I get a diferent error when I do this:
> >>> >> gnt-backup export -n node instance

> >>> >> I get the following errors with all the instances I try to backup.
> >>> >> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
> >>> >> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
> >>> >> xen-03.grupogaviglio.zp
> >>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
> >>> >> xen-03.grupogaviglio.zp: Cannot find block device
> >>> >> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
> >>> >> not visible, size=2048m)>
> >>> >> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
> >>> >> xen-03.grupogaviglio.zp
> >>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
> >>> >> xen-03.grupogaviglio.zp: Cannot find block device
> >>> >> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
> >>> >> not visible, size=4096m)>
> >>> >> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
> >>> >> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
> >>> >> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
> >>> >> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
> >>> >> Error while executing backend function: [Errno 2] No such file or
> >>> >> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
> >>> >> Failure: command execution error:
> >>> >> Export failed, errors in export finalization, disk export: disk(s) 0, 1

> >>> >> I think it is because i the logical volumens id had change.
> >>> >> Any ideas on this?

> >>> > Huh, why did the ID change?

> >>> > iustin

> >>> Sorry, my mistake. The id didn't change. If I do a ls /dev/xenvg the
> >>> volumes are there and they have the same name.
> >>> I don't know why the gnt-backup and gnt-cluster verify complains about
> >>> they don't exist.

> >> Are you by any chance running an old ganeti version which don't support
> >> newer LVM versions? Check the list arhives, there were a number of
> >> discussions; basically anything older than 2.6.0 _could_ have problems.

> >> regards,
> >> iustin

> > Ah, that's the problem. I am running Ganti 2.4.5
> > I'm going to update it as soon as posible.
> > Thanks a lot Iustin.
> > Mauro.

> Iustin, do you have any idea when will Ganeti 2.6.0 be available in
> the Debian Testing (Wheezy) repository?

> On Wed, Oct 03, 2012 at 08:26:59PM -0300, Mauro Sánchez wrote:
>> 2012/10/3 Mauro Sánchez <maur...@gmail.com>:
>> > 2012/10/3 Iustin Pop <ius...@google.com>:
>> >> On Wed, Oct 03, 2012 at 07:45:44PM -0300, Mauro Sánchez wrote:
>> >>> 2012/10/3 Iustin Pop <ius...@google.com>:
>> >>> > On Wed, Oct 03, 2012 at 01:18:44PM -0300, Mauro Sánchez wrote:
>> >>> >> 2012/9/25 Iustin Pop <ius...@google.com>:
>> >>> >> > On Tue, Sep 25, 2012 at 09:00:00AM -0300, Mauro Sánchez wrote:
>> >>> >> >> Hello.
>> >>> >> >> I have a cluster with two nodes and I have to update the lvm2 package.
>> >>> >> >> Should I remove the node from the cluster before updating the package
>> >>> >> >> or that is not necesary?

>> >>> >> > If there will be no "traffic" on the cluster at that moment, it's not
>> >>> >> > needed to do anything. Otherwise, it's best to drain the queue before
>> >>> >> > upgrading, but removing the node from the cluster is not needed.

>> >>> >> > iustin

>> >>> >> Hello, I have updated the package and everything went just fine.
>> >>> >> I did the update because I wanted to fix an error with the gnt-backup
>> >>> >> of some of the instances I had.
>> >>> >> The problem I have now is that I get a diferent error when I do this:
>> >>> >> gnt-backup export -n node instance

>> >>> >> I get the following errors with all the instances I try to backup.
>> >>> >> Wed Oct  3 13:05:03 2012 Shutting down instance inst6b.grupogaviglio.zp
>> >>> >> Wed Oct  3 13:05:14 2012 Creating a snapshot of disk/0 on node
>> >>> >> xen-03.grupogaviglio.zp
>> >>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/0 on node
>> >>> >> xen-03.grupogaviglio.zp: Cannot find block device
>> >>> >> <LogicalVolume(/dev/xenvg/4ae6a2f8-7465-4a85-b29c-f59267c9ea85.disk0_data,
>> >>> >> not visible, size=2048m)>
>> >>> >> Wed Oct  3 13:05:19 2012 Creating a snapshot of disk/1 on node
>> >>> >> xen-03.grupogaviglio.zp
>> >>> >> Wed Oct  3 13:05:19 2012  - WARNING: Could not snapshot disk/1 on node
>> >>> >> xen-03.grupogaviglio.zp: Cannot find block device
>> >>> >> <LogicalVolume(/dev/xenvg/d2f0be0e-7a9c-49d3-96c3-71de705673df.disk1_data,
>> >>> >> not visible, size=4096m)>
>> >>> >> Wed Oct  3 13:05:19 2012 Starting instance inst6b.grupogaviglio.zp
>> >>> >> Wed Oct  3 13:05:24 2012 Finalizing export on xen-03.grupogaviglio.zp
>> >>> >> Wed Oct  3 13:05:25 2012  - WARNING: Could not finalize export for
>> >>> >> instance inst6b.grupogaviglio.zp on node xen-03.grupogaviglio.zp:
>> >>> >> Error while executing backend function: [Errno 2] No such file or
>> >>> >> directory: '/var/lib/ganeti/export/inst6b.grupogaviglio.zp.new/config.iniFsW6IV.new'
>> >>> >> Failure: command execution error:
>> >>> >> Export failed, errors in export finalization, disk export: disk(s) 0, 1

>> >>> >> I think it is because i the logical volumens id had change.
>> >>> >> Any ideas on this?

>> >>> > Huh, why did the ID change?

>> >>> > iustin

>> >>> Sorry, my mistake. The id didn't change. If I do a ls /dev/xenvg the
>> >>> volumes are there and they have the same name.
>> >>> I don't know why the gnt-backup and gnt-cluster verify complains about
>> >>> they don't exist.

>> >> Are you by any chance running an old ganeti version which don't support
>> >> newer LVM versions? Check the list arhives, there were a number of
>> >> discussions; basically anything older than 2.6.0 _could_ have problems.

>> >> regards,
>> >> iustin

>> > Ah, that's the problem. I am running Ganti 2.4.5
>> > I'm going to update it as soon as posible.
>> > Thanks a lot Iustin.
>> > Mauro.

>> Iustin, do you have any idea when will Ganeti 2.6.0 be available in
>> the Debian Testing (Wheezy) repository?

> It will not land in Wheezy, as 2.6 was not ready in time for the freeze.
> But note that the 2.5 version in Wheezy works correctly with newer LVM:
> I've added some patches that make it work correctly.

> iustin