Thanks, I'll check it out.

Il giorno mar 22 set 2020 alle ore 14:36 Gorka Eguileor <geguileo@redhat.com> ha scritto:
On 21/09, Ignazio Cassano wrote:
> Hello stackers,
> I am using cinder netapp nfs driver and I have vm performances problems
> during the backup of volumes.
> My backup tool make snapshots of volumes, then, on compute notes, it use
> virt-convert for creating a volume backup on another nfs backend using a
> different vlan.
> But during the virt-convert a lot of read are done on netapp and my virtual
> machines become very slow.
> I am asking if volume_copy_bps_limit can help me to mitigate.
>
> Please, anyone could help me ?
> Ignazio

Hi,

The volume_copy_bps_limit is a Cinder configuration parameter, and it is
used only when Cinder does the copying, which doesn't seem to be your
case.

In your case it looks like the backup tool is accessing directly the
compute hosts, so the throttling configuration should be done in the
backup tool.

If the backup tool doesn't support throttling, you can try using
cgroups...

Cheers,
Gorka.