Re: [PATCH v2 2/2] storage: btrfs subvolumes for directory pools

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

 



On 10/10/2013 09:57 AM, Daniel P. Berrange wrote:

>> +                }
>> +                if (volv->backingStore.path == NULL) {
>> +                    /* backing store not in the pool, ignore it */
> 
> Backing stores for volumes are not required to be in the same pool as
> the source. For example it is valid to have a qcow2 file backed by
> a lvm volume.

What's more, qemu allows one to have qcow2 data atop a network device;
but right now libvirt forcefully assumes that all network devices (nbd,
ceph, sheepdog, gluster) contain only raw data rather than other
formats.  I'm trying to investigate what implications this has, but
among other things, I think we need to use more polymorphic callbacks
where for a given backing store string, we determine which storage
driver to use, then that storage driver tells us how to manage data from
that storage (including having qcow2 data atop network storage).

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature

--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list

[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]