Re: [Qemu-devel] [Qemu-block] [PATCH v3 07/38] block/quorum: Implement bdrv_is_inserted()

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1441 bytes --]

On 06/04/2015 06:37 AM, Alberto Garcia wrote:
> On Wed 03 Jun 2015 09:43:48 PM CEST, Max Reitz wrote:
>> bdrv_is_inserted() should be invoked recursively on the children of
>> quorum.
>>
>> Signed-off-by: Max Reitz <mreitz@redhat.com>
>> Reviewed-by: Eric Blake <eblake@redhat.com>
>> ---
> 
>> +static bool quorum_is_inserted(BlockDriverState *bs)
>> +{
>> +    BDRVQuorumState *s = bs->opaque;
>> +    int i;
>> +
>> +    for (i = 0; i < s->num_children; i++) {
>> +        if (!bdrv_is_inserted(s->bs[i])) {
>> +            return false;
>> +        }
>> +    }
>> +
>> +    return true;
>> +}
>> +
> 
> I wonder if it can actually happen that only some of the BDS are
> inserted :-?

Probably not possible while having a working quorum. But if I understand
the series correctly, there may be windows of time when building up or
hot-swapping a child within a quorum where things are not yet
consistent, but where the code can query the current state of the
quorum, so it's better to be prepared for those windows.  And while
unlikely, it is possible to build up a quorum that includes host cdrom
passthrough or other scenario where one child can independently fail to
be inserted.  Who knows - we may even take advantage of this in COLO
checkpointing where an NBD child is not yet inserted.

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


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]