Re: [Qemu-devel] [PATCH v3] block/vxhs: Add Veritas HyperScale VxHS block device support

[Daniel P. Berrange]

On Fri, Oct 28, 2016 at 02:30:47PM -0700, ashish mittal wrote:
> Hi Eric,
> 
> On Fri, Oct 28, 2016 at 7:44 AM, Eric Blake <address@hidden> wrote:
> > On 10/28/2016 02:44 AM, Ashish Mittal wrote:
> >> This patch adds support for a new block device type called "vxhs".
> >> Source code for the qnio library that this code loads can be downloaded 
> >> from:
> >> https://github.com/MittalAshish/libqnio.git
> >>
> >> Sample command line using the JSON syntax:
> >> ./qemu-system-x86_64 -name instance-00000008 -S -vnc 0.0.0.0:0 -k en-us
> >> -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5
> >> -msg timestamp=on
> >> 'json:{"driver":"vxhs","vdisk_id":"{c3e9095a-a5ee-4dce-afeb-2a59fb387410}",
> >> "server":{"host":"172.172.17.4","port":"9999"}}'
> >
> > Are the {} really part of vdisk_id?  That's not a usual portion of a
> > UUID.  And it should probably be spelled vdisk-id.
> >
> 
> The {} are not required for qemu/libqnio or the test server. It is
> just some artifact that our openstack components need. Then test
> server does not need {} at all. I will remove it from the sample
> command syntax in v4.
> 
> > It's a shame that port is a string and not an int, but it matches
> > existing practice, so that part is correct.
> >
> >>
> >> Sample command line using the URI syntax:
> >> qemu-img convert -f raw -O raw -n
> >> /var/lib/nova/instances/_base/0c5eacd5ebea5ed914b6a3e7b18f1ce734c386ad
> >> vxhs://192.168.0.1:9999/%7Bc6718f6b-0401-441d-a8c3-1f0064d75ee0%7D
> >>
> >
> > The URI syntax is potentially harder to use when compared to structured
> > usage (Kevin's work to add -blockdev, which depends on the QAPI definition).
> >
> 
> We decided to keep both after discussion with other reviewers. Do
> qemu-img and qemu-io binaries support json syntax similar to
> qemu-system-x86_64?
> 
> >> Signed-off-by: Ashish Mittal <address@hidden>
> >> ---
> >> v3 changelog:
> >> (1) Added QAPI schema for the VxHS driver.
> >>
> >
> > For this email, I'm focusing just on the QAPI.
> >
> >> +++ b/qapi/block-core.json
> >> @@ -1708,7 +1708,7 @@
> >>  ##
> >>  { 'enum': 'BlockdevDriver',
> >>    'data': [ 'archipelago', 'blkdebug', 'blkverify', 'bochs', 'cloop',
> >> -            'dmg', 'file', 'ftp', 'ftps', 'gluster', 'host_cdrom',
> >> +            'dmg', 'file', 'ftp', 'ftps', 'gluster', 'host_cdrom', 'vxhs',
> >>              'host_device', 'http', 'https', 'luks', 'null-aio', 'null-co',
> >
> > Missing mention of the new enum value in the comments above.
> >
> 
> Will fix in v4.
> 
> >>              'parallels', 'qcow', 'qcow2', 'qed', 'quorum', 'raw',
> >>           'replication', 'tftp', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
> >> @@ -2220,6 +2220,21 @@
> >>    'data': { 'filename': 'str' } }
> >>
> >>  ##
> >> +# @BlockdevOptionsVxHS
> >> +#
> >> +# Driver specific block device options for VxHS
> >> +#
> >> +# @vdisk_id:    UUID of VxHS volume
> >
> > Again, this should be vdisk-id.
> >
> Will fix in v4.
> 
> >> +#
> >> +# @server:      vxhs server IP, port
> >> +#
> >> +# Since: 2.7
> >
> > You missed 2.7; this should be 2.8 (if you make soft freeze) or even 2.9.
> >
> Will fix in v4.
> 
> >> +##
> >> +{ 'struct': 'BlockdevOptionsVxHS',
> >> +  'data': { 'vdisk_id': 'str',
> >> +            'server': 'InetSocketAddress' } }
> >
> > Is there any way to use a Unix socket, or is this server ONLY accessible
> > over IPv4/IPv6?
> >
> 
> Right now we support IPv4 only.

IMHO it should allow use of UNIX sockets, as its possible to have
SSH setup a tunnel to a IP server, and expose the endpoint via a
UNIX socket. So even if your reference server only supports IPv4,
users can conceivably connect with any sockets protocol.


Regards,
Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://entangle-photo.org       -o-    http://search.cpan.org/~danberr/ :|