On 03.08.2016 13:21, Kevin Wolf wrote: > There is no reason why an NBD server couldn't be started for any node, > even if it's not on the top level. This converts nbd-server-add to > accept a node-name. > > Note that there is a semantic difference between using a BlockBackend > name and the node name of its root: In the former case, the NBD server > is closed on eject; in the latter case, the NBD server doesn't drop its > reference and keeps the image file open this way. > > Signed-off-by: Kevin Wolf <kw...@redhat.com> > --- > blockdev-nbd.c | 21 +++++++++------------ > qapi/block.json | 4 ++-- > 2 files changed, 11 insertions(+), 14 deletions(-)
Well, while I concede that it's possible to create R/W NBD servers on basically any node even today, I share Eric's concern that we shouldn't make it so simple. On the other hand, given COLO, I can't think of any simple solution to this. As far as I'm aware, COLO will make use of R/W NBD servers on backing nodes, so we can't just disallow creation of R/W servers on non-root nodes. So I guess your take on this is that it was already possible to create NBD servers wherever you so pleased and that we should just put this off until we have better op blockers? Max
signature.asc
Description: OpenPGP digital signature
