On 13 February 2018 at 02:01, Terry Barnaby <ter...@beam.ltd.uk> wrote:
>> Yes: http://nfs.sourceforge.net/#faq_a8
> Quite right, it was network limited (disk vs network speed is about the
> same). Using a slower USB stick disk shows that fsync() is not working with
> a NFSv4 "async" export.
> But why is this ? It just doesn't make sense to me that fsync() should work
> this way even with an NFS "async" export ? Why shouldn't it do the right
> thing "synchronize a file's in-core state with storage device" (I don't
> consider an NFS server a storage device only the non volatile devices it
> uses). It seems it would be easy to flush the clients write buffer to the
> NFS server (as it does now) and then perform the fsync() on the server for
> the file in question. What am I missing ?
You seem to be missing the part where several people have told you
that the async option in the server is misnamed. NFS server async()
was named that ~20 years ago (?) to match requirements from sites that
wanted NFSv2 'look and feel' with NFSv3 and has been constantly called
that since because changing it would break people's setups.
What you are wanting may be useful as a renamed and different feature,
but it really needs to be done on the NFS kernel mailing list versus
here. While you have 2 NFS oriented kernel developers here, they are
only a subset of the people who would need to look at it and see if it
could be done.
Stephen J Smoogen.
devel mailing list -- firstname.lastname@example.org
To unsubscribe send an email to devel-le...@lists.fedoraproject.org