Am Freitag 10 September 2010, 11:53:48 schrieben Sie: > Hans Meine <meine@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote: > > all. At the server side, there is simply a hierarchy of ZFS partitions > > (because a per-user quota has to be mapped to per-partition quotas, so > > every user got a separate ZFS partition) mounted below the NFS-exported > > directory. > > That sounds like a fairly straightforward FSID transition. NFS creates a > new superblock for that, and should propagate the FSC option across. OK, that sounds fine. > What kernel are you using on the client? Citing my original post: I am using Gentoo's 2.6.31-gentoo-r6-default linux sources, with everything fscache-related (except the _DEBUG options) enabled. I can also try a vanilla kernel if you prefer, but I would not have installed the gentoo version if they deviated much from the original. (See http://dev.gentoo.org/~mpagano/genpatches/about.htm - they really don't add drivers or features except possibly for their live CDs.) Anyhow, 2.6.31 is not too new anymore, so updating sounds like a good option. Is the FSC option propagation independent of nfs-utils? Greetings, Hans -- Linux-cachefs mailing list Linux-cachefs@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cachefs
