Chris Csanady
2006-Jun-26 11:26 UTC
[zfs-discuss] Bandwidth disparity between NFS and ZFS (Solved)
I don''t know how I missed it, but there are periodic commit requests by the NFS client. These occur often enough that the data ends up being written twice as you have suggested. In any case, this is really annoying, as dd certainly isn''t requesting this behavior. Perhaps the clients are just stingy with their memory, but for whatever reason, it is unfortunate for ZFS on the server. Chris On 6/24/06, Chris Csanady <cc at 137.org> wrote:> On 6/24/06, Neil Perrin <Neil.Perrin at sun.com> wrote: > > > > The data will be written twice on ZFS using NFS. This is because NFS > > on closing the file internally uses fsync to cause the writes to be > > committed. This causes the ZIL to immediately write the data to the intent log. > > Later the data is also written committed as part of the pools transaction group > > commit, at which point the intent block blocks are freed. > > In this case though, the file is left open, so there should be no synchronous > I/O. (tcpdump -vv confirms that all writes are marked as unstable, and there > are no commits.) Perhaps the NFS server is issuing the I/O synchronously > when it should not be, thus causing the double writes?