----- Original Message -----
> From: "yiming xie" <plato...@gmail.com>
> To: ceph-users@lists.ceph.com
> Sent: Wednesday, September 21, 2016 3:53:35 AM
> Subject: [ceph-users] Help on RGW NFS function
> Hi,
> I have some question about rgw nfs.
> ceph release notes: You can now access radosgw buckets via NFS
> (experimental).
> In addition to the sentence, ceph documents does not do any explanation
> I don't understand the experimental implications.
> 1. RGW nfs functional integrity of it? If nfs function is not complete, which
> features missing?

NFSv4 only initially (but NFS3 support just added on master).  The I/O model is 
simplified.  Objects in RGW cannot be mutated in place, and the NFS client 
always overwrites.  Clients are currently expected to write sequentually from 
offset 0--on Linux, you should mount with -osync.

The RGW/S3 namespace is an emulation of a posix one using substring search, so 
we impose some limitations.  You cannot move directories, is one.  There are 
likely be published limits on bucket/object listing.

Some bugfixes are still in backport to Jewel.  That release supports NFSv4 and 
not NFS3.

> 2. How stable is the RGW nfs?

Some features are still being backported to Jewel.  I've submitted one 
important bugfix on master this week.  We are aiming for "general usability" in 
over the next 1-2 months (NFSv4).

> 3. RGW nfs latest version can be used in a production environment yet?

If you're conservative, it's probably not "ready."  Now would be a good time to 
experiment with the feature and see whether it is potentially useful to you.


> Please reply to my question as soon as possible. Very grateful, thank you!
> plato.xie
> _______________________________________________
> ceph-users mailing list
> ceph-users@lists.ceph.com
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Matt Benjamin
Red Hat, Inc.
315 West Huron Street, Suite 140A
Ann Arbor, Michigan 48103


tel.  734-707-0660
fax.  734-769-8938
cel.  734-216-5309
ceph-users mailing list

Reply via email to