I've been working on keeping the roadmap in the project tracker up to date
with the current release schedule and the features targetting each
version. That can be found at
http://tracker.newdream.net/projects/ceph/roadmap
The main exception is that many bugs end up accumulating with the next
stable release as a target (e.g., v0.21.4), and then (when they're not
resolved) get pushed off to the next release. And as releases get further
into the future the picture gets a bit for hazy.
Those caveats aside, the basic schedule and goals are:
v0.22 (1-2 weeks away)
- Stable and usable clustered MDS in non-failure conditions. We will
encourage anybody to test the clustered MDS for correctness, performance,
and stability, and to report any and all issue.
- Background OSD scrubbing. This will help us catch remaining object
replication/recovery problems sooner rather than later.
- Stable RBD.
v0.23 (6 weeks away)
- Stable clustered MDS failure recovery.
- Stable directory fragmentation and defragmetnation. (Fragmentation is
what faciliates very large directories and the distribution of large
directories over multiple nodes in the MDS cluster.)
- RBD image locking.
- Stable radosgw (S3 compatible REST proxy)
v1.0 (Christmas!)
- Some basic fsck functionality
- Cautious use in production setting.
This is all, of course, subject to change depending on what issues come up
as we ramp up our testing and QA.
I'm also interested in hearing from you what features or issues you are
most interested in seeing us focus on. What are your use cases? If you
have done any testing, what problems have you run into? If you haven't,
what is preventing you from testing (besides some promise of stability or
production readiness)?
sage
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
