On 05/21/2013 10:03 PM, Chip Childers wrote:
On Tue, May 21, 2013 at 10:21:14AM -0400, Chip Childers wrote:
On Tue, May 21, 2013 at 09:19:05AM -0500, Joe Brockmeier wrote:
Hi all,
We apparently have a regression in 4.0.2 that will break new deployments
on Ceph RDB, and I was also CC'ed on another fix last week that could
apply to the 4.0.x branch.
The thought was that we would not have another release in the 4.0.x
branch because we're so close to 4.1.0, but right now it's unclear how
quickly 4.1.0 will be released.
So the question is: Should we go ahead and do a 4.0.3 release?
It's relatively easy for me to pull in the patches and roll up a
release. What I'm concerned about is distracting folks from the 4.1.0
process to take time to vote - if done correctly, the VOTE can be a
time-consuming process.
Thoughts?
Don't forget that we need to add the appropriate DB update bits, change
all the version numbers in code / config and docs and do the release
notes. It's a little bigger than just the cherry-picks, but not
insurmountable.
I didn't answer your question directly. I'm -0 on a 4.0.3 release, but
that's because I'd like to see us focus on getting 4.1.0 done (resolving
the issues you note in 4.0.2). I'm not a -1 on it though.
The longer I think about it, the more I agree. Since these issues have
been resolved in 4.1 and we are so close, we'd better focus on 4.1 indeed.
Since I think 4.0.3 would only be a release with just this RBD fix.
Wido
