On Friday 10 July 2015 03:22 AM, John Gardeniers wrote:
Vijay,
Instead of thinking about the next version, shouldn't you be trying to
get at least one truly stable version? 3.5 still has problems and the
only things I've been reading about 3.6 and 3.7 are problems, so they
are far from stable. This idea of releasing a new version instead of
fixing the current one is merely driving people away from Gluster.
Our intention is to do both - have stability releases happening in the
3.5.x, 3.6.x and 3.7.x trains while we develop the next Gluster releases.
If we look at the previous 6-12 month period, there has been at least
one minor release per month as we released 3.6 and 3.7. If the minor
releases are not being very effective in addressing stability, we would
love to have some inputs on why that's not happening. Something like:
a> Minor releases do not have the content/bug fixes that you are looking for
b> Minor releases should happen even more frequently
c> Fix for my issue is present in master but has not been backported to
a particular release train
and more along these lines would surely help us in doing a course
correction.
OTOH, we are *extremely* serious about stability and there is an ever
increasing focus on testing & feedback from testing. Some of the recent
actions in this space include:
- Weekly IRC bug triage meetings for prioritizing bugs
- Addition of dedicated maintainers for tests in gluster
- More pre-commit & post-commit tests being integrated with Jenkins
Thanks for bringing this up. Feedback of this nature with more details
will help us determine the content for 3.8. If there are issues that you
would like to be addressed, please let us know. We can either fix them
in the appropriate release trains if the nature of changes is not
significant or if it happens to be so, we can address the issues as
enhancements in 3.8.
Regards,
Vijay
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users
