Hi Ceph,

The Ceph stable releases now have a home page : 
http://tracker.ceph.com/projects/ceph-releases/. It is intended to be a 
convenient landing page for people working on backports (listed as 
"Developers") and for people interested in following the progress of a given 
release. It contains links to the pending issues being backported to the next 
stable release:
        
    dumpling pending issues 
http://tracker.ceph.com/projects/ceph/issues?query_id=71
    firefly pending issues 
http://tracker.ceph.com/projects/ceph/issues?query_id=75
    giant pending issues 
http://tracker.ceph.com/projects/ceph/issues?query_id=68

If an issue is not listed where it should, feel free to comment on it and ask 
that it is included.

There is one issue per release 
(http://tracker.ceph.com/projects/ceph-releases/issues) and the description is 
updated when progress is made.

For instance, the latest Firefly being prepared (v0.80.10) is at 
http://tracker.ceph.com/issues/11090. The Workflow part of the description ( 
http://tracker.ceph.com/issues/11090#Workflow ) list the chronological steps 
and lines marked with IN PROGRESS show at which stage it is. The inventory ( 
http://tracker.ceph.com/issues/11090#Inventory ) groups all pull requests, 
issues, commits collected from the tracker, the git repository and git hub pull 
requests ( as cross referenced by 
http://workbench.dachary.org/dachary/ceph-workbench ). The inventory also lists 
the GO / NO GO decision of each Ceph component lead ( as listed at 
http://ceph.com/docs/master/dev/development-workflow/#resolving-bug-reports-and-implementing-features
 ). 

The bulk of the description is the analysis of teuthology integration tests, 
grouped by project. For instance the RADOS tests 
(http://tracker.ceph.com/issues/11090#rados) show two runs and no regression. 
The second run was necessary because external factors interfered with the tests 
(machines broke, DNS problems etc.). The RBD run was good the first time ( 
http://tracker.ceph.com/issues/11090#rbd ). And the RGW run revealed that a 
test had to be adapted ( http://tracker.ceph.com/issues/11090#rgw ). In a 
nutshell, the more lines there are for a given run, the more troubles had to be 
solved.

I'd be happy to answer any question you may have and hopefully clarify how 
backports are being done and what to expect for the next point release.

Cheers

-- 
Loïc Dachary, Artisan Logiciel Libre

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to