[Openstack-devel] dropping essex from wheezy?

[gustavo panizzo <gfa>]

hello folks

    i would like to propose the elimination of essex from wheezy, our
current stable release.

- it's security maintenance is a pain, i've tried to backport some fixes
but a year, or more, in openstack is a lot. many fixes are not easily
back ported.

if we chose the "make our own patch" way, is really hard to test it.
you can create a test for the patch. but real world testing is almost
impossible (does anybody run an essex install?)

- at this point in openstack development, nobody would run essex.
seriously, not even in test environments, much less in production
environments

- there is no upgrade path from essex to folsom, upgrade btw releases
are supported (and painful) starting at folsom.

- popcon counts are low
  nova-common has 120 installs, keystone 30


i think we should backport havana (when it reaches testing) to
wheezy-backports, then we could start to offer a real openstack solution
on debian official repositories.

this is just my opinion, i haven't done most of the work. but will be
easier to help to package/maintain a current openstack release. i may be
terribly wrong :P

toughs?

-- 
1AE0 322E B8F7 4717 BDEA BF1D 44BB 1BA7 9F6C 6333