hi jitu bhai,
thanks for nice topic, as we know potential shippable product must be
deployed after every sprint.
we keep tester and QA process with in the scrum team and sprint, so on
the last stand up meeting QA persons can speak out if anything (or any
bug) might mess up the whole system.
usually we keep separate release milestones, so our continuous or
after sprint deployments go to the test server.
we (in my new company tekSymmetry) are planning to give our client
access to the test server.
after few sprints when we are ready and confident enough to announce a
new version, we deploy them in production environment.
so you see our existing server are not getting unstable sprint outcome
rather they are getting when we are ready with stable release.
definitely if you want to deploy every outcome of the sprint to the
production server, i agree with you, that requires a lot of confidence
and verifications processes as well.
i'd suggest keep separate milestones for production environment (only
stable version will go to production) and let everyone see what's on
test environment.
something similar we have been following -
[sprint 1] [sprint 2] [sprint 3] [sprint 4] [sprint 5] [sprint 6]
[sprint 7] [sprint 8] [sprint 9] [sprint 10]
[ release 1 ] [ release
2 ] [ release 4 ]
thanks once again for nice topic :)_)
best wishes,