Description
Right now each role can define event handlers for bootstrap and configure. I suggest that the Whirr core should be aware of more service lifecycle events like: start, stop & cleanup. This semantic improvement should allow us to add more features like: cluster reconfiguration (e.g. update Hadoop settings and restart all the nodes - rolling restart), add / remove nodes etc.
Attachments
Attachments
Issue Links
- relates to
-
WHIRR-294 Separate Hadoop install and configuration scripts into more generic functions
- Open