The Ambari RPM location is now an ambari configuration property in
the blueprint. This will support debug/issue resolution scenarios
as well as the export of this information.
Fixes: bug #1208966
* Config objects returned contains default values, types and descriptions
* Introduced "importance" field, it help differentiate all configs in UI
Implements blueprint vanilla-plugin-configs-improvements
* We only support default configs now
* Username and Password for Swift access should be pointed during hadoop-job starting
* Tenant and Auth-Url is get from context
Implements blueprint savanna-swift-integration
* pluggable provisioning mechanism
* a lot of old code has been removed
* new version of REST API (1.0)
* image registry draft implemented as novaclient extension
* oslo updated
* using oslo.db instead of flask-sqlalchemy
* some hacking fixes
* using alembic for db migrations
Partially implements blueprint pluggable-cluster-provisioning.
Partially implements blueprint savanna-rest-api-1-0.
Implements blueprint hadoop-image-registry.
Implements blueprint fulfill-openstack-requirements.
Implements blueprint db-migrate-support.
* Xml config provisioning is done for core, hdfs, mapred xmls
* Removed xslt transformations from data nodes
* At now Savanna contains default configs for Hadoop in resources/
* This version of default configs is applied for Hadoop 1.1.1
* No needs to distribute new cloud image
Fixes: bug #1169900
In 02f59a0768a56c084fa72ac9c1e47f1a819baadc we removed:
from setup.py file and it broke tarball generation in that some files
are now missing.
Fixes: bug #1169295
* setup.py(setuptools.setup): Remove "Intended Audience :: BigDate"
since it's not in pypi's list of valid trove classifiers and
prevents successful upload of the package when present.