Mirko Kaempf created HDT-29:
-------------------------------
Summary: Dependency Management for different Clusters
Key: HDT-29
URL: https://issues.apache.org/jira/browse/HDT-29
Project: Hadoop Development Tools
Issue Type: Improvement
Components: Build and infrastructure
Reporter: Mirko Kaempf
A Hadoop Cluster should be represented by a copy of it's config files and all
libraries which are deployed to the cluster nodes within this cluster. Such a
setup is more or less static but not all the times. Therefore we can use
versioning (based on git, svn). That data should be imported if new files are
in the cluster (checks based on MD5Hash of files). Right now no EDIT
functionality just load as readonly and switch through different states.
The result will be an Eclipse-Project within the current developer workspace.
This type will be called "ClusterReferenceProject" and it is the data source
for the cluster connection Wizards and views we have in HDT. This way it is
much less work to connect to a cluster and to switch between different cluster
for testing new code.
Java Projects can reference such files directly even if no Maven is used.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
