* A memory leak in the data path between DataNodes and the cluster could have
  caused the HDFS process to unexpectedly retart. If this occurred, HDFS clients
  connected to an affected node were disconnected.

* Adding support for appendToFile to return the correct response depending on 
the client version

* Under some circumstances, the HDFS process on a node could have run out of 
  system memory while DataNodes were transferring data to the HDFS server. If 
  this occurred, the HDFS process on the affected node restarted, and HDFS 
  clients connected to that node were disconnected.

* Adding ambari server support for rolling or manual upgrades

* Add (and use) mutex callbacks so that sasl will run multithreaded.

* Previously, adding OneFS support for any Ambari release required a code change.
Customers will not have been able to deploy Ambari 2.1.2 or any release
of Ambari that we did not explicitly support.
After this patch, OneFS will explicitly support second digit (minor) versions
of Ambari, and be compatible with any dot release (third digit) version.
For example, declaring support for Ambari 2.1 means OneFS will support
2.1.0, 2.1.1, 2.1.2, and any future bugfix release in the 2.1 line.
