1.9.3
-----
o Fix for a race between Slave shutting down a channel and Master reaping it, which led to exceptions on the Master side

1.9.1 (2013-06-24)
------------------
o Client now uses heap buffers, reduces GC requirements and out of memory errors
o Master will now not timeout long running write operations, like store copying
o Redesigned ResourcePool, now it properly pools and reuses resources, should be more
  easy going on the garbage collector

1.9.RC2 (2013-04-30)
--------------------
o Upgrades to Netty 3.6.3

1.9.M05 (2013-03-05)
--------------------
o Shutting down Server now takes proper care of executors and works with multiple versions of netty

1.6 (2012-01-20)
---------------
o New HA request type allows for copy of transaction interval from master to slave, to recover from missing logical log files.

1.6.M02 (2011-12-16)
--------------------
o Server stops streaming to a client if the client should crash or disconnect meanwhile.
o Individually configurable read timeout for awaiting lock acquisition with ha.lock_read_timeout.
  Server "silent channel checker" uses that lock timeout threshold also.

1.5 (2011-10-18)
----------------
o Added com protocol version (one internal version and one application version) which must match in order
  for instances to communicate.
o Added a check to prevent "multiple roots" in ZooKeeper problem from occuring.

1.5.M02 (2011-10-10)
--------------------
o Server limits the number of unread chunks to send to the client, preventing unecessary memory hogging.
o Server-side exceptions are serialized to the client and thrown there instead of slave getting read timeouts.
o Changed configuration names to reduce confusion: 'zoo keeper' -> 'coordinator', 'machine' -> 'server'.
o Successful slave transactions are seen as committed, instead of rolled back.
o Slaves updates coordinator asynchronously (configurable to be synchronous) to avoid bottleneck.

1.5.M01 (2011-08-31)
--------------------
o Frame length configurable per communication interface.
o Fixed some issues regarding monitoring of open channels and what resources they map to and occupy.

1.4.M03 (2011-05-26)
--------------------
o Changed to always dispose channels after exception.
o Modified slave context to include a session id.

1.3.M05 (2011-03-24)
-------------------
o Fixed a bug that could occur when copying the store.

1.3.M04 (2011-03-10)
--------------------
o Backup disabled when run on Windows due to limitations in OS/filesystem.

1.3.M03 (2011-02-24)
--------------------
o Patched to work with new backup tool.
o Added a check client-side so that server store ID matches client store ID.

1.3.M02 (2011-02-10)
--------------------
o Initial release.

