JedisClusterCommand
* improvements on connection error handling
** if based on slot connection throws connection related exception, retry to random node
** if we retry with random node, but all nodes are unreachable, throw JedisConnectionException without retry
** try to release connection whether connection is broken or not
* bug fix : if asking flag is on, and success this time, set asking flag to off
JedisClusterConnectionHandler
* have flexibility on initializing slots cache
** allow some nodes connection failure - skip
** if current node is success initializing slots cache, skip other nodes
** if current node failed to initialize slots cache, discard all discovered nodes and slots
* set nodes if node does not exist in nodes
** it restricts JedisPool to replace - prevent IllegalStateException : Returned object not currently part of this pool
JedisSlotBasedConnectionGuaranteedConnectionHandler
* getConnection (random connection)
** check all connections by random sequence
** always return valid connection (able to ping-pong)
** throw exception if all connections are invalid
* some refactoring
* Implement Closeable from Jedis, ShardedJedis with Pooled
** resources from JedisPool, JedisSentinelPool, ShardedJedis, ShardedJedisPool
* Connection class : check whether Jedis Connection is broken
** when it's time to throw JedisConnectionException, mark Connection to broken
first if Response's dependency found and not built
* there's some dependency with exec response and command responses
within multi
* if command responses's get() called before exec response's build(), it
calls exec response's build() first
* unit test included
cursor
* overload all *scan method to change "int" cursor to "string" cursor
* set @Deprecated and leave reason to deprecate and when to remove to
current *scan method
* modify unit tests to make it work with new *scan method
This allows a Jedis object to participate in try-with-resources when
using Java 7+. This change is fully backwards compatible to Java 6 and
previous releases of the Jedis client.
* Implements new sentinel commands (failover, monitor, remove, set)
* unit test included
** added 2 redis-server and 1 sentinel for failover test
* with some refactoring
** SentinelCommands : refactor to have interface
** HostAndPortUtil : same format to cluster setup
If Thread A calls a subscribe method on Jedis it will block on a socket read
call waiting for messages or subscription notifications. Thread B is now free
to call additional methods on JedisPubSub to change the current subscriptions
that thread A is waiting for. Essentially Thread A will do reads on the
socket and Thread B will do writes.
An issue occurs in that while Thread A is doing reads, in the
getObjectMultiBulkReply() method there is an implicit flush() call. This
means both Thread A and Thread B may do a write to the socket. Under this
situation if Thread A does a flush while Thread B is writing the internal
buffer will be corrupted. The fix is to make thread A never call flush().
This allows Thread A to be solely reads and Thread B to be solely writes.
Additionally since Thread B is sending commands, the internal pipeline count
is incremented and never decremented. So when Thread A terminates it's read
it resets the pipeline count.
- Add asking to cluster commands
- Make jedis cluster return connection to original pool
- Add tests for MOVED and ASK cluster responses
- Refactor connection handler to recalculate connections based on slots
This commit makes the first usable version of Jedis along with Redis Cluster
multi)
* BinaryClient / BinaryJedis : added feature to reset its state
(watched, multi)
* JedisPool / JedisSentinelPool : calls new feature (reset state) when
Jedis object returns to pool
* Unit Test included
