ConnectionWatchdog throws NullPointerException when connection drops (Redisson 2.5.0)

[johnou]

When restarting the Redis server all connection threads logged the following exception.

2016-10-18 12:45:59,686 ERROR [redisson-netty-1-1] (org.redisson.client.handler.CommandsQueue) Exception occured. Channel: [id: 0xa23d4302, L:0.0.0.0/0.0.0.0:42732 ! R:xxxxxxxxxxxxxx]
java.lang.NullPointerException: null
    at org.redisson.client.handler.ConnectionWatchdog.reconnect(ConnectionWatchdog.java:88) ~[redisson-2.5.0.jar:?]
    at org.redisson.client.handler.ConnectionWatchdog.channelInactive(ConnectionWatchdog.java:75) ~[redisson-2.5.0.jar:?]
    at io.netty.channel.AbstractChannelHandlerContext.invokeChannelInactive(AbstractChannelHandlerContext.java:251) [netty-transport-4.0.42.Final.jar:4.0.42.Final]
    at io.netty.channel.AbstractChannelHandlerContext.invokeChannelInactive(AbstractChannelHandlerContext.java:237) [netty-transport-4.0.42.Final.jar:4.0.42.Final]
    at io.netty.channel.AbstractChannelHandlerContext.fireChannelInactive(AbstractChannelHandlerContext.java:230) [netty-transport-4.0.42.Final.jar:4.0.42.Final]
    at io.netty.channel.DefaultChannelPipeline$HeadContext.channelInactive(DefaultChannelPipeline.java:1289) [netty-transport-4.0.42.Final.jar:4.0.42.Final]

After the connection "interruption" the application starts logging the following..

2016-10-18 13:10:34,676 ERROR [globalEventExecutor-2-1] (org.redisson.connection.ElasticacheConnectionManager) Command execution timeout for xxxxxxxxxxxx
org.redisson.client.RedisTimeoutException: Command execution timeout for xxxxxxxxxxx
    at org.redisson.client.RedisConnection.await(RedisConnection.java:126) ~[redisson-2.5.0.jar:?]
    at org.redisson.client.RedisConnection.sync(RedisConnection.java:158) ~[redisson-2.5.0.jar:?]
    at org.redisson.client.RedisConnection.sync(RedisConnection.java:144) ~[redisson-2.5.0.jar:?]
    at org.redisson.connection.ElasticacheConnectionManager$1.run(ElasticacheConnectionManager.java:134) [redisson-2.5.0.jar:?]
    at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:1.8.0_102]
    at io.netty.util.concurrent.ScheduledFutureTask.run(ScheduledFutureTask.java:126) [netty-common-4.0.42.Final.jar:4.0.42.Final]
    at io.netty.util.concurrent.GlobalEventExecutor$TaskRunner.run(GlobalEventExecutor.java:237) [netty-common-4.0.42.Final.jar:4.0.42.Final]
    at io.netty.util.concurrent.DefaultThreadFactory$DefaultRunnableDecorator.run(DefaultThreadFactory.java:144) [netty-common-4.0.42.Final.jar:4.0.42.Final]
    at java.lang.Thread.run(Thread.java:745) [?:1.8.0_102]

The host resolves to the same IP so that isn't a problem and I am able to telnet and issue the replication command without problems (immediate response)..

Escape character is '^]'.
INFO replication
$178
# Replication
role:master
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

[johnou]

org.redisson.connection.MasterSlaveConnectionManager#init is called after org.redisson.connection.MasterSlaveConnectionManager#createClient(java.lang.String, int, int, int) so a null timer is passed right down the chain into ConnectionWatchdog..

[johnou]

Might be enough just to move init(this.config); in org.redisson.connection.ElasticacheConnectionManager#ElasticacheConnectionManager before the iteration over node addresses.

[mrniko]

Please rebase your fix on master and send PR for RedisClient again.

[johnou]

@mrniko PTAL #673