---》报错
SLF4J: Detected both log4j-over-slf4j.jar AND slf4j-log4j12.jar on the class path, preempting StackOverflowError.
SLF4J: See also http://www.slf4j.org/codes.html#log4jDelegationLoop for more details.
10059 [Thread-15-KafkaSpout] ERROR backtype.storm.util - Async loop died!
java.lang.NoClassDefFoundError: Could not initialize class org.apache.log4j.Log4jLoggerFactory
at org.apache.log4j.Logger.getLogger(Logger.java:39) ~[log4j-over-slf4j-1.6.6.jar:1.6.6]
at kafka.utils.Logging$class.logger(Logging.scala:24) ~[kafka_2.11-0.8.2.1.jar:na]
---》原因
log4j-over-slf4j.jar AND slf4j-log4j12.jar 循环调用冲突了，再进一步原因是kafka、hbase中用的是log4j。
---》解决方法
* 方案一：把storm中的log4j-over-slf4j 依赖排除；
org.apache.storm storm-core 0.9.5 provided org.slf4j log4j-over-slf4j org.apache.kafka kafka_2.11 0.8.2.1 org.apache.hbase hbase-client 1.0.1.1
* 方案二：把kafka和hbase中的slf4j-log4j12 依赖排除；
org.apache.storm storm-core 0.9.5 provided --> org.slf4j log4j-over-slf4j org.slf4j slf4j-api org.apache.kafka kafka_2.11 0.8.2.1 org.slf4j slf4j-log4j12 log4j log4j org.apache.hbase hbase-client 1.0.1.1 org.slf4j slf4j-log4j12 log4j log4j redis.clients jedis 2.7.0
* 采用方案二便于集群发布
* storm的日志时采用的 SLF4J和Logback日志框架，所以自己写的storm代码统一使用 slf4j 的包，可以避免大多数包冲突问题。
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
public class MyBolt {
private static final Logger LOG = LoggerFactory
.getLogger(MyBolt.class);
}
---》参考
http://www.tuicool.com/articles/aIzyqiy
http://stackoverflow.com/questions/21630860/storm-topology-not-submit
http://blog.csdn.net/gongmf/article/details/40379547