diff --git a/docs/content/ingestion/stream-pull.md b/docs/content/ingestion/stream-pull.md
index 10cee8dc209..1fed08d11c1 100644
--- a/docs/content/ingestion/stream-pull.md
+++ b/docs/content/ingestion/stream-pull.md
@@ -295,6 +295,8 @@ Otherwise, you can run real-time nodes without replication.
 There is now also an [experimental low level Kafka firehose](../development/kafka-simple-consumer-firehose.html) which 
 solves the issues described above with using the high level Kafka consumer.
 
+Please note that druid will skip over event that failed its checksum and it is corrupt.
+
 ### Locking
 
 Using stream pull ingestion with Realtime nodes together batch ingestion may introduce data override issues. For example, if you 
diff --git a/extensions/kafka-eight/src/main/java/io/druid/firehose/kafka/KafkaEightFirehoseFactory.java b/extensions/kafka-eight/src/main/java/io/druid/firehose/kafka/KafkaEightFirehoseFactory.java
index 6d84b6c4667..9bd4c5c1d42 100644
--- a/extensions/kafka-eight/src/main/java/io/druid/firehose/kafka/KafkaEightFirehoseFactory.java
+++ b/extensions/kafka-eight/src/main/java/io/druid/firehose/kafka/KafkaEightFirehoseFactory.java
@@ -34,6 +34,7 @@ import kafka.consumer.ConsumerConfig;
 import kafka.consumer.ConsumerIterator;
 import kafka.consumer.KafkaStream;
 import kafka.javaapi.consumer.ConsumerConnector;
+import kafka.message.InvalidMessageException;
 
 import java.io.IOException;
 import java.nio.ByteBuffer;
@@ -111,13 +112,24 @@ public class KafkaEightFirehoseFactory implements FirehoseFactory<ByteBufferInpu
       @Override
       public InputRow nextRow()
       {
-        final byte[] message = iter.next().message();
+        try {
+          final byte[] message = iter.next().message();
 
-        if (message == null) {
+          if (message == null) {
+            return null;
+          }
+
+          return theParser.parse(ByteBuffer.wrap(message));
+
+        }
+        catch (InvalidMessageException e) {
+          /*
+          IF the CRC is caused within the wire transfer, this is not the best way to handel CRC.
+          Probably it is better to shutdown the fireHose without commit and start it again.
+           */
+          log.error(e,"Message failed its checksum and it is corrupt, will skip it");
           return null;
         }
-
-        return theParser.parse(ByteBuffer.wrap(message));
       }
 
       @Override