diff --git a/.gitignore b/.gitignore
index 5dca467..c6d755d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -23,3 +23,17 @@ _testmain.go
 *.exe
 *.test
 *.prof
+
+# Created by https://www.gitignore.io/api/vim
+
+### Vim ###
+# swap
+[._]*.s[a-w][a-z]
+[._]s[a-w][a-z]
+# session
+Session.vim
+# temporary
+.netrwhist
+*~
+# auto-generated tag files
+tags
diff --git a/firehose.go b/firehose.go
index 40fb9af..5b7a1ca 100644
--- a/firehose.go
+++ b/firehose.go
@@ -143,6 +143,6 @@ func (p *Producer) sendFirehoseRecords(args *gokinesis.RequestArgs) {
 	}
 
 	if conf.Debug.Verbose && p.getMsgCount()%100 == 0 {
-		log.Println("Messages sent so far: " + strconv.Itoa(p.getMsgCount()))
+		log.Println("Messages sent so far: " + strconv.FormatInt(p.getMsgCount(), 10))
 	}
 }
diff --git a/kinesis.go b/kinesis.go
index a31ff22..5410947 100644
--- a/kinesis.go
+++ b/kinesis.go
@@ -3,6 +3,7 @@ package kinetic
 import (
 	"errors"
 	"sync"
+	"sync/atomic"
 	"time"
 
 	gokinesis "github.com/rewardStyle/go-kinesis"
@@ -57,11 +58,8 @@ type kinesis struct {
 
 	client gokinesis.KinesisClient
 
-	msgCount   int
-	msgCountMu sync.Mutex
-
-	errCount   int
-	errCountMu sync.Mutex
+	msgCount int64
+	errCount int64
 }
 
 func (k *kinesis) init(stream, shard, shardIteratorType, accessKey, secretKey, region string) (*kinesis, error) {
@@ -157,39 +155,27 @@ func (k *kinesis) refreshClient(accessKey, secretKey, region string) {
 }
 
 func (k *kinesis) decMsgCount() {
-	k.msgCountMu.Lock()
-	k.msgCount--
-	k.msgCountMu.Unlock()
+	atomic.AddInt64(&k.msgCount, -1)
 }
 
 func (k *kinesis) incMsgCount() {
-	k.msgCountMu.Lock()
-	k.msgCount++
-	k.msgCountMu.Unlock()
+	atomic.AddInt64(&k.msgCount, 1)
 }
 
-func (k *kinesis) getMsgCount() int {
-	k.msgCountMu.Lock()
-	defer k.msgCountMu.Unlock()
-	return k.msgCount
+func (k *kinesis) getMsgCount() int64 {
+	return atomic.LoadInt64(&k.msgCount)
 }
 
 func (k *kinesis) decErrCount() {
-	k.errCountMu.Lock()
-	k.errCount--
-	k.errCountMu.Unlock()
+	atomic.AddInt64(&k.errCount, -1)
 }
 
 func (k *kinesis) incErrCount() {
-	k.errCountMu.Lock()
-	k.errCount++
-	k.errCountMu.Unlock()
+	atomic.AddInt64(&k.errCount, 1)
 }
 
-func (k *kinesis) getErrCount() int {
-	k.errCountMu.Lock()
-	defer k.errCountMu.Unlock()
-	return k.errCount
+func (k *kinesis) getErrCount() int64 {
+	return atomic.LoadInt64(&k.errCount)
 }
 
 func getLock(sem chan bool) {
diff --git a/listener.go b/listener.go
index f29962e..ffa016e 100644
--- a/listener.go
+++ b/listener.go
@@ -206,11 +206,19 @@ retry:
 func (l *Listener) consume() {
 	l.setConsuming(true)
 
-	counter := 0
-	timer := time.Now()
+	readCounter := 0
+	readTimer := time.Now()
+
+	GsiCounter := 0
+	GsiTimer := time.Now()
 
 	for {
-		l.throttle(&counter, &timer)
+		if !l.shouldConsume() {
+			l.setConsuming(false)
+			break
+		}
+
+		l.throttle(&readCounter, &readTimer)
 
 		// args() will give us the shard iterator and type as well as the shard id
 		response, err := l.client.GetRecords(l.args())
@@ -241,9 +249,7 @@ func (l *Listener) consume() {
 
 				// If we received an error we should wait and attempt to
 				// refresh the shard iterator again
-				<-time.After(1 * time.Second)
-
-				log.Println("Retrying after waiting one second.")
+				l.throttle(&GsiCounter, &GsiTimer)
 
 				goto refresh_iterator
 			}
@@ -259,11 +265,6 @@ func (l *Listener) consume() {
 				}
 			}
 		}
-
-		if !l.shouldConsume() {
-			l.setConsuming(false)
-			break
-		}
 	}
 }
 
diff --git a/listener_test.go b/listener_test.go
index 8f3e3b6..a667276 100644
--- a/listener_test.go
+++ b/listener_test.go
@@ -49,7 +49,7 @@ func TestListenerError(t *testing.T) {
 			// Let the error propagate
 			<-time.After(1 * time.Second)
 
-			So(listener.errCount, ShouldNotEqual, 0)
+			So(listener.getErrCount(), ShouldNotEqual, 0)
 			So(listener.IsListening(), ShouldEqual, true)
 		})
 	})
diff --git a/producer.go b/producer.go
index 36c757b..8beccba 100644
--- a/producer.go
+++ b/producer.go
@@ -218,7 +218,7 @@ stop:
 			p.incMsgCount()
 
 			if conf.Debug.Verbose && p.getMsgCount()%100 == 0 {
-				log.Println("Received message to send. Total messages received: " + strconv.Itoa(p.getMsgCount()))
+				log.Println("Received message to send. Total messages received: " + strconv.FormatInt(p.getMsgCount(), 10))
 			}
 
 			kargs := p.args()
@@ -320,7 +320,7 @@ func (p *Producer) sendRecords(args *gokinesis.RequestArgs) {
 	}
 
 	if conf.Debug.Verbose && p.getMsgCount()%100 == 0 {
-		log.Println("Messages sent so far: " + strconv.Itoa(p.getMsgCount()))
+		log.Println("Messages sent so far: " + strconv.FormatInt(p.getMsgCount(), 10))
 	}
 }
 
diff --git a/producer_test.go b/producer_test.go
index a017922..a9d60be 100644
--- a/producer_test.go
+++ b/producer_test.go
@@ -43,8 +43,8 @@ func TestProducerError(t *testing.T) {
 		Convey("It should handle errors successfully", func() {
 			producer.errors <- errors.New("All your base are belong to us!")
 			// Let the error propagate
-			<-time.After(1 * time.Second)
-			So(producer.errCount, ShouldEqual, 1)
+			<-time.After(3 * time.Second)
+			So(producer.getErrCount(), ShouldEqual, 1)
 			So(producer.IsProducing(), ShouldEqual, true)
 		})
 	})