diff --git a/webrtc/modules/audio_coding/neteq4/neteq_impl.cc b/webrtc/modules/audio_coding/neteq4/neteq_impl.cc
index 4ffe3e79ea..aa3370f8c3 100644
--- a/webrtc/modules/audio_coding/neteq4/neteq_impl.cc
+++ b/webrtc/modules/audio_coding/neteq4/neteq_impl.cc
@@ -870,7 +870,7 @@ int NetEqImpl::GetDecision(Operations* operation,
   }
   const RTPHeader* header = packet_buffer_->NextRtpHeader();
 
-  if (decision_logic_->CngRfc3389On()) {
+  if (decision_logic_->CngRfc3389On() || last_mode_ == kModeRfc3389Cng) {
     // Because of timestamp peculiarities, we have to "manually" disallow using
     // a CNG packet with the same timestamp as the one that was last played.
     // This can happen when using redundancy and will cause the timing to shift.
@@ -878,7 +878,6 @@ int NetEqImpl::GetDecision(Operations* operation,
         decoder_database_->IsComfortNoise(header->payloadType) &&
         end_timestamp >= header->timestamp) {
       // Don't use this packet, discard it.
-      // TODO(hlundin): Write test for this case.
       if (packet_buffer_->DiscardNextPacket() != PacketBuffer::kOK) {
         assert(false);  // Must be ok by design.
       }
diff --git a/webrtc/modules/audio_coding/neteq4/neteq_unittest.cc b/webrtc/modules/audio_coding/neteq4/neteq_unittest.cc
index 32f5bf40b1..2e0f2d5255 100644
--- a/webrtc/modules/audio_coding/neteq4/neteq_unittest.cc
+++ b/webrtc/modules/audio_coding/neteq4/neteq_unittest.cc
@@ -220,6 +220,7 @@ class NetEqDecodingTest : public ::testing::Test {
                 bool expect_seq_no_wrap, bool expect_timestamp_wrap);
 
   void LongCngWithClockDrift(double drift_factor);
+  void DuplicateCng();
 
   NetEq* neteq_;
   FILE* rtp_fp_;
@@ -1143,7 +1144,7 @@ void NetEqDecodingTest::WrapTest(uint16_t start_seq_no,
   NetEqOutputType output_type;
   uint32_t receive_timestamp = 0;
 
-  // Insert speech for 1 second.
+  // Insert speech for 2 seconds.
   const int kSpeechDurationMs = 2000;
   int packets_inserted = 0;
   uint16_t last_seq_no;
@@ -1229,4 +1230,81 @@ TEST_F(NetEqDecodingTest, TimestampAndSequenceNumberWrap) {
   WrapTest(0xFFFF - 10, 0xFFFFFFFF - 5000, drop_seq_numbers, true, true);
 }
 
+void NetEqDecodingTest::DuplicateCng() {
+  uint16_t seq_no = 0;
+  uint32_t timestamp = 0;
+  const int kFrameSizeMs = 10;
+  const int kSampleRateKhz = 16;
+  const int kSamples = kFrameSizeMs * kSampleRateKhz;
+  const int kPayloadBytes = kSamples * 2;
+
+  // Insert three speech packet. Three are needed to get the frame length
+  // correct.
+  int out_len;
+  int num_channels;
+  NetEqOutputType type;
+  uint8_t payload[kPayloadBytes] = {0};
+  WebRtcRTPHeader rtp_info;
+  for (int i = 0; i < 3; ++i) {
+    PopulateRtpInfo(seq_no, timestamp, &rtp_info);
+    ASSERT_EQ(0, neteq_->InsertPacket(rtp_info, payload, kPayloadBytes, 0));
+    ++seq_no;
+    timestamp += kSamples;
+
+    // Pull audio once.
+    ASSERT_EQ(0,
+              neteq_->GetAudio(
+                  kMaxBlockSize, out_data_, &out_len, &num_channels, &type));
+    ASSERT_EQ(kBlockSize16kHz, out_len);
+  }
+  // Verify speech output.
+  EXPECT_EQ(kOutputNormal, type);
+
+  // Insert same CNG packet twice.
+  const int kCngPeriodMs = 100;
+  const int kCngPeriodSamples = kCngPeriodMs * kSampleRateKhz;
+  int payload_len;
+  PopulateCng(seq_no, timestamp, &rtp_info, payload, &payload_len);
+  // This is the first time this CNG packet is inserted.
+  ASSERT_EQ(0, neteq_->InsertPacket(rtp_info, payload, payload_len, 0));
+
+  // Pull audio once and make sure CNG is played.
+  ASSERT_EQ(0,
+            neteq_->GetAudio(
+                kMaxBlockSize, out_data_, &out_len, &num_channels, &type));
+  ASSERT_EQ(kBlockSize16kHz, out_len);
+  EXPECT_EQ(kOutputCNG, type);
+  EXPECT_EQ(timestamp - 10, neteq_->PlayoutTimestamp());
+
+  // Insert the same CNG packet again. Note that at this point it is old, since
+  // we have already decoded the first copy of it.
+  ASSERT_EQ(0, neteq_->InsertPacket(rtp_info, payload, payload_len, 0));
+
+  // Pull audio until we have played |kCngPeriodMs| of CNG. Start at 10 ms since
+  // we have already pulled out CNG once.
+  for (int cng_time_ms = 10; cng_time_ms < kCngPeriodMs; cng_time_ms += 10) {
+    ASSERT_EQ(0,
+              neteq_->GetAudio(
+                  kMaxBlockSize, out_data_, &out_len, &num_channels, &type));
+    ASSERT_EQ(kBlockSize16kHz, out_len);
+    EXPECT_EQ(kOutputCNG, type);
+    EXPECT_EQ(timestamp - 10, neteq_->PlayoutTimestamp());
+  }
+
+  // Insert speech again.
+  ++seq_no;
+  timestamp += kCngPeriodSamples;
+  PopulateRtpInfo(seq_no, timestamp, &rtp_info);
+  ASSERT_EQ(0, neteq_->InsertPacket(rtp_info, payload, kPayloadBytes, 0));
+
+  // Pull audio once and verify that the output is speech again.
+  ASSERT_EQ(0,
+            neteq_->GetAudio(
+                kMaxBlockSize, out_data_, &out_len, &num_channels, &type));
+  ASSERT_EQ(kBlockSize16kHz, out_len);
+  EXPECT_EQ(kOutputNormal, type);
+  EXPECT_EQ(timestamp + kSamples - 10, neteq_->PlayoutTimestamp());
+}
+
+TEST_F(NetEqDecodingTest, DiscardDuplicateCng) { DuplicateCng(); }
 }  // namespace webrtc