diff --git a/test/core/end2end/tests/max_concurrent_streams.c b/test/core/end2end/tests/max_concurrent_streams.c
index e6502a7b0cd..24b6c99a7c4 100644
--- a/test/core/end2end/tests/max_concurrent_streams.c
+++ b/test/core/end2end/tests/max_concurrent_streams.c
@@ -444,7 +444,7 @@ static void test_max_concurrent_streams(grpc_end2end_test_config config) {
   config.tear_down_data(&f);
 }
 
-static void test_max_concurrent_streams_with_timeout(
+static void test_max_concurrent_streams_with_timeout_on_first(
     grpc_end2end_test_config config) {
   grpc_end2end_test_fixture f;
   grpc_arg server_arg;
@@ -478,8 +478,8 @@ static void test_max_concurrent_streams_with_timeout(
   server_args.num_args = 1;
   server_args.args = &server_arg;
 
-  f = begin_test(config, "test_max_concurrent_streams_with_timeout", NULL,
-                 &server_args);
+  f = begin_test(config, "test_max_concurrent_streams_with_timeout_on_first",
+                 NULL, &server_args);
   cqv = cq_verifier_create(f.cq);
 
   grpc_metadata_array_init(&request_metadata_recv);
@@ -636,8 +636,198 @@ static void test_max_concurrent_streams_with_timeout(
   config.tear_down_data(&f);
 }
 
+static void test_max_concurrent_streams_with_timeout_on_second(
+    grpc_end2end_test_config config) {
+  grpc_end2end_test_fixture f;
+  grpc_arg server_arg;
+  grpc_channel_args server_args;
+  grpc_call *c1;
+  grpc_call *c2;
+  grpc_call *s1;
+  //  grpc_call *s2;
+  cq_verifier *cqv;
+  grpc_call_details call_details;
+  grpc_metadata_array request_metadata_recv;
+  grpc_metadata_array initial_metadata_recv1;
+  grpc_metadata_array trailing_metadata_recv1;
+  grpc_metadata_array initial_metadata_recv2;
+  grpc_metadata_array trailing_metadata_recv2;
+  grpc_status_code status1;
+  grpc_call_error error;
+  char *details1 = NULL;
+  size_t details_capacity1 = 0;
+  grpc_status_code status2;
+  char *details2 = NULL;
+  size_t details_capacity2 = 0;
+  grpc_op ops[6];
+  grpc_op *op;
+  int was_cancelled;
+
+  server_arg.key = GRPC_ARG_MAX_CONCURRENT_STREAMS;
+  server_arg.type = GRPC_ARG_INTEGER;
+  server_arg.value.integer = 1;
+
+  server_args.num_args = 1;
+  server_args.args = &server_arg;
+
+  f = begin_test(config, "test_max_concurrent_streams_with_timeout_on_second",
+                 NULL, &server_args);
+  cqv = cq_verifier_create(f.cq);
+
+  grpc_metadata_array_init(&request_metadata_recv);
+  grpc_metadata_array_init(&initial_metadata_recv1);
+  grpc_metadata_array_init(&trailing_metadata_recv1);
+  grpc_metadata_array_init(&initial_metadata_recv2);
+  grpc_metadata_array_init(&trailing_metadata_recv2);
+  grpc_call_details_init(&call_details);
+
+  /* perform a ping-pong to ensure that settings have had a chance to round
+     trip */
+  simple_request_body(config, f);
+  /* perform another one to make sure that the one stream case still works */
+  simple_request_body(config, f);
+
+  /* start two requests - ensuring that the second is not accepted until
+     the first completes , and the second request will timeout in the
+     concurrent_list */
+  c1 = grpc_channel_create_call(f.client, NULL, GRPC_PROPAGATE_DEFAULTS, f.cq,
+                                "/alpha", "foo.test.google.fr:1234",
+                                n_seconds_time(1000), NULL);
+  GPR_ASSERT(c1);
+  c2 = grpc_channel_create_call(f.client, NULL, GRPC_PROPAGATE_DEFAULTS, f.cq,
+                                "/beta", "foo.test.google.fr:1234",
+                                n_seconds_time(3), NULL);
+  GPR_ASSERT(c2);
+
+  GPR_ASSERT(GRPC_CALL_OK == grpc_server_request_call(
+                                 f.server, &s1, &call_details,
+                                 &request_metadata_recv, f.cq, f.cq, tag(101)));
+
+  memset(ops, 0, sizeof(ops));
+  op = ops;
+  op->op = GRPC_OP_SEND_INITIAL_METADATA;
+  op->data.send_initial_metadata.count = 0;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_SEND_CLOSE_FROM_CLIENT;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  error = grpc_call_start_batch(c1, ops, (size_t)(op - ops), tag(301), NULL);
+  GPR_ASSERT(GRPC_CALL_OK == error);
+
+  memset(ops, 0, sizeof(ops));
+  op = ops;
+  op->op = GRPC_OP_RECV_STATUS_ON_CLIENT;
+  op->data.recv_status_on_client.trailing_metadata = &trailing_metadata_recv1;
+  op->data.recv_status_on_client.status = &status1;
+  op->data.recv_status_on_client.status_details = &details1;
+  op->data.recv_status_on_client.status_details_capacity = &details_capacity1;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_RECV_INITIAL_METADATA;
+  op->data.recv_initial_metadata = &initial_metadata_recv1;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  error = grpc_call_start_batch(c1, ops, (size_t)(op - ops), tag(302), NULL);
+  GPR_ASSERT(GRPC_CALL_OK == error);
+
+  CQ_EXPECT_COMPLETION(cqv, tag(101), 1);
+  CQ_EXPECT_COMPLETION(cqv, tag(301), 1);
+  cq_verify(cqv);
+
+  memset(ops, 0, sizeof(ops));
+  op = ops;
+  op->op = GRPC_OP_SEND_INITIAL_METADATA;
+  op->data.send_initial_metadata.count = 0;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_SEND_CLOSE_FROM_CLIENT;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  error = grpc_call_start_batch(c2, ops, (size_t)(op - ops), tag(401), NULL);
+  GPR_ASSERT(GRPC_CALL_OK == error);
+
+  memset(ops, 0, sizeof(ops));
+  op = ops;
+  op->op = GRPC_OP_RECV_STATUS_ON_CLIENT;
+  op->data.recv_status_on_client.trailing_metadata = &trailing_metadata_recv2;
+  op->data.recv_status_on_client.status = &status2;
+  op->data.recv_status_on_client.status_details = &details2;
+  op->data.recv_status_on_client.status_details_capacity = &details_capacity2;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_RECV_INITIAL_METADATA;
+  op->data.recv_initial_metadata = &initial_metadata_recv1;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  error = grpc_call_start_batch(c2, ops, (size_t)(op - ops), tag(402), NULL);
+  GPR_ASSERT(GRPC_CALL_OK == error);
+
+  /* the second request is time out*/
+  CQ_EXPECT_COMPLETION(cqv, tag(401), 0);
+  CQ_EXPECT_COMPLETION(cqv, tag(402), 1);
+  cq_verify(cqv);
+
+  /* second request is finished because of time out, so destroy the second call
+   */
+  grpc_call_destroy(c2);
+
+  /* now reply the first call */
+  memset(ops, 0, sizeof(ops));
+  op = ops;
+  op->op = GRPC_OP_SEND_INITIAL_METADATA;
+  op->data.send_initial_metadata.count = 0;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_RECV_CLOSE_ON_SERVER;
+  op->data.recv_close_on_server.cancelled = &was_cancelled;
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  op->op = GRPC_OP_SEND_STATUS_FROM_SERVER;
+  op->data.send_status_from_server.trailing_metadata_count = 0;
+  op->data.send_status_from_server.status = GRPC_STATUS_UNIMPLEMENTED;
+  op->data.send_status_from_server.status_details = "xyz";
+  op->flags = 0;
+  op->reserved = NULL;
+  op++;
+  error = grpc_call_start_batch(s1, ops, (size_t)(op - ops), tag(102), NULL);
+  GPR_ASSERT(GRPC_CALL_OK == error);
+
+  CQ_EXPECT_COMPLETION(cqv, tag(302), 1);
+  CQ_EXPECT_COMPLETION(cqv, tag(102), 1);
+  cq_verify(cqv);
+
+  cq_verifier_destroy(cqv);
+
+  grpc_call_destroy(c1);
+  grpc_call_destroy(s1);
+
+  gpr_free(details1);
+  gpr_free(details2);
+  grpc_metadata_array_destroy(&initial_metadata_recv1);
+  grpc_metadata_array_destroy(&trailing_metadata_recv1);
+  grpc_metadata_array_destroy(&initial_metadata_recv2);
+  grpc_metadata_array_destroy(&trailing_metadata_recv2);
+  grpc_metadata_array_destroy(&request_metadata_recv);
+  grpc_call_details_destroy(&call_details);
+
+  end_test(&f);
+  config.tear_down_data(&f);
+}
+
 void max_concurrent_streams(grpc_end2end_test_config config) {
-  test_max_concurrent_streams_with_timeout(config);
+  test_max_concurrent_streams_with_timeout_on_first(config);
+  test_max_concurrent_streams_with_timeout_on_second(config);
   test_max_concurrent_streams(config);
 }