Merge pull request #6005 from sreecha/stress_exit_codes

Terminate with error exit code in case of any failure

test/cpp/interop/stress_test.cc

@@ -98,11 +98,11 @@ DEFINE_string(test_cases, "",
               " 'large_unary', 10% of the time and 'empty_stream' the remaining"
               " 70% of the time");
 
-DEFINE_int32(log_level, GPR_LOG_SEVERITY_DEBUG,
+DEFINE_int32(log_level, GPR_LOG_SEVERITY_INFO,
              "Severity level of messages that should be logged. Any messages "
              "greater than or equal to the level set here will be logged. "
              "The choices are: 0 (GPR_LOG_SEVERITY_DEBUG), 1 "
-             "(GPR_LOG_SEVERITY_INFO) and 2 (GPR_LOG_SEVERITY_ERROR.");
+             "(GPR_LOG_SEVERITY_INFO) and 2 (GPR_LOG_SEVERITY_ERROR)");
 
 using grpc::testing::kTestCaseList;
 using grpc::testing::MetricsService;

tools/run_tests/stress_test/README.md

@@ -63,8 +63,8 @@
 The stress tests are launched by the following script (path is relative to GRPC root directory) :
 `tools/run_tests/stress_test/run_stress_tests_on_gke.py`
 
-The script has several parameters and you can find out more details by using the `--help` flag.
-  - `<grpc_root_dir>$ tools/run_tests/stress_test/run_stress_tests_on_gke.py --help`
+You can find out more details by using the `--help` flag.
+  - `<grpc_root_dir>$ tools/run_tests/stress_test/run_on_gke.py --help`
 
 > **Example**
 > ```bash

tools/run_tests/stress_test/STRESS_CLIENT_SPEC.md

@@ -6,14 +6,14 @@ This document specifies the features a stress test client should implement in or
 --------------
 **1.** A stress test client should be able to repeatedly execute one or more of the existing 'interop test cases'. It may just be a wrapper around the existing interop test client. The exact command line arguments the client should support are listed in _Table 1_ below.
 
-**2.** The stress test client must implement a metrics server defined by _[metrics.proto](https://github.com/grpc/grpc/blob/master/src/proto/grpc/testing/metrics.proto)_ and must expose _qps_ as a long-valued Gauge. The client can track the overall _qps_ in one Gauge or in multiple Gagues (for example: One per Channel or Stub). 
+**2.** The stress test client must implement a metrics server defined by _[metrics.proto](https://github.com/grpc/grpc/blob/master/src/proto/grpc/testing/metrics.proto)_ and must expose _qps_ as a long-valued Gauge. The client can track the overall _qps_ in one Gauge or in multiple Gauges (for example: One per Channel or Stub).
  The framework periodically queries the _qps_ by calling the `GetAllGauges()` method (the framework assumes that all the returned Gauges are _qps_ Gauges) and uses this to determine if the stress test client is running or crashed or stalled.
 > *Note:* In this context, the term  _**qps**_  means _interop test cases per second_  (not _messages per second_ or _rpc calls per second_)
 
 
 **Table 1:** Command line arguments that should be supported by the stress test client.
 
->_**Note** The current C++ [stress client](https://github.com/grpc/grpc/blob/master/test/cpp/interop/stress_test.cc) supports more flags than those listed here but those flags will soon be deprecated_|
+>_**Note** The current C++ [stress client](https://github.com/grpc/grpc/blob/master/test/cpp/interop/stress_test.cc) supports more flags than those listed here but those flags will soon be deprecated._
 
 Parameter             |                    Description
 ----------------------|---------------------------------

tools/run_tests/stress_test/run_on_gke.py

@@ -633,4 +633,5 @@ if __name__ == '__main__':
       os.path.dirname(sys.argv[0]), '../../..'))
   os.chdir(grpc_root)
 
-  run_tests(config)
+  if not run_tests(config):
+    sys.exit(1)