diff --git a/CHANGELOG b/CHANGELOG
index 8275daed3..dd393842a 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -3,8 +3,10 @@ Changes between 1.2.1 and 1.2.0
 
 - fix issue67: new super-short traceback-printing option: "--tb=line" will print a single line for each failing (python) test indicating its filename, lineno and the failure value
 - fix issue78: always call python-level teardown functions even if the
-  according setup failed - but make sure that setup is called repeatedly 
-  and no teardown if the setup raises a Skipped (as sone by py.test.skip()).
+  according setup failed.  This includes refinements for calling setup_module/class functions 
+  which will now only be called once instead of the previous behaviour where they'd be called
+  multiple times if they raise an exception (including a Skipped exception).  Any exception
+  will be re-corded and associated with all tests in the according module/class scope.
 - fix issue63: assume <40 columns to be a bogus terminal width, default to 80
 - update apipkg.py to fix an issue where recursive imports might
   unnecessarily break importing 
diff --git a/py/_plugin/pytest_runner.py b/py/_plugin/pytest_runner.py
index 14fa38bb6..6814fb204 100644
--- a/py/_plugin/pytest_runner.py
+++ b/py/_plugin/pytest_runner.py
@@ -2,7 +2,7 @@
 collect and run test items and create reports. 
 """
 
-import py
+import py, sys
 from py._test.outcome import Skipped
 
 #
@@ -252,6 +252,10 @@ class SetupState(object):
             self.stack.append(col) 
             try:
                 col.setup() 
-            except Skipped:
-                self.stack.pop()
-                raise 
+            except Exception:
+                col._prepare_exc = sys.exc_info()
+                raise
+        # check if the last collection node has raised an error 
+        for col in self.stack:
+            if hasattr(col, '_prepare_exc'):
+                py.builtin._reraise(*col._prepare_exc) 
diff --git a/testing/plugin/test_pytest_runner.py b/testing/plugin/test_pytest_runner.py
index 4764de98d..963847d8f 100644
--- a/testing/plugin/test_pytest_runner.py
+++ b/testing/plugin/test_pytest_runner.py
@@ -34,6 +34,16 @@ class TestSetupState:
         ss.teardown_exact(item)
         ss.teardown_exact(item)
 
+    def test_setup_fails_and_failure_is_cached(self, testdir):
+        item = testdir.getitem("""
+            def setup_module(mod):
+                raise ValueError(42)
+            def test_func(): pass
+        """)
+        ss = runner.SetupState()
+        py.test.raises(ValueError, "ss.prepare(item)")
+        py.test.raises(ValueError, "ss.prepare(item)")
+
 class BaseFunctionalTests:
     def test_passfunction(self, testdir):
         reports = testdir.runitem("""
diff --git a/testing/plugin/test_pytest_runner_xunit.py b/testing/plugin/test_pytest_runner_xunit.py
index f95915a89..77ed35e9a 100644
--- a/testing/plugin/test_pytest_runner_xunit.py
+++ b/testing/plugin/test_pytest_runner_xunit.py
@@ -150,7 +150,7 @@ def test_failing_setup_calls_teardown(testdir):
         "*2 error*"
     ])
 
-def test_setup_that_skips_calledagain_and_no_teardown(testdir):
+def test_setup_that_skips_calledagain_and_teardown(testdir):
     p = testdir.makepyfile("""
         import py
         def setup_module(mod):
@@ -164,8 +164,27 @@ def test_setup_that_skips_calledagain_and_no_teardown(testdir):
     """)
     result = testdir.runpytest(p)
     result.stdout.fnmatch_lines([
-        "*2 skipped*",
+        "*ValueError*43*",
+        "*2 skipped*1 error*",
     ])
-    assert "43" not in result.stdout.str()
+
+def test_setup_fails_again_on_all_tests(testdir):
+    p = testdir.makepyfile("""
+        import py
+        def setup_module(mod):
+            raise ValueError(42)
+        def test_function1():
+            pass
+        def test_function2():
+            pass
+        def teardown_module(mod):
+            raise ValueError(43)
+    """)
+    result = testdir.runpytest(p)
+    result.stdout.fnmatch_lines([
+        "*3 error*"
+    ])
+    assert "passed" not in result.stdout.str()
+