diff --git a/rally/benchmark/validation.py b/rally/benchmark/validation.py
index 3d25f13644..8c0b1d7cae 100644
--- a/rally/benchmark/validation.py
+++ b/rally/benchmark/validation.py
@@ -371,3 +371,13 @@ def required_services(config, clients, task, *required_services):
         if service not in available_services:
             return ValidationResult(
                 False, _("Service is not available: %s") % service)
+
+
+@validator
+def required_contexts(config, clients, task, *context_names):
+    missing_contexts = set(context_names) - set(config.get("context", {}))
+    if missing_contexts:
+        message = (_("The following contexts are required but missing from "
+                     "the benchmark configuration file: %s") %
+                   ", ".join(missing_contexts))
+        return ValidationResult(False, message)
diff --git a/tests/benchmark/test_validation.py b/tests/benchmark/test_validation.py
index 668143bead..2071c4df96 100644
--- a/tests/benchmark/test_validation.py
+++ b/tests/benchmark/test_validation.py
@@ -80,6 +80,22 @@ class ValidationUtilsTestCase(test.TestCase):
         validator, = self._get_scenario_validators(func_failure, scenario)
         self.assertFalse(validator(None, None, None).is_valid)
 
+    def test_required_contexts(self):
+        config = {"context": {"context01": {}, "context02": {}}}
+
+        required_contexts = (lambda *contexts:
+                             validation.required_contexts(*contexts)
+                             (lambda: None).validators.pop()
+                             (config, None, None))
+
+        # All the required contexts are defined
+        result = required_contexts("context01", "context02")
+        self.assertTrue(result.is_valid)
+
+        # A required context is not defined
+        result = required_contexts("context03")
+        self.assertFalse(result.is_valid)
+
     def test_required_services(self):
         available_services = {
             consts.ServiceType.IDENTITY: consts.Service.KEYSTONE,