Merge pull request #50 from xlab-uiuc/trace-improve

Task desc && tool result adjustment

Author: yinfangchen

aiopslab/observer/trace_api.py

@@ -211,9 +211,9 @@ def extract_traces(
             )
             for trace in traces:
                 for span in trace["spans"]:
-                    span[
+                    span["serviceName"] = trace["processes"][span["processID"]][
                         "serviceName"
-                    ] = service  # Directly associate service name with each span
+                    ]
                 all_traces.append(trace)  # Collect the trace with service name included
         self.cleanup()
         print("Cleanup completed.")
@@ -251,7 +251,7 @@ def process_traces(self, traces) -> pd.DataFrame:
                 operation_name_list.append(span["operationName"])
                 start_time_list.append(span["startTime"])
                 duration_list.append(span["duration"])
-                
+
                 has_error = False
                 response = "Unknown"
                 for tag in span.get("tags", []):
@@ -261,7 +261,6 @@ def process_traces(self, traces) -> pd.DataFrame:
                         response = tag["value"]
                 error_list.append(has_error)
                 response_list.append(response)
-                    
 
         df = pd.DataFrame(
             {

aiopslab/orchestrator/problems/ad_service_failure/ad_service_failure.py

@@ -61,7 +61,6 @@ class AdServiceFailureLocalization(AdServiceFailureBaseTask, LocalizationTask):
     def __init__(self):
         AdServiceFailureBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the ad service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/ad_service_high_cpu/ad_service_high_cpu.py

@@ -61,7 +61,6 @@ class AdServiceHighCpuLocalization(AdServiceHighCpuBaseTask, LocalizationTask):
     def __init__(self):
         AdServiceHighCpuBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the ad service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/ad_service_manual_gc/ad_service_manual_gc.py

@@ -61,7 +61,6 @@ class AdServiceManualGcLocalization(AdServiceManualGcBaseTask, LocalizationTask)
     def __init__(self):
         AdServiceManualGcBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the ad service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/assign_non_existent_node/assign_non_existent_node_social_net.py

@@ -93,7 +93,6 @@ class AssignNonExistentNodeSocialNetLocalization(
     def __init__(self):
         AssignNonExistentNodeSocialNetBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -137,7 +136,6 @@ class AssignNonExistentNodeSocialNetAnalysis(
     def __init__(self):
         AssignNonExistentNodeSocialNetBaseTask.__init__(self)
         AnalysisTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -179,7 +177,6 @@ class AssignNonExistentNodeSocialNetMitigation(
     def __init__(self):
         AssignNonExistentNodeSocialNetBaseTask.__init__(self)
         MitigationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float) -> dict:
         print("== Evaluation ==")

aiopslab/orchestrator/problems/auth_miss_mongodb/auth_miss_mongodb.py

@@ -86,7 +86,6 @@ class MongoDBAuthMissingLocalization(MongoDBAuthMissingBaseTask, LocalizationTas
     def __init__(self):
         MongoDBAuthMissingBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -128,7 +127,6 @@ class MongoDBAuthMissingAnalysis(MongoDBAuthMissingBaseTask, AnalysisTask):
     def __init__(self):
         MongoDBAuthMissingBaseTask.__init__(self)
         AnalysisTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -169,7 +167,6 @@ class MongoDBAuthMissingMitigation(MongoDBAuthMissingBaseTask, MitigationTask):
     def __init__(self):
         MongoDBAuthMissingBaseTask.__init__(self)
         MitigationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     # TODO: this migigate eval should be a bit different.
     # The error will not be on the container/pod level but the app level,

aiopslab/orchestrator/problems/cart_service_failure/cart_service_failure.py

@@ -61,7 +61,6 @@ class CartServiceFailureLocalization(CartServiceFailureBaseTask, LocalizationTas
     def __init__(self):
         CartServiceFailureBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the cart service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/image_slow_load/image_slow_load.py

@@ -61,7 +61,6 @@ class ImageSlowLoadLocalization(ImageSlowLoadBaseTask, LocalizationTask):
     def __init__(self):
         ImageSlowLoadBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the frontend service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/k8s_target_port_misconfig/target_port.py

@@ -88,7 +88,6 @@ class K8STargetPortMisconfigLocalization(
     def __init__(self, faulty_service: str = "user-service"):
         K8STargetPortMisconfigBaseTask.__init__(self, faulty_service=faulty_service)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -133,7 +132,6 @@ class K8STargetPortMisconfigAnalysis(K8STargetPortMisconfigBaseTask, AnalysisTas
     def __init__(self, faulty_service: str = "user-service"):
         K8STargetPortMisconfigBaseTask.__init__(self, faulty_service=faulty_service)
         AnalysisTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -167,7 +165,6 @@ class K8STargetPortMisconfigMitigation(K8STargetPortMisconfigBaseTask, Mitigatio
     def __init__(self, faulty_service: str = "user-service"):
         K8STargetPortMisconfigBaseTask.__init__(self, faulty_service=faulty_service)
         MitigationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compose-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float) -> dict:
         print("== Evaluation ==")

aiopslab/orchestrator/problems/kafka_queue_problems/kafka_queue_problems.py

@@ -61,7 +61,6 @@ class KafkaQueueProblemsLocalization(KafkaQueueProblemsBaseTask, LocalizationTas
     def __init__(self):
         KafkaQueueProblemsBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the kafka service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/loadgenerator_flood_homepage/loadgenerator_flood_homepage.py

@@ -66,7 +66,6 @@ class LoadGeneratorFloodHomepageLocalization(
     def __init__(self):
         LoadGeneratorFloodHomepageBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the frontend service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/payment_service_failure/payment_service_failure.py

@@ -63,7 +63,6 @@ class PaymentServiceFailureLocalization(
     def __init__(self):
         PaymentServiceFailureBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the payment service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/payment_service_unreachable/payment_service_unreachable.py

@@ -65,7 +65,6 @@ class PaymentServiceUnreachableLocalization(
     def __init__(self):
         PaymentServiceUnreachableBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the payment service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/product_catalog_failure/product_catalog_failure.py

@@ -65,7 +65,6 @@ class ProductCatalogServiceFailureLocalization(
     def __init__(self):
         ProductCatalogServiceFailureBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the product catalog service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/recommendation_service_cache_failure/recommendation_service_cache_failure.py

@@ -67,7 +67,6 @@ class RecommendationServiceCacheFailureLocalization(
     def __init__(self):
         RecommendationServiceCacheFailureBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the reccomendation service."
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")

aiopslab/orchestrator/problems/scale_pod/scale_pod_social_net.py

@@ -92,7 +92,6 @@ class ScalePodSocialNetLocalization(ScalePodSocialNetBaseTask, LocalizationTask)
     def __init__(self):
         ScalePodSocialNetBaseTask.__init__(self)
         LocalizationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -134,7 +133,6 @@ class ScalePodSocialNetAnalysis(ScalePodSocialNetBaseTask, AnalysisTask):
     def __init__(self):
         ScalePodSocialNetBaseTask.__init__(self)
         AnalysisTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float):
         print("== Evaluation ==")
@@ -174,7 +172,6 @@ class ScalePodSocialNetMitigation(ScalePodSocialNetBaseTask, MitigationTask):
     def __init__(self):
         ScalePodSocialNetBaseTask.__init__(self)
         MitigationTask.__init__(self, self.app)
-        self.task_desc += "Start by investigating the `compost-post-service` pod"
 
     def eval(self, soln: Any, trace: list[SessionItem], duration: float) -> dict:
         print("== Evaluation ==")

aiopslab/orchestrator/tasks/localization.py

@@ -31,6 +31,7 @@ def __init__(self, app: Application):
             The service you are working with today is described below:
             {app_summary}
 
+            You are requested to identify the service(s) where the root cause of the fault lies.
             You will begin by analyzing the service's state and telemetry, and then submit one of two possible solutions:
             1. list[str]: list of faulty components (i.e., service names)
             2. str: `None` if no faults were detected