@@ -658,7 +658,9 @@ Resources:
658
658
Properties:
659
659
ActionsEnabled: true
660
660
AlarmActions:
661
- - !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-high-usage"]
661
+ # TODO: after we have run at high usage for a while, consider re-enabling this alarm. Right now it is too noisy
662
+ # - !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-high-usage"]
663
+ - !Ref AWS::NoValue
662
664
AlarmDescription: Send message if abnormally high Javabuilder usage detected.
663
665
Monitors usage across the HTTP API, WebSocket API, and all Build and Run
664
666
Lambdas.
@@ -760,133 +762,98 @@ Resources:
760
762
Type: AWS::CloudWatch::Alarm
761
763
Properties:
762
764
AlarmName: !Sub "${SubDomainName}_<%= name . downcase%> _slow_cleanup_time"
763
- AlarmDescription: p(95) cleanup time in Javabuilder's <%= name%> build and run lambda exceeded
764
- the normal cleanup time for 15 minutes. Investigate if there has been a performance regression.
765
+ AlarmDescription: Average cleanup time in Javabuilder's <%= name%> build and run lambda was high for at
766
+ least 15 out of the last 20 minutes. Investigate if there has been a performance regression.
765
767
ActionsEnabled: true
766
768
OKActions: []
767
769
AlarmActions:
768
770
- !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-slow-performance"]
769
771
InsufficientDataActions: []
772
+ MetricName: CleanupTime
773
+ Namespace: Javabuilder
774
+ Statistic: Average
775
+ Dimensions:
776
+ - Name: functionName
777
+ Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
770
778
EvaluationPeriods: 20
771
779
DatapointsToAlarm: 15
772
- ComparisonOperator: GreaterThanUpperThreshold
780
+ ComparisonOperator: GreaterThanThreshold
773
781
TreatMissingData: notBreaching
774
- Metrics:
775
- - Id: m1
776
- ReturnData: true
777
- MetricStat:
778
- Metric:
779
- Namespace: Javabuilder
780
- MetricName: CleanupTime
781
- Dimensions:
782
- - Name: functionName
783
- Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
784
- Period: 60
785
- Stat: p95
786
- - Id: ad1
787
- Label: CleanupTime (expected)
788
- ReturnData: true
789
- Expression: ANOMALY_DETECTION_BAND(m1, 5)
790
- ThresholdMetricId: ad1
782
+ Threshold: 200
783
+ Period: 60
791
784
792
785
<%= name%> SlowColdBootTimeAlarm:
793
786
Type: AWS::CloudWatch::Alarm
794
787
Properties:
795
788
AlarmName: !Sub "${SubDomainName}_<%= name . downcase%> _slow_cold_boot_time"
796
- AlarmDescription: p(95) cold boot time in Javabuilder's <%= name%> build and run lambda exceeded
797
- the normal cleanup time for 15 minutes. Investigate if there has been a performance regression.
789
+ AlarmDescription: Average cold boot time in Javabuilder's <%= name%> build and run lambda was high for at
790
+ least 15 out of the last 20 minutes. Investigate if there has been a performance regression.
798
791
ActionsEnabled: true
799
792
OKActions: []
800
793
AlarmActions:
801
794
- !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-slow-performance"]
802
795
InsufficientDataActions: []
796
+ MetricName: ColdBootTime
797
+ Namespace: Javabuilder
798
+ Statistic: Average
799
+ Dimensions:
800
+ - Name: functionName
801
+ Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
803
802
EvaluationPeriods: 20
804
803
DatapointsToAlarm: 15
805
- ComparisonOperator: GreaterThanUpperThreshold
804
+ ComparisonOperator: GreaterThanThreshold
806
805
TreatMissingData: notBreaching
807
- Metrics:
808
- - Id: m1
809
- ReturnData: true
810
- MetricStat:
811
- Metric:
812
- Namespace: Javabuilder
813
- MetricName: ColdBootTime
814
- Dimensions:
815
- - Name: functionName
816
- Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
817
- Period: 60
818
- Stat: p95
819
- - Id: ad1
820
- Label: ColdBootTime (expected)
821
- ReturnData: true
822
- Expression: ANOMALY_DETECTION_BAND(m1, 2)
823
- ThresholdMetricId: ad1
806
+ Threshold: 10500
807
+ Period: 60
824
808
825
809
<%= name%> SlowInitializationTimeAlarm:
826
810
Type: AWS::CloudWatch::Alarm
827
811
Properties:
828
812
AlarmName: !Sub "${SubDomainName}_<%= name . downcase%> _slow_initialization_time"
829
- AlarmDescription: p(95) initialization time in Javabuilder's <%= name%> build and run lambda exceeded
830
- the normal cleanup time for 15 minutes. Investigate if there has been a performance regression.
813
+ AlarmDescription: Average initialization time in Javabuilder's <%= name%> build and run lambda was high for at
814
+ least 15 out of the last 20 minutes. Investigate if there has been a performance regression.
831
815
ActionsEnabled: true
832
816
OKActions: []
833
817
AlarmActions:
834
818
- !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-slow-performance"]
835
819
InsufficientDataActions: []
820
+ MetricName: InitializationTime
821
+ Namespace: Javabuilder
822
+ Statistic: Average
823
+ Dimensions:
824
+ - Name: functionName
825
+ Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
836
826
EvaluationPeriods: 20
837
827
DatapointsToAlarm: 15
838
- ComparisonOperator: GreaterThanUpperThreshold
828
+ ComparisonOperator: GreaterThanThreshold
839
829
TreatMissingData: notBreaching
840
- Metrics:
841
- - Id: m1
842
- ReturnData: true
843
- MetricStat:
844
- Metric:
845
- Namespace: Javabuilder
846
- MetricName: InitializationTime
847
- Dimensions:
848
- - Name: functionName
849
- Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
850
- Period: 60
851
- Stat: p95
852
- - Id: ad1
853
- Label: InitializationTime (expected)
854
- ReturnData: true
855
- Expression: ANOMALY_DETECTION_BAND(m1, 5)
856
- ThresholdMetricId: ad1
830
+ Threshold: 5000
831
+ Period: 60
832
+
857
833
858
834
<%= name%> SlowTransitionTimeAlarm:
859
835
Type: AWS::CloudWatch::Alarm
860
836
Properties:
861
837
AlarmName: !Sub "${SubDomainName}_<%= name . downcase%> _slow_transition_time"
862
- AlarmDescription: p(95) transition time in Javabuilder's <%= name%> build and run lambda exceeded
863
- the normal cleanup time for 15 minutes. Investigate if there has been a performance regression.
838
+ AlarmDescription: Average transition time in Javabuilder's <%= name%> build and run lambda was high for at
839
+ least 15 out of the last 20 minutes. Investigate if there has been a performance regression.
864
840
ActionsEnabled: true
865
841
OKActions: []
866
842
AlarmActions:
867
843
- !If [SilenceAlertsCondition, !Ref AWS::NoValue, !Sub "arn:aws:sns:${AWS::Region}:${AWS::AccountId}:javabuilder-slow-performance"]
868
844
InsufficientDataActions: []
845
+ MetricName: TransitionTime
846
+ Namespace: Javabuilder
847
+ Statistic: Average
848
+ Dimensions:
849
+ - Name: functionName
850
+ Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
869
851
EvaluationPeriods: 20
870
852
DatapointsToAlarm: 15
871
- ComparisonOperator: GreaterThanUpperThreshold
853
+ ComparisonOperator: GreaterThanThreshold
872
854
TreatMissingData: notBreaching
873
- Metrics:
874
- - Id: m1
875
- ReturnData: true
876
- MetricStat:
877
- Metric:
878
- Namespace: Javabuilder
879
- MetricName: TransitionTime
880
- Dimensions:
881
- - Name: functionName
882
- Value: !Ref BuildAndRunJava<%= name%> ProjectFunction
883
- Period: 60
884
- Stat: p95
885
- - Id: ad1
886
- Label: TransitionTime (expected)
887
- ReturnData: true
888
- Expression: ANOMALY_DETECTION_BAND(m1, 5)
889
- ThresholdMetricId: ad1
855
+ Threshold: 2500
856
+ Period: 60
890
857
891
858
<%= name%> HighInvocationsAlarm:
892
859
Type: AWS::CloudWatch::Alarm
0 commit comments