AWS Cloudwatch Monitoring

// Spring Boot - Micrometer integration with CloudWatch
@Service
public class PoliticianMetricsService {
    private final MeterRegistry meterRegistry;
    
    @Autowired
    public PoliticianMetricsService(MeterRegistry meterRegistry) {
        this.meterRegistry = meterRegistry;
    }
    
    public void recordRiskScoreCalculation(String politicianId, double score) {
        // Custom metric: risk score distribution
        meterRegistry.gauge("cia.politician.risk_score", 
            Tags.of("politician_id", politicianId), 
            score);
            
        // Counter: total risk calculations
        meterRegistry.counter("cia.risk_calculations.total",
            Tags.of("result", score > 70 ? "high" : "normal"))
            .increment();
    }
    
    public void recordDataSourceRefresh(String source, boolean success) {
        // Timer: data source refresh duration
        Timer.builder("cia.datasource.refresh.duration")
            .tag("source", source)
            .tag("status", success ? "success" : "failure")
            .register(meterRegistry)
            .record(() -> {
                // Refresh logic here
            });
    }
}

# CloudFormation/Terraform - Critical Alarms
Resources:
  HighErrorRateAlarm:
    Type: AWS::CloudWatch::Alarm
    Properties:
      AlarmName: CIA-HighErrorRate
      AlarmDescription: Alert when error rate exceeds 5%
      MetricName: Errors
      Namespace: AWS/Lambda
      Statistic: Sum
      Period: 300
      EvaluationPeriods: 2
      Threshold: 50
      ComparisonOperator: GreaterThanThreshold
      AlarmActions:
        - !Ref SNSTopicARN
      
  DatabaseConnectionPoolExhausted:
    Type: AWS::CloudWatch::Alarm
    Properties:
      AlarmName: CIA-DBConnectionPoolExhausted
      MetricName: DatabaseConnectionsInUse
      Namespace: CIA/Database
      Statistic: Maximum
      Period: 60
      EvaluationPeriods: 1
      Threshold: 90  # 90% of max connections
      ComparisonOperator: GreaterThanThreshold
      
  RiskScoreCalculationLatency:
    Type: AWS::CloudWatch::Alarm
    Properties:
      AlarmName: CIA-RiskScoreHighLatency
      MetricName: RiskScoreCalculationDuration
      Namespace: CIA/Application
      Statistic: p99
      Period: 300
      EvaluationPeriods: 2
      Threshold: 5000  # 5 seconds
      ComparisonOperator: GreaterThanThreshold

{
  "widgets": [
    {
      "type": "metric",
      "properties": {
        "title": "CIA Platform Health",
        "metrics": [
          [ "CIA/Application", "RequestCount", { "stat": "Sum" } ],
          [ ".", "ErrorCount", { "stat": "Sum", "color": "#d62728" } ],
          [ ".", "ResponseTime", { "stat": "Average" } ]
        ],
        "period": 300,
        "region": "eu-north-1",
        "yAxis": {
          "left": { "min": 0 }
        }
      }
    },
    {
      "type": "metric",
      "properties": {
        "title": "Political Data Processing",
        "metrics": [
          [ "CIA/DataIngestion", "RiksdagenAPICallsTotal" ],
          [ ".", "ValDataRefreshSuccess" ],
          [ ".", "WorldBankDataSync" ]
        ]
      }
    },
    {
      "type": "log",
      "properties": {
        "title": "Recent Errors",
        "query": "SOURCE '/aws/lambda/cia-app' | fields @timestamp, @message | filter @message like /ERROR/ | sort @timestamp desc | limit 20",
        "region": "eu-north-1"
      }
    }
  ]
}

-- Top 10 slowest API endpoints
fields @timestamp, request.path, request.duration_ms
| filter request.duration_ms > 1000
| sort request.duration_ms desc
| limit 10

-- Error rate by endpoint
fields request.path, response.status_code
| filter response.status_code >= 500
| stats count() as error_count by request.path
| sort error_count desc

-- Risk score calculation performance
fields @timestamp, politician_id, calculation_duration_ms
| filter event_type = "risk_score_calculated"
| stats avg(calculation_duration_ms) as avg_duration, 
        max(calculation_duration_ms) as max_duration,
        count() as total_calculations
  by bin(5m)