API Error Rates and Metric Delays

6:38 AM PST We are investigating increased error rates for CloudWatch PutMetricData API in US-EAST-1 Region.

8:40 AM PST We can confirm elevated PutMetricData API error rates and some delayed metrics in US-EAST-1 Region. We are actively working to resolve the issue.

1:49 PM PST We are still seeing elevated PutMetricData API error rates and some delayed metrics in US-EAST-1 Region. We continue to work toward full recovery.

3:42 PM PST We continue to experience increased error rates impacting the PutMetricData API error rates, and delayed Cloudwatch metrics. We expect to begin seeing recovery once the on-going Kinesis issue is fully resolved. Once we have observed recovery, metrics will backfill if they were queued during the impact

5:31 PM PST We continue experience API error rates and delayed Cloudwatch metrics. We expect to begin seeing recovery once the on-going Kinesis issue is fully resolved. Once we have observed recovery, metrics will backfill if they were queued during the impact.

8:18 PM PST We continue to experience API error rates and delayed CloudWatch metrics. We expect to see recovery once the on-going Kinesis issue is fully resolved. Metrics will start to backfill once the Kinesis issue is resolved.

10:44 PM PST We are beginning to see recovery in API error rates and delayed metrics as Kinesis recovers.

10:56 PM PST Between 5:15 AM and 10:31 PM PST, some customers experienced elevated error rates when calling CloudWatch APIs in the US-EAST-1 Region. Metrics were delayed, and CloudWatch alarms on delayed metrics transitioned into the INSUFFICIENT_DATA state. We have resolved the issue. Delayed metrics are in the process of backfilling in CloudWatch console graphs and for API retrieval. The service is operating normally.