Increased Error Rates and Latencies

Final update

Between 11:49 PM PDT on October 19 and 2:24 AM PDT on October 20, we experienced increased error rates and latencies for AWS Services in the US-EAST-1 Region. Additionally, services or features that rely on US-EAST-1 endpoints such as IAM and DynamoDB Global Tables also experienced issues during this time. At 12:26 AM on October 20, we identified the trigger of the event as DNS resolution issues for the regional DynamoDB service endpoints. After resolving the DynamoDB DNS issue at 2:24 AM, services began recovering but we had a subsequent impairment in the internal subsystem of EC2 that is responsible for launching EC2 instances due to its dependency on DynamoDB. As we continued to work through EC2 instance launch impairments, Network Load Balancer health checks also became impaired, resulting in network connectivity issues in multiple services such as Lambda, DynamoDB, and CloudWatch. We recovered the Network Load Balancer health checks at 9:38 AM. As part of the recovery effort, we temporarily throttled some operations such as EC2 instance launches, processing of SQS queues via Lambda Event Source Mappings, and asynchronous Lambda invocations. Over time we reduced throttling of o

1. Resolved · Oct 20, 22:53 UTC
   Between 11:49 PM PDT on October 19 and 2:24 AM PDT on October 20, we experienced increased error rates and latencies for AWS Services in the US-EAST-1 Region. Additionally, services or features that rely on US-EAST-1 endpoints such as IAM and DynamoDB Global Tables also experienced issues during this time. At 12:26 AM on October 20, we identified the trigger of the event as DNS resolution issues for the regional DynamoDB service endpoints. After resolving the DynamoDB DNS issue at 2:24 AM, services began recovering but we had a subsequent impairment in the internal subsystem of EC2 that is responsible for launching EC2 instances due to its dependency on DynamoDB. As we continued to work through EC2 instance launch impairments, Network Load Balancer health checks also became impaired, resulting in network connectivity issues in multiple services such as Lambda, DynamoDB, and CloudWatch. We recovered the Network Load Balancer health checks at 9:38 AM. As part of the recovery effort, we temporarily throttled some operations such as EC2 instance launches, processing of SQS queues via Lambda Event Source Mappings, and asynchronous Lambda invocations. Over time we reduced throttling of operations and worked in parallel to resolve network connectivity issues until the services fully recovered. By 3:01 PM, all AWS services returned to normal operations. Some services such as AWS Config, Redshift, and Connect continue to have a backlog of messages that they will finish processing over
2. Update · Oct 20, 21:48 UTC
   We have restored EC2 instance launch throttles to pre-event levels and EC2 launch failures have recovered across all Availability Zones in the US-EAST-1 Regions. AWS services which rely on EC2 instance launches such as Redshift are working through their backlog of EC2 instance launches successfully and we anticipate full recovery of the backlog over the next two hours. We can confirm that Connect is handling new voice and chat sessions normally. There is a backlog of analytics and reporting data that we must process and anticipate that we will have worked through the backlog over the next two hours. We will provide an update by 3:30 PM PDT.
3. Update · Oct 20, 20:52 UTC
   We have continued to reduce throttles for EC2 instance launches in the US-EAST-1 Region and we continue to make progress toward pre-event levels in all Availability Zones (AZs). AWS services such as ECS and Glue, which rely on EC2 instance launches will recover as the successful instance launch rate improves. We see full recovery for Lambda invocations and are working through the backlog of queued events which we expect to be full processed in approximately in the next two hours. We will provide another update by 2:30 PM PDT.
4. Update · Oct 20, 20:03 UTC
   Service recovery across all AWS services continues to improve. We continue to reduce throttles for new EC2 Instance launches in the US-EAST-1 Region that were put in place to help mitigate impact. Lambda invocation errors have fully recovered and function errors continue to improve. We have scaled up the rate of polling SQS queues via Lambda Event Source Mappings to pre-event levels. We will provide another update by 1:45 PM PDT.
5. Update · Oct 20, 19:15 UTC
   We continue to observe recovery across all AWS services, and instance launches are succeeding across multiple Availability Zones in the US-EAST-1 Regions. For Lambda, customers may face intermittent function errors for functions making network requests to other services or systems as we work to address residual network connectivity issues. To recover Lambda’s invocation errors, we slowed down the rate of SQS polling via Lambda Event Source Mappings. We are now increasing the rate of SQS polling as we experience more successful invocations and reduced function errors. We will provide another update by 1:00 PM PDT.
6. Update · Oct 20, 18:22 UTC
   Our mitigations to resolve launch failures for new EC2 instances continue to progress and we are seeing increased launches of new EC2 instances and decreasing networking connectivity issues in the US-EAST-1 Region. We are also experiencing significant improvements to Lambda invocation errors, especially when creating new execution environments (including for Lambda@Edge invocations). We will provide an update by 12:00 PM PDT.
7. Update · Oct 20, 17:38 UTC
   Our mitigations to resolve launch failures for new EC2 instances are progressing and the internal subsystems of EC2 are now showing early signs of recovering in a few Availability Zones (AZs) in the US-EAST-1 Region. We are applying mitigations to the remaining AZs at which point we expect launch errors and network connectivity issues to subside. We will provide an update by 11:30 AM PDT.
8. Update · Oct 20, 17:03 UTC
   We continue to apply mitigation steps for network load balancer health and recovering connectivity for most AWS services. Lambda is experiencing function invocation errors because an internal subsystem was impacted by the network load balancer health checks. We are taking steps to recover this internal Lambda system. For EC2 launch instance failures, we are in the process of validating a fix and will deploy to the first AZ as soon as we have confidence we can do so safely. We will provide an update by 10:45 AM PDT.
9. Update · Oct 20, 16:13 UTC
   We have taken additional mitigation steps to aid the recovery of the underlying internal subsystem responsible for monitoring the health of our network load balancers and are now seeing connectivity and API recovery for AWS services. We have also identified and are applying next steps to mitigate throttling of new EC2 instance launches. We will provide an update by 10:00 AM PDT.
10. Update · Oct 20, 15:43 UTC
    We have narrowed down the source of the network connectivity issues that impacted AWS Services. The root cause is an underlying internal subsystem responsible for monitoring the health of our network load balancers. We are throttling requests for new EC2 instance launches to aid recovery and actively working on mitigations.
11. Update · Oct 20, 15:04 UTC
    We continue to investigate the root cause for the network connectivity issues that are impacting AWS services such as DynamoDB, SQS, and Amazon Connect in the US-EAST-1 Region. We have identified that the issue originated from within the EC2 internal network. We continue to investigate and identify mitigations.
12. Update · Oct 20, 14:29 UTC
    We have confirmed multiple AWS services experienced network connectivity issues in the US-EAST-1 Region. We are seeing early signs of recovery for the connectivity issues and are continuing to investigate the root cause.
13. Update · Oct 20, 14:14 UTC
    We can confirm significant API errors and connectivity issues across multiple services in the US-EAST-1 Region. We are investigating and will provide further update in 30 minutes or soon if we have additional information.
14. Update · Oct 20, 13:42 UTC
    We have applied multiple mitigations across multiple Availability Zones (AZs) in US-EAST-1 and are still experiencing elevated errors for new EC2 instance launches. We are rate limiting new instance launches to aid recovery. We will provide an update at 7:30 AM PDT or sooner if we have additional information.
15. Update · Oct 20, 12:48 UTC
    We are making progress on resolving the issue with new EC2 instance launches in the US-EAST-1 Region and are now able to successfully launch new instances in some Availability Zones. We are applying similar mitigations to the remaining impacted Availability Zones to restore new instance launches. As we continue to make progress, customers will see an increasing number of successful new EC2 launches. We continue to recommend that customers launch new EC2 Instance launches that are not targeted to a specific Availability Zone (AZ) so that EC2 has flexibility in selecting the appropriate AZ. We also wanted to share that we are continuing to successfully process the backlog of events for both EventBridge and Cloudtrail. New events published to these services are being delivered normally and are not experiencing elevated delivery latencies. We will provide an update by 6:30 AM PDT or sooner if we have additional information to share.
16. Update · Oct 20, 12:10 UTC
    We confirm that we have now recovered processing of SQS queues via Lambda Event Source Mappings. We are now working through processing the backlog of SQS messages in Lambda queues.