Is Amazon Managed Streaming for Apache Kafka down?
Current Amazon Managed Streaming for Apache Kafka status is UP

We continue to work toward resolving the error rates and increased delays in network propagation for newly launched EC2 instances in a single Availability Zone (use1-az2) in US-EAST-1 Region. At this time, we are focusing on iteratively restoring healthy capacity, ensuring that we do so safely. We can confirm that our mitigations are effective at reducing the propagation delay. We will provide further updates on our progress by 5:00 PM PDT, as well as a clear indication when we start seeing recovery in the affected cells.

We continue to make progress towards resolving the increased networking latencies and errors affecting Availability Zones (usw2-az1 and usw2-az2) in the US-WEST-2 Region. In the last 30 minutes, we’ve applied an update to address resource contention within the subsystem responsible for network mappings propagation and are seeing early signs of improvement. We will continue to monitor before deploying this change more broadly and will continue to provide updates.

We continue to make progress towards resolving the increased networking latencies and errors affecting Availability Zones (usw2-az1 and usw2-az2) in the US-WEST-2 Region. While we continue to make progress in addressing the issue, we wanted to provide some more details on the issue. Within Amazon Virtual Private Cloud (VPC) any changes to the network configuration - including launching an EC2 instance, attaching an Elastic IP address or Elastic Network Interface - needs to be propagated to the underlying hardware to ensure that network packets can flow between source and destination. We call this network configuration “network mappings”, as it contains information about network paths or mappings. Starting at 10:00 AM PDT this morning, we have been experiencing a delay in the propagation of these mappings within a single cell (part of the Availability Zone) in usw2-az1 and usw2-az2 Availability Zones. The root cause appears to be increased load to the subsystem responsible for the handling of these network mappings. We have been working to reduce the load on this service to improve propagation times, but while we have made some progress, mapping propagation latencies have not returned to normal levels. We continue to work to identify all forms of resource contention that could be leading to load, and have a few additional updates that we are currently working on.

Once we have reduced the load on the subsystem responsible for network mapping propagation, we would expect full recovery. Since the issue only affects networking mappings as a result of changes to the network configuration, avoiding any mutating changes to your network configuration in the affected Availability Zone would limit impact.

For customers looking to fail away from the affected Availability Zone or Region, please ensure that you are not relying on mutating network configuration in the affected Availability Zones as part of that process.

We will continue to keep you updated as we make progress towards full recovery.

We continue to make progress towards resolving the increased networking latencies and errors affecting Availability Zones (usw2-az1 and usw2-az2) in the US-WEST-2 Region. While we continue to make progress in addressing the issue, we wanted to provide some more details on the issue. Within Amazon Virtual Private Cloud (VPC) any changes to the network configuration - including launching an EC2 instance, attaching an Elastic IP address or Elastic Network Interface - needs to be propagated to the underlying hardware to ensure that network packets can flow between source and destination. We call this network configuration “network mappings”, as it contains information about network paths or mappings. Starting at 10:00 AM PDT this morning, we have been experiencing a delay in the propagation of these mappings within a single cell (part of the Availability Zone) in usw2-az1 and usw2-az2 Availability Zones. The root cause appears to be increased load to the subsystem responsible for the handling of these network mappings. We have been working to reduce the load on this service to improve propagation times, but while we have made some progress, mapping propagation latencies have not returned to normal levels. We continue to work to identify all forms of resource contention that could be leading to load, and have a few additional updates that we are currently working on.

Once we have reduced the load on the subsystem responsible for network mapping propagation, we would expect full recovery. Since the issue only affects networking mappings as a result of changes to the network configuration, avoiding any mutating changes to your network configuration in the affected Availability Zone would limit impact.

For customers looking to fail away from the affected Availability Zone or Region, please ensure that you are not relying on mutating network configuration in the affected Availability Zones as part of that process.

We will continue to keep you updated as we make progress towards full recovery.