Skip to main content

ECS container network latency

Last updated on

ECS container network latency is an AWS chaos fault that adds NETWORK_LATENCY milliseconds of one-way latency on the network interface NETWORK_INTERFACE inside a percentage of running ECS tasks (EC2 launch type) for a configurable duration. The fault uses AWS Systems Manager Run Command on the EC2 host of each selected task and applies the latency inside the affected container, scoped via ECS container metadata.

Use this fault to test how a workload behaves when network latency rises between an ECS task and its dependencies: whether timeouts and retries are sized correctly, whether circuit breakers engage, and whether latency-sensitive features (synchronous calls, streaming) degrade gracefully or fail outright.

Run your first experiment

If you have not configured the chaos infrastructure yet, go to Quickstart to install the chaos infrastructure and run an experiment end to end.

Use cases​

Run this fault when you want to answer concrete questions like:

  • Timeout sizing: When the network slows by NETWORK_LATENCY ms, do client timeouts fire correctly without cascading failures?
  • Retry storms: Do retries amplify the latency into an outage, or does an exponential backoff calm the system?
  • User-perceived latency: Does the application surface user-visible latency increases, and do dashboards reflect them?
  • Synchronous-call chains: For services with deep synchronous call graphs, does the added latency push end-to-end latency over SLO?
  • Streaming workloads: Do long-lived streams (gRPC, WebSocket) tolerate the latency change, or do they drop?

Prerequisites​

  • Kubernetes version: 1.21 or later for the chaos infrastructure cluster. Go to What's supported to confirm distribution support.
  • Target ECS service or cluster: CLUSTER_NAME exists in REGION and uses the EC2 launch type.
  • Container instances are SSM-managed: Every EC2 container instance is registered with AWS Systems Manager.
  • ECS container metadata enabled.
  • AWS credentials available: Either an AWS credentials file uploaded as a File Secret in Harness Secret Manager (see Authentication below) or an IAM role for service accounts (IRSA) bound to the chaos infrastructure service account.
  • IAM permissions granted: The credentials or role include the permissions listed below.

Supported environments​

PlatformSupport status
Amazon ECS on EC2 launch typeSupported
Amazon ECS on Fargate launch typeNot supported
Linux container instancesSupported
Windows container instancesNot supported
AWS regionsSupported in every commercial region; pass the region in REGION

Permissions required​

The IAM principal that the chaos pod uses (the credentials mounted from the Harness Secret Manager file secret, the IRSA role on the chaos service account, or the role assumed via ASSUME_ROLE_ARN) needs the following AWS actions.

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "ecs:DescribeClusters",
        "ecs:DescribeServices",
        "ecs:DescribeTasks",
        "ecs:ListTasks",
        "ecs:ListContainerInstances",
        "ecs:DescribeContainerInstances"
      ],
      "Resource": "*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "ssm:SendCommand",
        "ssm:CancelCommand",
        "ssm:GetCommandInvocation",
        "ssm:DescribeInstanceInformation"
      ],
      "Resource": "*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "ec2:DescribeInstances",
        "ec2messages:AcknowledgeMessage",
        "ec2messages:DeleteMessage",
        "ec2messages:FailMessage",
        "ec2messages:GetEndpoint",
        "ec2messages:GetMessages",
        "ec2messages:SendReply"
      ],
      "Resource": "*"
    }
  ]
}

Go to common policy for all AWS faults to use a single superset IAM policy across every AWS fault.

Authentication​

The fault supports three credential delivery models. Pick one based on how your chaos infrastructure is deployed.

MethodWhen to use itHow to configure
Harness Secret Manager file secretChaos infrastructure runs outside EKS, or you want explicit static credentialsUpload the AWS credentials file as a File Secret in Harness Secret Manager and reference its identifier via AWS_AUTHENTICATION_SECRET
IAM Roles for Service Accounts (IRSA)Chaos infrastructure runs in EKS and uses an OIDC-bound service accountNo tunable changes; the chaos pod inherits the role automatically. Go to AWS IAM integration to set it up
Assume roleThe fault needs to act in a different account or with elevated permissionsSet ASSUME_ROLE_ARN to the role ARN; the chaos pod assumes the role on top of its base credentials

When using the Harness Secret Manager method, the contents of the File Secret should be the AWS credentials file in the standard ~/.aws/credentials format:

[default]
aws_access_key_id = REPLACE_WITH_ACCESS_KEY_ID
aws_secret_access_key = REPLACE_WITH_SECRET_ACCESS_KEY

Upload this file as a File Secret in Harness Secret Manager (Project Setup → Secrets → New File Secret), and pass the secret identifier in AWS_AUTHENTICATION_SECRET when configuring the fault.

Go to AWS named profile for chaos to switch between profiles inside a single credentials file.

Fault tunables​

Configure the following fault parameters when you add ECS container network latency to an experiment in Chaos Studio. Defaults are shown for reference.

Required parameters

TunableDescriptionDefault
CLUSTER_NAMEName of the target ECS cluster.(required)
REGIONAWS region that hosts the ECS cluster (for example us-east-1).(required)

Targeting parameters

TunableDescriptionDefault
SERVICE_NAMEName of the target ECS service. When set, the fault selects TASK_REPLICA_AFFECTED_PERC of the service's running tasks.""
TASK_REPLICA_AFFECTED_PERCPercentage of running tasks to affect when SERVICE_NAME is set.100

Chaos parameters

TunableDescriptionDefault
NETWORK_LATENCYAmount of one-way latency to inject (in milliseconds).2000
NETWORK_INTERFACENetwork interface to apply the latency on, inside the container. auto discovers the primary egress interface automatically.auto
TOTAL_CHAOS_DURATIONDuration of the fault in seconds.30
INSTALL_DEPENDENCIESInstall the networking tooling on each container instance if missing.true
DEFAULT_HEALTH_CHECKWhen true, the fault runs additional checks against ECS to verify task health.false
SEQUENCEOrder in which multiple tasks are stressed: parallel issues the rule on all selected tasks at once; serial does so one at a time.parallel
RAMP_TIMEWait period in seconds before and after the fault. Go to ramp time to read how it is applied.0

Authentication

TunableDescriptionDefault
ASSUME_ROLE_ARNARN of an IAM role to assume on top of the base credentials. Leave empty to use the base credentials directly.""
AWS_AUTHENTICATION_SECRETIdentifier of the File Secret in Harness Secret Manager that contains the AWS credentials file. Not required when using IRSA.""

Tunables that apply to every fault are documented in common tunables for all faults. AWS-specific shared tunables are documented in common AWS fault tunables.

Fault execution in brief​

Resolves the running tasks for SERVICE_NAME (or all tasks in CLUSTER_NAME), picks TASK_REPLICA_AFFECTED_PERC of them, and dispatches a network-latency rule via AWS Systems Manager Run Command to the EC2 host of each selected task. The rule is applied inside the target container on NETWORK_INTERFACE and held for TOTAL_CHAOS_DURATION seconds, then reversed.

Expected behavior during fault execution​

  • New network packets leaving the affected containers via NETWORK_INTERFACE are delayed by NETWORK_LATENCY ms.
  • End-to-end request latency rises by approximately NETWORK_LATENCY ms per hop traversed by the affected interface.
  • Application timeouts that were sized below the added latency begin to fire.
  • CloudWatch latency metrics (where exported) rise on the affected service.
When the fault ends

The chaos pod removes the latency rule on each host. Network latency returns to baseline immediately for new packets.

Signals to watch​

Attach resilience probes to assert each layer:

  • Application latency: Use an HTTP probe and assert percentile latency rises by approximately NETWORK_LATENCY ms during the fault.
  • Timeout failures: Use a Prometheus probe on application-level timeout counters.
  • Network metrics: Use a Prometheus probe on aws_ecs_network_tx_bytes and aws_ecs_network_rx_bytes to confirm traffic continues but slowly.
  • Downstream dependency calls: Use a Prometheus probe on retry counters to detect retry storms.

Verify the fault execution effect​

While the experiment is running, confirm latency rose and recovered:

  1. Probe from a peer.

    curl -w "@curl-format.txt" -o /dev/null -s http://<service-endpoint>/healthz

    Request time should rise by approximately NETWORK_LATENCY ms during the chaos window.

  2. Inspect the latency rule on the host (via SSM).

    aws ssm send-command \
      --region <region> \
      --document-name AWS-RunShellScript \
      --instance-ids <container-instance-id> \
      --parameters 'commands=["tc qdisc show dev '"$NETWORK_INTERFACE"'"]'

    Output should show a netem rule during the chaos window and no chaos rule afterwards.

  3. Inspect SSM command status.

    aws ssm list-command-invocations --region <region> --details --filters "key=Status,value=InProgress"

Recovery and cleanup​

  • End of duration: The chaos pod removes the latency rule on each host.
  • Abort the experiment: Stopping the experiment from Chaos Studio cancels the SSM command and removes the rule.
  • Manual recovery: If the fault exits before the rule is removed, send an SSM command to clear the rule on the affected interface.
  • Workload recovery: Long-lived TCP connections may benefit from a reset after the latency disappears if congestion windows are stuck at the higher RTT estimate.

Limitations​

  • EC2 launch type only: The fault uses SSM Run Command against the underlying EC2 host. Fargate is not supported.
  • Container metadata must be enabled.
  • SSM-managed hosts only.
  • Linux-only.
  • One direction only: Latency is applied on egress from the container on NETWORK_INTERFACE; return-path delays from peers are unaffected. Combine with a peer-side fault for round-trip emulation.
  • Cross-region targeting: A single experiment targets one region (the value of REGION).

Troubleshooting​

ECS container network latency fails with AccessDeniedException in Harness Chaos Engineering

The credentials supplied to the chaos pod do not have the required ECS or SSM permissions. Confirm the IAM policy attached to the user, role, or IRSA service account includes ecs:DescribeServices, ecs:DescribeTasks, ssm:SendCommand, and ssm:GetCommandInvocation.

ECS container network latency reports interface not found

NETWORK_INTERFACE=auto attempts to find the container's primary egress interface; if discovery fails, supply an explicit name (typically 'eth0' inside the container). Confirm the interface exists with 'ip link show' inside the container (via SSM exec). For tasks with multiple ENIs, choose the correct interface for the traffic you want to slow.

Latency rule applied but client latency does not rise

The most common causes are: the affected interface is not the path the traffic takes (set NETWORK_INTERFACE explicitly); the rule was applied to a sibling task on the same host (verify with the SSM command output); the load balancer routed traffic away from the affected tasks because of failing health checks; or the application uses a long-lived TCP connection that was already established before the rule (new packets on the existing connection still see the latency, but the application may be CPU-bound on a single connection).

Latency rule remains after the chaos window

If the cleanup SSM command failed (timeout, transient SSM error), the rule may persist. Send an SSM AWS-RunShellScript command to the affected container instance that removes the tc qdisc rule from the root of the chosen network interface. The exact command is recorded in the chaos pod logs.