Ctrlk
Sign UpGet a DemoContact Us
For the complete documentation index, see llms.txt. This page is also available as Markdown.

EC2 memory hog

Consume a configurable amount of memory inside a target EC2 instance via AWS Systems Manager so you can test how the workload behaves when the host is starved of memory.

EC2 memory hog is an AWS chaos fault that consumes a configurable amount of memory (in MB or as a percentage of total) inside a target EC2 instance for a configurable duration. The fault dispatches the stress workload via AWS Systems Manager Run Command, so the target instance must have the SSM Agent installed and an IAM role that permits SSM messages.

Use this fault to test how a workload reacts when its host runs out of memory: do OOM kills land on the right processes, do JVM heap sizes hit hard limits, does autoscaling provision replacement capacity in time?

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:

  • OOM-killer behaviour: When the host is starved of memory, does the OOM-killer pick the right victim, or does it kill the application instead of a noisy neighbour?

  • JVM heap pressure: Does a Java application surface OutOfMemoryError cleanly, or does it freeze in GC?

  • Container memory limits: For containerized workloads, do cgroup memory limits trigger pod restarts as expected?

  • Autoscaling on memory: Do CloudWatch memory alarms (custom metric or CloudWatch Agent) trigger ASG scale-out in time?

  • Co-tenant isolation: When one workload consumes excess memory, do neighbouring workloads stay healthy?

Prerequisites

  • Kubernetes version: 1.21 or later for the chaos infrastructure cluster.

  • Target instance is reachable via SSM: The instance has the SSM Agent running and an instance profile with AmazonSSMManagedInstanceCore.

  • Selector provided: Either EC2_INSTANCE_ID or EC2_INSTANCE_TAG is set.

  • AWS credentials available: Either an AWS credentials file uploaded as a File Secret in Harness Secret Manager (see Authentication below) or IRSA on the chaos infrastructure service account.

Supported environments

Platform
Support status

Amazon EC2 (Linux instances with SSM Agent)

Supported

Amazon EKS managed worker nodes

Supported (if SSM Agent is installed)

Amazon EKS self-managed worker nodes

Supported (if SSM Agent is installed)

Targeting by tag

Supported via EC2_INSTANCE_TAG

Targeting by ID

Supported via EC2_INSTANCE_ID

Windows instances

Not supported (Linux-only stress payload)

Permissions required

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

Authentication

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

Method
When to use it
How to configure

Harness Secret Manager file secret

Chaos infrastructure runs outside EKS, or you want explicit static credentials

Upload 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 account

No tunable changes; the chaos pod inherits the role automatically. Go to AWS IAM integration to set it up

Assume role

The fault needs to act in a different account or with elevated permissions

Set 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 File Secret should contain an AWS credentials file in the standard ~/.aws/credentials format:

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.

Fault tunables

Required parameters

Tunable
Description
Default

REGION

AWS region that hosts the target instance.

(required)

EC2_INSTANCE_ID or EC2_INSTANCE_TAG

One of these must be set to select the target instance(s).

""

Chaos parameters

Tunable
Description
Default

MEMORY_CONSUMPTION

Memory to consume in MB. Mutually exclusive with MEMORY_PERCENTAGE.

100

MEMORY_PERCENTAGE

Percentage of total memory to consume (1-100). When set, overrides MEMORY_CONSUMPTION.

""

NUMBER_OF_WORKERS

Number of worker threads to spawn for memory stress.

1

TOTAL_CHAOS_DURATION

Duration of the fault in seconds.

60

INSTANCE_AFFECTED_PERC

Percentage of matching instances to target (only with EC2_INSTANCE_TAG). 0 targets one instance.

0

INSTALL_DEPENDENCIES

Install the in-instance stress tool if missing. Set to False to skip.

True

PROXY

HTTP/HTTPS proxy used by the in-instance installer (for example https_proxy=http://proxy.server:3128).

""

SEQUENCE

Order in which multiple instances are processed: parallel or serial.

parallel

RAMP_TIME

Wait period in seconds before and after the fault.

0

Authentication

Tunable
Description
Default

ASSUME_ROLE_ARN

ARN of an IAM role to assume on top of the base credentials.

""

AWS_AUTHENTICATION_SECRET

Identifier of the File Secret in Harness Secret Manager that contains the AWS credentials file. Not required when using IRSA.

""

CHOOSE MEMORY_PERCENTAGE OR MEMORY_CONSUMPTION

Use MEMORY_PERCENTAGE for instance-size-independent testing (for example "consume 80% of memory regardless of host size"). Use MEMORY_CONSUMPTION when you need an absolute MB target to hit a known OOM threshold.

Fault execution in brief

Sends an SSM Run Command to the selected instance(s) in REGION that allocates the requested memory (MEMORY_CONSUMPTION MB or MEMORY_PERCENTAGE of total) across NUMBER_OF_WORKERS worker threads for TOTAL_CHAOS_DURATION seconds.

Expected behavior during fault execution

  • Memory utilization on the target instance rises to the requested level.

  • If the requested allocation exceeds available memory, the OOM-killer activates and terminates the highest-scoring victim process (typically the stress workload itself, but possibly the application if the application has a higher OOM score).

  • Swap usage rises if swap is configured; otherwise allocations fail with ENOMEM.

  • Application latency typically rises due to memory pressure on the page cache.

WHEN THE FAULT ENDS

The chaos pod terminates the stress workload. Memory is released back to the kernel within seconds. Pages evicted from the page cache during the fault are repopulated lazily as the workload accesses them.

Signals to watch

  • Memory utilization: Use a Prometheus probe on the CloudWatch Agent's mem_used_percent metric (CloudWatch does not expose memory by default).

  • OOM events: Use a command probe that greps dmesg for Killed process lines.

  • Application availability: Use an HTTP probe to detect downtime if the application is OOM-killed.

Verify the fault execution effect

While the experiment is running:

  1. Check memory via SSM.

  2. Check for OOM-killer events.

Recovery and cleanup

  • End of duration: The chaos pod terminates the stress workload; memory is released to the kernel.

  • Abort the experiment: Stopping the experiment from Chaos Studio cancels the in-flight SSM command.

  • Manual recovery: If the stress workload outlives the experiment, kill it via SSM: pkill -f stress or pkill -f <stress-binary>.

  • Application restart: If the application was OOM-killed during the fault, restart it via your normal restart workflow (systemd, kubelet, supervisor).

Limitations

  • Linux-only payload: This fault stresses Linux instances. For Windows hosts, use windows-ec2-memory-hog.

  • SSM Agent required: Instances without the SSM Agent online cannot be targeted.

  • OOM victim is not deterministic: The OOM-killer scores victims based on oom_score, which depends on process size and oom_score_adj. The stress workload usually scores highest, but a misconfigured application can also be killed.

  • Swap interactions: If the instance has swap configured, the OOM threshold is delayed and the fault may appear "soft" until swap fills.

  • CloudWatch memory metric: AWS does not expose memory utilization by default. Install the CloudWatch Agent on the target instance to track memory metrics for probes.

Troubleshooting

EC2 memory hog experiment fails with InvalidInstanceId in Harness Chaos Engineering

The SSM Agent is not online for the target instance. Confirm with aws ssm describe-instance-information --filters 'Key=InstanceIds,Values='. If the instance is missing, install the SSM Agent and attach an instance profile that includes AmazonSSMManagedInstanceCore.

EC2 memory hog runs but no memory pressure is visible

The most common causes are: the in-instance stress tool failed to install (set INSTALL_DEPENDENCIES=True and check network egress from the instance, or use PROXY); MEMORY_CONSUMPTION is far below available memory; or you are reading default CloudWatch metrics which do not include memory. Install the CloudWatch Agent on the instance to expose mem_used_percent, or run 'free -h' via SSM during the fault.

EC2 memory hog killed the application instead of the stress workload

The OOM-killer chose your application because it scored highest on oom_score. Set the application's oom_score_adj to a negative value (for example -500) via systemd's OOMScoreAdjust, or run it under a cgroup with a memory.high limit lower than the host. Re-run with a smaller MEMORY_CONSUMPTION first to confirm the fix.

Related faults

Last updated

Was this helpful?