False Positive Alert for aws.lambda.enhanced.out_of_memory Metric

[mevlaneee]

Hi Team,

We have a general monitor configured to track all AWS Lambda functions using the metric aws.lambda.enhanced.out_of_memory. On 20 March, this monitor flagged one of our Lambda functions (webhooks) as experiencing an out-of-memory issue.
However, after investigating:
We have a dedicated monitor for the webhooks Lambda using the metric aws.lambda.enhanced.max_memory_used, and it has not indicated any abnormal memory usage.
We reviewed the logs in Amazon CloudWatch for that day and found no evidence of out-of-memory errors or related failures.
The Lambda function appears to be operating within expected memory limits.

On below screenshot is shown the errors coming up from the shared monitor for webhooks lambda,

While in the other hand the monitor that uses 'aws.lambda.enhanced.max_memory_used' metric shows that there were no spikes,

Also the same results coming from AWS Cloud Watch,

[litianningdatadog]

@mevlaneee Could you share a bit more detail about your setup?

• Datadog extension layer
• Runtime (Java, Go, Python, etc.)
• Environment variables in use
  Also,
• When did this issue begin—only after 3/20, or possibly earlier? If it started after 3/20, do you recall any changes made to your setup around that time?
• Is there any reproducible setup or scenario?

[mevlaneee]

• Datadog-Extension Layer version - 77, compatible architecture - x86_64,
• Runtime - .NET 8 (C#/F#/PowerShell)
• ENV Vars. - DD_API_KEY_SECRET_ARN, DD_CAPTURE_LAMBDA_PAYLOAD, DD_DATA_STREAMS_ENABLED, DD_DBM_PROPAGATION_MODE, DD_ENV, DD_FLUSH_TO_LOG, DD_MERGE_XRAY_TRACES, DD_SERVERLESS_LOGS_ENABLED, DD_SERVICE, DD_SiTE, DD_TAGS, DD_TRACE_ENABLED
• the issue began only in 3/20 and never reoccurred
• no there isn't any reproducible scenario

[litianningdatadog]

Thanks for the input. I highly recommend upgrading your Datadog extension to the latest version, which is v94 as of today.