René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Error logging is incompatible with Lambda Advanced Logging Controls","articleBody":"For Lambdas configured to log in JSON (`_AWS_LAMBDA_LOG_FORMAT == LogFormat.JSON`), error log messages contain [the property `\"log_level\": \"ERROR\"`](https://github.com/aws/aws-lambda-python-runtime-interface-client/blob/f11e7c5c5cc7caf7a44263d3978e0dfda4f15250/awslambdaric/bootstrap.py#L111). If the Lambda is configured to use [log-level filtering](https://docs.aws.amazon.com/lambda/latest/dg/monitoring-cloudwatchlogs-log-level.html) the filtering doesn't work as expected because the log-level filtering feature makes the assumption that the message contains the JSON property `level` (not `log_level`). Lambda will assign the log output the level `INFO` to messages with missing (or invalid) `level` property, i.e. all messages created by the awslambdaric function `log_error` are no longer considered errors because the property `log_level` is simply ignored by Lambda.\n\nHence, if we use log-level filtering and set the level to `WARN` (a pretty common config for production) **all error log messages created by awslambdaric (most notably uncaught handler errors) are missing in CloudWatch** because they are wrongly attributed the (default) log level `INFO`.\n\nWe should either add the property `level` to the `error_result` or replace the property `log_level` with `level` in the `error_result` created by [`log_error(...)` ](https://github.com/aws/aws-lambda-python-runtime-interface-client/blob/f11e7c5c5cc7caf7a44263d3978e0dfda4f15250/awslambdaric/bootstrap.py#L111).","author":{"url":"https://github.com/faberchri","@type":"Person","name":"faberchri"},"datePublished":"2026-04-27T20:54:25.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":1},"url":"https://github.com/202/aws-lambda-python-runtime-interface-client/issues/202"}