I have working AWS cloudwatch alarm form lambda configured in the way:
Namespace: AWS/Lambda
Metric name: Errors
Statistic: Sum
Error: 5min
For conditions with static threshold >= 1
Got the alarm triggered and I cannot find what caused it to rise in the logs as they are flooded with messages that contain an "Error" phrase (one of the properties that are logged is has "error" in the name).
Does anybody know what is considered an "Error" in the AWS Metric? or how can I search for that specific event that caused the metric to rise?
I have tried quering Logs Insights for that period of time with filtering
| sort @timestamp desc
| limit 20
| filter @message like /(?i)(error)/
but only got lots of messages that looks the same and if they would be counted in the metric then it would be thousands of errors not just 1