Recipes for S3 monitoring
0
Are there any recommendations / recipes for S3 Request metrics monitoring?
Customer wants to use CW alarms particularly for the PutRequests, GetRequests, 4xxErrors, and DeleteRequests metrics, the challenge is trying to avoid false positives.
Any advices are appreciated. Thank you
Language
English
asked 5 months ago218 views
1 Answer
- Newest
- Most votes
- Most comments
1
Are you considering S3 Storage Lens as a source for these metrics? Once you will be able to publish these statistics to CloudWatch for PutRequests, GetRequests, 4xxErrors, and DeleteRequests metrics, you will be able to create your alarms. What scenarios they want the Alarm to be triggered for on each of these API requests? Anomalies?
answered 5 months ago
Relevant content
- Accepted Answerasked a month ago
- asked 7 months ago
- asked 10 months ago
AWS OFFICIALUpdated 3 years ago- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated a year ago
Haven´t considered CW Storage Lens, only S3 Request Metrics, I will take a look on that. The main objective is to establish a baseline for normal 4xx error rates, and then investigate anomalies, which could be a sign of a configuration problem, networking issue, or other performance disruption. They are looking for tips to properly configure alarms thresholds in order to avoid false positives.
Sorry, I just updated that. I meant to say S3 Storage Lens. I will say explore anomaly detection as threshold for CloudWatch alarm if they can't come up with baseline threshold. For errors, I will start by visualizing the trend, Statistics and percentiles. They can consider metric math for that. I will say start by analyzing the current baseline as ideal performance, and then determine a value that drifts from the baseline. However, we have some recommendations for thresholds based on different metrics here: https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/Best_Practice_Recommended_Alarms_AWS_Services.html
Thanks, that helps a lot. I will take a look.