AppRunner showing increased CPU consumption

Worth taking a look at this as well:

https://docs.aws.amazon.com/apprunner/latest/dg/apprunner-availability-change.html

https://repost.aws/knowledge-center/elasticache-redis-high-cpu-usage

Based on the available information, there was a significant change to how CPU utilization metrics are calculated and reported in App Runner. Prior to August 23, 2023, CPU utilization metrics were based on vCPU units utilized rather than percentage utilization as they are calculated today. If you're viewing metrics from around that transition period, you may see display irregularities and metrics shown in both units.

However, regarding your current situation with increased CPU consumption this week, there are a few things to consider:

1. Metric Calculation Differences: There can be discrepancies between CloudWatch metrics (which are collected at the hypervisor level) and what you might observe at the application level. CloudWatch samples at 1-5 minute intervals, which could show different patterns than second-by-second data.

2. Recent Runtime Changes: There were performance changes announced for certain runtimes starting June 17, 2024, which primarily affected build times (approximately 240 seconds increase) for older runtimes like Node.js 12/14, Python 3.7/3.8, and Corretto 8/11. While this was specifically about build times, it's worth noting if you're using any of these runtimes.

If your application performance isn't actually degraded and response times remain normal, the increased CPU utilization readings might be related to how the metrics are being collected or reported rather than actual performance issues. You may want to check if there have been any underlying infrastructure changes or if other customers are reporting similar observations during the same timeframe.

Since you mentioned no configuration or deployment changes on your end, this could be worth raising with AWS Support to investigate if there have been any platform-level changes affecting metric reporting.
Sources
Viewing App Runner service metrics reported to CloudWatch - AWS App Runner
Troubleshoot high CPU utilization on my EC2 instance | AWS re:Post
Performance changes for code-based services on Jun 14, 2024 - AWS App Runner

Hi @meesk,

I’ve had the same issue this week. Today, I fixed it by pushing a minor change (just some comments in the code) to trigger a fresh deployment. That clean redeploy stabilized the CPU and Memory for me.

Hope it helps!