How to Troubleshoot Sudden RDS ACU Spike on AWS

Select Language:

If you’ve noticed a sudden spike in activity on your RDS Serverless v2 instance around mid-August and are unsure why it’s happening, you’re not alone. Even if your database usually sees light internal Amazon queries, an unexpected increase can cause high CPU utilization, making it seem like your server is overwhelmed.

Here’s a simple step-by-step approach to troubleshoot and handle this situation:

First, check the query logs. You might see multiple internal queries related to replica host status or heartbeat updates. These are normal in building or maintaining replica syncs, but sometimes, they can increase unexpectedly.

Next, try turning off specific monitoring tools like Enhanced Monitoring and Performance Insights to see if they are contributing to the load. If the CPU usage remains high—say above 50%—it’s a sign that the issue isn’t just monitoring overhead.

It’s also a good idea to manually inspect recent database logs around the time of the spike. Look for specific queries or processes that might have triggered increased activity. Sometimes, internal processes or Amazon’s maintenance tasks can cause short-term surges.

If after these steps the load remains high, consider scaling your capacity temporarily or configuring read replicas to distribute the load better. Also, review your application’s query patterns to ensure they are optimized and not causing unnecessary internal requests.

In some cases, the best course of action is reaching out to Amazon support to verify if there’s an ongoing incident or maintenance activity that coincides with your usage spike. They can provide insight into any internal changes or updates affecting your database.

Managing unexpected capacity changes can be challenging, but regular monitoring and understanding internal query patterns are key to maintaining a healthy RDS environment. Stay vigilant, and don’t hesitate to seek support when needed.