Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

CloudSQL instance High CPU Ussage after upgrade

High CPU usage (aprox 18% stable in southamerica-west1 and aprox 12% in us-central1) and high network egress after Cloud SQL MySQL instance upgrading from version 8.0.33 to 8.0.34 with low or none workload.
I used To have aroung 7% CPU usage in all instances before upgrading. I raised an issue and the assigned agent marked as WAI (even I provided screenshots with before and after upgrading)

Does anyone knows how to fix this?

Solved Solved
1 3 1,109
1 ACCEPTED SOLUTION

I understand your concern and the frustration this issue is causing. I've taken note of the steps you've already tried, including creating a new Cloud SQL MySQL instance from scratch.

Firstly, let's try a few more troubleshooting steps:

  1. Review the Cloud SQL Logs tab for any errors or warnings that could hint at the cause of the high CPU usage.
  2. Use the Cloud SQL Performance Insights tool for a performance analysis, which might highlight bottlenecks or other issues.
  3. Examine your database configuration to ensure it's optimized, considering parameters like innodb_buffer_pool_size and max_connections.

If these steps don't provide a solution, I recommend opening another support ticket with Google Cloud, especially given the new evidence you've gathered. A dedicated engineer should be able to delve deeper into the issue with the added information you can provide.

View solution in original post

3 REPLIES 3

 

To troubleshoot the issue, you might want to start by ensuring that your database configuration is optimized for your workload. For example, you may need to adjust the values of certain configuration parameters, such as innodb_buffer_pool_size and max_connections. Additionally, make sure to use indexes to improve the performance of your queries and avoid running long-running queries. Using a connection pool can also help reduce the number of connections to your database.

If you have tried the above and are still experiencing high CPU usage, you can run a performance analysis using the Cloud SQL Performance Insights tool. This tool can help you identify any bottlenecks or other performance issues. Additionally, check the Cloud SQL logs by going to the Logs tab in the Cloud SQL console. The logs may contain information about what is causing the high CPU usage and network egress.

Here are some additional tips for reducing CPU usage and network egress in Cloud SQL:

  • Use a caching layer to reduce the number of database queries.
  • Although it may not directly affect the CPU usage of your Cloud SQL instance, using a content delivery network (CDN) to serve static content, such as images and CSS files, can help reduce overall network egress.

Hi there, like I said, is not a configuration or high workload in those instances, in fact, they are almost iddle. I even tried to create a new instance from scratch and did not created any database, did not connected to the new instance at all, and since the beginning the CPU ussage was too high compared with the MySQL version 8.0.33. Please a trully would like to ask somebody to take a look on this, Is not working as expected (at least in southamerica-west1 or us-central1)

I understand your concern and the frustration this issue is causing. I've taken note of the steps you've already tried, including creating a new Cloud SQL MySQL instance from scratch.

Firstly, let's try a few more troubleshooting steps:

  1. Review the Cloud SQL Logs tab for any errors or warnings that could hint at the cause of the high CPU usage.
  2. Use the Cloud SQL Performance Insights tool for a performance analysis, which might highlight bottlenecks or other issues.
  3. Examine your database configuration to ensure it's optimized, considering parameters like innodb_buffer_pool_size and max_connections.

If these steps don't provide a solution, I recommend opening another support ticket with Google Cloud, especially given the new evidence you've gathered. A dedicated engineer should be able to delve deeper into the issue with the added information you can provide.