I’ve been using Gitlab for 4+ years now, with 100+ repositories. Due to the relatively high number of repositories, the memory usage of the Gitlab server has been increasing over time, reaching a maximum of 8GB of RAM.

I needed to find a way to reduce the memory usage. FAST!

I searched online and found a official Gitlab documentation that explains how to do it. But these changes come with a lot of comprimises, please read the documentation carefully, they may impact performance and stability, specially if you have a lot of traffic on your Gitlab server.

This post is an overview of another post and also my experience performing this migration.

https://medium.com/google-cloud/seamless-cloud-sql-migration-how-to-move-your-data-between-google-cloud-projects-with-ease-ee1592fd49c6

This process was performed around March/April of 2025. If you are reading this post well into 2025, you might want to check if GCP already has a direct way to do it, using Web UI or CLI, or they might have changed something since then.

For the past two years, I’ve been working with GCP and CloudSQL. I’ve had the experience of working with CloudSQL, more specifically with PostgreSQL, and it works pretty well.

If you have ever worked with a Postgres database, or any database for that matter, you know that migrating can be horrifying and give you physical pain :) (I’m exaggerating, but you get the point).

If you have do have to make a migration, on a Postgres database, there are a few ways to do it. I found a way to do it using docker containers, since I don’t want to install Postgres on my machine.

A guide on how to configure S3 backup for Longhorn

A guide on how to use Docker in WSL2