Database High Peak CPU utilization

Since the 25 of April, it seems we are facing a constant degradation of the database.

Currently, we are reaching around 90% of CPU utilization.

We need to execute a task force to understand what we changed or what we need to rollback to reduce this load to average 50%.

we are at risk of PostgreSQL having a poor performance that we will end up in a shutdown or really slow down all the queries.

Change-related on this date: https://gitlab.com/gitlab-com/gl-infra/chef-repo/-/merge_requests/1716

Production incident related: production#7034 (closed)

Acceptance criteria:

• Investigate and list the changes to revert or improve to reduce the high CPU utilization on the primary database