Mayra Cabrera authored 5 years ago and Achilleas Pipinellis committed 5 years ago

Specifies that background migrations should be scheduled as post
deployment migrations

Marcel Amirault authored 5 years ago and Achilleas Pipinellis committed 5 years ago

Clean up capitalization in /development /gitlab-basics and
/install

- Also makes other minor Markdown fixes that were near the main fixes.

Extending from ActiveRecord::Migration is deprecated, but was still used
in a bunch of places.

This changes the BackgroundMigration worker so it checks for the health
of the DB before performing a background migration. This in turn allows
us to reduce the minimum interval, without having to worry about blowing
things up if we schedule too many migrations.

In this setup, the BackgroundMigration worker will reschedule jobs as
long as the database is considered to be in an unhealthy state. Once the
database has recovered, the migration can be performed.

To determine if the database is in a healthy state, we look at the
replication lag of any replication slots defined on the primary. If the
lag is deemed to great (100 MB by default) for too many slots, the
migration is rescheduled for a later point in time.

The health checking code is hidden behind a feature flag, allowing us to
disable it if necessary.

1. We can't just steal from the queue, in case there was a problem with Sidekiq.
2. We need to consider import / export.

The introduction of background migrations means some of our requirements
for online upgrades have to be adjusted / clarified.

See https://gitlab.com/gitlab-org/gitlab-ce/issues/35939 for more
information.

Background migrations can be used to perform long running data
migrations without these blocking a deployment procedure.

See MR https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/11854 for
more information.