If the tracking database is not configured, the secondary attempts to read from
the Geo DB but fails spectacularly because the tables don't exist.

Fixes transient failure: waits for asyncronous update

Closes #2079

See merge request !1567

Port 'Refactor test_utils bundle' to EE

See merge request !1563

Fix sticking of the database load balancer

See merge request !1502

CE upstream: Tuesday

Closes gitaly#166 and gitaly#168

See merge request !1565

Port of 28732-expandable-folders to EE

See merge request !1538

Allow to edit build minutes per user ee

Closes gitlab-ce#30163

See merge request !1535

remove useless queries with false conditions (e.g 1=0)

Closes #29492

See merge request !10141

CE Upstream - Monday

Closes gitaly#148

See merge request !1560

Don't autofill kubernetes namespace

See merge request !10438

Backport differences in global search from EE to CE

See merge request !10445

Improves Jira integration documentation

Closes #29670

See merge request !10067

Add long polling diagrams

See merge request !10270

Update patch_versions.md, add `yarn:install` in `Clean up assets and cache` command.

See merge request !10442

Remove index for users.current sign in at

Closes #29523

See merge request !10401

Split status and confidentiality action

See merge request !10443

Fix elasticsearch global code search following https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/9655

There were 3 problems with sticking:

1. When sticking to a request (based on a previous request) we would
   always overwrite the WAL pointer, leading to a user being stuck to
   the primary for too long.

2. Sticking was not working for the Grape API. The API in particular is
   tricky because we have to stick either using a user, CI runner, or a
   CI build.

3. Refreshing of permissions did not lead to the refreshed users being
   stuck to the primary. This could result in users not being able to
   access new resources for a brief moment of time, or them still being
   able to see old resources.

To solve this some of the Grape related logic is handled by injecting EE
specific modules in the right places. This ensures we can stick as early
as possible, using the right data. The same technique is applied for
various parts of the CI codebase.

Backport API changes needed to fix sticking in EE

See merge request !10376

Ensure we generate unique usernames otherwise validations fail

See merge request !10436

Fix a Knapsack issue that would load support/capybara.rb before support/env.rb

See merge request !10430

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Rémy Coutable <remy@rymai.me>

Port of fix-github-importer-slowness to EE

See merge request !1561

Don't use FFaker in factories, use sequences instead

See merge request !1495

Don't use FFaker in factories, use sequences instead

Closes #29643

See merge request !10184

Ensure users have a short username otherwise a click event is triggered outside the search field

Closes #29241

See merge request !10411

Quiet pipeline emails

Closes #24845

See merge request !10333

Once https://github.com/ArturT/knapsack/issues/57 is solved and released
in a new gem version, we can remove the KNAPSACK_TEST_DIR in
.gitlab-ci.yml.
Signed-off-by: Rémy Coutable <remy@rymai.me>