So it's more modular and extensible

It takes a lot of lines to move a script

It assures that requests are aborted after 60 seconds, otherwise
an exception is raised. This exception is logged by Sentry, also
there is a Prometheus counter for measuring number of requests in each
state.

This sampler gathers Puma-specific metrics which can be used by
Prometheus then.

Resolves https://gitlab.com/gitlab-org/gitlab-ce/issues/60107

closes https://gitlab.com/gitlab-org/gitlab-ee/issues/11105

Before this fix, a suggestion which just removes
an empty line wasn't appliable

Storing this key in secrets.yml was a bad idea,
it would require users using HA setups to manually
replicate secrets across nodes during update,
it also needed support from omnibus package

* Revert "Generate Let's Encrypt private key"
  This reverts commit 444959bf.

* Add Let's Encrypt private key to settings
  as encrypted attribute

* Generate Let's Encrypt private key
  in database migration

If the postgres image version isn't passed to upgrades, helm will revert to the default postgres version. If it crosses incompatible version boundaries, this will break postgres horribly, as it won't be able to read the data files.

When user uses Pipelines for merge requests, the pipeline is a run on
a merge request ref instead of branch ref. However, we should send
source ref as a webhook in order to respect the original behavior.

As documented in
https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/api/project_level_variables.md#create-variable
and it's a boolean in `db/schema.rb`

When using the members/all api the same user was returned multiple times
when he was a member of the project/group and also of one of the
ancestor groups.
Now the member is returned only once giving priority to the membership
on the project and maintaining the same behaviour of the members UI.

- Add .no-tabs to login-box
- Externalize strings in common signup box
- Leverage render_if_exists
- Update PO file

Added a changelog entry for the feature

As documented in
https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/api/project_level_variables.md#create-variable
and it's a boolean in `db/schema.rb`

The circuit breaker itself was removed in 11.5, this removes the
corresponding API endpoints which returned empty data since then.

When a milestone name contained an HTML entity that would be escaped (&,
<, >), then it wasn't possible to refer to this milestone by name, or
use it in a quick action.

This already worked for labels, but not for milestones. We take care to
re-escape un-matched milestones, too.

This env was missing, causing the variable to not
be propagated to child containers and thus, be ineffective

Update documentation to reflect removal

Creates a new filename to register auth logs.
This change should allow SRE's queries to make better queries
through logging infrastructure.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/54528

In https://github.com/rails/rails/commit/83b767ce, Rails 5.1 removed
support for using a String to specify a middleware. When the
strategy_class argument is passed from the GitLab YAML config to Devise,
Devise passes the string value straight through to Rails, and GitLab
would crash with a NoMethodError inside ActionDispatch::MiddlewareStack.

To make this OmniAuth strategy work again, we normalize the arguments by
converting the strategy_class value into an actual Class.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/62216

This reverts commit 744f1f2e, reversing
changes made to c4d930e5.

This resolves an "ArgumentError: comparison of String with 0 failed"
issue where the visibility_level is stored as a string in the project
import data because the value comes directly from the Web form. This
problem happened upon creating a project from a template or restoring a
project.

We now cast the value to an integer to guard against these kinds of
failures.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/61692

This reverts merge request !27514

We kept it for smooth update only

Files in lib will be eager loaded and hence will require haml_lint to be
loaded. Since this is only a development dependency, we can't assume
this gem will be available in production, so it should never be loaded
in production.

The migration pod for the GitLab Helm Chart was failing because
of the NameError:

uninitialized constant HamlLint::Linter::NoPlainNodes::LinterRegistry

We now explicitly specify ::HamlLint::LinterRegistry to avoid resolution
errors.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/62125

When Kubernetes clusters were originally built they could only
exist at the project level, and so there was logic included
that assumed there would only ever be a single Kubernetes
namespace per cluster. We now support clusters at the group
and instance level, which allows multiple namespaces.

This change consolidates various project-specific fallbacks to
generate namespaces, and hands all responsibility to the
Clusters::KubernetesNamespace model. There is now no concept of
a single namespace for a Clusters::Platforms::Kubernetes; to
retrieve a namespace a project must now be supplied in all cases.

This simplifies upcoming work to use a separate Kubernetes
namespace per project environment (instead of a namespace
per project).

Jobs should be better isolated to avoid interference with other `image` or `before_script` statements.