[ci skip]

Revert "Merge branch 'fix_stale_last_owner' into 'master'"

This reverts merge request !529

Accidentally merged my own MR before review. Reverting and resubmitting. 

See merge request !530

This reverts merge request !529

Prevent stale data in LDAP group sync last owner check

Fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/17764

Customers were reporting that LDAP group sync would occasionally remove the last owner of a group. This didn't seem right (we have tests for this) so I dug in to see what was going on. Previously `owners` was a memoized method which was the reason for stale data. Now, it's an AR association and the same problem applies. The easiest solution is to reload the model before checking for last owner. 

See merge request !529

Check fast-forward more precisely

The original issue is that (#260), some people want to be able to merge
a fast-forward MR regardless it's rebased or not. The previously
check for this: `target_sha == source_sha_parent`, assumes that the
questioning MR has a linear history (i.e. it's rebased), therefore
we could do a fast-forward merge. However, we could also do fast-forward
merge if the MR has already merged target branch (i.e. it's merged).

This MR would allow them to merge any fast-forward MR regardless rebased
or not.

But actually this breaks some assumption. In the option of:

> Merge commit with semi-linear history
>> A merge commit is created for every merge, but merging is only
>> allowed if the branch has been rebased. This way you get a history
>> that reads linearly (as with fast-forward merges), with the addition
>> of merge commits.
>> When the branch has not been rebased, the user is given the option to do
>> so.

It clearly says: `merging is only allowed if the branch has been rebased`
This MR would break this assumption. The same applies to:

> Fast-forward merge
>> No merge commits are created and all merges are fast-forwarded, which
>> means that merging is only allowed if the branch has been rebased.
>> When the branch has not been rebased, the user is given the option to do
>> so.

This means that rebase and FF are two separated concept and we're
mixing them here. We should probably update the wordings and perhaps
provide another option if we want to keep all the workflows.

My own thought is actually, allowing only FF merge might make little
sense. Allowing only FF *and* rebased (linear history) would make more
sense. That means I think the original behaviour actually makes more
sense, just that the wordings are wrong. i.e. They could be FF merged,
but that's not the point, we still want you to rebase this branch!

On the other hand, I am not against adding a new option for this
behaviour either. People have different workflows, and I think it makes
sense to allow any workflows which Git provided.

Either way, I think some wordings should be changed.

By the way, could we write a test for this?

Fixes #260

/cc @DouweM 

See merge request !454

Don't expose repository credentials when raising error on mirror update.

REF: https://gitlab.com/gitlab-com/support-forum/issues/816

See merge request !527

Mark the import as failed if a temporary or unknown error happened.

https://gitlab.com/gitlab-com/support-forum/issues/816

I've seen some users have reported that the updated of their mirror
repositories is being stuck, it basically happens because a temporary or
unknown error happened and the import status was kept in `started`
instead of being marked as `failed`.

See merge request !526

Rename git hooks to push rules

Fixes https://gitlab.com/gitlab-org/gitlab-ee/issues/189

I also created an issue for removing deprecated API https://gitlab.com/gitlab-org/gitlab-ee/issues/755

See merge request !523

Add the update guide from 8.10 CE to 8.10 EE



See merge request !524

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

I've seen some users have reported that the updated of their mirror
repositories is being stuck, it basically happens because a temporary or
unknown error happened and the import status was kept in `started`
instead of being marked as `failed`.

I've seen some users have reported that the updated of their mirror
repositories is being stuck, it basically happens because a temporary or
unknown error happened and the import status was kept in `started`
instead of being marked as `failed`.

CE upstream



See merge request !521

Remove duplicated link



See merge request !522

Remove incorrect settings

These settings do not work, in fact they prevent GitLab from booting.
They are also not needed because Kerberos integration uses
/etc/krb5.conf on the GitLab server.

See merge request !514

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

Move global ci entries handling from legacy to new config

## What does this MR do?

This MR moves responsibility of handling global CI config entries (like `image`, `services`), from legacy `GitlabCiYamlProcessor` to new CI Config

## Why was this MR needed?

This is the next iteration of CI configuration refactoring

## What are the relevant issue numbers?

#15060

## Does this MR meet the acceptance criteria?

- Tests
  - [x] Added for this feature/bug
  - [x] All builds are passing
- [x] Conform by the [style guides](https://gitlab.com/gitlab-org/gitlab-ce/blob/master/CONTRIBUTING.md#style-guides)
- [x] Branch has no merge conflicts with `master` (if you do - rebase it please)

See merge request !4820

Assert against `ActionMailer::Base.deliveries` relatively

- Closes #19496 
- Fixes `master` build


See merge request !5082

- Look for a `change` in its size rather than asserting against an
  actual size.

- This previously failed because another spec had an email in
  `ActionMailer::Base.deliveries`, which failed this `be_nil` assertion.

Add troubleshooting section for SMTP settings



See merge request !5046

Added user avatar to header

## What does this MR do?

Adds  the users avatar to the header with a dropdown for profile & sign out.

## What are the relevant issue numbers?

Closes #18543 

## Screenshots (if relevant)

![Screen_Shot_2016-06-27_at_12.47.34](/uploads/852d0eb7075eefcf21f5cf5ff71a389a/Screen_Shot_2016-06-27_at_12.47.34.png)

![Screen_Shot_2016-06-27_at_12.47.36](/uploads/6cb29843d3a12bbf2d3ed2a7c9bde770/Screen_Shot_2016-06-27_at_12.47.36.png)

See merge request !4939

Handle custom Git hook result in GitLab UI

## What does this MR do?

This MR fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/19190
It allows to show messages from git custom hooks

## Screenshots (if relevant)

![joxi_screenshot_1467651998090](/uploads/7f814523c96712c07b4f4b52c843aab3/joxi_screenshot_1467651998090.png)

See merge request !5073

[ci skip]

Fix dead links in the docs (Jun 2016)

## What does this MR do?

Fix dead links in the docs found as of Jun 2016.

## Are there points in the code the reviewer needs to double check?

n/a

## Why was this MR needed?

This MR must improve UX on docs.gitlab.com.

## What are the relevant issue numbers?

Closes #19156 

## Does this MR meet the acceptance criteria?

- No [CHANGELOG](https://gitlab.com/gitlab-org/gitlab-ce/blob/master/CHANGELOG) updated
- [x] [Documentation created/updated](https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/development/doc_styleguide.md)
- [x] Conform by the [style guides](https://gitlab.com/gitlab-org/gitlab-ce/blob/master/CONTRIBUTING.md#style-guides)
- [x] Branch has no merge conflicts with `master` (if you do - rebase it please)
- [x] [Squashed related commits together](https://git-scm.com/book/en/Git-Tools-Rewriting-History#Squashing-Commits)

See merge request !4921

Upgrade rspec-rails from 3.4.2 to 3.5.0.

Also upgrade its dependencies. Includes Rails 5 support. Working toward #14286 

Changelogs: http://rspec.info/blog/2016/07/rspec-3-5-has-been-released/

See merge request !5051

Update README.md - SSH password can be changed with `ssh-keygen -p`

SSH password can be changed with `ssh-keygen -p`

See merge request !4152

Lazy-load the krb5_auth gem

Closes #723.

See merge request !520

Save artifacts sizes

## What does this MR do?

Introduce ci_builds.artifacts_size as an integer, so that it's easier to access than reading from the file again.

## What are the relevant issue numbers?

Closes #18869

See merge request !4964