- May 30, 2014
-
-
Marin Jankovski authored
-
- May 27, 2014
-
- May 22, 2014
-
- May 15, 2014
-
-
Jacob Vosmaer (GitLab) authored
-
Jacob Vosmaer (GitLab) authored
-
- May 13, 2014
-
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
- May 12, 2014
-
-
Sid Sijbrandij authored
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
- May 11, 2014
-
-
sue445 authored
-
- May 08, 2014
-
-
Jacob Vosmaer (GitLab) authored
This fixes a bug where wiki repositories for projects with an empty main repository would not get backed up.
-
Jacob Vosmaer (GitLab) authored
Fixes a bug with non-UTF8 locales introduced by 2b816075.
-
Jacob Vosmaer (GitLab) authored
-
- May 05, 2014
-
-
Drew Blessing authored
-
- Apr 29, 2014
-
-
Jacob Vosmaer (GitLab) authored
-
Dmitriy Zaporozhets authored
-
- Apr 24, 2014
-
-
Dmitriy Zaporozhets authored
-
- Apr 23, 2014
-
-
Jacob Vosmaer (GitLab) authored
-
Jacob Vosmaer (GitLab) authored
-
- Apr 16, 2014
- Apr 15, 2014
-
-
Jacob Vosmaer (GitLab) authored
Before this change, Rails cache data was stored in a global Redis namespace. As a consequence, clearing the Rails cache (`rake cache:clear`) would also delete all Sidekiq queue data and session storage. This change puts all Rails cache data in a `cache:gitlab` namespace, making `rake cache:clear` safe again.
-
- Apr 11, 2014
-
-
Jonathan A. Sternberg authored
This fixes issue #161.
-
- Apr 10, 2014
-
-
Mart Sõmermaa authored
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
- Apr 09, 2014
-
-
Marin Jankovski authored
-
- Apr 07, 2014
-
-
Job van der Voort authored
-
Marin Jankovski authored
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Jacob Vosmaer (GitLab) authored
On GitLab Cloud, MySQL connection reaping seemed to cause more problems (exceptions and segfaults in the native MySQL client) than it solved.
-
Jacob Vosmaer (GitLab) authored
The Satellite#create method no longer sets the `$?` global variable. Instead, we can use the method return value to test if the satellite was created successfully
-
- Apr 03, 2014
-
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
Sid Sijbrandij authored
-
- Apr 02, 2014
-
-
Marin Jankovski authored
-
- Apr 01, 2014
-
-
Dmitriy Zaporozhets authored
Signed-off-by:
Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>
-
- Mar 31, 2014
-
-
Jacob Vosmaer (GitLab) authored
-
- Mar 28, 2014
-
-
Jacob Vosmaer (GitLab) authored
Before there was a bug in omniauth-ldap which prevented samaccountname showing up as a possible username for new LDAP users. Thanks to upstream fixes, we no longer need to work around this bug.
-
Jacob Vosmaer (GitLab) authored
Invoking 'db:schema:load' turned out to be a bad idea: when downgrading an existing GitLab installation, the schema of the newer version would be preserved when trying to import the old version.
-
- Mar 26, 2014
-
-
Marin Jankovski authored
-
Marin Jankovski authored
-
- Mar 25, 2014
-
-
Marin Jankovski authored
-