This project is mirrored from https://:*****@gitlab.com/gitlab-org/omnibus-gitlab.git.
Pull mirroring failed .
Repository mirroring has been paused due to too many failed attempts. It can be resumed by a project maintainer or owner.
Last successful update .
Repository mirroring has been paused due to too many failed attempts. It can be resumed by a project maintainer or owner.
Last successful update .
- Oct 20, 2020
-
-
Ash McKenzie authored
-
- Oct 05, 2020
-
-
Robert Marshall authored
- Fixes a rubocop warning that popped up after merge conflit. Signed-off-by:
Robert Marshall <rmarshall@gitlab.com>
-
Robert Marshall authored
- Run the SELinux helper after GitLab Workhorse and other files are created ensuring their file contexts are restored - Place GitLab Workhorse socket files into a sockets directory allowing the SELinux file context to be inherited on service restarts - Add a GitLab Workhorse helper to ensure that new UNIX sockets for Workhorse always inherit the proper SELinux file context Signed-off-by:
Robert Marshall <rmarshall@gitlab.com>
-
- Sep 24, 2020
-
-
Stan Hu authored
-
- Aug 28, 2020
-
-
- May 07, 2020
-
-
Signed-off-by:
Balasankar "Balu" C <balasankarc@autistici.org>
-
- Nov 19, 2019
-
-
Balasankar C authored
CentOS 8 requires that pathnames be present when calling `restorecon`, but this was not a requirement in CentOS 7. We fix this by moving all SELinux-related changes to the gitlab::selinux recipe so that all the required files and directories can be made before any SELinux changes are applied. This makes sense from a code organization standpoint and also has the nice side effect of removing the need to ignore unknown files (-i option in restorecon) as well. This commit is essentially commit 5b992246 by Stan Hu, rebased against latest master Signed-off-by:
Balasankar "Balu" C <balasankar@gitlab.com>
-
- Sep 01, 2019
-
-
Patrick Bajao authored
-
- Apr 16, 2019
-
-
Patrick Bajao authored
-
- Jan 27, 2019
-
-
Stan Hu authored
Since https://gitlab.com/gitlab-org/gitlab-shell/merge_requests/181, gitlab-shell no longer can talk to Redis directly. All operations go through the internal API. Removing this config is useful for several reasons: 1. It removes dead code. 2. Since the gitlab-shell config is repeatedly parsed with every git action, removing a few lines can improve performance. 3. There is a bug in string handling when there are YAML characters included in the Redis password field.
-
- Dec 04, 2018
-
-
Balasankar C authored
Earlier, we replaced gitlab['redis_*'] settings with redis['master_*'] values in sentinel mode. However, this worked only for gitlab-rails and not for several other services like gitlab-monitor. This commit changes that and get gitlab-rails the necessary details based on the knowledge whether sentinels are present or not.
-
- Jun 14, 2018
-
-
Balasankar C authored
Change ownership to root:git and permission to 0640
-
- Mar 22, 2018
-
-
Jacob Vosmaer (GitLab) authored
-
- Jan 31, 2018
-
-
Stan Hu authored
For fast SSH key lookups to work (https://docs.gitlab.com/ee/administration/operations/fast_ssh_key_lookup.html), SELinux spawns `/opt/gitlab/embedded/service/gitlab-shell/bin/gitlab-shell-authorized-keys-check` and needs the following access: * Read: /var/opt/gitlab/gitlab-shell/config.yml * Read: /var/opt/gitlab/gitlab-rails/etc/gitlab_shell_secret * Write: /var/log/gitlab/gitlab-shell/gitlab-shell.log * Connect: unicorn (port 8080) Because the SELinux policy is a static policy, right now we don't support the ability to change internal unicorn ports. Admins would have to create a special .te file for the environment, or we'd have to dynamically generate it for them, which is it a bit tricky if they have changed their port contexts. Granting `http_cache_port_t` permissions also includes access to these ports: ``` http_cache_port_t tcp 8080, 8118, 8123, 10001-10010 http_cache_port_t udp 3130 ``` Closes #2855
-
- Jan 29, 2018
-
-
Balasankar C authored
-
- Jan 25, 2018
-
-
Balasankar C authored
-
- Nov 06, 2017
-
-
Nick Thomas authored
-
- Sep 05, 2017
-
-
Elliot Wright authored
-
- Aug 29, 2017
-
-
Elliot Wright authored
-
- Apr 04, 2017
-
-
Marin Jankovski authored
-
- Mar 07, 2017
-
-
- Feb 24, 2017
-
-
Pawel Chojnacki authored
-
- Dec 20, 2016
-
-
Gabriel Mazetto authored
-
- Dec 15, 2016
-
-
Ian Baum authored
* Using custom_hooks_dir, not custom_hook_dir * Fixed missed reference in gitlab.rb.template * Using %r for regex's in specs that reference a path
-
Ian Baum authored
* By default, do not populate this value * Change value name to be config_hook_dir
-
Ian Baum authored
Defaults to package provided directory, but allow the users to point to their own location. Only updates the config.yml file for gitlab-shell, assumes the user will manage the directory themselves.
-
- Dec 12, 2016
-
-
Ian Baum authored
* Using custom_hooks_dir, not custom_hook_dir * Fixed missed reference in gitlab.rb.template * Using %r for regex's in specs that reference a path
-
- Dec 09, 2016
-
-
Ian Baum authored
* By default, do not populate this value * Change value name to be config_hook_dir
-
- Dec 08, 2016
-
-
Ian Baum authored
Defaults to package provided directory, but allow the users to point to their own location. Only updates the config.yml file for gitlab-shell, assumes the user will manage the directory themselves.
-
- Dec 02, 2016
-
-
Balasankar C authored
-
- Nov 22, 2016
-
-
DJ Mountney authored
Without making it configurable, and always having it set, logging would always be used and would take up quite a bit of space.
-
- Nov 04, 2016
-
-
Marin Jankovski authored
-
- Oct 28, 2016
-
-
Marin Jankovski authored
-
Marin Jankovski authored
-
- Oct 27, 2016
-
-
Marin Jankovski authored
-
- Sep 09, 2016
-
-
Jacob Vosmaer (GitLab) authored
-
- Sep 02, 2016
-
-
Jacob Vosmaer (GitLab) authored
-
- Aug 24, 2016
-
-
DJ Mountney authored
And also simplify the recipes that make use of manage-storage-directories by moving the condition into the definition
-
DJ Mountney authored
-
DJ Mountney authored
-