- 24 Jun, 2016 1 commit
-
-
Jacob Vosmaer authored
-
- 23 Jun, 2016 8 commits
-
-
Stan Hu authored
Put Redis loading code in one place Before this change you could not run the gitlab_net specs in isolation. See merge request !64
-
Robert Speicher authored
ARGF does more than we want ARGF is a virtual IO stream that treats the elements of ARGV as paths to files to be opened. In practice, git-receive-pack will probably never pass arguments to pre-receive and post-receive because it is not supposed to (according to 'man 5 githooks'). In case it does, however our hooks should not have the surprise behavior of treating those arguments as files to be opened. See merge request !66
-
Jacob Vosmaer authored
ARGF is a virtual IO stream that treats the elements of ARGV as paths to files to be opened. In practice, git-receive-pack will probably never pass arguments to pre-receive and post-receive because it is not supposed to (according to 'man 5 githooks'). In case it does, however our hooks should not have the surprise behavior of treating those arguments as files to be opened.
-
Valery Sizov authored
Simplify the GitAccessStatus class Make the optional 'message' argument required. Remove unused 'to_json' method. See merge request !65
-
Jacob Vosmaer authored
Make the optional 'message' argument required. Remove unused 'to_json' method.
-
Jacob Vosmaer authored
-
Yorick Peterse authored
Update PostReceive worker to provide enqueued_at time See merge request !63
-
Paco Guzman authored
-
- 23 May, 2016 1 commit
-
-
Jacob Vosmaer authored
-
- 16 May, 2016 1 commit
-
-
Jacob Vosmaer (GitLab) authored
Use Redis Ruby client instead of shelling out to redis-cli Previously the post-receive hook fired redis-cli, but if the argument list was too long the hook would silently fail. Instead of shelling out to redis-cli, we use a Ruby client to send the same message. Closes gitlab-org/gitlab-ce#17329 See merge request !59
-
- 12 May, 2016 6 commits
- 10 May, 2016 2 commits
-
-
Rémy Coutable authored
Remove broken badge The state is shown on the repository anyway See merge request !39
-
Zeger-Jan van de Weg authored
-
- 09 May, 2016 1 commit
-
-
Stan Hu authored
Fix spelling mistakes in README.md See merge request !57
-
- 05 May, 2016 1 commit
-
-
Mark Campbell authored
-
- 28 Apr, 2016 3 commits
-
-
Jacob Vosmaer (GitLab) authored
Remove the update-head command since GitLab doesn't use it anymore It looks like we tend to remove commands that are not used by GitLab anymore (see gitlab-org/gitlab-shell!26) and the `update-head` command will soon be in that case (see https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/3705), so I think we should remove it from here too. Note: I'm setting this as a WIP to ensure this doesn't get merged since the GitLab MR is not merged at that moment. See merge request !51
-
Rémy Coutable authored
Signed-off-by: Rémy Coutable <remy@rymai.me>
-
Rémy Coutable authored
Revert "Remove unused #list_remote_tags" This reverts commit f5e87590 because `#list_remote_tags` is still used by EE:https://gitlab.com/gitlab-org/gitlab-ee/blob/2c364a79ee110a3adf49d6bd30be985ed13c3ec5/lib/gitlab/backend/shell.rb#L45. /cc @razer6 See merge request !56
-
- 25 Apr, 2016 1 commit
-
-
Rémy Coutable authored
This reverts commit f5e87590. Signed-off-by: Rémy Coutable <remy@rymai.me>
-
- 20 Apr, 2016 2 commits
-
-
Stan Hu authored
Update post receive worker so it logs a unique JID in sidekiq Taken from !50. Instead of ``` 2016-04-14T03:07:32.373Z 5285 TID-osycgmlyo PostReceive JID- INFO: start 2016-04-14T03:07:32.374Z 5285 TID-osycgmlyo PostReceive JID- INFO: arguments: [...] 2016-04-14T03:07:32.534Z 5285 TID-osycgmlyo PostReceive JID- INFO: done: 0.161 sec ``` Have this log ``` 2016-04-14T03:07:32.373Z 5285 TID-osycgmlyo PostReceive JID-54b0b2f6616cae37e3e87f8a INFO: start 2016-04-14T03:07:32.374Z 5285 TID-osycgmlyo PostReceive JID-54b0b2f6616cae37e3e87f8a INFO: arguments: [...] 2016-04-14T03:07:32.534Z 5285 TID-osycgmlyo PostReceive JID-54b0b2f6616cae37e3e87f8a INFO: done: 0.161 sec ``` This way sidekiq can Log a unique JID in the sidekiq.log for PostReceive. So when parsing the logs (with logstash for example) you know it belongs to that unique job. This puts the logs in a uniform manner like the other workers that are pushed to redis (which do have a JID) For example the ProjectWebHookWorker ``` 2016-04-14T03:13:07.917Z 5285 TID-osycsh7z0 ProjectWebHookWorker JID-800085fb3cf7241fdeecc6ec INFO: start 2016-04-14T03:13:07.918Z 5285 TID-osycsh7z0 ProjectWebHookWorker JID-800085fb3cf7241fdeecc6ec INFO: arguments: [...] 2016-04-14T03:13:12.500Z 5285 TID-osycsh7z0 ProjectWebHookWorker JID-800085fb3cf7241fdeecc6ec INFO: done: 4.583 sec ``` See merge request !55
-
aiionx authored
-
- 19 Apr, 2016 2 commits
-
-
Rémy Coutable authored
Remove branch functionality The branch functionality is already replaced with rugged inside GitLab See merge request !52
-
Robert Schilling authored
-
- 18 Apr, 2016 3 commits
-
-
Rémy Coutable authored
Remove rm-tag command Not needed anymore because gitlab-org/gitlab-ce!3748 switched the functionality to rugged. See merge request !54
-
Robert Schilling authored
-
Rémy Coutable authored
Remove unused #list_remote_tags This function is not used anymore inside GitLab. See merge request !53
-
- 16 Apr, 2016 1 commit
-
-
Robert Schilling authored
-
- 07 Apr, 2016 1 commit
-
-
Yorick Peterse authored
-
- 06 Apr, 2016 2 commits
-
-
Yorick Peterse authored
Prune during 'git gc' after all See merge request !48
-
Jacob Vosmaer authored
This reverts commit 1ca8a44f. Never pruning loose objects at all is not really a solution for anything, and it causes annoying `git gc --auto` warnings.
-
- 31 Mar, 2016 4 commits
-
-
Douwe Maan authored
Add new command to list tags from a remote repo. The output of the `git ls-remote` command is written to the STDOUT so the client can read and parse the list of tags. If there is an error it's also written to STDOUT. Closes #39 See merge request !47
-
Rubén Dávila authored
-
Rubén Dávila authored
The output of the `git ls-remote` command is written to the STDOUT so the client can read and parse the list of tags. If there is an error it's also written to STDOUT.
-
Yorick Peterse authored
[ci skip]
-