Styjun

we have a locally-hosted Enterprise Edition of GitLab at my place of employment (currently at v7.12.00-ee ceb5083). I can create a repository through the GUI without issue. But when I try to add anything to the repository, I get this error:

D:wstesting [master]> git push -u origin master
Counting objects: 3, done.
Writing objects: 100% (3/3), 219 bytes | 0 bytes/s, done.
Total 3 (delta 0), reused 0 (delta 0)
remote: GitLab: API is not accessible
To http://gitlab.ops.cld/duffrw/testing.git
 ! [remote rejected] master -> master (pre-receive hook declined)
error: failed to push some refs to 'http://gitlab.ops.cld/duffrw/testing.git'

I see in the /var/log/gitlab/gitlab-shell/gitlab-shell.log file that the API isn't responding, and is giving an "Internal Server Error":

E, [2015-07-22T16:05:51.812454 #15110] ERROR -- : API call <POST http://127.0.0.1:8080/api/v3/internal/allowed> failed: 500 => <"message":"500 Internal Server Error">.

A few questions... Where can I go from here? There must some sort of service provider that isn't running. Why not? What's it called? How do I start it?

I checked in /var/log/gitlab/nginx/gitlab_access.log and found these lines, but I didn't see anything interesting:

172.31.201.164 - - [22/Jul/2015:16:05:50 +0000] "GET /duffrw/testing.git/info/refs?service=git-receive-pack HTTP/1.1" 401 0 "-" "git/1.9.5.msysgit.1"
172.31.201.164 - - [22/Jul/2015:16:05:50 +0000] "GET /duffrw/testing.git/info/refs?service=git-receive-pack HTTP/1.1" 401 0 "-" "git/1.9.5.msysgit.1"
172.31.201.164 - duffrw [22/Jul/2015:16:05:51 +0000] "GET /duffrw/testing.git/info/refs?service=git-receive-pack HTTP/1.1" 200 179 "-" "git/1.9.5.msysgit.1"
172.31.201.164 - duffrw [22/Jul/2015:16:05:51 +0000] "POST /duffrw/testing.git/git-receive-pack HTTP/1.1" 200 124 "-" "git/1.9.5.msysgit.1"

There aren't any errors for the push of this repository in /var/log/gitlab/nginx/gitlab_error.log and /var/log/gitlab/nginx/error.log is completely empty.

What am I missing?

Update:

Using the gitlab-ctl tail command is very handy! It revealed a stack trace, which I sent to GitLab and they told me that this was a bug in 7.12.00. It is fixed in 7.13.00.

If using gitlab.com, the server is either being updated or having issues.

You can check https://status.gitlab.com for updates from the GitLab team.

There's also a Twitter account you can follow: https://twitter.com/gitlabstatus

You have to set the correct URL for the GitLab API in GitLab's configuration files.

As we can see in your gitlab-shell.log your API address is currently set to http://127.0.0.1:8080/. If you change the default URL where GitLab is running or if you changed its default port, you have to change it.

This parameter is configurable in /etc/gitlab/gitlab.rb :

## GitLab URL
##! URL on which GitLab will be reachable.
##! For more details on configuring external_url see:
##! https://docs.gitlab.com/omnibus/settings/configuration.html#configuring-the-external-url-for-gitlab
external_url 'http://127.0.0.1:3000'
[...]

This is the default value (as port 3000 is the default value in GitLab omnibus for unicorn). After you updated it you'll have to run a sudo gitlab-ctl reconfigure to apply this parameter.

It will override the file opt/gitlab/embedded/service/gitlab-shell/config.yml as mentioned in it's header comments.

We've encountered this issue when hosting our gitlab on a "compute engine" from google cloud.
The reason for this behaviour was not having a "swap" configured in our machine. Gitlab wasn't able to allocate additional memory.

We decided to scale up our memory instead of creating a swap partition but it's your choice... :)

I had this once in gitlab.com and it was not the reliability, just that I was trying to push with --force to the master branch, and it was protected (see "protected branches" in the settings of the repository).

We've encountered this frequently of late. It's definitely GitLab running out of memory. We are not yet able to determine if GitLab is leaking memory or if our server instance is under-provisioned.

Restart the gitlab service fix my problem.

sudo gitlab-ctl restart

sudo gitlab-ctl reconfigure

I got the same problems in Bitnami. I have restarted all things and solved this problem by using

/opt/bitnami/ctlscript.sh restart