502 Bad Gateway for gitlab after a yum update - nginx

Question

I had a gitlab environnement that was working fine on CentOS 6, using port 8081 (apache already uses 80), and a yum update passed and now, I can't access my gitlab : 502 Bad Gateway
The issue is... I don't understand much of what I'm doing. Configurating gitlab at first was not easy... But this time, I don't know what to do.
In nginx logs (nginx has been installed from gitlab), I have :
2016/01/28 18:37:37 [crit] 31765#0: *1 connect() to unix:localhost:8081 failed (2: No such file or directory) while connecting to upstream, client: 59.118.47.19, server: gitlab.mydomain.com, request: "GET / HTTP/1.1", upstream: "http://unix:localhost:8081:/", host: "gitlab.mydomain.com:8081"
I already tried to check the conf, to use gitlab-ctl reconfigure, restart, but this doesn't seem to be changing anything at all.
Here is the result from gitlab-ctl show-config
gitlab-ctl show-config
[2016-01-28T18:42:27+01:00] WARN: Ohai::Config[:disabled_plugins] is set. Ohai::Config[:disabled_plugins] is deprecated and will be removed in future releases of ohai. Use ohai.disabled_plugins in your configuration file to configure :disabled_plugins for ohai.
Starting Chef Client, version 12.5.1
resolving cookbooks for run list: ["gitlab::show_config"]
Synchronizing Cookbooks:
- package (0.0.0)
- runit (0.14.2)
- gitlab (0.0.1)
Compiling Cookbooks...
{
"gitlab": {
"bootstrap": {
},
"omnibus-gitconfig": {
},
"manage-accounts": {
},
"user": {
"git_user_email": "gitlab#gitlab.mydomain.com"
},
"redis": {
},
"ci-redis": {
},
"gitlab-rails": {
"secret_token": "2e6c9197f8a81829348e8a29d0608976189abe04b0e28384428c49abcb8c8c9a5b95cfeac5b5f3e59",
"gitlab_host": "gitlab.mydomain.com",
"gitlab_email_from": "gitlab#gitlab.mydomain.com",
"gitlab_https": false,
"gitlab_port": 8081,
"shared_path": "/var/opt/gitlab/gitlab-rails/shared",
"artifacts_path": "/var/opt/gitlab/gitlab-rails/shared/artifacts",
"lfs_storage_path": "/var/opt/gitlab/gitlab-rails/shared/lfs-objects",
"pages_path": "/var/opt/gitlab/gitlab-rails/shared/pages",
"db_username": "gitlab",
"db_host": null,
"db_port": 5432
},
"gitlab-ci": {
"secret_token": null,
"secret_key_base": "18f7e0e892c1cba94454a289abc285492afvd42909fbez364c9a4a45efc7c092b678e5b6bb986383",
"db_key_base": "3fd05caaefd2ez3d5c5abc184fezi71962480936de4674609a6802b",
"db_username": "gitlab_ci",
"db_host": null,
"db_port": 5432
},
"gitlab-shell": {
"secret_token": "95f9af3zff567dcd70fbbc847dc67679d5e3717498294fhe97f978756f14bcca26f94b48db43f85008c1fab280d1e1ce3c21d4156369b"
},
"unicorn": {
},
"ci-unicorn": {
},
"sidekiq": {
},
"ci-sidekiq": {
},
"gitlab-workhorse": {
"listen_addr": "localhost:8081",
"auth_socket": "/var/opt/gitlab/gitlab-rails/sockets/gitlab.socket"
},
"mailroom": {
},
"nginx": {
"listen_port": 8081
},
"ci-nginx": {
"listen_port": 80
},
"mattermost-nginx": {
"listen_port": null
},
"pages-nginx": {
"listen_port": null
},
"logging": {
},
"remote-syslog": {
},
"logrotate": {
},
"high-availability": {
},
"postgresql": {
},
"web-server": {
},
"mattermost": {
"email_invite_salt": "5723719e29c111gr11c34905c25e838da",
"file_public_link_salt": "ede2ceb23d88reg83cc7060852525050251",
"email_password_reset_salt": "9a79cff28e13fe248e53c3a13dc0d1f3aee",
"sql_at_rest_encrypt_key": "f57d4bf0284dff8db5691ffc489b3f8312a1",
"sql_data_source": "user=gitlab_mattermost host=/var/opt/gitlab/postgresql port=5432 dbname=mattermost_production",
"sql_data_source_replicas": [
"user=gitlab_mattermost host=/var/opt/gitlab/postgresql port=5432 dbname=mattermost_production"
]
},
"external-url": "http://gitlab.mydomain.com:8081",
"ci-external-url": null,
"mattermost-external-url": null,
"pages-external-url": null
}
}
Converging 0 resources
Running handlers:
Running handlers complete
Chef Client finished, 0/0 resources updated in 01 seconds
If you have any pointers to what I could do/check, that would be great because I am kind of stucked (and as I don't understand much of what I'm doing... That doesn't help...)
Thank you !