Docker 部署 Gitlab

在 Docker 中安装 Gitlab 教程,官方文档,如果你想使用原生安装,教程在这里:CentOS7安装维护Gitlab

下载镜像

docker pull gitlab/gitlab-ce

运行容器

sudo docker run \
  --hostname gitlab.example.com \
  --publish 8443:443 --publish 8081:80 -p 2222:22 \
  --name gitlab \
  --restart always \
  --volume $HOME/_docker/gitlab/config:/etc/gitlab \
  --volume $HOME/_docker/gitlab/logs:/var/log/gitlab \
  --volume $HOME/_docker/gitlab/data:/var/opt/gitlab \
  -v /etc/localtime:/etc/localtime \
  -d \
  gitlab/gitlab-ce:latest

由于端口冲突,重新映射了一个端口 2222,如果不想麻烦,可以事先将 ssh 端口号更改成别的端口号,修改ssh端口号的方法

# 要从之前的:
git clone git@gitlab.example.com:myuser/awesome-project.git
# 改为明确使用 `ssh://` 的 `URL` 方式。
git clone ssh://git@gitlab.example.com:2222/myuser/awesome-project.git

为了克隆不必麻烦,保留 gitlab22 端口映射,将主机的 sshd22 端口映射到容器中去。将主机的 sshd 端口更改为 2222

编辑文件 /etc/ssh/sshd_config,将其中的 #Port 22 注释去掉,将数字 22 更改为 2222,执行下面的命令重启 sshd 服务

systemctl restart sshd

防火墙的规则,添加开发 2222 端口

iptables -A INPUT -p tcp --dport 2222 -j ACCEPT
iptables -A OUTPUT -p tcp --sport 2222 -j ACCEPT
# 再查看下是否添加上去, 看到添加了
iptables -L -n

如果此容器由于权限问题而无法启动,请尝试通过执行以下操作来修复它:

docker exec -it gitlab update-permissions
docker restart gitlab

容器手动备份

# 第一种进行入容器执行命令的方法进行手工备份
docker exec -it 容器名或容器id bash  # 进入容器
gitlab-rake gitlab:backup:create   # 执行gitlab备份命令

# 第二种直接使用外部命令执行,一次完成
docker exec 容器名或容器id gitlab-rake gitlab:backup:create

自动备份

通过在宿主机上使用 crontab 使用备份命令实现自动备份

添加备份脚本 vi ~/_docker/gitlab/gitlab.backup.sh,将下面内容添加到脚本中,保存之后添加可执行权限 chmod +x gitlab.backup.sh

#!/bin/bash
case "$1" in 
  start)
    docker exec gitlab-ce11.2.3 gitlab-rake gitlab:backup:create
    ;;
esac

创建定时执行计划

crontab -e # 进入编辑,添加下面内容

# 每天2点备份 gitlab 数据
0 2 * * * $HOME/_docker/gitlab/gitlab.backup.sh start
# *  *  *  *  *  command
# 分  时  日  月  周  命令

# 其中,
# 第1列表示分钟,1~59,每分钟用*表示
# 第2列表示小时,1~23,(0表示0点)
# 第3列表示日期,1~31
# 第4列表示月份,1~12
# 第5列表示星期,0~6(0表示星期天)
# 第六列表示要运行的命令。

上面两行保存之后,重新载入配置

service crond reload
# or
systemctl reload crond.service

备份保留七天

设置只保存最近7天的备份,编辑 vi $HOME/_docker/gitlab/config/gitlab.rb 配置文件,找到如下代码,删除注释 # 保存

# /etc/gitlab/gitlab.rb 配置文件 修改下面这一行
gitlab_rails['backup_keep_time'] = 604800  

重新加载 gitlab 配置文件

docker exec 容器名或容器ID gitlab-ctl reconfigure  

容器管理

docker stop gitlab # 停止容器
docker rm gitlab # 删除容器
docker start gitlab # 启动容器
# 编辑 gitlab 容器配置
docker exec -it gitlab vim /etc/gitlab/gitlab.rb
# 重启 gitlab 容器
docker restart gitlab

通过 Docker Compose 按照

使用 Docker Compose,可以轻松配置,安装和升级基于 Docker 的 GitLab 安装,官方教程在这里

第一步: Docker 官方教程安装 Docker Compose。

第二步: 创建 docker-compose.yml 文件,将下面配置复制到文件中 (或者下载官方示例):

web:
  image: 'gitlab/gitlab-ce:latest'
  restart: always
  hostname: 'gitlab.example.com'
  environment:
    GITLAB_OMNIBUS_CONFIG: |
      external_url 'https://gitlab.example.com'
      gitlab_rails['time_zone'] = 'Asia/Shanghai'
      gitlab_rails['backup_keep_time'] = 259200 # 3 Day, 259200 seconds 
      registry_external_url 'http://192.168.188.222:5008'
  ports:
    - '8081:80'
    - '8443:443'
    - '22:22'
  volumes:
    - ./gitlab-data/config:/etc/gitlab
    - ./gitlab-data/logs:/var/log/gitlab
    - ./gitlab-data/data:/var/opt/gitlab
    - /etc/localtime:/etc/localtime

第三步: 确保与 docker-compose.yml 文件同一目录下运行 docker-compose up -d 启动 Gitlab

使用 Docker Swarm

官方教程 创建 docker-compose.yml 文件

version: "3.6"
services:
  gitlab:
    image: gitlab/gitlab-ce:latest
    container_name: gitlab
    ports:
      - "22:22"
      - "80:80"
      - "443:443"
    volumes:
      - /srv/gitlab/data:/var/opt/gitlab
      - /srv/gitlab/logs:/var/log/gitlab
      - /srv/gitlab/config:/etc/gitlab
      - /etc/localtime:/etc/localtime
    environment:
      GITLAB_OMNIBUS_CONFIG: "from_file('/omnibus_config.rb')"
    configs:
      - source: gitlab
        target: /omnibus_config.rb
    secrets:
      - gitlab_root_password
  gitlab-runner:
    image: gitlab/gitlab-runner:alpine
    container_name: gitlab-runner
    deploy:
      mode: replicated
      replicas: 4
configs:
  gitlab:
    file: ./gitlab.rb
secrets:
  gitlab_root_password:
    file: ./root_password.txt

创建 gitlab.rb 文件

external_url 'https://my.domain.com/'
gitlab_rails['initial_root_password'] = File.read('/run/secrets/gitlab_root_password')
gitlab_rails['backup_keep_time'] = 604800 
gitlab_rails['time_zone'] = 'Asia/Shanghai' # 中国的东八区时间

创建 root_password.txt 文件

MySuperSecretAndSecurePass0rd!

确保您与 docker-compose.yml 在同一目录中并运行:

docker stack deploy --compose-file docker-compose.yml gitlab

注册 runner

更新配置

如果您在 config.toml 中更改配置,则可能需要重新启动运行程序以应用更改。 确保重新启动整个容器,而不是使用 gitlab-runner restart

docker restart gitlab-runner

升级版本

Pull the latest version (or a specific tag):

docker pull gitlab/gitlab-runner:latest

Stop and remove the existing container:

docker stop gitlab-runner && docker rm gitlab-runner

Start the container as you did originally:

docker run -d --name gitlab-runner --restart always \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /home/www/gitlab/gitlab-runner/config:/etc/gitlab-runner \
  gitlab/gitlab-runner:latest

服务 gitlab-runner 跑起来之后可以注册对应的仓库

docker run --rm -it -v /home/www/gitlab/gitlab-runner/config:/etc/gitlab-runner gitlab/gitlab-runner register

# Runtime platform                                    arch=amd64 os=linux pid=8 revision=943fc252 version=13.7.0
# Running in system-mode.

# Enter the GitLab instance URL (for example, https://gitlab.com/):
# https://g.xxxxx.cn/
# Enter the registration token:
# ze9H4**********
# Enter a description for the runner:
# [7d0472a5e808]: web
# Enter tags for the runner (comma-separated):

# Registering runner... succeeded                     runner=ze9H44QH
# Enter an executor: docker-ssh+machine, docker-ssh, parallels, virtualbox, docker+machine, kubernetes, custom, docker, shell, ssh:
# shell
# Runner registered successfully. Feel free to start it, but if it's running already the config should be automatically reloaded!
concurrent = 1
check_interval = 0

[session_server]
  session_timeout = 1800

[[runners]]
  name = "项目名称"
  url = "https://g.xxxxx.cn/"
  token = "xxx-y1vb"
  executor = "docker"

  [runners.custom_build_dir]
  [runners.cache]
    [runners.cache.s3]
    [runners.cache.gcs]
    [runners.cache.azure]
  [runners.docker]
    environment = ['GIT_SSL_NO_VERIFY=true']
    tls_verify = false
    image = "node:12"
    privileged = false
    pull_policy = "if-not-present"
    disable_entrypoint_overwrite = false
    oom_kill_disable = false
    disable_cache = false
    volumes = ["/cache", "/var/run/docker.sock:/var/run/docker.sock"]
    shm_size = 0
  1. ⚠️ token 是生成的,必须通过 gitlab-runner 生成 token
  2. volumes = ["/cache", "/var/run/docker.sock:/var/run/docker.sock"] 配置添加很重要,解决下面错误:
ERROR: error during connect: Get http://docker:2375/v1.40/info: dial tcp: lookup docker on 8.8.8.8:53: no such host
  1. pull_policy = "if-not-present" 策略改为:镜像不存在时才拉取。
  2. Gitlab runner: This job is stuck because the project doesn't have any runners online assigned to it.

工作被卡住了,因为你的 runner 有标签,但你的 job 没有。 按照以下 4 个步骤,让您的 runner 在没有标签的情况下运行:

Gitlab runner

Gitlab runner

CI 中使用编译提交镜像

下面是 官方仓库 Docker.gitlab-ci.yml 模板

docker-build-master:
  # Official docker image.
  image: docker:latest
  stage: build
  services:
    - docker:dind
  before_script:
    - docker login -u "$CI_REGISTRY_USER" -p "$CI_REGISTRY_PASSWORD" $CI_REGISTRY
  script:
    - docker build --pull -t "$CI_REGISTRY_IMAGE" .
    - docker push "$CI_REGISTRY_IMAGE"
    # 运行服务
    - if [ $(docker ps -aq --filter name=docker-service-name) ]; then docker rm -rf docker-service-name;fi
    - docker run -itd -p 5000:5000 --name docker-service-name "$CI_REGISTRY_IMAGE":latest
  only:
    - master
docker-build:
  # Official docker image.
  image: docker:latest
  stage: build
  services:
    - docker:dind
  before_script:
    - docker login -u "$CI_REGISTRY_USER" -p "$CI_REGISTRY_PASSWORD" $CI_REGISTRY
  script:
    - docker build --pull -t "$CI_REGISTRY_IMAGE:$CI_COMMIT_REF_SLUG" .
    - docker push "$CI_REGISTRY_IMAGE:$CI_COMMIT_REF_SLUG"
  except:
    - master

错误处理

gitlab ci ERROR: Uploading artifacts to coordinator... too large archive

使用管理员帐户登陆 Gitlab -> Admin Area -> Settings 修改 Maximum artifacts size (MB) 值,然后保存,然而并没有解决,我的问题是 nginx 代理造成的,最终通过修改 nginx 代理配置解决问题:

client_max_body_size       10m;

升级

目标版本 你的版本 支持升级 路径 注意
14.1.6 13.9.2 13.9.2 -> 13.12.12 -> 14.0.11 -> 14.1.6 需要两个中间版本:13.1214.0,然后是 14.1
13.12.10 12.9.2 12.9.2 -> 12.10.14 -> 13.0.14 -> 13.1.11 -> 13.8.8 -> 13.12.10 需要四个中间版本:12.1013.013.113.8.8,然后是 13.12.10
13.2.10 11.5.0 11.5.0 -> 11.11.8 -> 12.0.12 -> 12.1.17 -> 12.10.14 -> 13.0.14 -> 13.1.11 -> 13.2.10 需要六个中间版本:11.1112.012.112.1013.013.1,然后是 13.2.10

假设我是 13.9.2 升级到 14.1.6,通过官方提供的升级路径 => 13.9.2 -> 13.12.12 -> 14.0.11 -> 14.1.6

docker pull gitlab/gitlab-ce:13.12.12-ce.0
docker pull gitlab/gitlab-ce:13.12.15-ce.0
docker pull gitlab/gitlab-ce:14.0.11-ce.0
docker pull gitlab/gitlab-ce:14.0.12-ce.0
docker pull gitlab/gitlab-ce:14.1.6-ce.0
docker pull gitlab/gitlab-ce:14.1.7-ce.0
docker pull gitlab/gitlab-ce:14.1.8-ce.0
docker pull gitlab/gitlab-ce:14.8.2-ce.0

我先将所有的版本下载到本地。先将 13.9.2 升级到 14.0.11,启动的时候会有提示升级需要更改配置:

There was an error running gitlab-ctl reconfigure:

Removed configurations found in gitlab.rb. Aborting reconfigure.

* unicorn['worker_processes'] has been deprecated since 13.10 and was removed in 14.0. Starting with GitLab 14.0, Unicorn is no longer supported and users must switch to Puma, following https://docs.gitlab.com/ee/administration/operations/puma.html.
* unicorn['worker_memory_limit_min'] has been deprecated since 13.10 and was removed in 14.0. Starting with GitLab 14.0, Unicorn is no longer supported and users must switch to Puma, following https://docs.gitlab.com/ee/administration/operations/puma.html.
* unicorn['worker_memory_limit_max'] has been deprecated since 13.10 and was removed in 14.0. Starting with GitLab 14.0, Unicorn is no longer supported and users must switch to Puma, following https://docs.gitlab.com/ee/administration/operations/puma.html.

Running handlers complete

Chef Infra Client failed. 0 resources updated in 12 seconds

通过回滚到 13.9.2 更改配置重新加载配置:

gitlab-ctl reconfigure

启动没有问题再升级到 14.0.11