问题解答

问题一：GitLab容器一直重启

使用docker logs命令查询

$ docker log CONTAINER_ID
...
...
System Info:
------------
chef_version=14.13.11
platform=ubuntu
platform_version=16.04
ruby=ruby 2.6.3p62 (2019-04-16 revision 67580) [x86_64-linux]
program_name=/opt/gitlab/embedded/bin/chef-client
executable=/opt/gitlab/embedded/bin/chef-client


Running handlers:
There was an error running gitlab-ctl reconfigure:

/etc/gitlab/gitlab.rb:1: unexpected fraction part after numeric literal
external_url = 192.168.0.144:7002
               ^~~~~~~
/etc/gitlab/gitlab.rb:1: syntax error, unexpected tINTEGER, expecting end-of-input
external_url = 192.168.0.144:7002
                         ^~~

Running handlers complete
Chef Client failed. 0 resources updated in 01 seconds

发现是external_url配置失误导致。解决方法如下：

停止gitlab容器
修改配置文件（在主机中）/srv/gitlab/config/gitlab.rb
重新启动容器

问题二：GitLab导致8080端口冲突

问题复现

安装完GitLab后，修改配置文件/etc/gitlab/gitlab.rb

##external_url 'http://gitlab.example.com'
external_url 'http://localhost:8800'

本以为这样就能修改GitLab端口号为8800了，没想到再次登录8080仍旧出现了GitLab页面

问题解析

查询哪个程序监听了8080端口

# netstat -lnp | grep 8080
tcp        0      0 127.0.0.1:8080          0.0.0.0:*               LISTEN      26436/config.ru

查询相应的进程

# netstat -lnp | grep 26436
tcp        0      0 127.0.0.1:8080          0.0.0.0:*               LISTEN      26436/config.ru 
unix  2      [ ACC ]     STREAM     LISTENING     343538   26436/config.ru     /var/opt/gitlab/gitlab-rails/sockets/gitlab.socket

仍然是GitLab在监听8080端口，参考gitlab 8.13 80 8080端口冲突问题，查看配置文件unicorn.rb

# This file is managed by gitlab-ctl. Manual changes will be
# erased! To change the contents below, edit /etc/gitlab/gitlab.rb
# and run `sudo gitlab-ctl reconfigure`.

# What ports/sockets to listen on, and what options for them.
listen "127.0.0.1:8080", :tcp_nopush => true

默认情况下unicorn同样监听8080端口，查询/etc/gitlab/gitlab.rb中相应的设置

# cat gitlab.rb | grep unicorn
#unicorn['port'] = 8800

解决方案

需要在gitlab.rb上同时修改unicorn监听端口号，修改配置文件/etc/gitlab/gitlab.rb如下

##external_url 'http://gitlab.example.com'
external_url 'http://localhost:8800'
unicorn['port'] = 8801

重新启动GitLab

# gitlab-ctl reconfigure
# gitlab-ctl restart

查询配置文件/var/opt/gitlab/gitlab-rails/etc/unicorn.rb

# cat unicorn.rb | grep listen
# What ports/sockets to listen on, and what options for them.
listen "127.0.0.1:8801", :tcp_nopush => true

测试端口号

$ curl localhost:8080
curl: (7) Failed to connect to localhost port 8080: Connection refused
$ curl localhost:8800
<!DOCTYPE html>
<html>
<head>
  <meta content="width=device-width, initial-scale=1, maximum-scale=1" name="viewport">
...
...
# curl localhost:8801
<html><body>You are being <a href="http://localhost:8801/users/sign_in">redirected</a>.</body></html>

问题三： [GitLab][Webhook]不允许本地连接

在gitlab仓库中设置Webhook，使用本地连接出现如下错误:

Url is blocked: Requests to the local network are not allowed

参考:gitlab使用webhook向jenkins发送请求，报错 Requests to the local network are not allowed

登录root账户，点击Configure Gitlab选项

进入Settings -> Network，展开Outbound requests，选中Allow requests to the local network from web hooks and services

问题四：GitLab: You are not allowed to force push code to a protected branch on this project

强制上传代码到Gitlab仓库的dev分支出错，提示如下：

GitLab: You are not allowed to force push code to a protected branch on this project

参考：解决 GitLab: You are not allowed to force push code to a protected branch on this project问题

是由于分支保护的原因，需要进入仓库setting -> Repository -> Protected Branches

允许dev分支能够强制推送

问题五：unicorn出错

问题描述

通过docker部署gitlab一段时间后，突然出现502错误

查看容器状态，显示不健康

CONTAINER ID        IMAGE                     COMMAND                  CREATED             STATUS                 PORTS                                                           NAMES
89481faa2ed1        gitlab/gitlab-ce:latest   "/assets/wrapper"        2 weeks ago         Up 7 hours (unhealthy)

浏览容器日志，发现如下错误：

==> /var/log/gitlab/unicorn/unicorn_stdout.log <==
bundler: failed to load command: unicorn (/opt/gitlab/embedded/bin/unicorn)

==> /var/log/gitlab/unicorn/unicorn_stderr.log <==
ArgumentError: Already running on PID:460 (or pid=/opt/gitlab/var/unicorn/unicorn.pid is stale)
  /opt/gitlab/embedded/lib/ruby/gems/2.6.0/gems/unicorn-5.4.1/lib/unicorn/http_server.rb:205:in `pid='
  /opt/gitlab/embedded/lib/ruby/gems/2.6.0/gems/unicorn-5.4.1/lib/unicorn/http_server.rb:137:in `start'
  /opt/gitlab/embedded/lib/ruby/gems/2.6.0/gems/unicorn-5.4.1/bin/unicorn:126:in `<top (required)>'
  /opt/gitlab/embedded/bin/unicorn:23:in `load'
  /opt/gitlab/embedded/bin/unicorn:23:in `<top (required)>'

==> /var/log/gitlab/unicorn/current <==
2019-12-18_12:27:47.34406 failed to start a new unicorn master
2019-12-18_12:27:47.35882 starting new unicorn master
2019-12-18_12:27:47.91735 master failed to start, check stderr log for details

问题解析

看样子是unicorn的问题，参考：

gitlab服务器502恢复过程

gitlab docker Web界面打开反应迟钝的解决办法

进入容器内部，查看unicorn状态

$ docker exec -it xxxx bash
# gitlab-ctl tail unicorn

发现每次unicorn显示的PID都不同，修改/etc/gitlab/gitlab.rb，添加

unicorn['listen'] = 'localhost'
unicorn['port'] = 8999

更新并重启后，gitlab服务恢复正常

# gitlab-ctl reconfigure
# gitlab-ctl restart

问题六：版本更新后database出错

问题描述

拉取了最新的Docker Gitlab镜像，部署时发现了如下错误：

gitlab     |     System Info:
gitlab     |     ------------
gitlab     |     chef_version=14.14.29
gitlab     |     platform=ubuntu
gitlab     |     platform_version=16.04
gitlab     |     ruby=ruby 2.6.6p146 (2020-03-31 revision 67876) [x86_64-linux]
gitlab     |     program_name=/opt/gitlab/embedded/bin/chef-client
gitlab     |     executable=/opt/gitlab/embedded/bin/chef-client
gitlab     |     
gitlab     | 
gitlab     | Running handlers:
gitlab     | There was an error running gitlab-ctl reconfigure:
gitlab     | 
gitlab     | bash[migrate gitlab-rails database] (gitlab::database_migrations line 55) had an error: Mixlib::ShellOut::ShellCommandFailed: Expected process to exit with [0], but received '1'
gitlab     | ---- Begin output of "bash"  "/tmp/chef-script20200618-23-sfpkhu" ----
gitlab     | STDOUT: rake aborted!
gitlab     | PG::ConnectionBad: could not connect to server: No such file or directory
gitlab     |    Is the server running locally and accepting
gitlab     |    connections on Unix domain socket "/var/opt/gitlab/postgresql/.s.PGSQL.5432"?
gitlab     | /opt/gitlab/embedded/service/gitlab-rails/lib/tasks/gitlab/db.rake:48:in `block (3 levels) in <top (required)>'
gitlab     | /opt/gitlab/embedded/bin/bundle:23:in `load'
gitlab     | /opt/gitlab/embedded/bin/bundle:23:in `<main>'
gitlab     | Tasks: TOP => gitlab:db:configure
gitlab     | (See full trace by running task with --trace)
gitlab     | STDERR: 
gitlab     | ---- End output of "bash"  "/tmp/chef-script20200618-23-sfpkhu" ----
gitlab     | Ran "bash"  "/tmp/chef-script20200618-23-sfpkhu" returned 1

问题解决

在网上查了一些资料，说是新旧版本的配置格式不一致，需要删除之前的配置数据。幸好在本地保存了之前使用的镜像，打包该镜像后上传到服务器重新加载即可

问题七：Prometheus吃磁盘空间

在云服务器上部署Gitlab，今天发现无法登录了。登上云服务器后发现磁盘容量没有了，通过查找发现Prometheus目录下占据了30GB的空间（总共50GB）

参考：

Monitoring GitLab with Prometheus

Prometheus eats disk space in /var/opt/gitlab/prometheus/data

GitLab 官方镜像内部集成 Prometheus 历史数据过大的问题处理

Prometheus是一个监控服务，会保存历史监控数据，下面尝试关闭该服务并删除之前的数据（在Docker Gitlab上操作）

修改配置文件/srv/gitlab/config/gitlab.rb，添加

prometheus_monitoring['enable'] = false

关闭gitlab并删除/srv/gitlab/data/prometheus文件夹数据
重启gitlab，问题解决

问题八：Access deined: You do not have permission push to this repository

在gitee上新建了一个账户，在上面新建一个仓库，将其ssh链接设置到本地已存在的仓库上，想要把本地仓库推送到远程仓库，出现如下错误：

$ git push origin master 
Access deined: You do not have permission push to this repository
fatal: 无法读取远程仓库。

请确认您有正确的访问权限并且仓库存在。

其原因在于本地新建了两个ssh密钥，设置在gitee的不同账户上，ssh使用之前的密钥读取gitee远程仓库，导致出错。参考如何使用特定的SSH Key提交GIT

在~/.ssh/config文件上添加如下内容

Host gitee.com
    HostName gitee.com
    User git
    IdentityFile 私钥文件路径
    IdentitiesOnly yes

重新提交即可

问题九：代码库大小限制

问题描述

上传代码到Gitee时，打印出如下日志：

 git push --force git@gitee.com:xx.xx.git master
remote: Powered by [01;33mGITEE.COM [0m[[01;35mGNK-3.8[0m][0m        
remote: Recommended to reduce the repository size        
remote: This repository(including wiki) size [01;33m581.64 MB[0m exceeds [01;33m500.00 MB[0m        
remote: HelpLink: https://gitee.com/help/articles/4232
To gitee.com:zjZSTU/xx.xxx.git

在网上查询发现是因为单个仓库大小超出了限制（500MB），经过一番查询发现Gitee和Coding的个人仓库都限制在500MB，Github对单个仓库大小的限制为1GB。

问题解决

阿里云的单个仓库限制为2GB，能够满足当前需求。

各个远程Git仓库服务器的官网地址如下：

Gitee
Coding
Github
Aliyun Code
相关阅读

Gitee 仓库体积过大，如何减小？

Tencent 仓库大小限制吗？

GitHub File and repository size limitations

Aliyun代码库对文件大小是否有限制？

问题十：publickey denied

ssh连接失败，这个问题很迷，耗了1天时间，网上找了很多资料没有结果，有一个博客主也遇到了同样的问题，他最后解决是因为突然就可以了，我也一样

问题十一：git is not in the sudoers file. This incident will be reported.

是因为git没有添加到sudo权限，参考xxx is not in the sudoers file.This incident will be reported.的解决方法

先切换到ubuntu用户
```
$ su ubuntu
```
赋予sudo文件写权限
```
$ sudo chmod u+w /etc/sudoers
```
修改文件/etc/sudoers
```
$ sudo vim /etc/sudoers
# 添加
youuser ALL=(ALL) ALL
```
表示允许用户youuser执行sudo命令(需要输入密码)
撤销sudo文件写权限
```
$ sudo chmod u-w /etc/sudoers
```
重新切换回原来用户
```
$ su git
```

问题十二：perl: warning: Falling back to a fallback locale ("en_US.utf8").

$ sudo adduser zhujian
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
    LANGUAGE = (unset),
    LC_ALL = (unset),
    LC_MEASUREMENT = "zh_CN.UTF-8",
    LC_PAPER = "zh_CN.UTF-8",
    LC_MONETARY = "zh_CN.UTF-8",
    LC_NAME = "zh_CN.UTF-8",
    LC_ADDRESS = "zh_CN.UTF-8",
    LC_NUMERIC = "zh_CN.UTF-8",
    LC_TELEPHONE = "zh_CN.UTF-8",
    LC_IDENTIFICATION = "zh_CN.UTF-8",
    LC_TIME = "zh_CN.UTF-8",
    LANG = "en_US.utf8"
    are supported and installed on your system.
perl: warning: Falling back to a fallback locale ("en_US.utf8").
Adding user `zhujian'ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQC+VVpw5kOZBmzJYz/vngYpAAV61Fq9oChSflQFkfzr1sKHRqq2/sqeZD3gzPQZbrKWcHbuGCWyQOvm1gH+67gW+TpUO9DWeeHqo3h5rlCW+ElJcL/q4b+ZBVEmGDjzE+Sg+6wM+izBl5xzHDFeLhN3Yw1OVc2rwQFQ/CD6FSKdL4b5bt0/5rpu65sv7haXjfDMSEsIVgPY5behLzZzoXy81iN4/tPF3cjDsn/x5Yywc60LdslJ5hW5wlozhq1LibUXk9JQu/+5DDZKi8ytMEoe1S7yROvaC/ofJQR22hINnFoLNBC8gSFM2YR+t9oBF0eiAaVwfgddA0+ScYrWA5Yr zj@zj-ThinkPad-T470p ...
Adding new group `zhujian' (1000) ...
Adding new user `zhujian' (1000) with group `zhujian' ...
Creating home directory `/home/zhujian' ...
...
...

提示语言包问题，参考perl: warning: Falling back to a fallback locale ("en_US.UTF-8")

$ sudo apt install locales-all

问题十三：`remote: error: cannot lock ref 'refs/heads/master': ref refs/heads/master`

在Travis CI上上传文件到远程仓库，出现如下问题

remote: error: cannot lock ref 'refs/heads/master': ref refs/heads/master

起因

我在上传提交时中间中断了

$ git push -u origin dev
Counting objects: 6, done.
Delta compression using up to 8 threads.
Compressing objects: 100% (6/6), done.
Writing objects: 100% (6/6), 556 bytes | 0 bytes/s, done.
Total 6 (delta 5), reused 0 (delta 0)
remote: Resolving deltas: 100% (5/5), completed with 5 local objects.
^C

然后再更新文章后再一次提交，就出现了远程仓库锁定的问题

解决

参考：git: error: cannot lock ref, error: cannot lock ref

将远程仓库下载到本地，然后执行如下命令

git remote prune origin

其作用是同步本地远程分支，删除本地已过时的远程分支

重新触发CI工具执行，此时能够成功更新了

问题十四：error: RPC failed; curl 56 GnuTLS recv error (-9)

问题描述

下载Github仓库时出现如下错误

remote: Enumerating objects: 150, done.
remote: Counting objects: 100% (150/150), done.
remote: Compressing objects: 100% (90/90), done.
error: RPC failed; curl 56 GnuTLS recv error (-9): A TLS packet with unexpected length was received.
fatal: The remote end hung up unexpectedly
fatal: 过早的文件结束符（EOF）
fatal: index-pack 失败

解析

参考git error: RPC failed; curl 56 GnuTLS和Git 克隆错误‘RPC failed; curl 56 Recv failure....’ 及克隆速度慢问题解决，有可能是http缓存不够或者网络不稳定问题

对于网络不稳定问题

Rebuilding git with openssl instead of gnutls fixed my problem.

对于http缓存不够

# httpBuffer加大    
$ git config --global http.postBuffer 524288000

解决

重新设置了http缓存，不过最后是通过码云的方式解决的

先将Github仓库拉取到码云仓库
在从码云上下载

问题十五：fatal: 无法读取远程仓库。

$ git clone ssh://git@gitlab.zhujian.tech:7020/zjykzj/hexo-blog.git
正克隆到 'hexo-blog'...
remote: 
remote: ========================================================================
remote: 
remote: The project you were looking for could not be found.
remote: 
remote: ========================================================================
remote: 
fatal: 无法读取远程仓库。

请确认您有正确的访问权限并且仓库存在。

参考：How to fix 'The project you were looking for could not be found' when using git clone

可以使用命令ssh-add -l查看是否已加载私钥，我遇到一种情况就是可能存在私钥冲突（多个私钥已加载）的情况，所以，所以需要重启ssh缓存代理，再加入合适的私钥

# 重启缓存代理
eval "$(ssh-agent -s)"
# 加入私钥
ssh-add ~/.ssh/<id_rsa>

问题十六：fatal: 无法读取远程仓库。

$ git push origin master 
ssh: connect to host ssh.github.com port 443: Connection timed out
fatal: 无法读取远程仓库。

请确认您有正确的访问权限并且仓库存在。

激活私钥
测试远程登录
代理（推荐）

参考：git上传代码报错ssh: connect to host github.com port 22: Connection timed out解决办法

问题十七：gnutls_handshake() failed: Error in the pull function

参考：使用curl出现gnutls_handshake() failed: Error in the pull function或者GnuTLS recv error: Error in the pull

方法一：sudo apt-get install libcurl4-openssl-dev
方法二：开启代理（推荐）

问题十八：Host key verification failed.

+ git push --force ssh://git@gitlab.zhujian.tech:7020/****/test.git master
Host key verification failed.
fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.

在网上找了很多资料，提出的解决方案是重新设置ssh公/私钥。当前我的问题不是这个，而是.ssh/config重新配置

Host <xxx/自定义gitlab地址>
        Port xxx                                   ---------------> 这个是关键，默认为22，当前应该设置为7020
        StrictHostKeyChecking no