前面已经介绍了Gitlab环境部署记录，这里简单说下Gitlab的备份和恢复操作记录：
1）Gitlab的备份目录路径设置
|
1
2
3
4
5
6
7
8
9
10
11
12
|
[root@code-server ~]# vim /etc/gitlab/gitlab.rbgitlab_rails['manage_backup_path'] = truegitlab_rails['backup_path'] = "/data/gitlab/backups" //gitlab备份目录gitlab_rails['backup_archive_permissions'] = 0644 //生成的备份文件权限gitlab_rails['backup_keep_time'] = 7776000 //备份保留天数为3个月（即90天，这里是7776000秒）[root@code-server ~]# mkdir -p /data/gitlab/backups[root@code-server ~]# chown -R git.git /data/gitlab/backups[root@code-server ~]# chmod -R 777 /data/gitlab/backups 如上设置了gitlab备份目录路径为/data/gitlab/backups，最后使用下面命令重载gitlab配置文件，是上述修改生效！root@code-server ~]# gitlab-ctl reconfigure |
2）GItlab备份操作（使用备份命令”gitlab-rake gitlab:backup:create”）
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
|
手动备份gitlab[root@code-server backups]# gitlab-rake gitlab:backup:createDumping database ...Dumping PostgreSQL database gitlabhq_production ... [DONE]doneDumping repositories ... * treesign/treesign ... [DONE] * gateway/gateway ... [DONE] * treesign/treesign-doc ... [SKIPPED] * qwsign/qwsign ... [DONE] * qwsign/qwsign-doc ... [DONE] * test/test ... [DONE]doneDumping uploads ...doneDumping builds ...doneDumping artifacts ...doneDumping pages ...doneDumping lfs objects ...doneDumping container registry images ...[DISABLED]Creating backup archive: 1510471890_2017_11_12_9.4.5_gitlab_backup.tar ... doneUploading backup archive to remote storage ... skippedDeleting tmp directories ... donedonedonedonedonedonedonedoneDeleting old backups ... done. (0 removed)然后查看下备份文件（文件权限是设定好的644）[root@code-server backups]# lltotal 244-rw-r--r-- 1 git git 245760 Nov 12 15:33 1510472027_2017_11_12_9.4.5_gitlab_backup.tar编写备份脚本，结合crontab实施自动定时备份，比如每天0点、6点、12点、18点各备份一次[root@code-server backups]# pwd/data/gitlab/backups[root@code-server backups]# vim gitlab_backup.sh#!/bin/bash/usr/bin/gitlab-rake gitlab:backup:create CRON=1注意：环境变量CRON=1的作用是如果没有任何错误发生时， 抑制备份脚本的所有进度输出[root@code-server backups]# crontab -l0 0,6,12,18 * * * /bin/bash -x /data/gitlab/backups/gitlab_backup.sh > /dev/null 2>&1 |
3）Gitlab恢复操作
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
|
GItlab只能还原到与备份文件相同的gitlab版本。假设在上面gitlab备份之前创建了test项目，然后不小心误删了test项目，现在就进行gitlab恢复操作： 1）停止相关数据连接服务[root@code-server backups]# gitlab-ctl stop unicornok: down: unicorn: 0s, normally up[root@code-server backups]# gitlab-ctl stop sidekiqok: down: sidekiq: 1s, normally up[root@code-server backups]# gitlab-ctl statusrun: gitaly: (pid 98087) 1883s; run: log: (pid 194202) 163003srun: gitlab-monitor: (pid 98101) 1883s; run: log: (pid 194363) 163002srun: gitlab-workhorse: (pid 98104) 1882s; run: log: (pid 194362) 163002srun: logrotate: (pid 98117) 1882s; run: log: (pid 5793) 160832srun: nginx: (pid 98123) 1881s; run: log: (pid 194359) 163002srun: node-exporter: (pid 98167) 1881s; run: log: (pid 194360) 163002srun: postgres-exporter: (pid 98173) 1881s; run: log: (pid 194204) 163003srun: postgresql: (pid 98179) 1880s; run: log: (pid 194365) 163002srun: prometheus: (pid 98187) 1880s; run: log: (pid 194364) 163002srun: redis: (pid 98230) 1879s; run: log: (pid 194358) 163002srun: redis-exporter: (pid 98234) 1879s; run: log: (pid 194208) 163003sdown: sidekiq: 8s, normally up; run: log: (pid 194437) 163001sdown: unicorn: 21s, normally up; run: log: (pid 194443) 163001s 2）现在通过之前的备份文件进行恢复[root@code-server backups]# pwd/data/gitlab/backups[root@code-server backups]# lltotal 244-rw-r--r-- 1 git git 245760 Nov 12 15:33 1510472027_2017_11_12_9.4.5_gitlab_backup.tar Gitlab的恢复操作会先将当前所有的数据清空，然后再根据备份数据进行恢复[root@code-server backups]# gitlab-rake gitlab:backup:restore BACKUP=1510472027_2017_11_12_9.4.5Unpacking backup ... doneBefore restoring the database we recommend removing all existingtables to avoid future upgrade problems. Be aware that if you havecustom tables in the GitLab database these tables and all data will beremoved. Do you want to continue (yes/no)?........ALTER TABLEALTER TABLEALTER TABLEALTER TABLEWARNING: no privileges were granted for "public"GRANT[DONE]doneRestoring repositories ... * treesign/treesign ... [DONE] * gateway/gateway ... [DONE] * treesign/treesign-doc ... [DONE] * qwsign/qwsign ... [DONE] * qwsign/qwsign-doc ... [DONE] * test/test ... [DONE]Put GitLab hooks in repositories dirs [DONE]doneRestoring uploads ...doneRestoring builds ...doneRestoring artifacts ...doneRestoring pages ...doneRestoring lfs objects ...doneThis will rebuild an authorized_keys file.You will lose any data stored in authorized_keys file.Do you want to continue (yes/no)? yes Deleting tmp directories ... donedonedonedonedonedonedonedone[root@code-server backups]# 最后再次启动Gitlab[root@code-server backups]# gitlab-ctl startok: run: gitaly: (pid 98087) 2138sok: run: gitlab-monitor: (pid 98101) 2138sok: run: gitlab-workhorse: (pid 98104) 2137sok: run: logrotate: (pid 98117) 2137sok: run: nginx: (pid 98123) 2136sok: run: node-exporter: (pid 98167) 2136sok: run: postgres-exporter: (pid 98173) 2136sok: run: postgresql: (pid 98179) 2135sok: run: prometheus: (pid 98187) 2135sok: run: redis: (pid 98230) 2134sok: run: redis-exporter: (pid 98234) 2134sok: run: sidekiq: (pid 104494) 0sok: run: unicorn: (pid 104497) 1s[root@code-server backups]# gitlab-ctl statusrun: gitaly: (pid 98087) 2142s; run: log: (pid 194202) 163262srun: gitlab-monitor: (pid 98101) 2142s; run: log: (pid 194363) 163261srun: gitlab-workhorse: (pid 98104) 2141s; run: log: (pid 194362) 163261srun: logrotate: (pid 98117) 2141s; run: log: (pid 5793) 161091srun: nginx: (pid 98123) 2140s; run: log: (pid 194359) 163261srun: node-exporter: (pid 98167) 2140s; run: log: (pid 194360) 163261srun: postgres-exporter: (pid 98173) 2140s; run: log: (pid 194204) 163262srun: postgresql: (pid 98179) 2139s; run: log: (pid 194365) 163261srun: prometheus: (pid 98187) 2139s; run: log: (pid 194364) 163261srun: redis: (pid 98230) 2138s; run: log: (pid 194358) 163261srun: redis-exporter: (pid 98234) 2138s; run: log: (pid 194208) 163262srun: sidekiq: (pid 104494) 4s; run: log: (pid 194437) 163260srun: unicorn: (pid 104497) 4s; run: log: (pid 194443) 163260s恢复命令完成后，可以check检查一下恢复情况[root@code-server backups]# gitlab-rake gitlab:check SANITIZE=trueChecking GitLab Shell ...GitLab Shell version >= 5.3.1 ? ... OK (5.3.1)Repo base directory exists?default... yesRepo storage directories are symlinks?default... noRepo paths owned by git:root, or git:git?default... yesRepo paths access is drwxrws---?default... yeshooks directories in repos are links: ...5/1 ... ok6/2 ... ok5/3 ... repository is empty12/4 ... ok12/5 ... okRunning /opt/gitlab/embedded/service/gitlab-shell/bin/checkCheck GitLab API access: OKAccess to /var/opt/gitlab/.ssh/authorized_keys: OKSend ping to redis server: OKgitlab-shell self-check successfulChecking GitLab Shell ... FinishedChecking Sidekiq ...Running? ... yesNumber of Sidekiq processes ... 1Checking Sidekiq ... FinishedChecking Reply by email ...Reply by email is disabled in config/gitlab.ymlChecking Reply by email ... FinishedChecking LDAP ...LDAP is disabled in config/gitlab.ymlChecking LDAP ... FinishedChecking GitLab ...Git configured correctly? ... yesDatabase config exists? ... yesAll migrations up? ... yesDatabase contains orphaned GroupMembers? ... noGitLab config exists? ... yesGitLab config up to date? ... yesLog directory writable? ... yesTmp directory writable? ... yesUploads directory exists? ... yesUploads directory has correct permissions? ... yesUploads directory tmp has correct permissions? ... yesInit script exists? ... skipped (omnibus-gitlab has no init script)Init script up-to-date? ... skipped (omnibus-gitlab has no init script)Projects have namespace: ...5/1 ... yes6/2 ... yes5/3 ... yes12/4 ... yes12/5 ... yesRedis version >= 2.8.0? ... yesRuby version >= 2.3.3 ? ... yes (2.3.3)Git version >= 2.7.3 ? ... yes (2.13.4)Active users: ... 11Checking GitLab ... Finished 然后稍等一会（如果启动gitlab后，访问出现500，这是因为redis等程序还没完全启动，等一会儿访问就ok了），再次登录Gitlab，就会发现之前误删除的test项目已经恢复了！ 另外：Gitlab迁移与恢复一样，但是要求两个GitLab版本号一致 |