@saltyang 2022-03-29T16:16:54.000000Z 字数 9400 阅读 1538

Local Deploy Webackup SOP

mbk docker deploy

Software Version

Centos 6.x

Docker Engine Version: 1.7.1

Docker Compose Version: 1.5.2

Docker Redis Image Version: 3.2

Docker Rabbitmq Image Version: 3.6.2

Docker Cassandra Image Version: 3.5

Docker Mysql Image Version: 5.6

Docker Centos Image Version: 6.7

Install Docker in Linux

#Install docker enginee

yum install docker-io or curl -fsSL https://get.docker.com/ | sh

#restart docker service

service docker start

#start docker deamon when machine power on

chkconfig docker on

Note: Centos7装docker

Load Docker image

# copy compose-mbk.tag.gz to your machine （compose-mbk.tag.gz 放在百度云盘我的应用数据>bypy>compose-mbk.tag.gz）

tar -xzvf compose-mbk.tag.gz

# load images

cd compose-mbk

docker load -i mbkcentos6.7.tar

docker load -i cassandra.tar

docker load -i mysql.tar

docker load -i rabbitmq.tar

docker load -i redis.tar

# Check whether load image successfylly or not. There are five images: mbkcentos6.7, cassandra, mysql, rabbitmq, redis

docker images

Run docker-compose

# Install docker-compose (Note : docker-compose version is 1.5.2)

curl -L https://get.daocloud.io/docker/compose/releases/download/1.5.2/\ docker-compose-`uname -s`-`uname -m` > /usr/local/bin/docker-compose (Note This is one command, not two commands)

chmod +x /usr/local/bin/docker-compose

docker-compose -v (Check the docker-compose version)

# Run docker-compose (Note: please stop mysqld/rabbitmq-server/redis/cassandra service on host machine to make sure the needed ports haven't used. )

docker-compose up -d

# Check whether docker container has started successfully or not. （There are five running containers: mbk_server, mbk_worker,mbk_cassandra, mbk_mysql, mbk_redis, the state of containers are "Up"）

docker-compose ps

Prepare db for deploy

Mysql:

# If you haven't mysql client , you can run this command to connect mbk_mysql container:

docker run -it --link mbk_mysql:mysql --rm mysql sh -c 'exec mysql -h"$MYSQL_PORT_3306_TCP_ADDR" -P"$MYSQL_PORT_3306_TCP_PORT" -uroot -p"$MYSQL_ENV_MYSQL_ROOT_PASSWORD"' (Note: this is one command, not three commands)

# If you have install mysql client, use mysql client connect mbk_mysql, you can get password from docker-compose.yml in compose-mbk folder

mysql -h you_host_ip -uroot -p

# after connect mbk_mysql, change default charset:

alter database mbackup character set utf8;

Cassandra:

# connect cassandra with cqlsh

docker run -it --link mbk_cassandra:cassandra --rm cassandra cqlsh cassandra

# create mbk keyspace

create keyspace mbackup with replication={'class': 'SimpleStrategy', 'replication_factor': 1};

Redis:

# connect redis with redis-cli

Port Hint: docker run -it --link mbk_redis:redis --rm redis redis-cli -h redis -p 6379 -a passwd

Port Expose: redis-cli -h host -a passwd

Deploy

 1> Clone code  (*If you have cloned, please pass this step*)
    <code>git clone git@www.cloudraid.com.cn:puya/mbackup.git</code>
 2> Enter into `mbackup/deploy` folder and Modify `deployConfig_local.json`
 Note :  The port of sshd is mbkserver: 50001 and mbkworker: 50002. The address of mysql/redis/rabbitmq/cassandra use service name in `docker-compose.yml`. For Example: mysql address should be `mbk_mysql` which is mapping to a ip address via host file in container.  
Example File Content:
    server
    "192.168.1.191:50001"
    worker
    "192.168.1.191:50002"
    portal
    "192.168.1.191:50001"
    mysql database address
    "mbk_mysql"
    mysql password:
    "puyacn#1.."
    redis address
    "mbk_redis"
    cassandra address
    "mbk_cassandra"
    rabbitmq-server
    "mbk_rabbitmq"
3> Deploy Step:
   a> Deploy Mbkserver
        python deployV2.py mbkserver -c deploy -a -l local
   b> Deploy Mbkworker
        python deployV2.py mbkworker -c deploy -a -l local
   c> Deploy Mbkportal
        python deployV2.py mbkportal -c deploy -a -l local
4> Finish
Client Slient Install Command Line:
    C:\Users\Salt\Desktop\wbksetup.exe /S /server_host=192.168.1.155 /ar=1

RPM OVA BUILD

Need modify CD-ROM ResourceSubType to vmware.cdrom.remotepassthrough

re-sha1sum ovf file and Change the value of sha1 in mf file
制作OVA注意事项

Note：

Volume写入不了，需要disabel Selinux：
- check Selinux的状态：/usr/sbin/sestatus -v
- disable Selinux ： vi /etc/selinux/config 将SELINUX=enforcing 改成SELINUX=disabled
- temporarily fix ： su -c "setenforce 0"
- set selinux rule to fix this： chcon -Rt svirt_sandbox_file_t /path/to/volume
Iptables 重启后，docker的规则被清除掉：
- iptables -t nat --list查看iptables中转换规则
- 修改iptables 配置文件： vim /etc/sysconfig/iptables-config 将no改为yes
copy文件到container中：
- 在container中安装scp: yum install openssh-clients
- Exapmle: scp -P 50002 -r storageManager.py hosted.py root@192.168.1.142:/opt/mbk/mbkworker/utils/
添加存储功能：
- docker-compose.yml: ./mbk/storage --> /mnt/storage
- 渠道用户登录后，添加本地存储：bucket0,bucket1,...
- 备份的数据存储在/mnt/storage
- 客户端支持http模式：mbk_storage_proto http;
- PTBUS 配置：
```
    {
        "access_key":"WLG36LHZ1HXERPTRH4H9",
        "access_secret":"qZXfctq9tMZfIJhcULiW4CnuqBqCejsVlEz2X2Du",
        "bucket":"bucket4",
        "url":"116.62.61.218"
    }
```
- 模板部署OVF：
  - 网卡修改 UUID
  - 设置上海时区： cp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
  - 同步时间： yum install ntp; ntpdate 1.cn.pool.ntp.org; systemctl enable ntpd

私密云添加存储配置：

私密云是标准的S3接口，经测试，只能通过http的方式进行上传和下载。所以需要在mbkclient.conf文件中添加一项：
mbk_storage_proto http;
例如:
location ~ ^/(task|user|sysinfo|device)/ {
        mbk_storage_proto http;
        mbk_backend_process;
        mbk_backend_debug on;
        mbk_backend_upstream https://192.168.1.189:443;
    }
Note: 在客户端build的时候需要加入此项

Centos7.2 init to deploy mbackup

Install mysql

wget http://repo.mysql.com/mysql-community-release-el7-5.noarch.rpm
rpm -ivh mysql-community-release-el7-5.noarch.rpm
yum install mysql-server
# 重置密码
$ mysql -u root
    mysql> use mysql
    mysql> update user set password=password('puyacn#1..') where user='root';
    mysql> flush privileges; 
    mysql> exit;
    // create database mbackup and set charset
    $ mysql -uroot -p
mysql> alter database mbackup character set utf8;

Config mysql master-slave

主数据库：192.168.1.177
从数据库：192.168.1.176
1 修改主数据库master
a> 创建用于同步的用户账户：
mysql> CREATE USER 'backup'@'192.168.1.176' IDENTIFIED BY 'backupcn1..';#创建用户
mysql> GRANT REPLICATION SLAVE ON *.* TO 'backup'@'192.168.1.176';#分配权限
mysql> flush privileges;   #刷新权限
b> 在配置文件：/etc/my.conf中插入如下两行：
[mysqld]
log-bin=mysql-bin
server-id=1
c> 重启mysql 查看master状态，并记录二进制文件名和位置。
mysql > SHOW MASTER STATUS;
+------------------+----------+--------------+------------------+
| File             | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000003 | 73       | test         | manual,mysql     |
+------------------+----------+--------------+------------------+
2 修改从数据库master
a> 导入主数据库的数据
    mysql> create database mbackup default charset utf8;
    mysql> use mbackup;
    mysql> source /root/backup.sql;
b> 修改mysql配置:/etc/my.cnf
    [mysqld]
    server-id=2 #设置server-id，必须唯一
c> 重启mysql，打开mysql会话，执行同步SQL语句
    mysql> CHANGE MASTER TO
    ->     MASTER_HOST='192.168.1.176',
    ->     MASTER_USER='backup',
    ->     MASTER_PASSWORD='backupcn1..',
    ->     MASTER_LOG_FILE='mysql-bin.000003',
    ->     MASTER_LOG_POS=73;
d> 启动slave同步进程
    mysql> start slave;
e> 查看slave状态
    show slave status\G;
    *************************** 1. row ***************************
                   Slave_IO_State: Waiting for master to send event
                      Master_Host: 182.92.172.80
                      Master_User: rep1
                      Master_Port: 3306
                    Connect_Retry: 60
                  Master_Log_File: mysql-bin.000013
              Read_Master_Log_Pos: 11662
                   Relay_Log_File: mysqld-relay-bin.000022
                    Relay_Log_Pos: 11765
            Relay_Master_Log_File: mysql-bin.000013
                 Slave_IO_Running: Yes   # 两者都为Yes是正常，如果Slave_IO_Running为Connecting
                Slave_SQL_Running: Yes   # 则检查网络，查看主机的3306端口是否开放
                  Replicate_Do_DB: 
              Replicate_Ignore_DB:

Intstall redis

yum install epel-release
yum install redis
//change redis conf file
vim   /etc/redis.conf
将bind 127.0.0.1注释掉
//启动服务
service redis start
//修改密码
requirepass 111111
//重启服务
service redis restart

Install rabbitmq

wget http://www.rabbitmq.com/releases/rabbitmq-server/v3.6.6/rabbitmq-server-3.6.6-1.el6.noarch.rpm
yum -y install rabbitmq-server-3.6.6-1.el6.noarch.rpm
rabbitmq-server start
rabbitmqctl add_user mbk mbkpwd
rabbitmqctl add_vhost vhost_mbk
rabbitmqctl set_permissions -p vhost_mbk mbk ".*" ".*" ".*"

Install cassandra
```
pass
```

Install ElasticSearch


# download elasticsearch and install 
wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-6.4.1.rpm
rpm -ivh elasticsearch-6.4.1.rpm
service elasticsearch start
# (Optional) you can downlaod kibana to post api with dev-tool
wget https://artifacts.elastic.co/downloads/kibana/kibana-6.4.1-x86_64.rpm
rpm -ivh kibana-6.4.1-x86_64.rpm
service kibana start
# config elasticsearch yml
vim /etc/elasticsearch/elasticsearch.yml
'''
cluster.name: mbk
http.port: 9200
network.host: 0.0.0.0
'''
service elasticsearch restart
# config kibana yml
vim /etc/kibana/kibana.yml
'''
server.host: "0.0.0.0"
server.port: 5601
elasticsearch.url: "http://localhost:9200"
'''
service kibana restart
# open firewall to enable port
firewall-cmd --zone=public --add-port=9200/tcp --permanent
firewall-cmd --zone=public --add-port=5601/tcp --permanent
service firewalld reload
# add elasticsearch to pip 
# test
curl http://192.168.1.144:9200/
curl curl http://192.168.1.144:5601/app/kibana

Deploy code

copy env and 3rdParty to /opt/mbk
python deployV2.py mbkserver -a -c deploy -l vpc
Note: libmysqlclient_r.so.16: cannot open shared object file: No such file or directory
cp libmysqlclient_r.so.16 to /usr/lib64
Note: ngix start -> libpcre.so.0: cannot open shared object file: No such file or directory
ln -s  libpcre.so.0 /lib64/libpcre.so
开放443端口
yum install firewalld firewalld-config
firewall-cmd --zone=public --add-port=443/tcp --permanent
service firewalld restart

存储配置

系统盘和数据盘分离（至少需要添加两块磁盘）
fdisk -l  #查看磁盘是否已分配
fdisk /dev/sdb #发现有磁盘，路径为/dev/sdb。然后使用fdisk命令进行建立分区
输入命令依次是：n -> p -> 1 -> 默认 -> 默认 -> w 
fdisk -l #查看一下，应该已经有了分区
mkfs.xfs -f /dev/sdb1 #建好分区后要格式化分区，建立文件系统，最好和其他几个分区保持一致
mount /dev/sdb1 /opt/mbk/storage # 文件系统建好后，选择挂载到/opt/mbk/storage 下
chown -R mbk:mbk /opt/mbk/storage #修改目录所属权限

\# 系统安装好后，发现home分区过大，想从home分区中拿出100G给／分区
\# 针对XFS系统
umount /home/
lvreduce -L -100G /dev/mapper/centos-home
mkfs.xfs /dev/mapper/centos-home -f
mount /dev/mapper/centos-home /home/
df -hT # 再次查看分区，发现home分区已经减小了100G,只不过这个分区里之前的数据都没有了
vgdisplay # #然后将上面从home分区拿出的100G放到／分区下
lvextend -L +100G /dev/mapper/centos-root
xfs_growfs /dev/mapper/centos-root
df -hT

\# 针对ext2、ext3、ext4文件系统
umount /home/
resize2fs -p /dev/mapper/vg_weidianserver2-lv_home 20G
mount /home
df -h
lvextend -L +812G /dev/mapper/vg_weidianserver2-lv_root
resize2fs -p /dev/mapper/vg_weidianserver2-lv_root
df -h
[解决linux系统CentOS下调整home和根分区大小的方法][3]

mount cifs 开机自启动
//192.168.1.33/webackup_share /opt/mbk/test_storage/bucket0/1 cifs user=Everyone,password=,vers=2.0,gid=994,uid=996 0 0