部署规划

按照官方的要求，测试环境要求最低2C8G（实测可用内存最少10g）我准备了几台centos7.9的虚拟机，配置如下

名词解析

中控机：存储Oceanbase数据库安装包和集群配置信息的机器
目标机器：安装OceanBase数据库的机器
OBD：Oceanbase Deployer，Oceanbase 开源软件的安装部署工具，简称OBD
OBProxy：Oceanbase Database Proxy, OceanBase高性能反向代理服务器，简称OBProxy
OCP: OceanBase cloud Platform OceanBase运维管理工具，简称OCP

note: 1，ob数据库的安装需要至少50g的磁盘，如果你计划将ob安装到指定目录，必须保证磁盘空间大于50g，我使用虚拟机，手动分区；2，部署ob虚拟机设置内存最少12g（可用内存的10g，cache不算进去）；

机器准备

说明几点：

我物理机48G内存，使用VMware虚拟机；
服务统一使用centos 7.9;
虚拟机创建时，最大磁盘填写100g，将虚拟机磁盘存储为单个文件
内存设置12g，11g内存ob在操作时只识别到5.7，最低要求可用8g内存(4.0应该是至少10G内存)；

通过echo 3 > /proc/sys/vm/drop_caches 释放缓存

cpu设置为2个处理器，每个处理器2核
网络模式为桥接模式

特别提醒：安装操作系统时磁盘要选手动设置，如果不选手动设置，自动分区，会平分，导致不够50g，手动分区如下：

统一配置

note: 先在一个虚拟机里把所有的配置都配置好，等配置完成后，再复制出来3台，个性化的修改下，省的一个个的配置了。

hosts设置

[root@dev ~]# vim /etc/hosts # 添加的内容192.168.1.153 dev3192.168.1.154 dev4192.168.1.155 dev5192.168.1.156 dev6  # 保存并退出

设置机器名和静态ip

#修改主机名,先统一为dev[root@dev ~]# vim /etc/hostname
# 修改为静态ip[root@dev ~]#  vim /etc/sysconfig/network-scripts/ifcfg-ens33#替换为以下内容TYPE=EthernetPROXY_METHOD=noneBROWSER_ONLY=noBOOTPROTO=staticDEFROUTE=yesIPV4_FAILURE_FATAL=noIPV6INIT=noIPV6_AUTOCONF=noIPV6_DEFROUTE=noIPV6_FAILURE_FATAL=noIPV6_ADDR_GEN_MODE=stable-privacyNAME=ens33UUID=7ab82905-739a-4a1a-8784-57f62d25ef32DEVICE=ens33ONBOOT=yesIPADDR=192.168.1.153NETMASK=255.255.255.0GATEWAY=192.168.1.1DNS1=114.114.114.114DNS2=8.8.8.8

‍

关闭大页

#关闭透明大页echo never > /sys/kernel/mm/transparent_hugepage/enabled

配置信息查看

# 查看内存[root@dev ~]# free -g              total        used        free      shared  buff/cache   availableMem:             10           1           8           0           1           5Swap:             5           0           5#查看磁盘[root@dev ~]# df -h文件系统                 容量  已用  可用 已用% 挂载点devtmpfs                 5.3G     0  5.3G    0% /devtmpfs                    5.3G     0  5.3G    0% /dev/shmtmpfs                    5.3G   13M  5.3G    1% /runtmpfs                    5.3G     0  5.3G    0% /sys/fs/cgroup/dev/mapper/centos-root   84G  4.2G   80G    5% //dev/mapper/centos-home   10G   33M   10G    1% /home/dev/sda1               1014M  185M  830M   19% /boottmpfs                    1.1G  4.0K  1.1G    1% /run/user/42tmpfs                    1.1G   24K  1.1G    1% /run/user/0/dev/sr0                 4.4G  4.4G     0  100% /run/media/root/CentOS 7 x86_64
#查看网卡,网卡为ens33[root@dev ~]# ifconfigens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500        inet 192.168.1.153  netmask 255.255.255.0  broadcast 192.168.1.255

时钟源ntp配置（需要所有机器root）

先把所有虚拟机的ntp的server都设置为192.168.1.156，等到虚拟机配置调整完毕，复制出几份以后，再调整156的配置

OceanBase集群中的服务器时间必须保持一致。

不一致会导致无法启动，运行时无主
误差在50ms以下认为是同步状态
最大容忍误差不超过100ms，超过100ms会出现无主情况；

#首先在所有的机器上安装ntp服务(默认有)[root@dev ~]# yum install ntp ntpdate -y
[root@dev ~]# vim /etc/ntp.conf #修改内容如下#默认Centos提供了4个时间同步服务器，都注释掉#server 0.centos.pool.ntp.org iburst#server 1.centos.pool.ntp.org iburst#server 2.centos.pool.ntp.org iburst#server 3.centos.pool.ntp.org iburst#以156的ntp服务，作为其他机器的ntp源，prefer优先本地主机server 192.168.1.156　　prefer　　# 保存并退出
# 重启ntp服务端[root@dev ~]# service ntpd restart
# 设置ntp开启自动启动[root@dev ~]# chkconfig ntpd on注意：正在将请求转发到“systemctl enable ntpd.service”。Created symlink from /etc/systemd/system/multi-user.target.wants/ntpd.service to /usr/lib/systemd/system/ntpd.service.

‍磁盘规划

生产环境做好划分，测试环境，就自己玩了；

/data/ob 为数据盘
/redo/ob 存放redo日志(可以不配置，不配置就和数据一起了，注意：配置的话不能放到/data/ob目录下）
/app/observer 存放ob数据库的二进制文件和运行日志

创建用户

个人自己玩，直接root即可，如果是生产，最好创建一个账户,并把ob相关的目录授权给该用户；

注意：我直接用root，没有创建用户

# 创建appuser账户[root@dev ~]# useradd -U appuser -d /home/appuser -s /bin/bash[root@dev ~]# mkdir -p /home/appuser[root@dev ~]# sudo chown -R appuser:appuser /home/appuser# 更改密码[root@dev ~]# passwd appuser101790# 为appuser设置sudo 权限[root@dev ~]# vim /etc/sudoers#修改为sudo 的时候不需要密码%wheel  ALL=(ALL)       NOPASSWD:ALL   # 修改appuser ALL=(ALL)       NOPASSWD:ALL   #添加
# 查看目录权限[root@dev ~]# ls -al /home/总用量 0drwxr-xr-x.  3 root    root     21 11月 27 19:40 .dr-xr-xr-x. 19 root    root    267 11月 27 19:24 ..drwx------   5 appuser appuser 128 11月 27 19:43 appuser# 将ob的运行目录的所有权改为appuser[root@dev ~]# chown -R appuser.appuser /data /app /redo

配置limits.conf

调整最大栈空间大小、可打开的最大文件句柄数、core文件大小；（不需要压测的话，不需要调整）

[root@dev ~]# vim /etc/security/limits.conf# 添加的内容root soft nofile 655350root hard nofile 655350* soft nofile 655350* hard nofile 655350* soft stack 20480* hard stack 20480* soft nproc 655360* hard nproc 655360* soft core unlimited* hard core unlimited
# 注意下limits.d 目录下的文件，这里的默认值4096，不更改，这个appuser不会生效[root@dev ~]#  vim /etc/security/limits.d/20-nproc.conf # 所有用户默认为4096，我改成了10240*          soft    nproc     4096
# 切换到appuser账户，查看下max user processes  是否已经变了[appuser@dev root]$ ulimit -a

配置sysctl.conf

优化服务器，调整服务器的sysctl.conf,以提高linux系统的性能；（不需要压测的话，开发环境不需要调整）

[root@dev ~]# vim /etc/sysctl.conf# for oceanbase## 修改内核异步 I/O 限制 (测试环境，可以只设置这里)fs.aio-max-nr=1048576
## 网络优化net.core.somaxconn = 2048net.core.netdev_max_backlog = 10000 net.core.rmem_default = 16777216 net.core.wmem_default = 16777216 net.core.rmem_max = 16777216 net.core.wmem_max = 16777216
net.ipv4.ip_local_port_range = 3500 65535 net.ipv4.ip_forward = 0 net.ipv4.conf.default.rp_filter = 1 net.ipv4.conf.default.accept_source_route = 0 net.ipv4.tcp_syncookies = 0 net.ipv4.tcp_rmem = 4096 87380 16777216 net.ipv4.tcp_wmem = 4096 65536 16777216 net.ipv4.tcp_max_syn_backlog = 16384 net.ipv4.tcp_fin_timeout = 15 net.ipv4.tcp_max_syn_backlog = 16384 net.ipv4.tcp_tw_reuse = 1 net.ipv4.tcp_tw_recycle = 1 net.ipv4.tcp_slow_start_after_idle=0
vm.swappiness = 0vm.min_free_kbytes = 2097152# 此处为 OceanBase 数据库的 data 目录kernel.core_pattern = /data/ob/core-%e-%p-%t# 保存退出
# 加载配置[root@dev ~]# sysctl -pfs.aio-max-nr = 1048576net.core.somaxconn = 2048net.core.netdev_max_backlog = 10000net.core.rmem_default = 16777216net.core.wmem_default = 16777216net.core.rmem_max = 16777216net.core.wmem_max = 16777216net.ipv4.ip_local_port_range = 3500 65535net.ipv4.ip_forward = 0net.ipv4.conf.default.rp_filter = 1net.ipv4.conf.default.accept_source_route = 0net.ipv4.tcp_syncookies = 0net.ipv4.tcp_rmem = 4096 87380 16777216net.ipv4.tcp_wmem = 4096 65536 16777216net.ipv4.tcp_max_syn_backlog = 16384net.ipv4.tcp_fin_timeout = 15net.ipv4.tcp_max_syn_backlog = 16384net.ipv4.tcp_tw_reuse = 1net.ipv4.tcp_tw_recycle = 1net.ipv4.tcp_slow_start_after_idle = 0vm.swappiness = 0vm.min_free_kbytes = 209715kernel.core_pattern = /data/ob/core-%e-%p-%t

关闭防火墙

#依次执行以下命令关闭防火墙systemctl disable firewalld systemctl stop firewalldsystemctl status firewalld
[root@dev ~]# systemctl disable firewalld Removed symlink /etc/systemd/system/multi-user.target.wants/firewalld.service.Removed symlink /etc/systemd/system/dbus-org.fedoraproject.FirewallD1.service.[root@dev ~]# systemctl stop firewalld[root@dev ~]# systemctl status firewalld● firewalld.service - firewalld - dynamic firewall daemon   Loaded: loaded (/usr/lib/systemd/system/firewalld.service; disabled; vendor preset: enabled)   Active: inactive (dead)     Docs: man:firewalld(1)

关闭SELinux

[root@dev ~]# vim /etc/selinux/config 将配置SELINUX=enforcing 改为 SELINUX=disabled# 保存并退出
# 更改前的状态[root@dev ~]# sestatusSELinux status:                 enabledSELinuxfs mount:                /sys/fs/selinuxSELinux root directory:         /etc/selinuxLoaded policy name:             targetedCurrent mode:                   enforcingMode from config file:          disabledPolicy MLS status:              enabledPolicy deny_unknown status:     allowedMax kernel policy version:      31
# 使生效（重启服务器，或执行以下命令）[root@dev ~]# setenforce 0# 重启[root@dev ~]# reboot# 重启后的状态[root@dev ~]# sestatusSELinux status:                 disabled

克隆虚拟机

关闭虚拟机，将配置好的机器再克隆出来3份（注意，是完整克隆）

差异化配置

每次开启一台虚拟机（ip相同，别同时起,这样一个ssh就搞定了），调整主机名和ip地址，一次设置为153、154、155、156

#修改主机名[root@dev ~]# vim /etc/host改为对应的dev3、dev4、dev5# 修改静态ip[root@dev ~]# vim /etc/sysconfig/network-scripts/ifcfg-ens33 # 修改IPADDRIPADDR=192.168.1.156

中控机设置无密码SSH登录目标机器

ps: 需要拥有目标机器的root用户权限。原理，就是通过ssh-copy-id将本机的密钥复制到目标机器的authorized_keys文件中，让目标机器信任中控机；

可以跳过这3步，复制下面的脚本

检查中控机上是否有密钥存在（新机器铁定没有）

ls ~/.ssh/id_rsa.pub

中控机生成SSH公钥和私钥（执行命令后后一直回车即可）

[root@dev6 ~]# ssh-keygen -t rsa

在中控机器上登录目标机器（需要输入密码，输入密码后，以后就不用再输密码了）

[root@dev6 ~]# ssh-copy-id root@192.168.1.153/usr/bin/ssh-copy-id: INFO: Source of key(s) to be installed: "/root/.ssh/id_rsa.pub"The authenticity of host '192.168.1.153 (192.168.1.153)' can't be established.ECDSA key fingerprint is SHA256:/3F5WaADETxAuquKv63YuXs/bAWZu50NOHetNZCKa9o.ECDSA key fingerprint is MD5:f8:c0:de:e0:e8:46:80:2f:af:56:fc:14:d7:78:29:7a.Are you sure you want to continue connecting (yes/no)? yes/usr/bin/ssh-copy-id: INFO: attempting to log in with the new key(s), to filter out any that are already installed/usr/bin/ssh-copy-id: INFO: 1 key(s) remain to be installed -- if you are prompted now it is to install the new keysroot@192.168.1.153's password: 
Number of key(s) added: 1
Now try logging into the machine, with:   "ssh 'root@192.168.1.153'"and check to make sure that only the key(s) you wanted were added.# 完成以后，就可以直接免密登录了[root@dev6 ~]# ssh root@192.168.1.153Last failed login: Sun Nov 27 10:33:46 CST 2022 from 192.168.1.156 on ssh:nottyThere was 1 failed login attempt since the last successful login.Last login: Sat Nov 26 21:27:10 2022 from 192.168.1.108
#在中控机上做appuser免密登录[root@dev6 ~]# ssh-copy-id appuser@192.168.1.153[root@dev6 ~]# ssh-copy-id appuser@192.168.1.154[root@dev6 ~]# ssh-copy-id appuser@192.168.1.155[root@dev6 ~]# ssh-copy-id appuser@192.168.1.156

注意，用什么用户部署，就需要这个用户做免密登录，执行类似的ssh-copy-id appuser@192.168.1.156

以上的三步可以封装成脚本

我的root和appuser的密码都一样，shell没有进行id_rsa.pub 校验，每次执行都是重新生成

#!/usr/bin/bash   SERVERS=("root@192.168.1.153" "root@192.168.1.154" "root@192.168.1.155" "appuser@192.168.1.153" "appuser@192.168.1.154" "appuser@192.168.1.155" "appuser@192.168.1.156")   PASSWORD="101790"   keygen() {   yum -y install expect   expect -c "      spawn ssh-keygen -t rsa      expect {         *(/root/.ssh/id_rsa):* { send -- \r;exp_continue}         *(y/n)* { send -- y\r;exp_continue}         *Enter* { send -- \r;exp_continue}         *(y/n)* { send -- y\r;exp_continue}         *Enter* { send -- \r;exp_continue}         eof {exit 0}      }      expect eof   "   }   copy(){   expect -c "      set timeout -1      spawn ssh-copy-id $1      expect {         *(yes/no)* { send -- yes\r; exp_continue }         *password:* { send -- $PASSWORD\r; exp_continue}         eof {exit 0}      }      expect eof   "   }   ssh_copy_id_to_all(){   keygen ;   for host in ${SERVERS[@]}   do         copy $host   done   }   ssh_copy_id_to_all

时钟源ntp配置（需要所有机器root）

修改中控的ntp源

[root@dev6 ~]# vim /etc/ntp.conf #将下面一行server  192.168.1.156# 修改为 阿里云的ntp服务，作为中控机的ntp源server  ntp.aliyun.com　　　　# 保存并退出
# 重启ntp服务端[root@dev6 ~]# service ntpd restart# 查看ntp是否启动成功[root@dev6 ~]# netstat -anp | grep   ntpdudp        0      0 192.168.122.1:123       0.0.0.0:*                           12498/ntpd          udp        0      0 192.168.1.156:123       0.0.0.0:*                           12498/ntpd          udp        0      0 127.0.0.1:123           0.0.0.0:*                           12498/ntpd          udp        0      0 0.0.0.0:123             0.0.0.0:*                           12498/ntpd          udp6       0      0 fe80::20c:29ff:fea2:123 :::*                                12498/ntpd          udp6       0      0 ::1:123                 :::*                                12498/ntpd          udp6       0      0 :::123                  :::*                                12498/ntpd          unix  2      [ ]         DGRAM                    140639   12498/ntpd    
#我们看下ntp.aliyun.com的ip地址是什么[root@dev6 ~]# ping ntp.aliyun.comPING ntp.aliyun.com (203.107.6.88) 56(84) bytes of data.64 bytes from 203.107.6.88 (203.107.6.88): icmp_seq=1 ttl=53 time=17.1 ms#我们看下ntp 服务的源也是203.107.6.88 ，说明配置成功[root@dev6 ~]# ntpq -p     remote           refid      st t when poll reach   delay   offset  jitter============================================================================== 203.107.6.88    100.107.25.114   2 u    3   64    3   20.746  -145.90   0.787
#刚配置完，并没有同步[root@dev6 ~]# ntpstatunsynchronised  time server re-starting   polling server every 8 s#同步成功以后   [root@dev6 ~]# ntpstatsynchronised to NTP server (203.107.6.88) at stratum 3   time correct to within 26 ms   polling server every 64 s

在目标机器开启NTP客户端服务

ps: NTP 服务端服务启动后，您需要等待 5 分钟再开启 NTP 客户端服务。否则系统会提示no server suitable for synchronization found。

使用ntpdate手动立即同步

# 修改客户端时间[root@dev3 ~]# date -s "2022-10-30 10:27:16"2022年 10月 30日 星期日 10:27:16 CST# 查看修改后的时间[root@dev3 ~]# date2022年 10月 30日 星期日 10:27:32 CST#使用ntpdate立即同步ntp服务器时间[root@dev3 ~]# ntpdate 192.168.1.15627 Nov 15:51:50 ntpdate[14021]: step time server 192.168.1.156 offset 2438624.076604 sec# 同步后的时间[root@dev3 ~]# date2022年 11月 27日 星期日 15:52:25 CST

注意：以上只是同步一次时间，可以设置为定时任务，每天进行同步

[root@dev3 ~]# crontab -e0 12 * * * /usr/sbin/ntpdate 192.168.1.156

自动同步（以ntp作为客户端）

[root@dev3 ~]#  vim /etc/ntp.conf#修改内容如下server 192.168.1.156　　prefer　# 保存并退出
# 启动ntpd服务#[root@dev3 ~]# service ntpd restart# 查看ntp的状态，可以看到已经指向了192.168.1.156[root@dev3 ~]# ntpstatsynchronised to NTP server (192.168.1.156) at stratum 4   time correct to within 37 ms   polling server every 128 s

部署

从4.0开始，ob提供了统一的安装包all-in-one package。我们可以通过这个包一次性完成OBD、OceanBase数据库、OBProxy、obagent等的安装。

下载&解压

在dev6上下载

# 下载[root@dev6 ~]# cd /app/[root@dev6 app]# wget https://obbusiness-private.oss-cn-shanghai.aliyuncs.com/download-center/opensource/oceanbase-all-in-one/7/x86_64/oceanbase-all-in-one.4.0.0.0-beta-100120221102135736.el7.x86_64.tar.gz
#解压[root@dev6 app]# tar zxvf oceanbase-all-in-one.4.0.0.0-beta-100120221102135736.el7.x86_64.tar.gz

安装OBD

[root@dev6 app]# cd oceanbase-all-in-one/bin/[root@dev6 bin]# lsenv.sh  install.sh  uninstall.sh#安装[root@dev6 bin]# ./install.sh # 检验obd[root@dev6 bin]# which obd/root/.oceanbase-all-in-one/obd/usr/bin/obd[root@dev6 bin]# which obclient/root/.oceanbase-all-in-one/obclient/u01/obclient/bin/obclient

安装数据库

安装成功以后，在~/.oceanbase-all-in-one/conf/autodeploy目录下会有相关的配置文件，我们重点看下autodeploy目录下的配置文件。

single-example.yaml单机版部署配置文件
default-example.yaml集群版部署配置文件

我部署的是集群版

[root@dev6 autodeploy]# cd ~/.oceanbase-all-in-one/conf/autodeploy[root@dev6 autodeploy]# vim default-example.yaml # 说明##  如果使用root部署，就不需要再改user这里了,我用appuser部署，就改了（root之前已经做了免密登录）user:   username: appuser   password: 101790## 修改部署的服务器和相关目录oceanbase-ce:  servers:    - name: server1 # 起的名字，在后面可以根据名字配置不同的配置，后面只配置了zone      # Please don't use hostname, only IP can be supported      ip: 192.168.1.153    - name: server2      ip: 192.168.1.154    - name: server3      ip: 192.168.1.155  global:    # ob 部署的目录    home_path: /app/observer    # 数据目录,数据目录一定要是空的，存储的是sstable的文件    data_dir: /data/ob    # ob的 clog, ilog, and slog. 不配置在data目录，测试就放一起了    #redo_dir: /redo/ob    # 前面找的网卡名称，跟servers里指定的ip对应的网卡，如果网卡不同，可以参考zone那里，每个server填写    devname: ens33        #进程observer对应的链接端口，默认2181    #mysql_port: 2881     #observer跟其他进程之间的rpc通信端口，默认2182    #rpc_port: 2882     #ob设置的最大内存，最小不少于8G    #memory_limit: 8G     # OB留给集群内部用的保留内存，会占用memory_limit    #system_memory: 4G    #运行日志的级别，默认日志级别是INFO，支持 DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR 六种级别    #syslog_level: WARN    # 日志是否覆盖，最多保留max_syslog_file_count 数量的日志文件    #enable_syslog_recycle: TRUE    #日志文件数量，这里默认保留最多 10 个历史运行日志文件    #max_syslog_file_count: 10    #OB集群的超级管理员root@sys的密码，默认为空    #root_password: 密码    #OBProxy连接OB集群使用的账户名（proxyro）的密码,默认为空    #proxyro_password: 密码  server1:    zone: zone1 # 逻辑机房的概念  server2:    zone: zone2  server3:    zone: zone3obproxy-ce:  depends:    - oceanbase-ce  servers:    - 192.168.1.156  global:    # The working directory for obproxy. Obproxy is started under this directory. This is a required field.    home_path: /app/obproxy    skip_proxy_sys_private_check: true    enable_strict_kernel_release: falseobagent:  depends:    #依赖的配置，会自动复用集群的`proxyro`密码、集群名等    - oceanbase-ce   servers:    - name: server1      # Please don't use hostname, only IP is supported.      ip: 192.168.1.153    - name: server2      ip: 192.168.1.154    - name: server3      ip: 192.168.1.155  global:    # The working directory for obagent. obagent is started under this directory. This is a required field.    home_path: /app/obagent#prometheus:#  servers:#    - 192.168.1.156#  global:#    home_path: /app/prometheus    #grafana:#  servers:#    - 192.168.1.156#  depends:#    - prometheus#  global:#    home_path: /app/grafana#    login_password: oceanbase

部署

#在  ~/.oceanbase-all-in-one/conf/autodeploy  目录下执行[root@dev6 autodeploy]$ obd cluster autodeploy obtest -c default-example.yaml

友情提示：deploy name 使用_命名不要用-命名，否则在ocp中无法显示，ob的命名规范是_

# 如果部署过程中失败，[ERROR] Deploy "obdev" is deployed. You could not deploy an deployed cluster. # 通过obd cluster list 查看部署的集群，然后，删除对应的配置[root@dev6 autodeploy]# obd cluster list+------------------------------------------------------+|                     Cluster List                     |+--------+---------------------------+-----------------+| Name   | Configuration Path        | Status (Cached) |+--------+---------------------------+-----------------+| obtest | /root/.obd/cluster/obtest | deployed        || obdev  | /root/.obd/cluster/obdev  | deployed        |+--------+---------------------------+-----------------+#删除[root@dev6 autodeploy]# rm -rf ~/.obd/cluster/obtest

部署成功后，default-example.yaml会复制到/root/.obd/cluster/obtest 目录下，并改名，以后集群读取的就是这个文件了

我们看下obd的语法

#我们可以一层层的使用 -help获取对应命令的帮助文档[root@dev6 autodeploy]# obd cluster autodeploy -helpUsage: obd cluster autodeploy <deploy name> [options]- deploy name 为集群名称，友情提示下，如果Options:  -c CONFIG, --config=CONFIG  # 指定配置文件                        Path to the configuration file.  -f, --force           Force autodeploy, overwrite the home_path.  -C, --clean           Clean the home path if the directory belong to you.  -U, --unuselibrepo, --ulp                        Disable OBD from installing the libs mirror                        automatically.  -A, --auto-create-tenant, --act                        Automatically create a tenant named `test` by using                        all the available resource of the cluster.  --force-delete        Force delete, delete the registered cluster.  -s, --strict-check    Throw errors instead of warnings when check fails.  -h, --help            Show help and exit.  -v, --verbose         Activate verbose output.

到此，安装完成；

安装完毕在observer中

[appuser@dev3 root]$ tree /app/observer//app/observer/├── admin│   ├── dbms_application_body_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_application_body_mysql.sql│   ├── dbms_application_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_application_mysql.sql│   ├── dbms_monitor_body_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_monitor_body_mysql.sql│   ├── dbms_monitor_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_monitor_mysql.sql│   ├── dbms_scheduler_mysql_body.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_scheduler_mysql_body.sql│   ├── dbms_scheduler_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_scheduler_mysql.sql│   ├── dbms_session_body_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_session_body_mysql.sql│   ├── dbms_session_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_session_mysql.sql│   ├── dbms_stats_body_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_stats_body_mysql.sql│   ├── dbms_stats_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./dbms_stats_mysql.sql│   ├── __dbms_upgrade_body_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./__dbms_upgrade_body_mysql.sql│   └── __dbms_upgrade_mysql.sql -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/admin/./__dbms_upgrade_mysql.sql├── audit│   └── observer_6888_20221202163057106846170.aud├── bin│   ├── import_time_zone_info.py -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/bin/./import_time_zone_info.py│   └── observer -> /home/appuser/.obd/repository/oceanbase-ce/4.0.0.0/42611dc51ca9bb28f36e60e4406ceea4a74914c7/bin/./observer├── etc│   ├── fill_help_tables-ob.sql│   ├── mysql_to_ora_priv.py│   ├── observer.config.bin│   ├── observer.config.bin.history│   ├── priv_checker.py│   ├── timezone_V1.log│   ├── upgrade_checker.py│   ├── upgrade_cluster_health_checker.py│   ├── upgrade_post_checker.py│   ├── upgrade_post.py│   ├── upgrade_pre.py│   ├── upgrade_rolling_post.py│   └── upgrade_rolling_pre.py├── etc2│   ├── observer.conf.bin│   └── observer.conf.bin.history├── etc3│   ├── observer.conf.bin│   └── observer.conf.bin.history├── lib│   ├── libaio.so -> /home/appuser/.obd/repository/oceanbase-ce-libs/4.0.0.0/188919f8128394bf9b62e3989220ded05f1d14da/libaio.so│   ├── libaio.so.1 -> /home/appuser/.obd/repository/oceanbase-ce-libs/4.0.0.0/188919f8128394bf9b62e3989220ded05f1d14da/libaio.so.1│   ├── libaio.so.1.0.1 -> /home/appuser/.obd/repository/oceanbase-ce-libs/4.0.0.0/188919f8128394bf9b62e3989220ded05f1d14da/libaio.so.1.0.1│   ├── libmariadb.so -> /home/appuser/.obd/repository/oceanbase-ce-libs/4.0.0.0/188919f8128394bf9b62e3989220ded05f1d14da/libmariadb.so│   └── libmariadb.so.3 -> /home/appuser/.obd/repository/oceanbase-ce-libs/4.0.0.0/188919f8128394bf9b62e3989220ded05f1d14da/libmariadb.so.3├── log│   ├── election.log│   ├── election.log.wf│   ├── observer.log│   ├── observer.log.20221203133323515│   ├── observer.log.20221203150336649│   ├── observer.log.20221203163403613│   ├── observer.log.20221203180408998│   ├── observer.log.wf│   ├── observer.log.wf.20221203133323515│   ├── observer.log.wf.20221203150336649│   ├── observer.log.wf.20221203163403613│   ├── observer.log.wf.20221203180408998│   ├── rootservice.log│   ├── rootservice.log.20221203033815206│   ├── rootservice.log.20221203072725586│   ├── rootservice.log.20221203111625926│   ├── rootservice.log.20221203150526647│   ├── rootservice.log.wf│   ├── rootservice.log.wf.20221203033815206│   ├── rootservice.log.wf.20221203072725586│   ├── rootservice.log.wf.20221203111625926│   ├── rootservice.log.wf.20221203150526647│   ├── trace.log│   ├── trace.log.20221202213749679│   ├── trace.log.20221203030453184│   ├── trace.log.20221203082749178│   ├── trace.log.20221203135455886│   ├── trace.log.wf│   ├── trace.log.wf.20221203030453184│   ├── trace.log.wf.20221203082749178│   └── trace.log.wf.20221203135455886├── run│   ├── lua.sock│   ├── mysql.sock│   └── observer.pid└── store -> /data/ob

常用操作

查看集群状态

# 查看有哪些集群，以及集群在obd缓存的状态[root@dev6 autodeploy]# obd cluster list+------------------------------------------------------+|                     Cluster List                     |+--------+---------------------------+-----------------+| Name   | Configuration Path        | Status (Cached) |+--------+---------------------------+-----------------+| obtest | /root/.obd/cluster/obtest | running         |+--------+---------------------------+-----------------+
# 通过集群名称，查看集群相关组件的状态[root@dev6 autodeploy]# obd cluster display obtestGet local repositories and plugins okOpen ssh connection okCluster status check okConnect to observer okWait for observer init ok+-------------------------------------------------+|                     observer                    |+---------------+---------+------+-------+--------+| ip            | version | port | zone  | status |+---------------+---------+------+-------+--------+| 192.168.1.153 | 4.0.0.0 | 2881 | zone1 | ACTIVE || 192.168.1.154 | 4.0.0.0 | 2881 | zone2 | ACTIVE || 192.168.1.155 | 4.0.0.0 | 2881 | zone3 | ACTIVE |+---------------+---------+------+-------+--------+obclient -h192.168.1.153 -P2881 -uroot -Doceanbase
Connect to obproxy ok+-------------------------------------------------+|                     obproxy                     |+---------------+------+-----------------+--------+| ip            | port | prometheus_port | status |+---------------+------+-----------------+--------+| 192.168.1.156 | 2883 | 2884            | active |+---------------+------+-----------------+--------+obclient -h192.168.1.156 -P2883 -uroot -Doceanbase+---------------------------------------------------+|                      obagent                      |+---------------+-------------+------------+--------+| ip            | server_port | pprof_port | status |+---------------+-------------+------------+--------+| 192.168.1.153 | 8088        | 8089       | active || 192.168.1.154 | 8088        | 8089       | active || 192.168.1.155 | 8088        | 8089       | active |+---------------+-------------+------------+--------+

集群操作

我们先来看下集群操作的命令

[root@dev6 autodeploy]# obd cluster -hUsage: obd cluster <command> [options]
Available commands:# 自动部署集群，我们用的就是这种方式，指定了配置文件autodeploy     Deploy a cluster automatically by using a simple configuration file.
check4ocp      Check Whether OCP Can Take Over Configurations in Use
chst           Change Deployment Configuration Style
deploy         Deploy a cluster by using the current deploy configuration or a deploy yaml file.#销毁集群destroy        Destroy a deployed cluster.# 显示集群信息display        Display the information for a cluster.
edit-config    Edit the configuration file for a specific deployment.#查看集群状态list           List all the deployments.
redeploy       Redeploy a started cluster.
reinstall      Reinstall a deployed component#重新加载集群reload         Reload a started cluster.#重启集群，后面跟集群名称restart        Restart a started cluster.#启动集群，后面跟集群名称，第一次运行start会对集群进行初始化，再使用start只会启动集群命令节点start          Start a deployed cluster.# 停止集群stop           Stop a started cluster.#租户操作tenant         Create or drop a tenant.# 集群升级upgrade        Upgrade a cluster.

查看集群状态

# 查看集群状态obd cluster display obtest;# 重启集群对应的obagentobd cluster restart obtest -c obagent;

手动重启OB节点(OBD是整个集群都重启了)

#以appuser用户登录dev3#获取当前observer的进程id[appuser@dev3 app]$ ps -ef | grep observer | grep -v grepappuser    6889      1 74 12月02 ?      20:07:20 /app/observer/bin/observer -r 192.168.1.153:2882:2881;192.168.1.154:2882:2881;192.168.1.155:2882:2881 -p 2881 -P 2882 -z zone1 -n obcluster -c 1 -d /data/ob -i ens33 -l INFO -o __min_full_resource_pool_memory=2147483648,enable_syslog_recycle=True,enable_syslog_wf=True,max_syslog_file_count=4,memory_limit=6G,system_memory=1G,cpu_count=16,datafile_size=44G,log_disk_size=20G
# kill 进程，集群会转移，如果直接kill -9 ,测试环境无所谓，生产就呵呵了[appuser@dev3 app]$ kill 6889#  等待 60s，等进程完全退出sleep 60#  反复确认进程完全退出ps -ef | grep observer#  配置 LIBRARY PATH（部署目录的lib下）export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/app/observer/lib/#  进入部署目录，启动进程cd /app/observer && bin/observer#  等待 10s 进程启动sleep 10#  反复确认进程启动时没有退出ps -ef | grep observer | grep -v grep#  等待 60s，等进程完全启动并恢复完毕sleep 60#  查看进程监听成功（默认监听 2881 和  2882 端口）netstat -ntlp|grep 288*

连接

MySQL 5.5/5.6/5.7可以直接链接OB的社区版；也可以用obclient连接；安装后的Ob 端口是2881，用户名是root@sys，不设置密码默认为空,安装完obd以后自动安装上了obclient；

#注意，obclient的帮助命令是obclient -help# -h 指定host,我在156上部署了obproxy# -P 大写P,断口# -p 小写p,密码# -u 用户名# -D 使用哪个库[root@dev6 autodeploy]# obclient -h192.168.1.156 -P2883  -uroot@sys Welcome to the OceanBase.  Commands end with ; or \g.Your OceanBase connection id is 3Server version: OceanBase_CE 4.0.0.0obclient [(none)]> show databases;+--------------------+| Database           |+--------------------+| information_schema || LBACSYS            || mysql              || oceanbase          || ORAAUDITOR         || SYS                || test               |+--------------------+7 rows in set (0.027 sec)

查看和修改集群参数

通过show parameters like ‘%参数名特征%’查看OB集群参数，不带like 会查看所有;

可通过命令 alter system set 参数名='参数值' [ server = '节点IP:节点RPC端口' ] ; 进行修改。不指定 server 子句表示参数修改应用于所有 OceanBase 集群节点。

# 查看所有obclient [oceanbase]> show parameters;# 查看系统日志级别obclient [oceanbase]> show parameters like 'syslog_level'\G;*************************** 1. row ***************************      zone: zone1  svr_type: observer    svr_ip: 192.168.1.153  svr_port: 2882      name: syslog_level data_type: NULL     value: INFO      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE*************************** 2. row ***************************      zone: zone2  svr_type: observer    svr_ip: 192.168.1.154  svr_port: 2882      name: syslog_level data_type: NULL     value: INFO      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE*************************** 3. row ***************************      zone: zone3  svr_type: observer    svr_ip: 192.168.1.155  svr_port: 2882      name: syslog_level data_type: NULL     value: INFO      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE3 rows in set (0.212 sec)
ERROR: No query specified# 修改日志级别为DEBUGobclient [oceanbase]> alter system set syslog_level = 'DEBUG' server='192.168.1.153:2882' ;Query OK, 0 rows affected (0.131 sec)#查看，发现已经生效obclient [oceanbase]> show parameters like 'syslog_level'\G;*************************** 1. row ***************************      zone: zone2  svr_type: observer    svr_ip: 192.168.1.154  svr_port: 2882      name: syslog_level data_type: NULL     value: INFO      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE*************************** 2. row ***************************      zone: zone3  svr_type: observer    svr_ip: 192.168.1.155  svr_port: 2882      name: syslog_level data_type: NULL     value: INFO      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE*************************** 3. row ***************************      zone: zone1  svr_type: observer    svr_ip: 192.168.1.153  svr_port: 2882      name: syslog_level data_type: NULL     value: DEBUG      info: specifies the current level of logging. There are DEBUG, TRACE, INFO, WARN, USER_ERR, ERROR, six different log levels.   section: OBSERVER     scope: CLUSTER    source: DEFAULTedit_level: DYNAMIC_EFFECTIVE3 rows in set (0.019 sec)

ITPub博客

oceanbase-安装

部署规划

机器准备

hosts设置

设置机器名和静态ip

关闭大页

配置信息查看

时钟源ntp配置（需要所有机器root）

‍磁盘规划

创建用户

配置limits.conf

配置sysctl.conf

关闭防火墙

关闭SELinux

克隆虚拟机

差异化配置

中控机设置无密码SSH登录目标机器

时钟源ntp配置（需要所有机器root）

部署

下载&解压

安装OBD

安装数据库

常用操作

查看集群状态

集群操作

连接

查看和修改集群参数

最新文章