Centos7.6安装MogDB

最新推荐文章于 2024-07-02 18:00:00 发布

jgmydsai

最新推荐文章于 2024-07-02 18:00:00 发布

阅读量927

点赞数 28

分类专栏： MogDB 文章标签： MogDB 安装

本文链接：https://blog.csdn.net/jgmydsai/article/details/137959570

版权

MogDB 专栏收录该内容

3 篇文章 0 订阅

订阅专栏

一、环境配制

1、确认OS版本

安装依赖包

yum -y install redhat-lsb

查看版本

[root@CentOS ~]# lsb_release -d
Description:	CentOS Linux release 7.6.1810 (Core)

2、为配制本地源挂载光驱

root下执行

mkdir -p /media/cdrom/
mount /dev/cdrom /media/cdrom

或使用iso文件

mkdir -p /media/cdrom/
mount -o loop /root/CentOS-7.6-x86_64-DVD-1804.iso /media/cdrom

3、配制本地源

root下执行

整理/etc/yum.repos.d/ 下的文件，可以只保留一个，内容改为

[base]
name=base
baseurl=file:///media/cdrom
enabled=1
gpgcheck=0

root下执行

yum clean all
yum makecache

4、安装python3

root下执行

1).安装依赖环境

yum -y install zlib-devel bzip2-devel openssl-devel ncurses-devel sqlite-devel readline-devel tk-devel gdbm-devel db4-devel libpcap-devel xz-devel gcc-c++ lrzsz

2).下载Python3

地址 https://www.python.org/downloads/

wget https://www.python.org/ftp/python/3.6.15/Python-3.6.15.tgz

3).安装python3

mkdir -p /usr/local/python3
tar -zxvf Python-3.6.15.tgz -C /usr/local/python3

4).进入解压后的目录，编译安装。

安装在 /usr/local/python3

cd /usr/local/python3/Python-3.6.15
./configure --prefix=/usr/local/python3 --enable-shared CFLAGS=-fPIC
make && make install

5).建立python3的软链

ln -s /usr/local/python3/bin/python3.6 /usr/bin/python3.6 -f
ln -s /usr/local/python3/bin/python3 /usr/bin/python3 -f
ln -s /usr/local/python3/bin/pip3.6 /usr/bin/pip3.6 -f
ln -s /usr/local/python3/bin/pip3 /usr/bin/pip3 -f

6).复制so

cp /usr/local/python3/Python-3.6.15/libpython3.6m.so.1.0 /usr/lib64/

7).并将/usr/local/python3/bin加入PATH(OMM下也要配制)

配制环境变量
vim ~/.bash_profile

# User specific environment and startup programs
PATH=$PATH:/usr/local/python3/bin
export PATH

8).验正版本

# python3 -V
Python 3.6.15

# pip3 -V
pip 18.1 from /usr/local/python3/lib/python3.6/site-packages/pip (python 3.6)

5、启动命令界面

systemctl set-default multi-user.target

6、禁止 SELINUX、防火墙

systemctl disable firewalld.service
systemctl stop firewalld.service

sed -i '/^SELINUX=/c'SELINUX=disabled /etc/selinux/config

重启后确认更改正确

[root@mogdb-02 ~]# getenforce 
Disabled

7、安装依赖包

yum install bzip2-x86_64 libaio-devel flex bison ncurses-devel glibc-devel patch redhat-lsb-core readline-devel lrzsz tigervnc-server

8、设置编码

echo "export LANG=en_US.UTF-8" >> ~/.bash_profile

9、关闭swap交换内存

cp /etc/fstab  /etc/fstab.bak
sed -i '/swap/s/^/#/' /etc/fstab
cat /etc/fstab|grep -v ^#|grep -v '^$'

关闭swap

swapoff -a

10、MTU使用默认值

[root@CentOS-7 ~]# ifconfig | grep mtu
enp0s3: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
enp0s8: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
virbr0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500

11、设置root用户远程登录

sed -i 's/^Banner .*/Banner none/' /etc/ssh/sshd_config
sed -i 's/^#PermitRootLogin .*/PermitRootLogin yes/' /etc/ssh/sshd_config
sed -i 's/^PermitRootLogin no/PermitRootLogin yes/'  /etc/ssh/sshd_config
systemctl restart sshd

[root@CentOS yum.repos.d]# egrep  'Banner|PermitRootLogin' /etc/ssh/sshd_config
PermitRootLogin yes
# the setting of "PermitRootLogin without-password".
#Banner none

12、关闭透明大页

验正是否关闭

[root@mogdb-304 ~]# cat /proc/sys/vm/nr_hugepages
0
[root@mogdb-304 ~]# grep -i HugePages_Total /proc/meminfo
HugePages_Total:       0

关闭：更改 /etc/default/grub，增加 transparent_hugepage=never
[root@mogdb-304 ~]# cat /etc/default/grub
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL_OUTPUT="console"
GRUB_CMDLINE_LINUX="crashkernel=auto rd.lvm.lv=centos_centos-76/root rd.lvm.lv=centos_centos-76/swap rd.lvm.lv=centos_centos-76/usr rhgb quiet transparent_hugepage=never"
GRUB_DISABLE_RECOVERY="true"

13、更改文件系统参数

vi /etc/security/limits.conf

root       soft    as     unlimited
omm        soft    as     unlimited
root       hard    as     unlimited
omm        hard    as     unlimited
root       soft    nproc  unlimited
omm        soft    nproc  unlimited
root       hard    nproc  unlimited
omm        hard    nproc  unlimited
*          soft    nofile 1000000
*          hard    nofile 1000000
*          stack   size   3072

14、准备安装目录

mkdir -p /opt/mogdb/db1
mkdir -p /opt/mogdb
mkdir -p /opt/omm/mogdb
cd /opt/omm/mogdb

15、手动配制ssh互信

PTK可自动配置互信，如果遇到问题可以手动配制

ssh-keygen -t rsa
ssh-copy-id mogdb-02

二、PTK安装

参见： https://docs.mogdb.io/zh/ptk/v1.1/install

1. 安装

curl --proto '=https' --tlsv1.2 -sSf https://cdn-mogdb.enmotech.com/ptk/install.sh | sh

安装后，重连ssh，或重启或执行下面命令

export PATH=/root/.ptk/bin:$PATH

2. 升级

ptk self upgrade

3. 查看下载链接

无外网环境下可以用下面的命令查看对应的MogDB下载链接，根据得到的地址去下载

./ptk download --dryrun

[root@localhost omm]# ./ptk download --dryrun
INFO[2023-10-23T18:47:55.460] PTK Version: 1.0.9 release                   
INFO[2023-10-23T18:47:55.460] detect current os...                         
INFO[2023-10-23T18:47:55.536] target url: https://cdn-mogdb.enmotech.com/mogdb-media/5.0.2/Kylin_x86_64/MogDB-5.0.2-Kylin-x86_64-all.tar.gz

三、单机安装

单机安装不支持CM，配制CM会有提示

[PTK-4039] cm_servers requires at least 2 hosts

1、配制安装参数

参见： https://docs.mogdb.io/zh/ptk/v1.1/usage-config

[root@mogdb-02 software]# cat config.yaml

global:
    cluster_name: dbCluster
    user: omm
    group: dbgrp
    base_dir: /opt/mogdb
    data_dir: /opt/mogdb/db1
    tool_dir: /opt/mogdb/tool
    tmp_dir:  /opt/mogdb/tmp
    log_dir:  /opt/mogdb/log
db_servers:
    - host: 192.168.56.66
      db_port: 5432

2、checkos

check时增加参数 gen-warning-fix # 输出 root-fix 脚本的时候，将 warning 级别的检查项也生成修复指令

[root@mogdb-02 software]# ptk checkos -f config.yaml --gen-warning-fix
INFO[2022-12-02T10:59:20.876] prechecking dependent tools...               
INFO[2022-12-02T10:59:21.223] [192.168.56.61][omm] check port 5432         
INFO[2022-12-02T10:59:22.093] [192.168.56.61][omm] port 5432 is free       
ERRO[2022-12-02T10:59:42.772] [192.168.56.61][omm] not found package: numactl 
INFO[2022-12-02T10:59:42.790] [192.168.56.61][omm] platform: centos_7_64bit 
INFO[2022-12-02T10:59:42.796] [192.168.56.61][omm] kernel version: 3.10.0-957.el7.x86_64 
INFO[2022-12-02T10:59:42.868] [192.168.56.61][omm] locale: LANG=en_US.UTF-8 
INFO[2022-12-02T10:59:42.883] [192.168.56.61][omm] timezone: +0800         
WARN[2022-12-02T10:59:42.906] [192.168.56.61][omm] device(dm-1) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.906] [192.168.56.61][omm] device(dm-8) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.906] [192.168.56.61][omm] device(sda) 'IO Request'=32768, expect 256 
WARN[2022-12-02T10:59:42.906] [192.168.56.61][omm] device(sr0) 'IO Request'=32768, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-0) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-2) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-3) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-4) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-5) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-6) 'IO Request'=128, expect 256 
WARN[2022-12-02T10:59:42.907] [192.168.56.61][omm] device(dm-7) 'IO Request'=128, expect 256 
INFO[2022-12-02T10:59:42.924] [192.168.56.61][omm] swap memory 10485756kB, total memory 8008932kB 
WARN[2022-12-02T10:59:42.931] [192.168.56.61][omm] net.ipv4.tcp_retries1=3, expect 5 
WARN[2022-12-02T10:59:43.033] [192.168.56.61][omm] net.ipv4.tcp_syn_retries=6, expect 5 
INFO[2022-12-02T10:59:43.114] [192.168.56.61][omm] check kernel.core_pattern 
INFO[2022-12-02T10:59:43.126] [192.168.56.61][omm] check removeIPC value   
INFO[2022-12-02T10:59:43.199] all checkers finished                        
INFO[2022-12-02T10:59:43.200] write fix os script to root_fix_os.2022.1202.105943.sh success 
INFO[2022-12-02T10:59:43.200] Time elapsed: 22s                            
# Check Results
                Item                |  Level    
------------------------------------+-----------
  A1.Check_OS_Version               | OK        
  A2.Check_Kernel_Version           | OK        
  A3.Check_Unicode                  | OK        
  A4.Check_TimeZone                 | OK        
  A5.Check_Swap_Memory_Configure    | Warning   
  A6.Check_SysCtl_Parameter         | Abnormal  
  A7.Check_FileSystem_Configure     | OK        
  A8.Check_Disk_Configure           | OK        
  A9.Check_BlockDev_Configure       | OK        
  A9.Check_Logical_Block            | OK        
  A10.Check_IO_Request              | Warning   
  A10.Check_Asynchronous_IO_Request | OK        
  A10.Check_IO_Configure            | OK        
  A11.Check_Network_Configure       | OK        
  A12.Check_Time_Consistency        | OK        
  A13.Check_Firewall_Service        | OK        
  A14.Check_THP_Service             | OK        
  A15.Check_Dependent_Package       | Abnormal  
  A16.Check_CPU_Instruction_Set     | OK        
  A17.Check_Port                    | OK        
Total count 20, abnormal count 2, warning count 2
Failed to check os, can’t perform installation unless fix all the abnormal items
You can use 'ptk checkos -i ITEM --detail' to get detail message
Please check root_fix_os.[TIMESTAMP].sh for commands to resolve.

3、通过前面生成的脚本 Fix 问题

需要保证yum正常，执行脚本后要重启

[root@mogdb-02 software]# ll
total 16
drwxr-xr-x 3 54332 54331 4096 Jun 16 17:26 compat-tools-master
-rw-r--r-- 1 root  root   208 Dec  2 10:58 config.yaml
-rw------- 1 54332 54331 1872 Jun 27 21:47 dbSingle.xml
drwxr-xr-x 2 root  root    46 Dec  2 10:35 mogdb
-rwxr--r-- 1 root  root   438 Dec  2 10:59 root_fix_os.2022.1202.105943.sh

[root@mogdb-02 software]# sh root_fix_os.2022.1202.105943.sh 
Loaded plugins: fastestmirror, langpacks
Determining fastest mirrors
epel/x86_64/metalink                                                                                                                    | 7.0 kB  00:00:00     
 * base: mirrors.aliyun.com
 * epel: mirrors.bfsu.edu.cn
 * extras: mirrors.aliyun.com
 * updates: mirrors.aliyun.com
base                                                                                                                                    | 3.6 kB  00:00:00     
epel                                                                                                                                    | 4.7 kB  00:00:00     
extras                                                                                                                                  | 2.9 kB  00:00:00     
updates                                                                                                                                 | 2.9 kB  00:00:00     
(1/4): epel/x86_64/group_gz                                                                                                             |  98 kB  00:00:00     
(2/4): epel/x86_64/updateinfo                                                                                                           | 1.0 MB  00:00:00     
(3/4): epel/x86_64/primary_db                                                                                                           | 7.0 MB  00:00:00     
(4/4): updates/7/x86_64/primary_db                                                                                                      |  18 MB  00:00:19     
Resolving Dependencies
--> Running transaction check
---> Package numactl.x86_64 0:2.0.12-5.el7 will be installed
--> Finished Dependency Resolution

Dependencies Resolved

===============================================================================================================================================================
 Package                              Arch                                Version                                      Repository                         Size
===============================================================================================================================================================
Installing:
 numactl                              x86_64                              2.0.12-5.el7                                 base                               66 k

Transaction Summary
===============================================================================================================================================================
Install  1 Package

Total download size: 66 k
Installed size: 141 k
Downloading packages:
numactl-2.0.12-5.el7.x86_64.rpm                                                                                                         |  66 kB  00:00:00     
Running transaction check
Running transaction test
Transaction test succeeded
Running transaction
  Installing : numactl-2.0.12-5.el7.x86_64                                                                                                                 1/1 
  Verifying  : numactl-2.0.12-5.el7.x86_64                                                                                                                 1/1 

Installed:
  numactl.x86_64 0:2.0.12-5.el7                                                                                                                                

Complete!
fs.file-max = 6815744
kernel.shmmni = 4096
net.ipv4.conf.all.rp_filter = 2
net.ipv4.conf.default.rp_filter = 2
fs.aio-max-nr = 1048576
net.ipv4.ip_local_port_range = 9000 65500
net.ipv4.tcp_max_tw_buckets = 10000
net.ipv4.tcp_tw_reuse = 1
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_keepalive_time = 30
net.ipv4.tcp_keepalive_intvl = 30
net.ipv4.tcp_retries2 = 12
net.ipv4.ip_local_reserved_ports = 5432-5439,20050-20057
net.core.wmem_max = 21299200
net.core.rmem_max = 21299200
net.core.wmem_default = 21299200
net.core.rmem_default = 21299200
kernel.sem = 250 6400000 1000 25600
net.ipv4.tcp_rmem = 8192 250000 16777216
net.ipv4.tcp_wmem = 8192 250000 16777216
vm.min_free_kbytes = 400469
net.core.netdev_max_backlog = 65535
net.ipv4.tcp_max_syn_backlog = 65535
net.core.somaxconn = 65535
kernel.shmall = 1152921504606846720
kernel.shmmax = 18446744073709551615
kernel.core_pattern = /var/log/coredump/%e.%p.%u.%t.core

执行完之后要重启

4、Install

Fix 之后要重启，使参数生效后再安装

cd /opt/omm/
ptk install -f config.yaml
录入初始用户密码，本例中为 Enmotech@123

[root@mogdb-02 software]# ptk install -f config.yaml --pkg MogDB-3.0.4-CentOS-x86_64.tar.gz 
INFO[2022-12-02T11:03:28.593] PTK Version: 0.5.4 release                   
INFO[2022-12-02T11:03:28.593] load config from config.yaml                 
If you continue to install the software,
you are accepting the license agreement of the software.

  [Y]: Accept and continue
  [C]: Show content of licence agreement
  [N]: Abort installation and exit

Please enter (Default [Y]): y
Cluster Name: "dbCluster"
+--------------+---------------+-------------+------+---------+------------------+----------+
| az(priority) |      ip       | user(group) | port |  role   |     data dir     | upstream |
+--------------+---------------+-------------+------+---------+------------------+----------+
| AZ1(1)       | 192.168.56.61 | omm(omm)    | 5432 | primary | /opt/mogdb/db1/db1 | -        |
+--------------+---------------+-------------+------+---------+------------------+----------+
Is cluster config correct?[Y/N](default=Y) y
INFO[2022-12-02T11:03:44.835] start check os ...                           
INFO[2022-12-02T11:03:44.847] prechecking dependent tools...               
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-0) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-2) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-6) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-8) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(sda) 'IO Request'=32768, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-1) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-3) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-4) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-5) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(dm-7) 'IO Request'=128, expect 256 
WARN[2022-12-02T11:03:44.924] [192.168.56.61][omm] device(sr0) 'IO Request'=32768, expect 256 
INFO[2022-12-02T11:03:44.962] [192.168.56.61][omm] timezone: +0800         
INFO[2022-12-02T11:03:44.976] [192.168.56.61][omm] swap memory 10485756kB, total memory 8008932kB 
INFO[2022-12-02T11:03:45.087] [192.168.56.61][omm] platform: centos_7_64bit 
INFO[2022-12-02T11:03:45.091] [192.168.56.61][omm] kernel version: 3.10.0-957.el7.x86_64 
INFO[2022-12-02T11:03:45.142] [192.168.56.61][omm] locale: LANG=en_US.UTF-8 
WARN[2022-12-02T11:03:45.167] [192.168.56.61][omm] net.ipv4.tcp_syn_retries=6, expect 5 
WARN[2022-12-02T11:03:45.200] [192.168.56.61][omm] net.ipv4.tcp_retries1=3, expect 5 
INFO[2022-12-02T11:03:45.350] [192.168.56.61][omm] check kernel.core_pattern 
INFO[2022-12-02T11:03:45.372] [192.168.56.61][omm] check removeIPC value   
INFO[2022-12-02T11:04:06.478] [192.168.56.61][omm] check port 5432         
INFO[2022-12-02T11:04:07.435] [192.168.56.61][omm] port 5432 is free       
INFO[2022-12-02T11:04:07.435] all checkers finished                        
INFO[2022-12-02T11:04:07.435] Time elapsed: 22s                            
INFO[2022-12-02T11:04:07.436] check os success                             
INFO[2022-12-02T11:04:07.436] start check distro ...                       
INFO[2022-12-02T11:04:07.436] check distro success                         
INFO[2022-12-02T11:04:07.436] start check user ...                         
INFO[2022-12-02T11:04:07.450] check user success                           
INFO[2022-12-02T11:04:07.450] start check port ...                         
INFO[2022-12-02T11:04:09.331] check port success                           
INFO[2022-12-02T11:04:09.331] start prepare installation package ...       
INFO[2022-12-02T11:04:09.331] installation package is ready                
INFO[2022-12-02T11:04:09.331] start validate installation package ...      
INFO[2022-12-02T11:04:14.074] parsed kernel package name: MogDB-3.0.4-CentOS-64bit.tar.gz 
INFO[2022-12-02T11:04:14.131] db product: MogDB, version: 3.1.0, number: 92.781, commit_id: 2aa5b715 
INFO[2022-12-02T11:04:14.131] validate installation package success        
INFO[2022-12-02T11:04:14.131] start complete the configuration ...
Please enter db password (8~16 characters): Enmotech@123
Retype db password: Enmotech@123
INFO[2022-12-02T11:12:06.633] complete the configuration success           
INFO[2022-12-02T11:12:06.633] start distribute installation package ...    
INFO[2022-12-02T11:12:06.805] distrubite packages success                  
INFO[2022-12-02T11:12:06.805] start setup db ...                           
WARN[2022-12-02T11:12:06.817] [192.168.56.61][omm] not found recommended config for version 3.1.0, fallback to use 3.0.3 recommended config 
INFO[2022-12-02T11:12:06.818] [192.168.56.61][omm] create os user omm, group omm 
INFO[2022-12-02T11:12:06.881] [192.168.56.61][omm] set ulimit configs      
INFO[2022-12-02T11:12:06.886] [192.168.56.61][omm] set user omm profiles   
INFO[2022-12-02T11:12:07.009] [192.168.56.61][omm] check or set C library  
INFO[2022-12-02T11:12:07.024] [192.168.56.61][omm] make dir(s): /opt/mogdb/tool,/opt/mogdb/tool/ptk_tool/bin 
INFO[2022-12-02T11:12:07.968] [192.168.56.61][omm] decompress MogDB-3.0.4-CentOS-x86_64.tar.gz to dir /opt/mogdb/tool 
INFO[2022-12-02T11:12:10.440] [192.168.56.61][omm] decompress *-om.tar.gz to dir /opt/mogdb/tool 
INFO[2022-12-02T11:12:11.728] [192.168.56.61][omm] try to fix psutil python lib 
INFO[2022-12-02T11:12:11.789] [192.168.56.61][omm] write file /opt/mogdb/tool/ptk_tool/bin/README.md 
INFO[2022-12-02T11:12:11.852] [192.168.56.61][omm] write file /opt/mogdb/tool/ptk_tool/bin/pscp 
INFO[2022-12-02T11:12:11.920] [192.168.56.61][omm] write file /opt/mogdb/tool/ptk_tool/bin/pssh 
INFO[2022-12-02T11:12:11.964] [192.168.56.61][omm] write file /opt/mogdb/tool/ptk_tool/bin/task_pool.py 
INFO[2022-12-02T11:12:11.987] [192.168.56.61][omm] change /opt/mogdb/tool owner to omm 
INFO[2022-12-02T11:12:11.997] [192.168.56.61][omm] make dir(s): /opt/mogdb/app 
INFO[2022-12-02T11:12:12.229] [192.168.56.61][omm] decompress MogDB-3.0.4-CentOS-64bit.tar.gz to dir /opt/mogdb/app 
INFO[2022-12-02T11:12:17.506] [192.168.56.61][omm] fix dynamic linker for gs_initdb 
INFO[2022-12-02T11:12:17.590] [192.168.56.61][omm] fix dynamic linker for mogdb 
INFO[2022-12-02T11:12:17.685] [192.168.56.61][omm] change /opt/mogdb/app owner to omm 
INFO[2022-12-02T11:12:17.706] [192.168.56.61][omm] make dir(s): /opt/mogdb/db1/log/gs_profile,/opt/mogdb/db1/log/pg_log,/opt/mogdb/db1/log/pg_audit,/opt/mogdb/db1/log/bin,/opt/mogdb/db1/log/pg_log/dn_6001,/opt/mogdb/db1/log/pg_audit/dn_6001 
INFO[2022-12-02T11:12:18.751] [192.168.56.61][omm] make dir(s): /opt/mogdb/tmp 
INFO[2022-12-02T11:12:18.840] [192.168.56.61][omm] save version to upgrade_version 
INFO[2022-12-02T11:12:18.890] [192.168.56.61][omm] create file cluster_manual_start 
INFO[2022-12-02T11:12:18.931] [192.168.56.61][omm] generate cluster_static_config file 
INFO[2022-12-02T11:12:18.938] [192.168.56.61][omm] change /opt/mogdb/app/bin/cluster_static_config owner to omm 
INFO[2022-12-02T11:12:18.942] [192.168.56.61][omm] make dir(s): /opt/mogdb/db1/db1 
INFO[2022-12-02T11:12:19.031] [192.168.56.61][omm] change /opt/mogdb/db1/db1 owner to omm 
INFO[2022-12-02T11:12:19.034] [192.168.56.61][omm] the database is initializing, please wait a few minutes 
INFO[2022-12-02T11:12:36.690] [192.168.56.61][omm] config postgresql.conf  
INFO[2022-12-02T11:12:36.791] [192.168.56.61][omm] config pg_hba.conf      
INFO[2022-12-02T11:12:36.843] setup db success                             
INFO[2022-12-02T11:12:36.843] generate certification files                 
INFO[2022-12-02T11:12:37.177] download certification files                 
INFO[2022-12-02T11:12:37.267] certification files setup success            
INFO[2022-12-02T11:12:37.267] start build ssh trust ...                    
INFO[2022-12-02T11:12:37.364] ssh trust build success                      
INFO[2022-12-02T11:12:37.364] start launch db ...                          
INFO[2022-12-02T11:12:37.364] [192.168.56.61][omm] start db by gs_ctl ...  
INFO[2022-12-02T11:12:37.364] [192.168.56.61][omm] the database is starting, please wait a few minutes 
INFO[2022-12-02T11:12:38.450] [192.168.56.61][omm] start db success        
INFO[2022-12-02T11:12:38.450] [192.168.56.61][omm] start alter db initial password ... 
INFO[2022-12-02T11:12:38.576] [192.168.56.61][omm] alter db password success 
INFO[2022-12-02T11:12:38.576] launch db success                            
INFO[2022-12-02T11:12:38.576] clear temp dirs ...                          
INFO[2022-12-02T11:12:38.596] clear temp dirs success                      
INFO[2022-12-02T11:12:38.596] Time elapsed: 9m4s                           

  cluste_name |     host      | user | port |    status     | message  
--------------+---------------+------+------+---------------+----------
  dbCluster   | 192.168.56.61 | omm  | 5432 | start_success | success  


Successfully set finish flag.
Preinstallation succeeded.

5、gen-om-xml

可以通过gen-om-xml命令生成 gs_om 的 XML 配置文件

ptk gen-om-xml -f config.yaml 
/opt/omm/config_20221202111918.xml

[root@mogdb-02 software]# cat config_20221202111918.xml 
<?xml version="1.0" encoding="UTF-8"?>
  <ROOT>
      <CLUSTER>
          <PARAM name="clusterName" value="dbCluster"></PARAM>
          <PARAM name="nodeNames" value="mogdb-02"></PARAM>
          <PARAM name="gaussdbLogPath" value="/opt/mogdb/db1/log"></PARAM>
          <PARAM name="gaussdbAppPath" value="/opt/mogdb/app"></PARAM>
          <PARAM name="gaussdbToolPath" value="/opt/mogdb/tool"></PARAM>
          <PARAM name="backIP1s" value="192.168.56.61"></PARAM>
          <PARAM name="clusterType" value="single-inst"></PARAM>
      </CLUSTER>
      <DEVICELIST>
          <DEVICE sn="mogdb-02">
              <PARAM name="name" value="mogdb-02"></PARAM>
              <PARAM name="azName" value="AZ1"></PARAM>
              <PARAM name="azPriority" value="1"></PARAM>
              <PARAM name="backIp1" value="192.168.56.61"></PARAM>
              <PARAM name="sshIp1" value="192.168.56.61"></PARAM>
              <PARAM name="dataNum" value="1"></PARAM>
              <PARAM name="dataPortBase" value="5432"></PARAM>
              <PARAM name="dataNode1" value="/opt/mogdb/db1/db1"></PARAM>
          </DEVICE>
      </DEVICELIST>
  </ROOT>

6、根据需要设置IP白名单

gs_guc set -I all -N all -h “host all all 0.0.0.0/0 md5”

四、安装扩展

1、安装所有扩展

ptk cluster -n dbCluster install-plugin --pkg

2、安装指定扩展

ptk cluster -n dbCluster install-plugin -p whale-5.0.0-01-CentOS-x86_64.tar.gz

五、compat-tools

compat-tools是一个兼容工具集合，旨在为从其他异构数据库，迁移到 openGauss 之后的系统，创建必要的函数，以及系统视图的兼容。为后续的系统运维与应用改造提供便利。

脚本中带有版本控制，直接运行脚本时，会根据如下三种情况进行处理：

若待创建对象在目标数据库中不存在，则直接进行创建
若待创建对象版本高于目标数据库中的对象版本，则进行升级重建
若待创建对象版本不高于目标数据库中的对象版本，则跳过创建

注意

1、如果要安装扩展 whale，需要在compat-tools之前安装
2、安装compat-tools，需要确认参数 behavior_compat_options=‘display_leading_zero’，如果参数不对，设置后重启再验正

cmsdb=# show behavior_compat_options;
 behavior_compat_options 
-------------------------
 display_leading_zero
(1 row)

3、为避免超时，建议延长超时设置
更改文件 Oracle_Packages.sql ，设置如下参数
set statement_timeout = 6000000;

4、安装完后要验正 dbms_lob功能是否正常安装

select dbms_lob.substr('abc',1,2);

1、下载

git clone https://gitee.com/enmotech/compat-tools.git

2、安装

注意：安装tools之前，需要先创建whale，假设目标DB为omm

gsql -d omm -r

omm=# create extension whale;
CREATE EXTENSION

开始安装tools

[omm@mogdb-02 compat-tools]$ gsql -d omm -f runMe.sql

3、卸载

[omm@mogdb-02 compat-tools]$ gsql -d omm -f uninstall.sql

六、卸载MogDB

1、查看当前已经安装的MogDB

[root@mogdb-02 ~]# ptk ls
  cluster_name |  id  |        addr        | user |    data_dir     |           db_version           |     create_time     | comment  
---------------+------+--------------------+------+-----------------+--------------------------------+---------------------+----------
  dbCluster    | 6001 | 192.168.56.62:5432 | omm  | /opt/mogdb/db1/db1 | MogDB 3.0.3.6 (build fffcf561) | 2023-03-01 10:27:46 |

2、停止DB

根据前面查看到的信息，操作对应的cluster

[root@mogdb-02 ~]# ptk cluster stop -n dbCluster
==================== stop cluster: "dbCluster" ====================
INFO[2023-05-08T11:29:27.982] operation: stop                              
INFO[2023-05-08T11:29:27.982] ========================================     
INFO[2023-05-08T11:29:27.982] stop db [192.168.56.62:5432] ...             
INFO[2023-05-08T11:29:29.035] stop db [192.168.56.62:5432] successfully    
INFO[2023-05-08T11:29:29.035] ========================================     
INFO[2023-05-08T11:29:29.035] stop successfully

3、uninstall

根据前面查看到的信息，操作对应的cluster

ptk uninstall -n dbCluster

4、手动清理残留信息

手动删除残留的安装目录

rm /opt/mogdb/* -rf
rm /opt/mogdb/db1/* -rf
rm /var/spool/mail/omm -rf

八、集群安装

1、添加vip

ifconfig enp0s8:5432 192.168.56.69 netmask 255.255.255.0 up

生成加密密码

[root@ha-01 mogdb]# ptk encrypt Enmotech@123
Enmotech@123: pTk6MmYyYmNlOTc8Qz5DPUM8PHlTWDFCUkZGYnljRVU5MWJ1NzdWcktaRU80XzZfMERFMDNPZnVIaV9TVzQ=

2、创建配置文件

2.1 一主一备

# 集群级别的配置信息，其中 `db_port` 和 `ssh_option` 字段是 **可重用字段**
global:
  cluster_name: dbCluster # 集群名称，PTK管理的集群的唯一标识
  user: omm # 运行数据库的操作系统用户名
  group: dbgrp # 运行数据库的操作系统用户组
  db_port: 5432 # 数据库端口
  ha_port: 5433 # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
  enable_cm: yes
  db_password: "pTk6ZTQ5NTlhOTA8Qz5DPUQ/QmFWbUlNNGRwLU5jQlBkUEtLY2pfY0wtcUN4Tm5aajBFcFJHQWpJQWViOVk="
  base_dir: /opt/mogdb # 数据库安装的基础目录，一般建议指定到配置文件中 ssh 用户可访问的目录下的路径，让PTK自动创建。
  # # gs_initdb 工具支持参数列表，安装时会追加到指令中
  gs_initdb_opts:
  - "--encoding=UTF-8"
  - "--dbcompatibility=A"
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  cm_option:
    dir: /opt/mogdb/cm # CM 安装目录
    cm_server_port: 15300 # cm server 监听端口
    db_service_vip: "192.168.56.69" # 数据库提供服务的虚拟IP
  # SSH 登录信息，登录用户需为 **root** 或者具有 sudo 权限的用户
  # SSHOption SSH 登录认证信息，支持密码和密钥两种方式，二者需至少设置一种
  ssh_option:
    port: 22 # SSH 服务端口
    user: root # SSH 登录用户
    password: "pTk6NjI1MjNlZmY8Qz5DPUQ/Q1EtSkcxQVpFOWFxUnlzX3lnZFNRa0paS0lJQnByY2VDY1pNTG4tdGVFRXc=" # SSH 登录用户密码
    conn_timeout: 5m0s # SSH 登录连接超时时间，支持单位 m(分),s(秒)
    exec_timeout: 5m0s # SSH 执行单条命令时的超时时间，支持单位 m(分),s(秒)
# 集群内的数据库实例服务器配置信息
db_servers:
  - host: 192.168.56.66 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.66
    role: primary # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高
    
  - host: 192.168.56.67 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.67
    role: standby # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高

# 集群内的CM服务部署机器列表
cm_servers:
  - host: 192.168.56.66 # CM Server 服务器 IP (仅支持 IPv4)
    port: 0 # CM Server 监听端口
    role: ""
  - host: 192.168.56.67 # CM Server 服务器 IP (仅支持 IPv4)
    port: 25300 # CM Server 监听端口
    role: ""

2.2 一主一备一级联

[root@mogdb-02 software]# cat config_cm.yaml

# 集群级别的配置信息，其中 `db_port` 和 `ssh_option` 字段是 **可重用字段**
global:
  cluster_name: dbCluster # 集群名称，PTK管理的集群的唯一标识
  user: omm # 运行数据库的操作系统用户名
  group: dbgrp # 运行数据库的操作系统用户组
  db_port: 5432 # 数据库端口
  ha_port: 5433 # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
  enable_cm: yes
  db_password: "pTk6ZTQ5NTlhOTA8Qz5DPUQ/QmFWbUlNNGRwLU5jQlBkUEtLY2pfY0wtcUN4Tm5aajBFcFJHQWpJQWViOVk="
  base_dir: /opt/mogdb # 数据库安装的基础目录，一般建议指定到配置文件中 ssh 用户可访问的目录下的路径，让PTK自动创建。
  # # gs_initdb 工具支持参数列表，安装时会追加到指令中
  gs_initdb_opts:
  - "--encoding=UTF-8"
  - "--dbcompatibility=A"
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  cm_option:
    dir: /opt/mogdb/cm # CM 安装目录
    cm_server_port: 15300 # cm server 监听端口
    db_service_vip: "192.168.56.69" # 数据库提供服务的虚拟IP
  # SSH 登录信息，登录用户需为 **root** 或者具有 sudo 权限的用户
  # SSHOption SSH 登录认证信息，支持密码和密钥两种方式，二者需至少设置一种
  ssh_option:
    port: 22 # SSH 服务端口
    user: root # SSH 登录用户
    password: "pTk6NjI1MjNlZmY8Qz5DPUQ/Q1EtSkcxQVpFOWFxUnlzX3lnZFNRa0paS0lJQnByY2VDY1pNTG4tdGVFRXc=" # SSH 登录用户密码
    conn_timeout: 5m0s # SSH 登录连接超时时间，支持单位 m(分),s(秒)
    exec_timeout: 5m0s # SSH 执行单条命令时的超时时间，支持单位 m(分),s(秒)
# 集群内的数据库实例服务器配置信息
db_servers:
  - host: 192.168.56.66 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.66
    role: primary # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高
    
  - host: 192.168.56.67 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.67
    role: standby # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高
    
  - host: 192.168.56.68 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.68
    role: cascade_standby # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    upstream_host: 192.168.56.67 # 当实例角色为 cascade_standby 时，该字段表示要跟随的上游备库 IP
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高
# 集群内的CM服务部署机器列表
cm_servers:
  - host: 192.168.56.66 # CM Server 服务器 IP (仅支持 IPv4)
    port: 0 # CM Server 监听端口
    role: ""
  - host: 192.168.56.67 # CM Server 服务器 IP (仅支持 IPv4)
    port: 25300 # CM Server 监听端口
    role: ""

3、checkos

[root@mogdb-02 software]# ptk checkos -f config_cm.yaml --gen-warning-fix

4、通过前面生成的脚本 Fix 问题

需要保证yum正常，执行脚本后要重启

[root@mogdb-02 software]# ll
-rwxr--r-- 1 root  root   438 Dec  2 10:59 root_fix_os.2022.1202.105943.sh

[root@mogdb-02 software]# sh root_fix_os.2022.1202.105943.sh

执行完之后要重启

5、Install

Fix 之后要重启，使参数生效后再安装

cd /opt/omm/
ptk install -f config_cm.yaml
录入初始用户密码，本例中为 Enmotech@123

[root@mogdb-02 software]# ptk install -f config_cm.yaml --pkg MogDB-5.0.1-CentOS-x86_64-all.tar.gz

6、配制vip

增加vip启停权限

把命令加入sudoers，各节点都要加

echo "Cmnd_Alias COMMAND_FLAG = /usr/sbin/ifconfig * netmask * up, /usr/sbin/ifconfig * down" >> /etc/sudoers
echo "omm ALL=(root) NOPASSWD:COMMAND_FLAG" >> /etc/sudoers

omm下测试权限

[omm@ha-01 ~]$ sudo /usr/sbin/ifconfig enp0s8:5432 192.168.56.69 netmask 255.255.255.0 up
[omm@ha-01 ~]$ ifconfig 
enp0s8: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.56.66  netmask 255.255.255.0  broadcast 192.168.56.255
        inet6 fe80::a00:27ff:fe77:bb6b  prefixlen 64  scopeid 0x20<link>
        ether 08:00:27:77:bb:6b  txqueuelen 1000  (Ethernet)
        RX packets 242  bytes 20806 (20.3 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 163  bytes 25373 (24.7 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

enp0s8:5432: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.56.69  netmask 255.255.255.0  broadcast 192.168.56.255
        ether 08:00:27:77:bb:6b  txqueuelen 1000  (Ethernet)

设置vip资源文件

增加vip资源

cm_ctl res --add --res_name="VIP_az" --res_attr="resources_type=VIP, float_ip=192.168.56.69"

增加vip资源与各节点映射

cm_ctl res --edit --res_name="VIP_az" --add_inst="node_id=1, res_instance_id=6001" --inst_attr="base_ip=192.168.56.66"
cm_ctl res --edit --res_name="VIP_az" --add_inst="node_id=2, res_instance_id=6002" --inst_attr="base_ip=192.168.56.67"

复制资源配制到各节点

建议使用cm_ctl res --check命令进行检查。自定义资源文件要求每个节点都要有且一致，检查完成没有错误后，需要手动将该文件分发到其他节点，分发完成后需要重启集群才能生效。
将在cm_agent目录下生成cm_resource.json文件

[omm@ha-01 ~]$ cat /opt/mogdb/cm/cm_agent/cm_resource.json 
{
	"resources":	[{
			"name":	"VIP_az",
			"resources_type":	"VIP",
			"instances":	[{
					"node_id":	1,
					"res_instance_id":	6001,
					"inst_attr":	"base_ip=192.168.56.66"
				}, {
					"node_id":	2,
					"res_instance_id":	6002,
					"inst_attr":	"base_ip=192.168.56.67"
				}],
			"float_ip":	"192.168.56.69"
		}]

scp /opt/mogdb/cm/cm_agent/cm_resource.json omm@192.168.56.67:/opt/mogdb/cm/cm_agent

8、根据需要设置IP白名单

gs_guc set -I all -N all -h “host all all 0.0.0.0/0 md5”

九、集群管理

1、查看集群VIP

[omm@ha-01 ~]$ cm_ctl show

[  Network Connect State  ]

Network timeout:       6s
Current CMServer time: 2023-10-07 11:16:10
Network stat('Y' means connected, otherwise 'N'):
|  \  |  Y  |  Y  |
|  Y  |  \  |  Y  |
|  Y  |  Y  |  \  |


[  Node Disk HB State  ]

Node disk hb timeout:    200s
Current CMServer time: 2023-10-07 11:16:11
Node disk hb stat('Y' means connected, otherwise 'N'):
|  N  |  N  |  N  |

[  FloatIp Network State  ]

node     instance base_ip       float_ip_name float_ip      
------------------------------------------------------------
1  ha-01 6001     192.168.56.66 VIP_az        192.168.56.69

2、启动集群

[omm@ha-01 ~]$ cm_ctl start
cm_ctl: checking cluster status.
cm_ctl: checking cluster status.
cm_ctl: checking finished in 638 ms.
cm_ctl: start cluster. 
cm_ctl: start nodeid: 1
cm_ctl: start nodeid: 2
cm_ctl: start nodeid: 3
.
cm_ctl: start cluster successfully.

3、切换到备机

[omm@ha-01 ~]$ cm_ctl switchover -A
cm_ctl: cmserver is switching over all the master and standby pairs.
...
cm_ctl: switchover -A successfully.

查看状态

[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Normal
redistributing  : No
balanced        : No
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Standby Normal
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Primary Normal
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Normal

4、切换到原始状态

[omm@ha-01 ~]$ cm_ctl switchover -a
cm_ctl: cmserver is rebalancing the cluster automatically.
...
cm_ctl: switchover successfully.

查看状态

[omm@ha-01 ~]$ gs_om -t status --detail
[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Normal
redistributing  : No
balanced        : Yes
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Primary Normal
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Standby Normal
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Normal

5、 build备库

删除备库文件

[omm@ha-02 data]$ rm $PGDATA/global -rf
[omm@ha-02 data]$ ll global
ls: cannot access global: No such file or directory

查看备库状态

[omm@ha-01 ~]$ gs_om -t status --detail
[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Unavailable
redistributing  : No
balanced        : Yes
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Primary Normal
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Down    Unknown
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Need repair(Connecting)
[omm@ha-01 ~]$

build

[omm@ha-01 ~]$ cm_ctl build -n 2 -D $PGDATA -f
.......................
cm_ctl: build successfully.

查看节点状态

[omm@ha-01 ~]$ gs_om -t status --detail
[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Normal
redistributing  : No
balanced        : Yes
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Primary Normal
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Standby Normal
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Normal

查看文件

[omm@ha-02 data]$ ll $PGDATA/global | wc
    102     911    5443

6、 build主库

[omm@ha-01 data]$ ll global | wc
    102     911    5443
[omm@ha-01 data]$ rm $PGDATA/global -rf

查看集群状态（要等待自动切换）

[omm@ha-01 ~]$ gs_om -t status --detail
[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Unavailable
redistributing  : No
balanced        : No
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Down    Unknown
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Primary Normal
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Need repair(Connecting)

重建节点，与上节命令一样

[omm@ha-01 ~]$ cm_ctl build -n 1 -D $PGDATA -f -b full
.......................
cm_ctl: build successfully.

查看状态

[omm@ha-01 ~]$ gs_om -t status --detail
[  CMServer State   ]

node     node_ip         instance                     state
-------------------------------------------------------------
1  ha-01 192.168.56.66   1    /opt/mogdb/cm/cm_server Primary
2  ha-02 192.168.56.67   2    /opt/mogdb/cm/cm_server Standby

[   Cluster State   ]

cluster_state   : Normal
redistributing  : No
balanced        : No
current_az      : AZ_ALL

[  Datanode State   ]

node     node_ip         instance             state            
---------------------------------------------------------------
1  ha-01 192.168.56.66   6001 /opt/mogdb/db1 P Standby Normal
2  ha-02 192.168.56.67   6002 /opt/mogdb/db1 S Primary Normal
3  ha-03 192.168.56.68   6003 /opt/mogdb/db1 C Cascade Standby Normal

7、查看集群状态

cm_ctl query -Cvi

[omm@node109 dn1]$ cm_ctl query -Cvi
[  CMServer State   ]

node       node_ip         instance state
-------------------------------------------
1  node109 172.16.108.109  1        Standby
2  node110 172.16.108.110  2        Primary


[ Defined Resource State ]

node       node_ip         res_name instance  state
-----------------------------------------------------
1  node109 172.16.108.109  dms_res  6001      OnLine
2  node110 172.16.108.110  dms_res  6002      OnLine
1  node109 172.16.108.109  dss      20001     OnLine
2  node110 172.16.108.110  dss      20002     OnLine

[   Cluster State   ]

cluster_state   : Normal
redistributing  : No
balanced        : Yes
current_az      : AZ_ALL

[  Datanode State   ]

node       node_ip         instance state            | node       node_ip         instance state
--------------------------------------------------------------------------------------------------------------
1  node109 172.16.108.109  6001     P Primary Normal | 2  node110 172.16.108.110  6002     S Standby Normal

8、关闭集群

[omm@ha-01 ~]$ cm_ctl stop
cm_ctl: stop cluster. 
cm_ctl: stop nodeid: 1
cm_ctl: stop nodeid: 2
cm_ctl: stop nodeid: 3
..........
cm_ctl: stop cluster successfully.

十、集群升级

注意升级前尽量备份

tar -cvf mogdb_bak_202402221723.tar  db1 log mogdb tmp tool

升级

/opt/omm 为插件包所在目录，需要升级的插件都放在该目录下，PTK会自动查找

ptk cluster -n dbCluster upgrade -p MogDB-5.0.5-CentOS-x86_64-all.tar.gz --plugin-dir /opt/omm

出错后回滚

ptk cluster -n dbCluster upgrade-rollback

十一、集群缩容

参考： https://docs.mogdb.io/zh/ptk/v1.0/usage-scale

[root@ha-01 ~]# ptk cluster -n dbCluster scale-in -H 192.168.56.68
INFO[2023-10-31T14:50:27.556] PTK Version: 1.0.10 release                  
The cluster will contains 2 cm nodes, so please confirm following cms configs:
- (Optional) db_service_vip=""
- (Required) third_party_gateway_ip=""
- (Optional) cms_enable_failover_on2nodes="True"
- (Optional) cms_enable_db_crash_recovery="False"

✔ The 'third_party_gateway_ip' is not configed, please enter: 192.168.56.67
Now, these configs are:
- db_service_vip=""
- third_party_gateway_ip="192.168.56.67"
- cms_enable_failover_on2nodes="True"
- cms_enable_db_crash_recovery="False"

✔ Do you want to modify them (default=n) [y/n]: n
INFO[2023-10-31T14:50:42.059] start scale-in ...                           
INFO[2023-10-31T14:50:43.564] start precheck ...                           
INFO[2023-10-31T14:50:43.856] precheck success                             
Cluster Name: "dbCluster"
+--------------+---------------+-------------+----------+--------------------+--------------------+---------------+
| az(priority) |      ip       | user(group) |   port   |        role        |      data dir      |   upstream    |
+--------------+---------------+-------------+----------+--------------------+--------------------+---------------+
| AZ1(1)       | 192.168.56.66 | omm(dbgrp)  | db:5432  | db:standby         | db:/opt/mogdb/db1 | -             |
|              |               |             | cm:15300 |                    | cm:/opt/mogdb/cm   |               |
|              | 192.168.56.67 | omm(dbgrp)  | db:5432  | db:primary         | db:/opt/mogdb/db1 | -             |
|              |               |             | cm:25300 |                    | cm:/opt/mogdb/cm   |               |
|              | 1̶9̶2̶.̶1̶6̶8̶.̶5̶6̶.̶6̶8̶ | o̶m̶m̶(̶d̶b̶g̶r̶p̶)̶  | d̶b̶:̶5̶4̶3̶2̶  | d̶b̶:̶c̶a̶s̶c̶a̶d̶e̶_̶s̶t̶a̶n̶d̶b̶y̶ | d̶b̶:̶/̶o̶p̶t̶/̶m̶o̶g̶d̶b̶/̶d̶a̶t̶a̶ | 1̶9̶2̶.̶1̶6̶8̶.̶5̶6̶.̶6̶7̶ |
+--------------+---------------+-------------+----------+--------------------+--------------------+---------------+
✔ Please confirm cluster after scale-in (default=n) [y/n]: y
INFO[2023-10-31T14:50:48.399] start drop nodes ...                         
INFO[2023-10-31T14:50:48.399] [192.168.56.68][omm] remove cron task: om_monitor 
INFO[2023-10-31T14:50:48.561] [192.168.56.68][omm] stop user "omm" db process 
INFO[2023-10-31T14:50:50.844] [192.168.56.68][omm] remove files: /opt/mogdb/app,/opt/mogdb/tool,/opt/mogdb/db1,/opt/mogdb/log,/opt/mogdb/tmp 
INFO[2023-10-31T14:50:51.182] [192.168.56.68][omm] remove user "omm" from cron.allow 
INFO[2023-10-31T14:50:51.254] [192.168.56.68][omm] remove user "omm" ulimits 
INFO[2023-10-31T14:50:51.331] [192.168.56.68][omm] remove base dir if empty: /opt/mogdb 
INFO[2023-10-31T14:50:51.398] [192.168.56.68][omm] kill user "omm" process 
INFO[2023-10-31T14:50:51.495] [192.168.56.68][omm] delete user "omm"       
INFO[2023-10-31T14:50:51.654] drop nodes success                           
INFO[2023-10-31T14:50:51.654] start update remaining cluster config ...    
INFO[2023-10-31T14:50:51.654] db instance count reduce from 3 to 2         
INFO[2023-10-31T14:50:51.654] [192.168.56.66][omm] update config ...       
INFO[2023-10-31T14:50:51.654] [192.168.56.66][omm] config postgresql.conf  
INFO[2023-10-31T14:50:51.654] [192.168.56.67][omm] update config ...       
INFO[2023-10-31T14:50:51.654] [192.168.56.67][omm] config postgresql.conf  
INFO[2023-10-31T14:50:51.654] [192.168.56.66][omm] update cm config file: /opt/mogdb/cm/cm_server/cm_server.conf 
INFO[2023-10-31T14:50:51.655] [192.168.56.67][omm] update cm config file: /opt/mogdb/cm/cm_server/cm_server.conf 
INFO[2023-10-31T14:50:51.859] [192.168.56.66][omm] clear hba               
INFO[2023-10-31T14:50:51.864] [192.168.56.67][omm] clear hba               
INFO[2023-10-31T14:50:51.983] [192.168.56.66][omm] generate cluster_static_config file 
INFO[2023-10-31T14:50:51.995] [192.168.56.67][omm] generate cluster_static_config file 
INFO[2023-10-31T14:50:52.137] [192.168.56.66][omm] change /opt/mogdb/app/bin/cluster_static_config owner to omm 
INFO[2023-10-31T14:50:52.137] [192.168.56.67][omm] change /opt/mogdb/app/bin/cluster_static_config owner to omm 
INFO[2023-10-31T14:50:52.285] [192.168.56.66][omm] reload db ...           
INFO[2023-10-31T14:50:52.290] [192.168.56.67][omm] reload db ...           
INFO[2023-10-31T14:50:52.393] update cluster config success                
INFO[2023-10-31T14:50:52.393] restart cm cluster ...                       
INFO[2023-10-31T14:50:53.558] restart cm cluster success                   
INFO[2023-10-31T14:50:53.558] reload cm cluster ...                        
WARN[2023-10-31T14:50:53.653] cm_ctl reload failed: Process exited with status 255: cm_ctl: send reload msg to cm_server, connect fail. 
INFO[2023-10-31T14:50:53.655] scale-in success                             
INFO[2023-10-31T14:50:53.655] time elapsed: 12s                            
[root@ha-01 ~]# ptk cluster -n dbCluster status
[   Cluster State   ]
cluster_name				: dbCluster
cluster_state   			: Normal
database_version			: openGauss 5.1.0 (build b5a8d5b0)

[  CMServer State   ]
  id |      ip       | port  | hostname |  role    
-----+---------------+-------+----------+----------
  1  | 192.168.56.66 | 15300 | ha-01    | primary  
  2  | 192.168.56.67 | 25300 | ha-02    | standby  

[  Datanode State   ]
  cluster_name |  id  |      ip       | port | user | nodename | db_role | state  | upstream  
---------------+------+---------------+------+------+----------+---------+--------+-----------
  dbCluster    | 6001 | 192.168.56.66 | 5432 | omm  | dn_6001  | standby | Normal | -         
               | 6002 | 192.168.56.67 | 5432 | omm  | dn_6002  | primary | Normal | -

十二、集群扩容

扩容要求，现有环境需要是CM，否则会有提示

not support scale-out cm server on cluster which cm not installed

1、扩容前状态

[root@mogdb-01 omm]# ptk cluster -n dbCluster status
[   Cluster State   ]
cluster_name                   : dbCluster
cluster_state                  : Normal
database_version               : MogDB 5.0.1.6 (build d78fb13b)

[  CMServer State   ]
  id |      ip       | port  | hostname |  role    
-----+---------------+-------+----------+----------
  1  | 192.168.56.66 | 15300 | mogdb-01 | primary  
  2  | 192.168.56.67 | 25300 | mogdb-02 | standby  

[  Datanode State   ]
  cluster_name |  id  |      ip       | port | user | nodename | db_role | state  |  uptime  | upstream  
---------------+------+---------------+------+------+----------+---------+--------+----------+-----------
  dbCluster    | 6001 | 192.168.56.66 | 5432 | omm  | dn_6001  | primary | Normal | 00:06:07 | -         
               |      | 192.168.10.66 |      |      |          |         |        |          |           
               | 6002 | 192.168.56.67 | 5432 | omm  | dn_6002  | standby | Normal | 00:05:27 | -         
               |      | 192.168.10.67 |      |      |          |         |        |          |           
[root@mogdb-01 omm]#

2、编辑配制文件

该文件参考一主一备一级联，去掉其中主与备，只保留本次要扩容的级联信息

# 集群级别的配置信息，其中 `db_port` 和 `ssh_option` 字段是 **可重用字段**
global:
  cluster_name: dbCluster # 集群名称，PTK管理的集群的唯一标识
  user: omm # 运行数据库的操作系统用户名
  group: dbgrp # 运行数据库的操作系统用户组
  db_port: 5432 # 数据库端口
  ha_port: 5433 # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
  enable_cm: yes
  db_password: "pTk6ZTQ5NTlhOTA8Qz5DPUQ/QmFWbUlNNGRwLU5jQlBkUEtLY2pfY0wtcUN4Tm5aajBFcFJHQWpJQWViOVk="
  base_dir: /opt/mogdb # 数据库安装的基础目录，一般建议指定到配置文件中 ssh 用户可访问的目录下的路径，让PTK自动创建。
  # # gs_initdb 工具支持参数列表，安装时会追加到指令中
  gs_initdb_opts:
  - "--encoding=UTF-8"
  - "--dbcompatibility=A"
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  # MogDB 高可用组件 CM 配置信息，若不使用 CM 组件，则无需填写该字段
  cm_option:
    dir: /opt/mogdb/cm # CM 安装目录
    cm_server_port: 15300 # cm server 监听端口
    db_service_vip: "192.168.56.69" # 数据库提供服务的虚拟IP
  # SSH 登录信息，登录用户需为 **root** 或者具有 sudo 权限的用户
  # SSHOption SSH 登录认证信息，支持密码和密钥两种方式，二者需至少设置一种
  ssh_option:
    port: 22 # SSH 服务端口
    user: root # SSH 登录用户
    password: "pTk6NjI1MjNlZmY8Qz5DPUQ/Q1EtSkcxQVpFOWFxUnlzX3lnZFNRa0paS0lJQnByY2VDY1pNTG4tdGVFRXc=" # SSH 登录用户密码
    conn_timeout: 5m0s # SSH 登录连接超时时间，支持单位 m(分),s(秒)
    exec_timeout: 5m0s # SSH 执行单条命令时的超时时间，支持单位 m(分),s(秒)
# 集群内的数据库实例服务器配置信息
db_servers:

  - host: 192.168.56.68 # 数据库实例服务器 IP (仅支持 IPv4)
    # 数据库主备间日志复制IP，默认和数据库监听地址相同
    ha_ips:
      - 192.168.10.68
    role: cascade_standby # 数据库实例角色，若不填，PTK 会随机选择一个实例作为主库，其余自动为备库
    upstream_host: 192.168.56.67 # 当实例角色为 cascade_standby 时，该字段表示要跟随的上游备库 IP
    az_name: AZ1 # Available Zone (可用区) 名称
    az_priority: 1 # 可用区使用的优先级，数值越小优先级越高

3、扩容

ptk cluster -n dbCluster scale-out -c config_cm_3.yaml

扩容后状态查询

[root@mogdb-01 omm]# ptk cluster -n dbCluster status
[   Cluster State   ]
cluster_name                   : dbCluster
cluster_state                  : Normal
database_version               : MogDB 5.0.1.6 (build d78fb13b)

[  CMServer State   ]
  id |      ip       | port  | hostname |  role    
-----+---------------+-------+----------+----------
  1  | 192.168.56.66 | 15300 | mogdb-01 | primary  
  2  | 192.168.56.67 | 25300 | mogdb-02 | standby  

[  Datanode State   ]
  cluster_name |  id  |      ip       | port | user | nodename |     db_role     | state  |  uptime  |   upstream     
---------------+------+---------------+------+------+----------+-----------------+--------+----------+----------------
  dbCluster    | 6001 | 192.168.56.66 | 5432 | omm  | dn_6001  | primary         | Normal | 00:21:43 | -              
               |      | 192.168.10.66 |      |      |          |                 |        |          |                
               | 6002 | 192.168.56.67 | 5432 | omm  | dn_6002  | standby         | Normal | 00:21:03 | -              
               |      | 192.168.10.67 |      |      |          |                 |        |          |                
               | 6003 | 192.168.56.68 | 5432 | omm  | dn_6003  | cascade_standby | Normal | 00:01:11 | 192.168.56.67  
               |      | 192.168.10.68 |      |      |          |                 |        |          |                
[root@mogdb-01 omm]#

附、问题处理

附1、 version `OPENSSL_1_1_1f’ not found

发现环境

[sqd@node152 ~]$ cat /etc/os-release 
NAME="Kylin Linux Advanced Server"
VERSION="V10 (Sword)"
ID="kylin"
VERSION_ID="V10"
PRETTY_NAME="Kylin Linux Advanced Server V10 (Sword)"
ANSI_COLOR="0;31"

[sqd@node152 ~]$ uname -a
Linux node152 4.19.90-24.4.v2101.ky10.aarch64 #1 SMP Mon May 24 14:45:37 CST 2021 aarch64 aarch64 aarch64 GNU/Linux

处理方法

unset LD_LIBRARY_PATH && ssh-keygen

附2、 initialize the memory protect for g_instance.attr.attr_storage.cstore_buffers (1024 Mbytes) or shared memory (4250 Mbytes) is larger.

虚拟机内存太少引起，更改如下参数

gs_guc set -N all -I all -c "max_process_memory = 4GB"
gs_guc set -N all -I all -c "shared_buffers = 512MB"
gs_guc set -N all -I all -c "cstore_buffers = 16MB"
gs_om -t stop
gs_om -t start

附3、配制 oracle_fdw 环境变量


解压Oracle客户端,配制环境变量

export ORACLE_HOME=/opt/instantclient_19_19
export PATH= $ORACLE_HOME:$ PATH
export TNS_ADMIN= $ORACLE_HOME/network/admin export LD_LIBRARY_PATH=$ ORACLE_HOME:$LD_LIBRARY_PATH
export NLS_LANG=‘AMERICAN_AMERICA.AL32UTF8’
export PATH



## 附4、 国产系统中文环境

yum -y install glibc-locale-source glibc-all-langpacks

locale -a | grep zh_CN


## 附5、 root 密码变动

运行命令时提供新的密码即可

[root@ha-01 ~]# ptk cluster -n dbCluster status
ERRO[2023-11-06T11:03:43.625] [192.168.56.67][omm] failed to connect to root@192.168.56.67: SSHConnectError: ssh: handshake failed: ssh: unable to authenticate, attempted methods [none password], no supported methods remain
✔ password
✔ please enter the password for connecting root@192.168.56.67: ********
[ Cluster State ]
cluster_name : dbCluster
cluster_state : Normal
database_version : openGauss 5.1.0 (build b5a8d5b0)

[ CMServer State ]
id | ip | port | hostname | role
-----±--------------±------±---------±---------
1 | 192.168.56.66 | 15300 | ha-01 | primary
2 | 192.168.56.67 | 25300 | ha-02 | standby

[ Datanode State ]
cluster_name | id | ip | port | user | nodename | db_role | state | upstream
---------------±-----±--------------±-----±-----±---------±--------±-------±----------
dbCluster | 6001 | 192.168.56.66 | 5432 | omm | dn_6001 | primary | Normal | -
| 6002 | 192.168.56.67 | 5432 | omm | dn_6002 | standby | Normal | -
[root@ha-01 ~]#


## 附6、 更改hostname

同步更改 /etc/hosts文件即可

jgmydsai

关注

28
点赞
踩
12

收藏

觉得还不错? 一键收藏
0
评论
Centos7.6安装MogDB

自己记录的MogDB安装过程，供参考
复制链接

扫一扫

专栏目录

Centos7.6安装MogDB

一、环境配制

1、确认OS版本

2、为配制本地源挂载光驱

3、配制本地源

4、安装python3

1).安装依赖环境

2).下载Python3

3).安装python3

4).进入解压后的目录，编译安装。

5).建立python3的软链

6).复制so

7).并将/usr/local/python3/bin加入PATH(OMM下也要配制)

8).验正版本

5、启动命令界面

6、禁止 SELINUX、防火墙

7、安装依赖包

8、设置编码

9、关闭swap交换内存

10、MTU使用默认值

11、设置root用户远程登录

12、关闭透明大页

13、更改文件系统参数

14、准备安装目录

15、手动配制ssh互信

二、PTK安装

1. 安装

2. 升级

3. 查看下载链接

三、单机安装

1、配制安装参数

2、checkos

3、通过前面生成的脚本 Fix 问题

4、Install

5、gen-om-xml

6、根据需要设置IP白名单

四、安装扩展

1、安装所有扩展

2、安装指定扩展

五、compat-tools

注意

1、下载

2、安装

3、卸载

六、卸载MogDB

1、查看当前已经安装的MogDB

2、停止DB

3、uninstall

4、手动清理残留信息

八、集群安装

1、添加vip

2、创建配置文件

2.1 一主一备

2.2 一主一备一级联

3、checkos

4、通过前面生成的脚本 Fix 问题

5、Install

6、配制vip

增加vip启停权限

设置vip资源文件

增加vip资源

增加vip资源与各节点映射

复制资源配制到各节点

8、根据需要设置IP白名单

九、集群管理

1、查看集群VIP

2、启动集群

3、切换到备机

4、切换到原始状态

5、 build备库

6、 build主库

7、 查看集群状态

8、 关闭集群

十、集群升级

升级

出错后回滚

十一、 集群缩容

十二、 集群扩容

1、 扩容前状态

2、编辑配制文件

7、查看集群状态

8、关闭集群

十一、集群缩容

十二、集群扩容

1、扩容前状态

附3、配制 oracle_fdw 环境变量