PostgreSQL高可用部署配置

原創 咖啡那么浓 2020-06-20 13:47
機器 操作系統 IP 安裝組件 版本
psql1 CentOS 7.3 10.10.0.1 postgresql/etcd/patroni 12.1/3.3.11/1.6.1
psql2 CentOS 7.3 10.10.0.2 postgresql/etcd/patroni 12.1/3.3.11/1.6.1
psql3 CentOS 7.3 10.10.0.3 postgresql/etcd/patroni 12.1/3.3.11/1.6.1
haproxy1 CentOS 7.3 10.10.0.4 haproxy/keepalived 1.5.18/2.0.20
haproxy2 CentOS 7.3 10.10.0.5 haproxy/keepalived 1.5.18/2.0.20

本文用到的所有文件

鏈接：https://pan.baidu.com/s/1lwaX_DuTcJwegLuJFZTH5w
提取碼：5615

1、基礎環境（所有節點）

1.1、修改主機名及hosts

hostnamectl set-hostname psql1
hostnamectl set-hostname psql2
hostnamectl set-hostname psql3
hostnamectl set-hostname haproxy1
hostnamectl set-hostname haproxy2
...

cat >> /etc/hosts <<EOF
10.10.0.1 psql1
10.10.0.2 psql2
10.10.0.3 psql3
10.10.0.4 haproxy1
10.10.0.5 haproxy2
EOF

1.2、修改系統進程打開最大文件數

echo "*   soft    nofile  655350" >> /etc/security/limits.conf
echo "*   hard    nofile  655350" >> /etc/security/limits.conf

1.3、關閉防火牆及selinux

systemctl stop firewalld && systemctl disable firewalld

setenforce 0

sed -i 's/^SELINUX=.*/SELINUX=disabled/' /etc/selinux/config

1.4、配置yum源

此處配置離線yum源安裝，如果是外網環境直接yum安裝即可

unzip pgsql_yum.zip && rm -rf pgsql_yum.zip && mv pgsql_yum /

mkdir -p /etc/yum.repos.d/yum.bak && mv /etc/yum.repos.d/*.repo /etc/yum.repos.d/yum.bak

cat > /etc/yum.repos.d/local.repo <<EOF
[local]
name=local
enabled=1
baseurl=file:///pgsql_yum/
gpgcheck=0
EOF

yum clean all && yum makecache

2、PostgreSQL（psql節點）

2.1、安裝依賴

yum -y install readline readline-devel zlib zlib-devel vim

2.2、安裝PostgreSQL

選擇空間較大的磁盤

tar -xvf postgresql-12.1.tar.gz && rm -rf postgresql-12.1.tar.gz && cd postgresql-12.1

./configure --prefix=/usr/local/pgsql

make && make install

3、Etcd（etcd節點）

3.1、安裝chrony同步集羣系統時間

建議集羣所有節點均安裝配置

  • 安裝chrony
yum -y install chrony
  • 啓動並加入開機自啓
systemctl start chronyd && systemctl enable chronyd && systemctl status chronyd
  • 系統時鐘同步，以psql1機器爲時鐘服務器
vim /etc/chrony.conf
  • 將以下內容註釋掉：

在這裏插入圖片描述

  • 添加時鐘服務器配置，即psql1機器
server 10.10.0.1 iburst
  • 強制同步時間
chronyc -a makestep
  • 重啓service
systemctl daemon-reload && systemctl restart chronyd

3.2、安裝Etcd

yum -y install etcd

3.3、配置Etcd

vim /etc/etcd/etcd.conf
  • etcd1(10.10.0.1)
ETCD_DATA_DIR="/var/lib/etcd/etcd1.etcd"
ETCD_LISTEN_PEER_URLS="http://10.10.0.1:2380"
ETCD_LISTEN_CLIENT_URLS="http://10.10.0.1:2379,http://127.0.0.1:2379"
ETCD_NAME="etcd1"
ETCD_INITIAL_ADVERTISE_PEER_URLS="http://10.10.0.1:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://10.10.0.1:2379"
ETCD_INITIAL_CLUSTER="etcd1=http://10.10.0.1:2380,etcd2=http://10.10.0.2:2380,etcd3=http://10.10.0.3:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"
  • etcd2(10.10.0.2)
ETCD_DATA_DIR="/var/lib/etcd/etcd2.etcd"
ETCD_LISTEN_PEER_URLS="http://10.10.0.2:2380"
ETCD_LISTEN_CLIENT_URLS="http://10.10.0.2:2379,http://127.0.0.1:2379"
ETCD_NAME="etcd2"
ETCD_INITIAL_ADVERTISE_PEER_URLS="http://10.10.0.2:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://10.10.0.2:2379"
ETCD_INITIAL_CLUSTER="etcd1=http://10.10.0.1:2380,etcd2=http://10.10.0.2:2380,etcd3=http://10.10.0.3:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"
  • etcd3(10.10.0.3)
ETCD_DATA_DIR="/var/lib/etcd/etcd3.etcd"
ETCD_LISTEN_PEER_URLS="http://10.10.0.3:2380"
ETCD_LISTEN_CLIENT_URLS="http://10.10.0.3:2379,http://127.0.0.1:2379"
ETCD_NAME="etcd3"
ETCD_INITIAL_ADVERTISE_PEER_URLS="http://10.10.0.3:2380"
ETCD_ADVERTISE_CLIENT_URLS="http://10.10.0.3:2379"
ETCD_INITIAL_CLUSTER="etcd1=http://10.10.0.1:2380,etcd2=http://10.10.0.2:2380,etcd3=http://10.10.0.3:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"

3.4、修改etcd.service

vim /usr/lib/systemd/system/etcd.service

直接刪除原有內容，替換爲以下配置

[Unit]
Description=Etcd Server
After=network.target
After=network-online.target
Wants=network-online.target

[Service]
Type=notify
WorkingDirectory=/var/lib/etcd/
EnvironmentFile=/etc/etcd/etcd.conf
User=etcd
# set GOMAXPROCS to number of processors
ExecStart=/bin/bash -c "GOMAXPROCS=$(nproc) /usr/bin/etcd \
--name=\"${ETCD_NAME}\" \
--data-dir=\"${ETCD_DATA_DIR}\" \
--listen-peer-urls=\"${ETCD_LISTEN_PEER_URLS}\" \
--listen-client-urls=\"${ETCD_LISTEN_CLIENT_URLS}\" \
--initial-advertise-peer-urls=\"${ETCD_INITIAL_ADVERTISE_PEER_URLS}\" \
--advertise-client-urls=\"${ETCD_ADVERTISE_CLIENT_URLS}\" \
--initial-cluster=\"${ETCD_INITIAL_CLUSTER}\"  \
--initial-cluster-token=\"${ETCD_INITIAL_CLUSTER_TOKEN}\" \
--initial-cluster-state=\"${ETCD_INITIAL_CLUSTER_STATE}\""
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target

3.5、啓動Etcd

systemctl daemon-reload && systemctl enable etcd && systemctl start etcd

3.6、驗證Etcd

  • 任意節點查看集羣狀態
etcdctl cluster-health

在這裏插入圖片描述

etcdctl member list

在這裏插入圖片描述

4、Patroni（patroni節點）

4.1、安裝setuptools

unzip setuptools-42.0.2.zip && rm -rf setuptools-42.0.2.zip && cd setuptools-42.0.2

python setup.py install

4.2、安裝pip

tar -xvf pip-19.3.1.tar.gz && rm -rf pip-19.3.1.tar.gz && cd pip-19.3.1

python setup.py install
  • 查看pip版本
pip -V

在這裏插入圖片描述

4.3、安裝patroni

  • 如果安裝過程中提示缺少依賴可去pypi下載安裝即可

https://pypi.org/

yum -y install python-devel

pip install *.whl

pip install psutil-5.6.7.tar.gz
pip install cdiff-1.0.tar.gz
pip install python-etcd-0.4.5.tar.gz
pip install PyYAML-5.3.tar.gz
pip install prettytable-0.7.2.tar.gz
pip install patroni-1.6.1.tar.gz

4.4、查看patroni版本

patronictl version

在這裏插入圖片描述

4.5、配置patroni

4.5.1、集羣psql1節點配置

  • 創建配置文件目錄
mkdir -p /data/patroni/conf
  • 拷貝相關文件
unzip patroni.zip

cp -r patroni /data/patroni/conf/
cp /data/patroni/conf/patroni/postgres0.yml /data/patroni/conf/
  • 編輯配置文件
vim /data/patroni/conf/postgres0.yml

scope: batman
#namespace: /service/
name: postgresql0
 
restapi:
  listen: 10.10.0.1:8008
  connect_address: 10.10.0.1:8008
#  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
#  authentication:
#    username: username
#    password: password
 
# ctl:
#   insecure: false # Allow connections to SSL sites without certs
#   certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#   cacert: /etc/ssl/certs/ssl-cacert-snakeoil.pem
 
etcd:
  host: 10.10.0.1:2379
 
bootstrap:
  # this section will be written into Etcd:/<namespace>/<scope>/config after initializing new cluster
  # and all other cluster members will use it as a `global configuration`
  dcs:
    ttl: 30
    loop_wait: 10
    retry_timeout: 10
    maximum_lag_on_failover: 1048576
#    master_start_timeout: 300
    synchronous_mode: false
    #standby_cluster:
      #host: 127.0.0.1
      #port: 1111
      #primary_slot_name: patroni
    postgresql:
      use_pg_rewind: true
      use_slots: true
      parameters:
         wal_level: logical
         hot_standby: "on"
         max_connections: 5000
         wal_keep_segments: 1000
         max_wal_senders: 10
         max_replication_slots: 10
         wal_log_hints: "on"
         archive_mode: "on"
         archive_timeout: 1800s
         archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
      recovery_conf:
         restore_command: cp ../wal_archive/%f %p
 
  # some desired options for 'initdb'
  initdb:  # Note: It needs to be a list (some options need values, others are switches)
  - encoding: UTF8
  - data-checksums
 
  pg_hba:  # Add following lines to pg_hba.conf after running 'initdb'
  # For kerberos gss based connectivity (discard @.*$)
  #- host replication replicator 127.0.0.1/32 gss include_realm=0
  #- host all all 0.0.0.0/0 gss include_realm=0
  - host replication replicator 0.0.0.0/0 md5
  - host all admin 0.0.0.0/0 md5
  - host all all 0.0.0.0/0 md5
 
  # Additional script to be launched after initial cluster creation (will be passed the connection URL as parameter)
# post_init: /usr/local/bin/setup_cluster.sh
 
  # Some additional users users which needs to be created after initializing new cluster
  users:
    admin:
      password: postgres
      options:
        - createrole
        - createdb
    replicator:
      password: replicator
      options:
        - replication
postgresql:
  listen: 0.0.0.0:5432
  connect_address: 10.10.0.1:5432
  data_dir: /data/postgres
  bin_dir: /usr/local/pgsql/bin
#  config_dir:
#  pgpass: /tmp/pgpass0
  authentication:
    replication:
      username: replicator
      password: replicator
    superuser:
      username: admin
      password: postgres
#    rewind:  # Has no effect on postgres 10 and lower
#      username: rewind_user
#      password: rewind_password
  # Server side kerberos spn
#  krbsrvname: postgres
  parameters:
    # Fully qualified kerberos ticket file for the running user
    # same as KRB5CCNAME used by the GSS
#   krb_server_keyfile: /var/spool/keytabs/postgres
    unix_socket_directories: '.'
 
#watchdog:
#  mode: automatic # Allowed values: off, automatic, required
#  device: /dev/watchdog
#  safety_margin: 5
 
tags:
    nofailover: false
    noloadbalance: false
    clonefrom: false
    nosync: false

4.5.2、集羣psql2節點配置

  • 創建配置文件目錄
mkdir -p /data/patroni/conf
  • 拷貝相關文件
unzip patroni.zip

cp -r patroni /data/patroni/conf/
cp /data/patroni/conf/patroni/postgres1.yml /data/patroni/conf/
  • 編輯配置文件
vim /data/patroni/conf/postgres1.yml

scope: batman
#namespace: /service/
name: postgresql1
 
restapi:
  listen: 10.10.0.2:8008
  connect_address: 10.10.0.2:8008
#  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
#  authentication:
#    username: username
#    password: password
 
# ctl:
#   insecure: false # Allow connections to SSL sites without certs
#   certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#   cacert: /etc/ssl/certs/ssl-cacert-snakeoil.pem
 
etcd:
  host: 10.10.0.2:2379
 
bootstrap:
  # this section will be written into Etcd:/<namespace>/<scope>/config after initializing new cluster
  # and all other cluster members will use it as a `global configuration`
  dcs:
    ttl: 30
    loop_wait: 10
    retry_timeout: 10
    maximum_lag_on_failover: 1048576
#    master_start_timeout: 300
    synchronous_mode: false
    #standby_cluster:
      #host: 127.0.0.1
      #port: 1111
      #primary_slot_name: patroni
    postgresql:
      use_pg_rewind: true
      use_slots: true
      parameters:
         wal_level: logical
         max_connections: 5000
         hot_standby: "on"
         wal_keep_segments: 1000
         max_wal_senders: 10
         max_replication_slots: 10
         wal_log_hints: "on"
         archive_mode: "on"
         archive_timeout: 1800s
         archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
      recovery_conf:
         restore_command: cp ../wal_archive/%f %p
 
  # some desired options for 'initdb'
  initdb:  # Note: It needs to be a list (some options need values, others are switches)
  - encoding: UTF8
  - data-checksums
 
  pg_hba:  # Add following lines to pg_hba.conf after running 'initdb'
  # For kerberos gss based connectivity (discard @.*$)
  #- host replication replicator 127.0.0.1/32 gss include_realm=0
  #- host all all 0.0.0.0/0 gss include_realm=0
  - host replication replicator 0.0.0.0/0 md5
  - host all admin 0.0.0.0/0 md5
  - host all all 0.0.0.0/0 md5
 
  # Additional script to be launched after initial cluster creation (will be passed the connection URL as parameter)
# post_init: /usr/local/bin/setup_cluster.sh
 
  # Some additional users users which needs to be created after initializing new cluster
  users:
    admin:
      password: postgres
      options:
        - createrole
        - createdb
    replicator:
      password: replicator
      options:
        - replication
postgresql:
  listen: 0.0.0.0:5432
  connect_address: 10.10.0.2:5432
  data_dir: /data/postgres
  bin_dir: /usr/local/pgsql/bin
#  config_dir:
#  pgpass: /tmp/pgpass0
  authentication:
    replication:
      username: replicator
      password: replicator
    superuser:
      username: admin
      password: postgres
#    rewind:  # Has no effect on postgres 10 and lower
#      username: rewind_user
#      password: rewind_password
  # Server side kerberos spn
#  krbsrvname: postgres
  parameters:
    # Fully qualified kerberos ticket file for the running user
    # same as KRB5CCNAME used by the GSS
#   krb_server_keyfile: /var/spool/keytabs/postgres
    unix_socket_directories: '.'
 
#watchdog:
#  mode: automatic # Allowed values: off, automatic, required
#  device: /dev/watchdog
#  safety_margin: 5
 
tags:
    nofailover: false
    noloadbalance: false
    clonefrom: false
    nosync: false

4.5.3、集羣psql3節點配置

  • 創建配置文件目錄
mkdir -p /data/patroni/conf
  • 拷貝相關文件
unzip patroni.zip

cp -r patroni /data/patroni/conf/
cp /data/patroni/conf/patroni/postgres2.yml /data/patroni/conf/
  • 編輯配置文件
vim /data/patroni/conf/postgres2.yml

scope: batman
#namespace: /service/
name: postgresql2
 
restapi:
  listen: 10.10.0.3:8008
  connect_address: 10.10.0.3:8008
#  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
#  authentication:
#    username: username
#    password: password
 
# ctl:
#   insecure: false # Allow connections to SSL sites without certs
#   certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
#   cacert: /etc/ssl/certs/ssl-cacert-snakeoil.pem
 
etcd:
  host: 10.10.0.3:2379
 
bootstrap:
  # this section will be written into Etcd:/<namespace>/<scope>/config after initializing new cluster
  # and all other cluster members will use it as a `global configuration`
  dcs:
    ttl: 30
    loop_wait: 10
    retry_timeout: 10
    maximum_lag_on_failover: 1048576
#    master_start_timeout: 300
    synchronous_mode: false
    #standby_cluster:
      #host: 127.0.0.1
      #port: 1111
      #primary_slot_name: patroni
    postgresql:
      use_pg_rewind: true
      use_slots: true
      parameters:
         wal_level: logical
         max_connections: 5000
         hot_standby: "on"
         wal_keep_segments: 1000
         max_wal_senders: 10
         max_replication_slots: 10
         wal_log_hints: "on"
         archive_mode: "on"
         archive_timeout: 1800s
         archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
      recovery_conf:
         restore_command: cp ../wal_archive/%f %p
 
  # some desired options for 'initdb'
  initdb:  # Note: It needs to be a list (some options need values, others are switches)
  - encoding: UTF8
  - data-checksums
 
  pg_hba:  # Add following lines to pg_hba.conf after running 'initdb'
  # For kerberos gss based connectivity (discard @.*$)
  #- host replication replicator 127.0.0.1/32 gss include_realm=0
  #- host all all 0.0.0.0/0 gss include_realm=0
  - host replication replicator 0.0.0.0/0 md5
  - host all admin 0.0.0.0/0 md5
  - host all all 0.0.0.0/0 md5
 
  # Additional script to be launched after initial cluster creation (will be passed the connection URL as parameter)
# post_init: /usr/local/bin/setup_cluster.sh
 
  # Some additional users users which needs to be created after initializing new cluster
  users:
    admin:
      password: postgres
      options:
        - createrole
        - createdb
    replicator:
      password: replicator
      options:
        - replication
postgresql:
  listen: 0.0.0.0:5432
  connect_address: 10.10.0.3:5432
  data_dir: /data/postgres
  bin_dir: /usr/local/pgsql/bin
#  config_dir:
#  pgpass: /tmp/pgpass0
  authentication:
    replication:
      username: replicator
      password: replicator
    superuser:
      username: admin
      password: postgres
#    rewind:  # Has no effect on postgres 10 and lower
#      username: rewind_user
#      password: rewind_password
  # Server side kerberos spn
#  krbsrvname: postgres
  parameters:
    # Fully qualified kerberos ticket file for the running user
    # same as KRB5CCNAME used by the GSS
#   krb_server_keyfile: /var/spool/keytabs/postgres
    unix_socket_directories: '.'
 
#watchdog:
#  mode: automatic # Allowed values: off, automatic, required
#  device: /dev/watchdog
#  safety_margin: 5
 
tags:
    nofailover: false
    noloadbalance: false
    clonefrom: false
    nosync: false

4.6、修改目錄權限

  • 記下data_dir上述yml配置文件中的值。該目錄需要確保postgres用戶具備寫入的權限。如果此目錄不存在，則創建，在所有patroni節點分別進行如下操作
groupadd postgres
useradd -g postgres postgres
chown -R postgres /usr/local/pgsql

mkdir -p /data/postgres
chown -Rf postgres:postgres /data/postgres
chmod 700 /data/postgres

4.7、啓動patroni

在psql1節點執行

chown -Rf postgres:postgres /data/patroni/conf
  • 創建service文件，修改其中的可執行目錄及配置文件
cat > /etc/systemd/system/patroni.service <<EOF
[Unit]
Description=Runners to orchestrate a high-availability PostgreSQL
After=network.target

[Service]
Type=simple

User=postgres
Group=postgres

ExecStart=/usr/bin/patroni /data/patroni/conf/postgres0.yml

KillMode=process

TimeoutSec=30

Restart=no

[Install]
WantedBy=multi-user.target
EOF
  • 啓動patroni初始化數據庫
systemctl daemon-reload && systemctl start patroni && systemctl enable patroni
  • 切換到postgres用戶查看patroni是否託管數據庫
su postgres

/usr/local/pgsql/bin/psql -h 127.0.0.1 -U admin postgres

在這裏插入圖片描述
在psql2節點執行

chown -Rf postgres:postgres /data/patroni/conf
  • 創建service文件，修改其中的可執行目錄及配置文件
cat > /etc/systemd/system/patroni.service <<EOF
[Unit]
Description=Runners to orchestrate a high-availability PostgreSQL
After=network.target

[Service]
Type=simple

User=postgres
Group=postgres

ExecStart=/usr/bin/patroni /data/patroni/conf/postgres1.yml

KillMode=process

TimeoutSec=30

Restart=no

[Install]
WantedBy=multi-user.target
EOF
  • 啓動patroni初始化數據庫
systemctl daemon-reload && systemctl start patroni && systemctl enable patroni
  • 切換到postgres用戶查看patroni是否託管數據庫
su postgres

/usr/local/pgsql/bin/psql -h 127.0.0.1 -U admin postgres

在這裏插入圖片描述

在psql3節點執行

chown -Rf postgres:postgres /data/patroni/conf
  • 創建service文件，修改其中的可執行目錄及配置文件
cat > /etc/systemd/system/patroni.service <<EOF
[Unit]
Description=Runners to orchestrate a high-availability PostgreSQL
After=network.target

[Service]
Type=simple

User=postgres
Group=postgres

ExecStart=/usr/bin/patroni /data/patroni/conf/postgres2.yml

KillMode=process

TimeoutSec=30

Restart=no

[Install]
WantedBy=multi-user.target
EOF
  • 啓動patroni初始化數據庫
systemctl daemon-reload && systemctl start patroni && systemctl enable patroni
  • 切換到postgres用戶查看patroni是否託管數據庫
su postgres

/usr/local/pgsql/bin/psql -h 127.0.0.1 -U admin postgres

在這裏插入圖片描述

4.8、查看集羣

  • 任意節點查看即可
patronictl -c /data/patroni/conf/postgres0.yml list

在這裏插入圖片描述

  • 如果需要切換master，運行如下命令即可
patronictl -c /data/patroni/conf/postgres0.yml switchover

在這裏插入圖片描述

5、HAProxy（HAProxy節點）

5.1、安裝haproxy

各節點配置相同

yum -y install haproxy

5.2、修改配置文件

  • 備份原有配置文件
cp -r /etc/haproxy/haproxy.cfg /etc/haproxy/haproxy.cfg_bak
  • 修改配置文件，直接替換爲新的配置，修改其中的IP、端口、頁面用戶名密碼
cat > /etc/haproxy/haproxy.cfg <<EOF
#---------------------------------------------------------------------
# 全局定義
global
    # log語法：log [max_level_1]
    # 全局的日誌配置，使用log關鍵字，指定使用127.0.0.1上的syslog服務中的local0日誌設備，
    # 記錄日誌等級爲info的日誌
#   log         127.0.0.1 local0 info
    log         127.0.0.1 local1 notice
    chroot      /var/lib/haproxy
    pidfile     /var/run/haproxy.pid
     
    # 定義每個haproxy進程的最大連接數 ，由於每個連接包括一個客戶端和一個服務器端，
    # 所以單個進程的TCP會話最大數目將是該值的兩倍。
    maxconn     4096
     
    # 用戶，組
    user        haproxy
    group       haproxy
     
    # 以守護進程的方式運行
    daemon
 
    # turn on stats unix socket
    stats socket /var/lib/haproxy/stats
 
#---------------------------------------------------------------------
# 默認部分的定義
defaults    
    # mode語法：mode {http|tcp|health} 。http是七層模式，tcp是四層模式，health是健康檢測，返回OK
    mode tcp   
    # 使用127.0.0.1上的syslog服務的local3設備記錄錯誤信息
    log 127.0.0.1 local3 err
 
    #if you set mode to http,then you nust change tcplog into httplog
    option     tcplog
     
    # 啓用該項，日誌中將不會記錄空連接。所謂空連接就是在上游的負載均衡器或者監控系統爲了
    #探測該服務是否存活可用時，需要定期的連接或者獲取某一固定的組件或頁面，或者探測掃描
    #端口是否在監聽或開放等動作被稱爲空連接；官方文檔中標註，如果該服務上游沒有其他的負
    #載均衡器的話，建議不要使用該參數，因爲互聯網上的惡意掃描或其他動作就不會被記錄下來
    option     dontlognull
     
    # 定義連接後端服務器的失敗重連次數，連接失敗次數超過此值後將會將對應後端服務器標記爲不可用      
    retries    3
     
    # 當使用了cookie時，haproxy將會將其請求的後端服務器的serverID插入到cookie中，以保證
    #會話的SESSION持久性；而此時，如果後端的服務器宕掉了，但是客戶端的cookie是不會刷新的
    #，如果設置此參數，將會將客戶的請求強制定向到另外一個後端server上，以保證服務的正常
    option redispatch
 
    #等待最大時長  When a server's maxconn is reached, connections are left pending in a queue  which may be server-specific or global to the backend.
    timeout queue           1m
     
    # 設置成功連接到一臺服務器的最長等待時間，默認單位是毫秒
    timeout connect         1m
     
    # 客戶端非活動狀態的超時時長   The inactivity timeout applies when the client is expected to acknowledge or  send data.
    timeout client          15m
     
    # Set the maximum inactivity time on the server side.The inactivity timeout applies when the server is expected to acknowledge or  send data.
    timeout server          15m
    timeout check           30s
    maxconn                 5120   
 
#---------------------------------------------------------------------
# 配置haproxy web監控，查看統計信息
listen status
    bind 0.0.0.0:1080   
    mode http   
    log global
     
    stats enable
    # stats是haproxy的一個統計頁面的套接字，該參數設置統計頁面的刷新間隔爲30s
    stats refresh 30s   
    stats uri /haproxy-stats
    # 設置統計頁面認證時的提示內容
    stats realm Private lands
    # 設置統計頁面認證的用戶和密碼，如果要設置多個，另起一行寫入即可
    stats auth admin:Gsld1234!
    # 隱藏統計頁面上的haproxy版本信息
#    stats hide-version
     
#---------------------------------------------------------------------
listen master
    bind *:5000
        mode tcp
        option tcplog
        balance roundrobin
    option httpchk OPTIONS /master
    http-check expect status 200
    default-server inter 3s fall 3 rise 2 on-marked-down shutdown-sessions
        server node1 10.10.0.1:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
        server node2 10.10.0.2:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
        server node3 10.10.0.3:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
listen replicas
    bind *:5001
        mode tcp
        option tcplog
        balance roundrobin
    option httpchk OPTIONS /replica
    http-check expect status 200
    default-server inter 3s fall 3 rise 2 on-marked-down shutdown-sessions
        server node1 10.10.0.1:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
        server node2 10.10.0.2:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
        server node3 10.10.0.3:5432 maxconn 1500 check port 8008 inter 5000 rise 2 fall 2
EOF

5.3、啓動

systemctl start haproxy && systemctl enable haproxy && systemctl status haproxy

5.4、頁面訪問

瀏覽器訪問http://10.10.0.4:1080/haproxy-stats，輸入前面配置文件中設置的用戶名密碼即可進入，這裏是admin/Gsld1234!

在這裏插入圖片描述

  • 頁面中我們可以看到當前的主節點及從節點

在這裏插入圖片描述

  • 我們通過5000端口和5001端口分別來提供寫服務和讀服務,如果需要對數據庫寫入數,只需要對外提供10.10.0.4:5000即可,可以模擬主庫故障，即關閉其中的master節點來驗證是否會進行自動主從切換

6、Keepalived（Keepalived節點）

6.1、安裝依賴

yum -y install openssl-devel

6.2、獲取Keepalived

https://www.keepalived.org/download.html

6.3、安裝

tar -xvf keepalived-2.0.20.tar.gz && rm -rf keepalived-2.0.20.tar.gz && cd keepalived-2.0.20

./configure --prefix=/usr/local/keepalived

make && make install

6.4、版本查看

/usr/local/keepalived/sbin/keepalived -v

[外鏈圖片轉存失敗,源站可能有防盜鏈機制,建議將圖片保存下來直接上傳(img-5vcA8WZE-1583740056005)(FE2C4B0EF49342B199BBE1A0D7CBE60E)]

6.5、配置文件

  • 配置文件參數詳解

https://www.cnblogs.com/arjenlee/p/9258188.html

  • 創建配置文件目錄及配置文件
mkdir -p /etc/keepalived

自帶配置文件：/usr/local/keepalived/etc/keepalived/keepalived.conf，可作參考，這裏直接創建新的配置文件即可

  • 主服務器，即haproxy1
vim /etc/keepalived/keepalived.conf


global_defs {
    router_id haproxy1
}

vrrp_script haproxy_check {
   script "/usr/local/keepalived/check.sh"
   interval 2
   weight -20
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 80
    advert_int 1
    
    track_script {
       haproxy_check
    }
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        10.10.0.8
    }
}
  • 從服務器，即haproxy2
vim /etc/keepalived/keepalived.conf


global_defs {
    router_id haproxy2
}

vrrp_script haproxy_check {
   script "/usr/local/keepalived/check.sh"
   interval 2
   weight：-20
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id 51
    priority 70
    advert_int 1
    
    track_script {
       haproxy_check
    }
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        10.10.0.8
    }
}
  • vrrp_script：檢查腳本，檢查haproxy狀態，如果掛掉，VIP漂移
  • script：腳本命令
  • interval：檢查間隔
  • weight：權重
  • state：標記該節點是master還是backup
  • interface：配置VIP綁定的網卡，這裏使用和外網通信的網卡
  • virtual_router_id：取1-255之間的值，主備需要相同，這樣才能成爲一個組
  • priority：權重，數值高的主機是master，所以主節點要比從節點大，這是影響主備的關鍵
  • advert_int：主備之間通訊的間隔秒數，用於判斷主節點是否存活
  • auth_type：進行安全認證的方式，PASS或者AH方式，推薦PASS
  • auth_pass：PASS的密碼
  • virtual_ipaddress：VIP地址，最多可以寫20個，keepalive啓動後會自動配置該VIP

6.6、檢測腳本

放置到配置文件所指定目錄並添加權限

vim /usr/local/keepalived/check.sh

#!/bin/bash
count=`ps aux | grep -v grep | grep haproxy | wc -l`
if [ $count -eq 0 ]; then
    exit 1
else
    exit 0
fi

chmod +x /usr/local/keepalived/check.sh

6.7、啓動

  • 創建service文件
cat > /etc/systemd/system/keepalived.service <<EOF
[Unit]
Description=LVS and VRRP High Availability Monitor
After=syslog.target network-online.target

[Service]
Type=forking
KillMode=process
ExecStart=/usr/local/keepalived/sbin/keepalived

[Install]
WantedBy=multi-user.target
EOF

systemctl daemon-reload

systemctl enable keepalived && systemctl start keepalived && systemctl status keepalived

6.8、主從查看

在主節點查看網絡信息，可以發現我們設置的VIP，當主節點、haproxy、keepalived任意一個掛掉時，VIP都會自動漂移到從節點實現高可用

在這裏插入圖片描述

發表評論
所有評論
還沒有人評論，想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.
相關文章

pgsql 創建dp_link

#創建dplink create extension dblink; #創建連接 select dblink_connect('test1_dblink','host=localhost port=5432 dbname=test use

有时间织个毛衣 2020-07-05 02:26:12

pgsql常用操作

登錄： psql -h localhost -p 5432 postgres postgres 創建表 CREATE TABLE "public"."t_user" ( "id" int8 NOT NULL ,"phone_inf

有时间织个毛衣 2020-07-05 02:26:02

pgsql 工具

--查看數據庫中有哪些序列 --r =普通表， i =索引，S =序列，v =視圖，m =物化視圖， c =複合類型，t = TOAST表，f =外部表 select * from pg_class where relkind='S'

有时间织个毛衣 2020-07-05 02:26:02

pgsql的備份與恢復

在window上備份：  （1）：在pgsql的bin目錄在打開命令窗口。 （2）：執行備份 ./pg_dump -h ip地址  -p 端口號 -U 用戶名 -d 數據庫 -Fc  壓縮備份 -f c:/databasename2.ba

有时间织个毛衣 2020-07-05 02:26:02

pgsql添加自增序列

1.創建一張表 CREATE TABLE "public"."tt" ( "name" varchar(128),  "status" int4 DEFAULT 2,  "id" int8 NOT NULL,  CONSTRAINT "t

有时间织个毛衣 2020-07-05 02:26:02

PostgreSQL如何建立自增主鍵

文章目錄一、 準備工作二、 創建序列達到自增的效果1. 創建序列2. 序列的相關方法3. 插入數據4. 設置默認值三、使用SERIAL1. 建表語句2. 插入數據              以前建立自增主鍵都是用mysql建的，只

慢点走 2020-07-04 09:46:10

pgsql 字符串按照指定字符分割,取最後一個

需求: 數據庫存的字段類似: 0000,1111,2222     0000,1111    0000 要求取出以逗號分割後,的最後一位. split_data(field, ',' length(replace(field, ',',

Brico_csdn 2020-07-03 08:29:32

解決：pgsql多模式下數據源配置問題

問題：數據庫（bidb）存在多個模式（bidb、dwcdm、mnt、public），springboot無法連接到想要的bidb模式下的表。 原因：用以上的配置語句，會默認連接到public模式下的表 解決：url後加上【?curre

陈暗暗 2020-06-29 07:42:02

pgsql，mysql case when

case when 有兩種用法: 一種是when後面是boolean值，一種是when 後面是值   布爾值語法: select CASE WHEN yhsm='' THEN '非即徵即退' WHEN position('即徵即退'

我是大头鸟 2020-06-27 02:47:36

pg,gp創建分區表、計算經緯度距離，轉換漂移經緯度

------------------------------------------------------------------------------------------------------ 1.計算經緯度距離 CREAT

yostkevin 2020-06-24 06:10:33

PostgreSQL查詢 動態輸入參數

工作中會碰到臨時幾天查詢相關報表的情況，sql寫好後每天只需改日期再執行一次就可以，但是一個個替換或者批量替換不僅耗時耗力，還有可能出錯，所以想要能夠動態輸入查詢參數並且同時改變成這個參數的功能。 之前使用pl/sql時，學會了查詢時使用

chengsi101 2020-06-24 04:16:25

postgresql 中 uuid 的使用

本文總共介紹兩種方法 ： 1.使用create extension命令 create extension "uuid-ossp" 安裝擴展成功以後，就可以通過uuid_generate_v4()或uuid_generate_v1()

Allen Leung 2020-06-23 21:05:21

4.pgsql各數據類型測試

#############################1.數字類型測試############################### insert into t_num_test values(1,2,12,3.141592653)

Lei_Da_Gou 2020-06-22 17:59:28

17.pgsql中的數據庫物理存儲結構-分頁文件（1）

1.表和索引所使用的頁面格式的概述 項指的是存儲在一個頁面裏的獨立數據值。在一個表裏，一個項是一個行；在一個索引裏，一個項是一條索引記錄。 每個表和索引都以固定尺寸(通常是 8K ，但也可以在編譯時選擇其它尺寸)的頁面數組存儲。在表裏，所

Lei_Da_Gou 2020-06-22 16:59:08

13.pgsql中的用戶、角色、權限

一、用戶/角色 1.查看原始數據角色 查看命令： \du 執行結果： 默認只有一個超級用戶，賬號密碼都爲postgres。 2.添加一個角色 添加sql： -- 添加角色role1 create role role1; 執行效果：

Lei_Da_Gou 2020-06-22 16:06:06