基于Docker的PostgreSQL高可用部署（支持离线部署、读写分离）

部署架构

架构图

一、物料准备

1、Docker安装包：https://download.docker.com/linux/static/stable/x86_64/docker-29.7.2.tgz

2、Docker Compose可执行文件：https://github.com/docker/compose/releases/download/v5.4.0/docker-compose-linux-x86_64

3、etcd镜像：gcr.io/etcd-development/etcd:v3.7.1

4、Patroni+PostgreSQL镜像：ghcr.io/zalando/spilo-18:4.1-p2

更多PostgreSQL版本：https://github.com/orgs/zalando/packages?repo_name=spilo

5、HAProxy镜像：haproxy:3.4.3

6、Keepalived镜像：osixia/keepalived:2.3.4

二、设置环境

注意修改下方hostname及ip配置。

hostnamectl set-hostname postgres_ha_104
echo '192.168.162.104 postgres_ha_104' >> /etc/hosts
echo '192.168.162.105 postgres_ha_105' >> /etc/hosts
echo '192.168.162.106 postgres_ha_106' >> /etc/hosts

三、安装Docker、Docker Compose

1、Docker：

tar xf docker-29.7.2.tgz
mv docker/* /usr/bin/
rm -rf docker docker-29.7.2.tgz
cat > /etc/systemd/system/docker.service << EOF
[Unit]
Description=Docker Application Container Engine
Documentation=https://docs.docker.com
After=network-online.target firewalld.service containerd.service time-set.target
Wants=network-online.target containerd.service

[Service]
Type=notify
# the default is not to use systemd for cgroups because the delegate issues still
# exists and systemd currently does not support the cgroup feature set required
# for containers run by docker
ExecStart=/usr/bin/dockerd
ExecReload=/bin/kill -s HUP $MAINPID
TimeoutStartSec=0
RestartSec=2
Restart=always

# Note that StartLimit* options were moved from "Service" to "Unit" in systemd 229.
# Both the old, and new location are accepted by systemd 229 and up, so using the old location
# to make them work for either version of systemd.
StartLimitBurst=3

# Note that StartLimitInterval was renamed to StartLimitIntervalSec in systemd 230.
# Both the old, and new name are accepted by systemd 230 and up, so using the old name to make
# this option work for either version of systemd.
StartLimitInterval=60s

# Having non-zero Limit*s causes performance problems due to accounting overhead
# in the kernel. We recommend using cgroups to do container-local accounting.
LimitNPROC=infinity
LimitCORE=infinity
# Older systemd versions default to a LimitNOFILE of 1024:1024, which is insufficient for many
# applications including dockerd itself and will be inherited. Raise the hard limit, while
# preserving the soft limit for select(2).
LimitNOFILE=1024:524288

# Comment TasksMax if your systemd version does not support it.
# Only systemd 226 and above support this option.
TasksMax=infinity

# set delegate yes so that systemd does not reset the cgroups of docker containers
Delegate=yes

# kill only the docker process, not all processes in the cgroup
KillMode=process
OOMScoreAdjust=-500

[Install]
WantedBy=multi-user.target

EOF

设置docker存储位置（注意修改存储位置）

mkdir -p /etc/docker
cat > /etc/docker/daemon.json << EOF
{
  "data-root": "/data/docker",
  "log-driver": "json-file",
  "log-opts": {
    "max-size": "100m",
    "max-file": "5"
  }
}
EOF
systemctl daemon-reload
systemctl start docker
systemctl enable docker

2、Docker Compose：

mv docker-compose-linux-x86_64 /usr/bin/docker-compose
chmod +x /usr/bin/docker-compose

四、部署etcd集群

注意修改下方ETCD_NAME、ETCD_ADVERTISE_CLIENT_URLS、ETCD_INITIAL_ADVERTISE_PEER_URLS、ETCD_INITIAL_CLUSTER配置。

mkdir -p /data/etcd
cat > /data/etcd/docker-compose.yml << EOF
services:
  etcd:
    image: gcr.io/etcd-development/etcd:v3.7.1
    container_name: etcd
    restart: unless-stopped
    network_mode: host
    environment:
      ETCD_NAME: etcd1
      ETCD_DATA_DIR: /etcd-data
      ETCD_LISTEN_CLIENT_URLS: http://0.0.0.0:2379
      ETCD_ADVERTISE_CLIENT_URLS: http://192.168.162.104:2379
      ETCD_LISTEN_PEER_URLS: http://0.0.0.0:2380
      ETCD_INITIAL_ADVERTISE_PEER_URLS: http://192.168.162.104:2380
      ETCD_INITIAL_CLUSTER: etcd1=http://192.168.162.104:2380,etcd2=http://192.168.162.105:2380,etcd3=http://192.168.162.106:2380
      ETCD_INITIAL_CLUSTER_TOKEN: etcd-cluster
      ETCD_INITIAL_CLUSTER_STATE: new
      ETCD_LOG_LEVEL: info
      ETCD_LOGGER: zap
      ETCD_LOG_OUTPUTS: stderr
    volumes:
      - /data/etcd/etcd_data:/etcd-data

EOF
cd /data/etcd && docker-compose up -d

五、部署Patroni管理的PostgreSQL集群

注意修改下方ETCD3_HOSTS、RESTAPI_CONNECT_ADDRESS、connect_address配置。

mkdir -p /data/postgres
cat > /data/postgres/docker-compose.yml << EOF
services:
  patroni:
    image: ghcr.io/zalando/spilo-18:4.1-p2
    container_name: patroni
    restart: unless-stopped
    shm_size: 2gb
    network_mode: host
    environment:
      SPILO_PROVIDER: local
      SCOPE: postgres-ha
      ETCD3_HOSTS: 192.168.162.104:2379,192.168.162.105:2379,192.168.162.106:2379
      RESTAPI_CONNECT_ADDRESS: 192.168.162.104:8008
      SPILO_CONFIGURATION: |
        postgresql:
          listen: 0.0.0.0:15432
          connect_address: 192.168.162.104:15432
          parameters:
            shared_buffers: 1GB
      PGUSER_SUPERUSER: postgres
      PGPASSWORD_SUPERUSER: postgres_pass
      PGUSER_ADMIN: admin
      PGPASSWORD_ADMIN: admin_pass
      PGUSER_STANDBY: standby
      PGPASSWORD_STANDBY: standby_pass
    volumes:
      - /data/postgres/pgdata:/home/postgres/pgdata

EOF
cd /data/postgres && docker-compose up -d

六、部署HAProxy集群

注意修改下方server ip配置。

mkdir -p /data/haproxy
cat > /data/haproxy/haproxy.cfg << EOF
global
    log stdout format raw local0
    maxconn 10000

defaults
    log global
    mode tcp
    option tcplog
    timeout connect 5s
    timeout client 1h
    timeout server 1h

frontend postgres_write
    bind *:5432
    mode tcp
    default_backend postgres_primary

backend postgres_primary
    mode tcp
    option httpchk GET /primary
    http-check expect status 200

    default-server inter 3s fall 3 rise 2 on-marked-down shutdown-sessions

    server pg1 192.168.162.104:15432 check port 8008
    server pg2 192.168.162.105:15432 check port 8008
    server pg3 192.168.162.106:15432 check port 8008

frontend postgres_read
    bind *:5433
    mode tcp
    default_backend postgres_replicas

backend postgres_replicas
    mode tcp
    option httpchk GET /replica
    http-check expect status 200
    balance roundrobin

    default-server inter 3s fall 3 rise 2 on-marked-down shutdown-sessions

    server pg1 192.168.162.104:15432 check port 8008
    server pg2 192.168.162.105:15432 check port 8008
    server pg3 192.168.162.106:15432 check port 8008

EOF
cat > /data/haproxy/docker-compose.yml << EOF
services:
  haproxy:
    image: haproxy:3.4.3
    container_name: haproxy
    restart: unless-stopped
    network_mode: host
    volumes:
      - /data/haproxy/haproxy.cfg:/usr/local/etc/haproxy/haproxy.cfg:ro

EOF
cd /data/haproxy && docker-compose up -d

七、部署Keepalived集群

注意修改router_id、interface、unicast_src_ip、unicast_peer、virtual_ipaddress配置，注意三个节点的priority值应分别为150、140、130。

mkdir -p /data/keepalived
cat > /data/keepalived/check_script.sh << EOF
#!/bin/bash
 
nc -z -w 2 127.0.0.1 5432
 
EOF
# 查看网卡名称
ifconfig
cat > /data/keepalived/keepalived.conf << EOF
global_defs {
    router_id PG_HA_104
}
 
vrrp_script chk_script {
    script "/usr/local/bin/check_script.sh"
    interval 3
    timeout 5
    fall 3
    rise 2
    init_fail
}
 
vrrp_instance VI_PG {
    state BACKUP
    priority 150
    nopreempt
    interface eth0
    virtual_router_id 51
    advert_int 1
    unicast_src_ip 192.168.162.104
 
    unicast_peer {
        192.168.162.105
        192.168.162.106
    }
 
    authentication {
        auth_type PASS
        auth_pass postgres
    }
 
    virtual_ipaddress {
        192.168.162.107/24
    }
 
    track_script {
        chk_script
    }
}
 
EOF
cat > /data/keepalived/docker-compose.yml << EOF
services:
  keepalived:
    image: osixia/keepalived:2.3.4
    container_name: keepalived
    restart: unless-stopped
    network_mode: host
    cap_add:
      - NET_ADMIN
      - NET_BROADCAST
      - NET_RAW
    volumes:
      - /data/keepalived/keepalived.conf:/etc/keepalived/keepalived.conf:ro
      - /data/keepalived/check_script.sh:/usr/local/bin/check_script.sh:ro
 
EOF
cd /data/keepalived && docker-compose up -d

八、不包含HAProxy的配置

HAProxy组件在整个集群中起到了读写分离，以及分离VIP和postgres主节点的功能。

然而在很多情况下，额外的HAProxy组件会带来许多意想不到的异常。

因此，如果要去除HAProxy组件，请按照如下操作。

1、停止HAProxy

cd /data/haproxy
docker-compose down

2、调整Keepalived检查脚本，改为由Keepalived直接检查主节点：

cd /data/keepalived
cat > /data/keepalived/check_script.sh << EOF
#!/bin/bash
 
wget --spider -q http://127.0.0.1:8008/primary
 
EOF
docker-compose restart

3、将postgres端口由15432改为5432

cd /data/postgres
vim /data/postgres/docker-compose.yml
           listen: 0.0.0.0:5432
          connect_address: 192.168.162.104:5432
docker-compose up -d

posted @ 2026-08-18 14:50 hackyo 阅读(22) 评论(0) 收藏举报

刷新页面返回顶部

Loading

hackyo