HGHAC集群版本升级

瀚高PG实验室2026-11-03 11:04

文章目录

文档用途

1、文档用于指导HGHAC集群版本升级。 2、此文档以V4.2.3升级到V4.2.7.4为示例。

详细信息

1、升级前检查etcd和hghac集群状态

复制代码 
1.查看etcd集群，Highgo@123为集群密码
[highgo@localhost ~]$ etcdctl --user=root:Highgo@123 endpoint status --write-out=table
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
|          ENDPOINT          |        ID        | VERSION | DB SIZE | IS LEADER | IS LEARNER | RAFT TERM | RAFT INDEX | RAFT APPLIED INDEX | ERRORS |
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
| http://192.168.80.142:2379 |  2f477b081f877ca |   3.5.9 |   20 kB |      true |      false |         2 |         25 |                 25 |        |
| http://192.168.80.143:2379 | e16a54f0b1466abc |   3.5.9 |   20 kB |     false |      false |         2 |         26 |                 26 |        |
| http://192.168.80.144:2379 | f5c09494b0f85dc0 |   3.5.9 |   20 kB |     false |      false |         2 |         27 |                 27 |        |
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
[highgo@localhost ~]$ etcdctl --user=root:Highgo@123 endpoint health --write-out=table
+----------------------------+--------+------------+-------+
|          ENDPOINT          | HEALTH |    TOOK    | ERROR |
+----------------------------+--------+------------+-------+
| http://192.168.80.142:2379 |   true | 6.279926ms |       |
| http://192.168.80.144:2379 |   true | 5.718948ms |       |
| http://192.168.80.143:2379 |   true | 2.640563ms |       |
+----------------------------+--------+------------+-------+
#2.查看hghac集群
[highgo@localhost ~]$ hghactl list
+ Cluster: highgo-ee-cluster --+--------------+-----------+----+-----------+
| Member | Host                | Role         | State     | TL | Lag in MB |
+--------+---------------------+--------------+-----------+----+-----------+
| ee_01  | 192.168.80.142:5866 | Leader       | running   |  2 |           |
| ee_02  | 192.168.80.143:5866 | Replica      | streaming |  2 |         0 |
| ee_03  | 192.168.80.144:5866 | Sync Standby | streaming |  2 |         0 |
+--------+---------------------+--------------+-----------+----+-----------+

2、暂停集群

复制代码 
[highgo@localhost ~]$ hghactl pause
Success: cluster management is paused
#暂停集群管理即进入维护模式，hghac仍会更新DCS中的集群状态信息，允许手动主备切换，但不会自动启动、停止或重启数据库，也不会触发自动故障切换，从而防止升级过程中发生不必要的自动故障转移。

3、停掉备节点服务

复制代码 
systemctl stop etcd
systemctl stop hghac
systemctl stop vip-manager
#由于集群处于维护状态，停止hghac服务不会停止数据库进程，数据库服务保持原状态。

4、备份可执行文件

复制代码 
#1.hghac主程序目录
mv /usr/local/hghac/hac/hghac /usr/local/hghac/hac/hghac.bak
#2.hghactl客户端目录
mv /usr/local/hghac/hac/hghactl /usr/local/hghac/hac/hghactl.bak
#3.etcd二进制
mv /usr/local/hghac/etcd/etcd /usr/local/hghac/etcd/etcd.bak
mv /usr/local/hghac/etcd/etcdctl /usr/local/hghac/etcd/etcdctl.bak
#4.vip-manager二进制
mv /usr/local/hghac/vip-manager/vip-manager /usr/local/hghac/vip-manager/vip-manager.bak

5、解压新版本

复制代码 
tar -xf /data/hgupload/hghac-4.2.7.4-2-v9-5841-6474b12-20260825.x86_64.tar.gz  -C /usr/local
chown highgo. /usr/local/hghac-4.2.7.4 -R
ll /usr/local/hghac-4.2.7.4/

6、新版本配置文件复制到对应目录

复制代码 
#1.hghac主程序目录
cp -rp /usr/local/hghac-4.2.7.4/hac/hghac    /usr/local/hghac/hac
#2.hghactl客户端目录
cp -rp /usr/local/hghac-4.2.7.4/hac/hghactl  /usr/local/hghac/hac
#3.etcd二进制
cp -rp /usr/local/hghac-4.2.7.4/etcd/etcd     /usr/local/hghac/etcd
cp -rp /usr/local/hghac-4.2.7.4/etcd/etcdctl  /usr/local/hghac/etcd
#4.vip-manager二进制
cp -rp /usr/local/hghac-4.2.7.4/vip-manager/vip-manager   /usr/local/hghac/vip-manager

7、查看备份文件和新文件

复制代码 
ll /usr/local/hghac/hac/hghac*
ll /usr/local/hghac/hac/hghactl*
ll /usr/local/hghac/etcd/etcd*
ll /usr/local/hghac/etcd/etcdctl*
ll /usr/local/hghac/vip-manager/vip-manager*

8、启动备节点各服务

复制代码 
#1.启动etcd服务，并查看服务状态
systemctl start etcd
systemctl status etcd
#2.查看etcd集群
etcdctl --user=root:Highgo@123 endpoint status --write-out=table
etcdctl --user=root:Highgo@123 endpoint health --write-out=table
#3.启动hghac服务，并查看服务状态
systemctl start hghac
systemctl status hghac
#4.查看hghac集群
hghactl list
#5.启动vip-manger服务，并查看服务状态
systemctl start vip-manager
systemctl status vip-manager

9、升级主节点

复制代码 
#1.主备切换
hghactl  switchover
#2.重复备节点步骤进行升级
10、检查集群状态
#1.查看etcd集群，Highgo@123为集群密码
[highgo@localhost ~]$ etcdctl --user=root:Highgo@123 endpoint status --write-out=table
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
|          ENDPOINT          |        ID        | VERSION | DB SIZE | IS LEADER | IS LEARNER | RAFT TERM | RAFT INDEX | RAFT APPLIED INDEX | ERRORS |
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
| http://192.168.80.142:2379 |  2f477b081f877ca |   3.5.9 |   20 kB |      true |      false |         2 |         25 |                 25 |        |
| http://192.168.80.143:2379 | e16a54f0b1466abc |   3.5.9 |   20 kB |     false |      false |         2 |         26 |                 26 |        |
| http://192.168.80.144:2379 | f5c09494b0f85dc0 |   3.5.9 |   20 kB |     false |      false |         2 |         27 |                 27 |        |
+----------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
[highgo@localhost ~]$ etcdctl --user=root:Highgo@123 endpoint health --write-out=table
+----------------------------+--------+------------+-------+
|          ENDPOINT          | HEALTH |    TOOK    | ERROR |
+----------------------------+--------+------------+-------+
| http://192.168.80.142:2379 |   true | 6.279926ms |       |
| http://192.168.80.144:2379 |   true | 5.718948ms |       |
| http://192.168.80.143:2379 |   true | 2.640563ms |       |
+----------------------------+--------+------------+-------+
#2.查看hghac集群状态
[highgo@localhost ~]$ hghactl list
+ Cluster: highgo-ee-cluster (7682598096742189505) --+----+-------------+-----+------------+-----+
| Member | Host                | Role    | State     | TL | Receive LSN | Lag | Replay LSN | Lag |
+--------+---------------------+---------+-----------+----+-------------+-----+------------+-----+
| ee_01  | 192.168.80.142:5866 | Replica | streaming | 11 |   0/4020208 |   0 |  0/4020208 |   0 |
| ee_02  | 192.168.80.143:5866 | Leader  | running   | 11 |             |     |            |     |
| ee_03  | 192.168.80.144:5866 | Replica | streaming | 11 |   0/4020208 |   0 |  0/4020208 |   0 |
+--------+---------------------+---------+-----------+----+-------------+-----+------------+-----+

相关文档

相关推荐
品牌常新2 小时前
室外巡检机器人推荐：园区与厂区怎么选
数据库·人工智能·机器人
snow@li2 小时前
redis：RedisDesktopManager（RDM/RESP.app）Redis 数据可视化全景深度梳理
数据库·redis·缓存
lqbzlll2 小时前
MySQL 慢查询日志开了但看不懂？从 slow_query_log 到 EXPLAIN ANALYZE 一步步定位
数据库·mysql·adb
code_slave(码畜)2 小时前
微服务架构落地：慢 SQL 治理与连接池调优 —— 守住数据库的最后一道防线
数据库·spring boot·sql·spring cloud·微服务·架构
IvorySQL2 小时前
PostgreSQL 日报｜子表 CHECK 约束恢复标记错误（11 月 2 日）
数据库·postgresql
你看我名字3 小时前
Redis 高可用、持久化与故障切换：从缓存组件到云上基础设施
数据库·redis·缓存
风翼靓崽3 小时前
centos7以mysql为例自建yum源
数据库·mysql·adb
码农阿豪4 小时前
MongoDB迁移到金仓数据库：商品属性继续灵活存，订单报表交给 SQL
数据库·sql·mongodb
热门推荐
01国内可直接用、免费额度/永久免费的大模型API清单（含 SiliconFlow、火山、阿里、智谱、百度、Kimi、DeepSeek、DMXAPI 等）02历年考研数学一、数学二、数学三真题试卷及答案PDF03GitHub 镜像站点04几个好用的ip纯净度检测网站05苹果iPA游戏软件资源下载网站06如何新建文件夹？ 电脑新建文件夹的4种方法07蜘蛛磁力 搜索引擎大全，如何使用蜘蛛磁力查找磁力链接08昇腾910B系列芯片完全指南：四款型号对比、服务器选型与大模型部署实战09Dify：一站式AI应用开发平台全解析与Windows部署指南10微信历史版本含下载地址（ Windows PC | 安卓 | MAC ）及设置微信不更新