pd多套集群部署同一机器的问题

                                         --2019-04-25 春雷

1、汇总

1.1、问题

多套tidb集群的pd  部署在同样的机器，pd的服务相同，导致pd无法启动

版本：2.1.2

1.2、问题及解决

修改相关文件的端口部分解决

2、具体

2.1、具体问题

2.1.1、系统服务

/etc/systemd/system

pd.service

2.1.2、pd的启停脚本

【${deploy_dir}/scripts/start_pd.sh】

#!/bin/bash
set -e

# WARNING: This file was auto-generated. Do not edit!
# All your edit might be overwritten!
sudo systemctl start  pd.service

【 ${deploy_dir} /scripts/stop_pd.sh】

#!/bin/bash
set -e

# WARNING: This file was auto-generated. Do not edit!
# All your edit might be overwritten!
sudo systemctl stop  pd.service

2.2、修复

tidb中控机：

【1、更改部署的】

/work/tidb/tidb-ansible-2.1/roles/pd/tasks/ systemd_deployment.yml

更改：

service_name: pd- {{ pd_client_port }}

【2、滚动升级的】

/work/tidb/tidb-ansible-2.1/ rolling_update.yml

更改：

- name: stop PD by systemd
systemd: name=pd -{{ pd_client_port }} .service state=stopped

- name: start PD by systemd
systemd: name=pd -{{ pd_client_port }} .service state=started

【3、更改start的】

/work/tidb/tidb-ansible-2.1/ start.yml

- name: start PD by systemd
systemd: name=pd- {{ pd_client_port }} .service state=started

2.3、修复后结果

手动删除目标pd机器的：

${deploy_dir}/scripts/start_pd.sh

${ deploy_dir } /scripts/stop_pd.sh

${ deploy_dir } /scripts/run_pd.sh

中控机重新部署：

ansible-playbook deploy.yml  -l  pd机器IP

检查：

start_pd.sh 

#!/bin/bash
set -e

# WARNING: This file was auto-generated. Do not edit!
# All your edit might be overwritten!
sudo systemctl start pd-10000 .service

stop_pd.sh

#!/bin/bash
set -e

# WARNING: This file was auto-generated. Do not edit!
# All your edit might be overwritten!
sudo systemctl stop pd-10000 .service

cd /etc/systemd/system

pd-10000 .service

开启：

ansible-playbook start.yml