CentOS 7 部署TiDB 2.1.4集群
2024-04-12 09:55:26
$ cat /etc/centos-release
CentOS Linux release 7.4.1708 (Core) 安装部署：
## yum -y install epel-release git curl sshpass python-pip git
# useradd -m -d /home/tidb tidb
# passwd tidb
# visudo
tidb ALL=(ALL) NOPASSWD: ALL生成 ssh key: 执行 su 命令从 root 用户切换到 tidb 用户下。
#ssh key
# su - tidb$ ssh-keygen -t rsa
提示 Enter passphrase 时直接回车即可。执行成功后，ssh 私钥文件为 /home/tidb/.ssh/id_rsa， ssh 公钥文件为 /home/tidb/.ssh/id_rsa.pub$ cd /home/tidb/tidb-ansible
$ sudo pip install -r ./requirements.txt
--ansible版本确认：
$ ansible --version
ansible 2.7.7config file = /etc/ansible/ansible.cfgconfigured module search path = [u'/home/tidb/.ansible/plugins/modules', u'/usr/share/ansible/plugins/modules']ansible python module location = /usr/lib/python2.7/site-packages/ansibleexecutable location = /bin/ansiblepython version = 2.7.5 (default, Aug 4 2017, 00:39:18) [GCC 4.8.5 20150623 (Red Hat 4.8.5-16)]下载2.1版本：
$ git clone -b release-2.1 https://github.com/pingcap/tidb-ansible.git-- 编辑主机文件：
$ vim /home/tidb/tidb-ansible/hosts.ini
[servers]
172.16.4.171
172.16.4.172
172.16.4.173
172.16.4.174
172.16.4.175[all:vars]
username = tidb
#ntp_server = pool.ntp.org
ntp_server = 172.16.4.168-- 配置主机互信
执行以下命令，按提示输入部署目标机器 root 用户密码。该步骤将在部署目标机器上创建 tidb 用户，并配置 sudo 规则，配置中控机与部署目标机器之间的 ssh 互信。
$ cd /home/tidb/tidb-ansible
$ ansible-playbook -i hosts.ini create_users.yml -u root -k
SSH password: PLAY [all] *******************************TASK [create user] ***********************
ok: [172.16.4.173]
ok: [172.16.4.174]
ok: [172.16.4.175]
ok: [172.16.4.172]
ok: [172.16.4.171]TASK [set authorized key] ****************
ok: [172.16.4.171]
ok: [172.16.4.172]
ok: [172.16.4.174]
ok: [172.16.4.173]
ok: [172.16.4.175]TASK [update sudoers file] **************
ok: [172.16.4.171]
ok: [172.16.4.173]
ok: [172.16.4.172]
ok: [172.16.4.174]
ok: [172.16.4.175]PLAY RECAP **************
172.16.4.171 : ok=3 changed=0 unreachable=0 failed=0
172.16.4.172 : ok=3 changed=0 unreachable=0 failed=0
172.16.4.173 : ok=3 changed=0 unreachable=0 failed=0
172.16.4.174 : ok=3 changed=0 unreachable=0 failed=0
172.16.4.175 : ok=3 changed=0 unreachable=0 failed=0 Congrats! All goes well. :-)目标主机上部署NTP服务：
你的部署目标机器时间、时区设置一致，已开启 NTP 服务且在正常同步时间，此步骤可忽略.
$ ansible-playbook -i hosts.ini deploy_ntp.yml -u tidb -b
...
PLAY RECAP ***********************
172.16.4.171 : ok=6 changed=3 unreachable=0 failed=0
172.16.4.172 : ok=6 changed=3 unreachable=0 failed=0
172.16.4.173 : ok=6 changed=3 unreachable=0 failed=0
172.16.4.174 : ok=6 changed=3 unreachable=0 failed=0
172.16.4.175 : ok=6 changed=3 unreachable=0 failed=0 Congrats! All goes well. :-)在部署目标机器上批量设置：
$ ansible -i hosts.ini all -m shell -a "cpupower frequency-set --governor performance" -u tidb -b各个目标主机采用ext4 文件格式。配置信息：
组件规划：$ vim inventory.ini
## TiDB Cluster Part
[tidb_servers]
172.16.4.172
172.16.4.173[tikv_servers]
172.16.4.173
172.16.4.174
172.16.4.175[pd_servers]
172.16.4.173
172.16.4.174
172.16.4.175[spark_master]
172.16.4.174[spark_slaves]
172.16.4.175[lightning_server]
172.16.4.174
172.16.4.175
[importer_server]
172.16.4.174
172.16.4.175## Monitoring Part
# prometheus and pushgateway servers
[monitoring_servers]
172.16.4.171
[grafana_servers]
172.16.4.171# node_exporter and blackbox_exporter servers
[monitored_servers]
172.16.4.171
172.16.4.172
172.16.4.173
172.16.4.174
172.16.4.175[alertmanager_servers]
172.16.4.171[kafka_exporter_servers]
172.16.4.171
## Binlog Part
[pump_servers]
pump1 ansible_host=172.16.4.173 deploy_dir=/data/tidb/pump
pump2 ansible_host=172.16.4.174 deploy_dir=/data/tidb/pump
pump3 ansible_host=172.16.4.175 deploy_dir=/data/tidb/pump
[drainer_servers]
172.16.4.173
172.16.4.174
172.16.4.175
## Group variables
[pd_servers:vars]
# location_labels = ["zone","rack","host"]## Global variables
[all:vars]
deploy_dir = /data/tidb/tidb_deploy## Connection
# ssh via normal user
ansible_user = tidbcluster_name = pre-tidbtidb_version = v2.1.4 # process supervision, [systemd, supervise]
process_supervision = systemdtimezone = Asia/Shanghaienable_firewalld = False
# check NTP service
enable_ntpd = True
set_hostname = False## binlog trigger
enable_binlog = True# kafka cluster address for monitoring, example:
# kafka_addrs = "192.168.0.11:9092,192.168.0.12:9092,192.168.0.13:9092"
kafka_addrs = "172.16.4.173:9092,172.16.4.174:9092,172.16.4.175:9092"# zookeeper address of kafka cluster for monitoring, example:
# zookeeper_addrs = "192.168.0.11:2181,192.168.0.12:2181,192.168.0.13:2181"
zookeeper_addrs = "172.16.4.173:2181,172.16.4.174:2181,172.16.4.175:2181"# store slow query log into seperate file
enable_slow_query_log = True# enable TLS authentication in the TiDB cluster
enable_tls = False# KV mode
deploy_without_tidb = False# Optional: Set if you already have a alertmanager server.
# Format: alertmanager_host:alertmanager_port
alertmanager_target = ""grafana_admin_user = "admin"
grafana_admin_password = "admin"### Collect diagnosis
collect_log_recent_hours = 2enable_bandwidth_limit = True
# default: 10Mb/s, unit: Kbit/s
collect_bandwidth_limit = 10000注意：若所有 server 返回 tidb 表示 ssh 互信配置成功：
$ ansible -i inventory.ini all -m shell -a 'whoami'
172.16.4.172 | CHANGED | rc=0 >>
tidb172.16.4.173 | CHANGED | rc=0 >>
tidb172.16.4.174 | CHANGED | rc=0 >>
tidb172.16.4.171 | CHANGED | rc=0 >>
tidb172.16.4.175 | CHANGED | rc=0 >>
tidbpump1 | CHANGED | rc=0 >>
tidbpump2 | CHANGED | rc=0 >>
tidbpump3 | CHANGED | rc=0 >>
tidb
若所有 server 返回 root 表示 tidb 用户 sudo 免密码配置成功：
$ ansible -i inventory.ini all -m shell -a 'whoami' -b
172.16.4.171 | CHANGED | rc=0 >>
root172.16.4.173 | CHANGED | rc=0 >>
root172.16.4.172 | CHANGED | rc=0 >>
root172.16.4.174 | CHANGED | rc=0 >>
root172.16.4.175 | CHANGED | rc=0 >>
rootpump2 | CHANGED | rc=0 >>
rootpump1 | CHANGED | rc=0 >>
rootpump3 | CHANGED | rc=0 >>
root执行 local_prepare.yml playbook，联网下载 TiDB binary 到中控机：
$ ansible-playbook local_prepare.yml初始化系统环境，修改内核参数：
$ ansible-playbook bootstrap.yml
... 软件下载信息 ...
TASK [local : download tidb binary] ************
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/tidb-v2.1.4-linux-amd64.tar.gz', u'version': u'v2.1.4', u'name': u'tidb'})TASK [local : download common binary] *************
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/fio-3.8.tar.gz', u'checksum': u'sha256:15739abde7e74b59ac59df57f129b14fc5cd59e1e2eca2ce37b41f8c289c3d58', u'version': 3.8, u'name': u'fio'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/grafana_collector-latest-linux-amd64.tar.gz', u'version': u'latest', u'name': u'grafana_collector'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/kafka_exporter-1.1.0.linux-amd64.tar.gz', u'version': u'1.1.0', u'name': u'kafka_exporter'})TASK [local : download diagnosis tools] **************
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/tidb-insight-v0.2.5-1-g99b8fea.tar.gz', u'version': u'v0.2.5-1-g99b8fea', u'name': u'tidb-insight'})TASK [local : download cfssl binary] *************TASK [local : download cfssljson binary] **************TASK [local : include_tasks] *************
included: /home/tidb/tidb-ansible/roles/local/tasks/binary_deployment.yml for localhostTASK [local : download other binary] ************TASK [local : download other binary under gfw] ************
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/prometheus-2.2.1.linux-amd64.tar.gz', u'version': u'2.2.1', u'name': u'prometheus'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/alertmanager-0.14.0.linux-amd64.tar.gz', u'version': u'0.14.0', u'name': u'alertmanager'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/node_exporter-0.15.2.linux-amd64.tar.gz', u'version': u'0.15.2', u'name': u'node_exporter'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/pushgateway-0.4.0.linux-amd64.tar.gz', u'version': u'0.4.0', u'name': u'pushgateway'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/grafana-4.6.3.linux-x64.tar.gz', u'version': u'4.6.3', u'name': u'grafana'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/blackbox_exporter-0.12.0.linux-amd64.tar.gz', u'version': u'0.12.0', u'name': u'blackbox_exporter'})TASK [local : download TiSpark packages] *************
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/spark-2.3.2-bin-hadoop2.7.tgz', u'checksum': u'sha256:6246b20d95c7596a29fb26d5b50a3ae3163a35915bec6c515a8e183383bedc43', u'version': u'2.3.2', u'name': u'spark-2.3.2-bin-hadoop2.7.tgz'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/tispark-latest-linux-amd64.tar.gz', u'version': u'latest', u'name': u'tispark-latest.tar.gz'})
ok: [localhost] => (item={u'url': u'http://download.pingcap.org/tispark-sample-data.tar.gz', u'version': u'latest', u'name': u'tispark-sample-data.tar.gz'})TASK [local : unarchive third party binary] *************
changed: [localhost] => (item={u'url': u'https://github.com/prometheus/prometheus/releases/download/v2.2.1/prometheus-2.2.1.linux-amd64.tar.gz', u'version': u'2.2.1', u'name': u'prometheus'})
changed: [localhost] => (item={u'url': u'https://github.com/prometheus/alertmanager/releases/download/v0.14.0/alertmanager-0.14.0.linux-amd64.tar.gz', u'version': u'0.14.0', u'name': u'alertmanager'})
changed: [localhost] => (item={u'url': u'https://github.com/prometheus/node_exporter/releases/download/v0.15.2/node_exporter-0.15.2.linux-amd64.tar.gz', u'version': u'0.15.2', u'name': u'node_exporter'})
changed: [localhost] => (item={u'url': u'https://github.com/prometheus/blackbox_exporter/releases/download/v0.12.0/blackbox_exporter-0.12.0.linux-amd64.tar.gz', u'version': u'0.12.0', u'name': u'blackbox_exporter'})
changed: [localhost] => (item={u'url': u'https://github.com/prometheus/pushgateway/releases/download/v0.4.0/pushgateway-0.4.0.linux-amd64.tar.gz', u'version': u'0.4.0', u'name': u'pushgateway'})
changed: [localhost] => (item={u'url': u'https://s3-us-west-2.amazonaws.com/grafana-releases/release/grafana-4.6.3.linux-x64.tar.gz', u'version': u'4.6.3', u'name': u'grafana'})TASK [local : unarchive tispark] *************
changed: [localhost]TASK [local : unarchive tispark-sample-data] *************
changed: [localhost]TASK [local : cp monitoring binary] *************
changed: [localhost] => (item=alertmanager)
changed: [localhost] => (item=prometheus)
changed: [localhost] => (item=node_exporter)
changed: [localhost] => (item=pushgateway)
changed: [localhost] => (item=blackbox_exporter)TASK [local : cp tispark] ***************************************************************************************************************************************************************************************************************
changed: [localhost]TASK [local : cp tispark-sample-data] *************
changed: [localhost]TASK [local : unarchive tidb binary] *************
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/tidb-v2.1.4-linux-amd64.tar.gz', u'version': u'v2.1.4', u'name': u'tidb'})TASK [local : unarchive common binary] *************
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/fio-3.8.tar.gz', u'checksum': u'sha256:15739abde7e74b59ac59df57f129b14fc5cd59e1e2eca2ce37b41f8c289c3d58', u'version': 3.8, u'name': u'fio'})
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/grafana_collector-latest-linux-amd64.tar.gz', u'version': u'latest', u'name': u'grafana_collector'})
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/kafka_exporter-1.1.0.linux-amd64.tar.gz', u'version': u'1.1.0', u'name': u'kafka_exporter'})TASK [local : cp tidb binary] *************
changed: [localhost] => (item={u'url': u'http://download.pingcap.org/tidb-v2.1.4-linux-amd64.tar.gz', u'version': u'v2.1.4', u'name': u'tidb'})
....
... 内存检查报错：
初始化系统环境，修改内核参数,此外还会检查内存和磁盘等信息：
若内存不足则会报错：fatal: [172.16.4.173]: FAILED! => {"changed": false, "msg": "This machine does not have sufficient RAM to run TiDB, at least 16000 MB."}
各个指标会检查是否达到最小的运行环境的数据。
若报错则注释掉：
$$ vim bootstrap.yml
# This play book is intend for one pass execution- name: initializing deployment targethosts: localhostgather_facts: falseroles:- check_config_static- name: check node confighosts: allgather_facts: falsebecome: trueroles:- pre-ansible- bootstrap- name: check systemhosts: allany_errors_fatal: trueroles:- check_system_static
# - { role: check_system_optional, when: not dev_mode|default(false) }- name: tikv_servers machine benchmarkhosts: tikv_serversgather_facts: falseroles:- { role: machine_benchmark, when: not dev_mode|default(false) }- name: create ops scriptshosts: localhostconnection: localgather_facts: falseroles:- ops-- SSD 硬盘检查报错：
TASK [machine_benchmark : Preflight check - Does fio randread iops of tikv_data_dir disk meet requirement] ******************************************************************************************************************************
fatal: [172.16.4.173]: FAILED! => {"changed": false, "msg": "fio: randread iops of tikv_data_dir disk is too low: 64 < 40000, it is strongly recommended to use SSD disks for TiKV and PD, or there might be performance issues."}
fatal: [172.16.4.174]: FAILED! => {"changed": false, "msg": "fio: randread iops of tikv_data_dir disk is too low: 58 < 40000, it is strongly recommended to use SSD disks for TiKV and PD, or there might be performance issues."}
fatal: [172.16.4.175]: FAILED! => {"changed": false, "msg": "fio: randread iops of tikv_data_dir disk is too low: 391 < 40000, it is strongly recommended to use SSD disks for TiKV and PD, or there might be performance issues."}$$ vim bootstrap.yml
- name: tikv_servers machine benchmarkhosts: tikv_serversgather_facts: falseroles:
# - { role: machine_benchmark, when: not dev_mode|default(false) }$ansible-playbook bootstrap.yml
...
PLAY RECAP ******************
172.16.4.171 : ok=28 changed=5 unreachable=0 failed=0
172.16.4.172 : ok=28 changed=5 unreachable=0 failed=0
172.16.4.173 : ok=28 changed=5 unreachable=0 failed=0
172.16.4.174 : ok=28 changed=5 unreachable=0 failed=0
172.16.4.175 : ok=28 changed=5 unreachable=0 failed=0
localhost : ok=3 changed=2 unreachable=0 failed=0
pump1 : ok=28 changed=5 unreachable=0 failed=0
pump2 : ok=28 changed=5 unreachable=0 failed=0
pump3 : ok=28 changed=5 unreachable=0 failed=0 Congrats! All goes well. :-)部署grafana的监控主机则需要部署依赖的软件：
#sudo yum -y install fontconfig open-sans-fonts部署tidb：
$ansible-playbook deploy.yml
...
PLAY RECAP ****************
172.16.4.171 : ok=114 changed=61 unreachable=0 failed=0
172.16.4.172 : ok=52 changed=22 unreachable=0 failed=0
172.16.4.173 : ok=96 changed=36 unreachable=0 failed=0
172.16.4.174 : ok=111 changed=43 unreachable=0 failed=0
172.16.4.175 : ok=112 changed=44 unreachable=0 failed=0
localhost : ok=1 changed=0 unreachable=0 failed=0
pump1 : ok=20 changed=9 unreachable=0 failed=0
pump2 : ok=20 changed=9 unreachable=0 failed=0
pump3 : ok=20 changed=9 unreachable=0 failed=0 Congrats! All goes well. :-)
-- 上面的步骤可简单记为：
初始化系统环境，修改内核参数：
$ ansible-playbook bootstrap.yml部署tidb：
$ ansible-playbook deploy.yml
启动tidb：
$ ansible-playbook start.ymltidb集群的关闭：
$ ansible-playbook stop.yml部署Tispark集群需要在所在的主机安装部署JDK8版本：登录TiDB数据库,登录数据库只能使用MySQL5.7的客户端版本，使用MySQL8.0则报错:
# cat /etc/yum.repos.d/mysql-community.repo
[mysql57-community]
name=MySQL 5.7 Community Server
baseurl=http://repo.mysql.com/yum/mysql-5.7-community/el/7/$basearch/
enabled=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-mysql# rpm -qa | grep -i mysql | grep -i 5.7
mysql-community-libs-5.7.24-1.el7.x86_64
mysql-community-devel-5.7.24-1.el7.x86_64
mysql-community-libs-compat-5.7.24-1.el7.x86_64
mysql-community-common-5.7.24-1.el7.x86_64
mysql-community-client-5.7.24-1.el7.x86_64
mysql-community-server-5.7.24-1.el7.x86_64TiDB的监控登录：
http：//172.16.4.171:3000
登录的用户名和密码为admin/admin.CentOS 7 部署TiDB 2.1.4集群相关推荐
- Centos 7 部署 OpenStack_Rocky版高可用集群3-1
Centos 7 部署 OpenStack_Rocky版高可用集群3-1 文章目录 Centos 7 部署 OpenStack_Rocky版高可用集群3-1 1.前期准备 1.1 软硬件准备 1.1. ...
- 项目四 CentOS使用kubeadm部署工具部署测试环境的K8s集群---Kubectl命令使用以及安装dashboard界面
大家好,我是SuieKa.在之前呢有幸学习了马哥教育提供的K8s入门指南以及视频.初来乍到,写一篇关于K8s的介绍以及部署测试环境使用的K8s集群. 树 @·K8s入门简单介绍 一.K8s(Kuber ...
- 部署Ceph分布式高可用集群中篇
前言 如何快速部署Ceph分布式高可用集群 Ceph分布式存储底层实现原理 上文介绍了ceph的原理以及ceph的部署(部署了一个mon角色)本文继续介绍下ceph的部署 部署OSD 查看磁盘使用情况 ...
- kubeadm部署k8s1.9高可用集群--1集群概述
前言 k8s部署的方式多种多样,除去各家云厂商提供的工具,在bare metal中,也有二进制部署和一系列的自动化部署工具(kubeadm,kubespary,rke等).具体二进制部署大家可以参考宋 ...
- 使用二进制包在生产环境部署 Kubernetes v1.13.2 集群
文章目录 使用二进制包在生产环境部署 Kubernetes v1.13.2 集群 一 背景 二 环境及架构图 2.1 软件环境 2.2 服务器规划 2.3 节点或组件功能简介 2.4 Kubernet ...
- CentOS7 使用二进制部署 Kubernetes v1.15.3集群
组件版本 && 集群环境 组件版本: Kubernetes v1.15.3 Etcd v3.3.10 Flanneld v0.11.0 服务器IP 角色 192.168.1.241 m ...
- Kubespray v2.22.1 在线部署 kubernetes v1.26.5 集群
文章目录 1. 介绍 2. 预备条件 3. 配置 hostname 4. yum 5. 下载 kubespray 6. 编写 inventory.ini 7. 配置互信 8. 安装 ansible 9 ...
- 如何在Rancher 2.2 Preview2上部署和管理多K8s集群应用
Rancher 2.2 Preview2于近日全面发布,这一版本包含了许多K8S集群操作的强大特性.本文将详细介绍多集群应用这一特性,让您可以在短时间内更新集群,大大提升工作效率. 近日,全球领先的容 ...
- 需要单机还是集群部署_5000W如何玩转Filecoin市场 部署最符合企业的集群模式
5000W如何玩转Filecoin市场 部署最符合企业的集群模式 对 Filecoin集群的研究,其实一直以来都没有停止过. 如果我们将 Filecoin挖矿比作建高楼大厦.那么,集群就像打地基,地基 ...
最新文章
- NeuIPS | 蛋白设计新方法？基于图的蛋白生成模型
- 终结谷歌AutoML的真正杀手！Saleforce开源TransmogrifAI
- apache2配置和使用
- 学军中学推理社2017届招新试题
- 密织“地网” 南充“试水”智慧安防
- 从客户端检测到有潜在危险的Request.Form值的asp.net代码
- 常见的 Error 和 Exception
- yml php,使用 docker-compose.yml 快速搭建php开发环境
- rstudio 修改代码间距_第一章 R和RStudio
- RHEL下安装配置基于2台服务器的MYSQL集群
- 阿里巴巴牵手上汽集团：成立汽车科技公司
- 2021年2月十大热门报告盘点（附百大热门报告列表及下载链接）
- LeetCode-326. Power of Three
- SQL注入如何进行攻击
- c51语言语句 指令集,MCU
- APP的UI自动化测试框架及平台化探索
- 解决 xshell 上面 zsh 的 home end 无效问题
- Java并发编程实战读书笔记合集
- 慕课版软件质量保证与测试（第四章.课后作业）
- asp.net通用Web2.0仿淘宝脚本验证组件