{"id":13753588,"url":"https://github.com/Neeky/mysqltools","last_synced_at":"2025-05-09T21:35:36.998Z","repository":{"id":38705417,"uuid":"86786470","full_name":"Neeky/mysqltools","owner":"Neeky","description":"一个用于快速构建大规模，高质量，全自动化的 mysql分布式集群环境的工具；包含mysql 安装、备份、监控、高可用、读写分离、优化、巡检、自行化运维","archived":false,"fork":false,"pushed_at":"2023-03-20T03:52:39.000Z","size":517703,"stargazers_count":1064,"open_issues_count":0,"forks_count":465,"subscribers_count":68,"default_branch":"master","last_synced_at":"2024-11-12T18:02:03.545Z","etag":null,"topics":["ansible","atlas","dble","dev-ops","innodb","innodb-cluster","master-slave","mgr","mycat","mysql","mysql-backup","mysql-dba","mysql-group-replicaion","mysql-maintance","mysql-monitor","mysql-tools","python","zabbix"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Neeky.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2017-03-31T06:37:22.000Z","updated_at":"2024-11-11T08:21:12.000Z","dependencies_parsed_at":"2024-10-29T23:25:18.820Z","dependency_job_id":null,"html_url":"https://github.com/Neeky/mysqltools","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Neeky%2Fmysqltools","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Neeky%2Fmysqltools/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Neeky%2Fmysqltools/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Neeky%2Fmysqltools/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Neeky","download_url":"https://codeload.github.com/Neeky/mysqltools/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224884615,"owners_count":17386121,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ansible","atlas","dble","dev-ops","innodb","innodb-cluster","master-slave","mgr","mycat","mysql","mysql-backup","mysql-dba","mysql-group-replicaion","mysql-maintance","mysql-monitor","mysql-tools","python","zabbix"],"created_at":"2024-08-03T09:01:25.228Z","updated_at":"2025-05-09T21:35:36.938Z","avatar_url":"https://github.com/Neeky.png","language":"Python","readme":"\n# mysqltools 权威指南\n\n官方微信公众平台 \n\n![官方微信公众平台](docs/mp-wechat.jpg)\n\n---\n\n## 文档结构\n- [概要](#概要)\n - [质量](#质量)\n - [效率](#效率)\n - [经济](#经济)\n - [技术介绍](#技术介绍)\n- [安装mysqltools](#安装mysqltools)\n - [安装前的准备](#安装前的准备)\n - [下载并解压](#下载并解压)\n - [安装Python](#安装Python)\n - [安装ansible](#安装ansible)\n - [配置ansible和mysqltools](#配置ansible和mysqltools)\n- [mysqltools快速开始](#mysqltools快速开始)\n - [mysqltools目录介绍](#mysqltools目录介绍)\n - [自动化mysql单实例安装](#自动化mysql单实例安装)\n - [进入mysql功能目录](#进入mysql功能目录)\n - [指定安装的目标主机](#指定安装的目标主机)\n - [执行自动化安装](#执行自动化安装)\n - [谈谈mysqltools如何实现高效](#谈谈mysqltools如何实现高效)\n - [谈谈mysqltools如何实现高质量](#谈谈mysqltools如何实现高质量)\n- [赞助/交流/技术支持](#联系我)\n- [mysql原生环境安装](#mysql原生环境安装)\n - [mysql单机](#mysql单机)\n - [mysql主从复制](#mysql主从复制)\n - [mysql多源复制](#mysql多源复制)\n - [mysql组复制](#mysql组复制)\n - [mysql小版本升级](#mysql小版本升级)\n- [读写分离](#读写分离)\n - [mycat读写分离](#mycat读写分离)\n- [高可用](#高可用)\n - [mha](#mha)\n - [环境规划](#环境规划)\n - [配置mha的一些前置备件](#配置mha的一些前置备件)\n - [配置ansible的hosts文件](#配置ansible的hosts文件)\n - [配置mysqltools中mha相关的配置项](#配置mysqltools中mha相关的配置项)\n - [配置mha的相关信息](#配置mha的相关信息)\n - [准备安装配置mha](#准备安装配置mha)\n - [安装配置mha](#安装配置mha)\n - [验证是否成功完成](#验证是否成功完成)\n- [备份](#备份)\n - [mysqltools备份相关的实现细节](#mysqltools备份相关的实现细节)\n - [实施备份计划的前期准备](#实施备份计划的前期准备)\n - [配置备份](#配置备份)\n - [检查配置是否成功的几个点](#检查配置是否成功的几个点)\n - [注意事项](#注意事项)\n- [巡检](#巡检)\n - [开发中]\n- [监控](#监控)\n - [目前已经实现的监控项](#目前已经实现的监控项)\n - [监控项的人肉使用方法](#监控项的人肉使用方法)\n - [zabbix监控环境介绍](#zabbix监控环境介绍)\n - [zabbix监控环境安装规划](#zabbix监控环境安装规划)\n - [安装用于保存监控数据的mysql数据库](#安装用于保存监控数据的mysql数据库)\n - [改配置文件中zabbix_server_ip这个配置项](#改配置文件中zabbix_server_ip这个配置项)\n - [安装httpd](#安装httpd)\n - [安装zabbix服务端](#安装zabbix服务端)\n - [安装zabbix客户端](#安装zabbix客户端)\n - [通过mysqltools中给出的模板来监控mysql](#通过mysqltools中给出的模板来监控mysql)\n - [mysqltools中定义的MySQL的监控模板](#mysqltools中定义的MySQL的监控模板)\n- [lnmp](#lnmp)\n - [安装mysql单机](#安装mysql单机)\n - [安装python](#安装python)\n - [安装nginx](#安装nginx)\n - [查看效果](#查看效果)\n\n---\n\n如果你使用 mysql-8.0 那么 https://github.com/Neeky/dbm-agent 在功能上会更加强大\n\n---\n\n\n## 概要\n\n 总的来说mysqltools源自于工作、一个dba的日常大概包括 数据库安装，读写分离、高可用、负载均衡等环境的配置，\n 数据库备份策略拟定与实施，数据库相关的监控，数据库优化，故障分析，也有可能参与到数据库建模，SQL的编写。\n \n 这样我们就面临两个问题 1、**质量** 质量表现在解决问题的深度(类似问题还会再出现吗？) 2、**效率** 效率表现在你单位时间内解决问题的数量(安装一百个库的用时是一个库的100倍 吗？)； 通常这两个目标并不是互斥的，也就是说我们可以两个都做到。\n\n ---\n\n1. ## 质量\n **KFC** vs **学校后街的蛋炒饭**\n\n KFC根据既定的流程生产每一个汉堡，假设这个流程下公众对汉堡给出的评分是80分，那么不管哪个KFC的店它生产出来的汉堡都稳定在80分；一段时间后它发现这个流程中可以改进的项，把汉堡的质量提升到81分，那么它就能做到所有的店里的汉堡都能打81分。\n\n 学校后街的蛋炒饭 好不好吃这个事难说；因为好多事都影响到它，有可能老板今天心情不好，也有可能是今天客人太多他比较急，这些都会影响到炒饭的质量。有一次我要买两盒，由于去的比较晚，老板只有一个鸡蛋了，你没有猜错！ 他就只放了一个蛋，按常理是要一盒一个的。\n\n 表面上看**KFC** 流程化生产的好处在于它的东西质量有保障，**最要命的是KFC只做加法，它可以不断提升自己，学校后街的蛋炒饭上周一，做的好吃，我们没办法确认那是不是超水平发挥，蛋炒饭质量的方差太大了。**\n\n \n **对于DBA来说可以专门针对自己的日常工作开发一款工具，这样做的好处有 1:)由于工具已经把流程固定下来了所以“产出的质量”有保证 2:)随着自己技术的进步自己工作的输出也可以稳步提高。** **这样我们在质量这个目标上就只做加法了。**\n \n ---\n\n2. ## 效率\n \n **流水线** vs **手工作坊**\n\n 流水线相对于手工作坊，那是生产力的巨大提升。我为什么要说这个？因为在MySQL的使用中可能会遇到一些场景，比如说“分库分表”，“高可用+读写分离”；特别是前者通常就是一个MyCAT后面有好几十个分片，上百个MySQL实例(通常它们会为一个分片做一主两从并加上高可用)，装100+个MySQL今晚加班不？ 配100+个主从今晚加班不？ 不要忘记还要给它们加\n 高可用呢？ 好吧这只是测试环境生产环境和测试环境是1:1的，那接下来几天加班不？ 对于生产通常还要加备份，监控那接下来几天加班不？\n\n **DBA的工具不应该只是能输出高质量的交付物，更应该要解放生产力----有批量管理的能力。**\n\n ---\n\n3. ## 经济\n **mysqltools 是开源的\u0026免费的\u0026高质量的MySQL数据管理工具**\n\n ---\n\n4. ## 技术介绍\n 1、mysqltools的高质量源自于**蒋乐兴**也就是我写出来的高质量的playbook\n\n 2、mysqltools的高效源自于**ansible**这个批量管理工具\n\n 3、1 , 2 基本上解决了原生MySQL的环境(**单机**、**master --\u003eslave**、**mysql-group-replication** 、**multi-source-replication**)的安装部署\n\n 4、mysqltools在为MySQL做**高可用**时采用的是**MHA**这个方案，**读写分离用**的是**MyCAT**\n\n 5、mysqltools在**备份**时支持到了 **xtrabackup**,**mysql enterprise backup**,**mysqldump**\n\n 6、mysqltools在**监控**MySQL时用的是**zabbix**\n\n 7、其它开源工具无法满足的功能通过自己编写程序实现。由于这个程序的许多功能是通用的，所以我把它单独了出来成为一个新的开源项目，这样它就与mysqltools解耦了。详见\u003ca href=\"https://github.com/Neeky/mysqltools-python\"\u003emysqltools-python \u003c/a\u003e\n\n mysqltools的定位是一个**集中化管理平台**你只要**在一台主机上安装好mysqltools就可以了**，其它主机作为被管理都的角色。由于mysqltools是基于**Python-3.x**开发\n 出来的所以你的**主控机**上应该事先安装好python-3.x、还要安装上ansible。 好在mysqltools已经包含有所有**Python**和**ansible**所有的包。\n\n ---\n\n## 安装mysqltools\n\n 假设我们有如下一套环境、把**172.16.192.131**这台主机作为主控机.\n \n **角色** | **ip地址** | **系统版本** |\n -----------:|:-------------------|--------------|\n 主控机 | 172.16.192.131 |centos-7.4 |\n 被控机 | 172.16.192.132 |centos-7.4 |\n ... | ... |centos-7.x |\n\n ---\n\n 1. ### 安装前的准备\n 1): **你的主控机上要配置有yum**、因为mysqltools要源码编译安装Python-3.6.2、这就涉及gcc ... 等依赖\n \n 2): **有主控机的root账号(安装软件时会用到)**\n \n 3): **被控机上也要配置好yum**\n\n ---\n\n 2. ### 下载并解压\n **mysqltools是开源在github上的、下载地址如下：https://github.com/Neeky/mysqltools/archive/master.zip** \n \n linux可以直接执行如下命令完成下载并解压到/usr/local/\n \n ```bash\n cd /tmp/\n wget https://github.com/Neeky/mysqltools/archive/master.zip \u0026\n \n ll -h /tmp/ \n -rwxr-xr-x. 1 root root 194M 3月 23 11:52 master.zip\n \n unzip master.zip\n \n mv mysqltools-master /usr/local/mysqltools\n ```\n \n ---\n\n 3. ### 安装Python\n **mysqltools包含了Python的自动化安装脚本、前提是yum已经可用**\n \n ```bash\n cd /usr/local/mysqltools/deploy/packages/python/\n bash install.sh\n \n ```\n 安装成功后的最后几行输出如下：\n ```bash\n Collecting setuptools\n Collecting pip\n Installing collected packages: setuptools, pip\n Successfully installed pip-9.0.1 setuptools-28.8.0\n ```\n \n 检查python3是否安装成功\n ```bash\n source /etc/profile\n \n python3 --version\n Python 3.6.2\n ```\n \n ---\n \n 4. ### 安装ansible\n **ansible和它相关的依赖我都打包到mysqltools中了、也和上面安装python一样一行命令就行**\n \n ```bash\n source /etc/profile\n cd /usr/local/mysqltools/deploy/packages/ansible\n bash install.sh \n \n ```\n 安装成功后可以看到如下输出\n ```bash\n Using /usr/local/python-3.6.2/lib/python3.6/site-packages\n Finished processing dependencies for ansible==2.4.0.0\n ```\n \n ---\n\n 5. ### 配置ansible和mysqltools\n 1): **增加ansible的配置文件**\n\n ```bash\n # 增加ansible的配置文件\n mkdir -p /etc/ansible\n touch /etc/ansible/hosts\n ```\n /etc/ansible/hosts文件如下：\n ```\n host_131 ansible_user=root ansible_host=172.16.192.131\n host_132 ansible_user=root ansible_host=172.16.192.132\n ```\n\n ---\n\n 2): **配置主控机与被控机之间的ssh信任**\n\n ```bash\n ssh-keygen\n ssh-copy-id root@172.16.192.131\n ssh-copy-id root@172.16.192.132\n ```\n 命令输出大致如下：\n ```bash\n ssh-keygen # 一直回车就能生成钥匙对了\n Generating public/private rsa key pair.\n Enter file in which to save the key (/root/.ssh/id_rsa): \n Enter passphrase (empty for no passphrase): \n Enter same passphrase again: \n Your identification has been saved in /root/.ssh/id_rsa.\n Your public key has been saved in /root/.ssh/id_rsa.pub.\n The key fingerprint is:\n SHA256:D9kR6/ehu5O99p/LRJlZWNqwZ0tzU4+jvPegq7j/Pq8 root@studio2018\n The keys randomart image is:\n +---[RSA 2048]----+\n | . . o|\n | o Oo|\n | o *+B|\n | + o ..+X|\n | S o + .* |\n | o . +.. |\n | . oo+. |\n | . ++=o.|\n | ooo+EOo**|\n +----[SHA256]-----+\n \n \n ssh-copy-id root@172.16.192.131 # 回答yes、然后输入目标主机的root密码\n /usr/bin/ssh-copy-id: INFO: Source of key(s) to be installed: \"/root/.ssh/id_rsa.pub\"\n The authenticity of host '172.16.192.131 (172.16.192.131)' can't be established.\n ECDSA key fingerprint is SHA256:qdoqi3B2aqO3ssOIphwOiWLywSlAoflX2YH+LCG7T/E.\n ECDSA key fingerprint is MD5:8f:78:6e:20:ab:d0:2a:6b:c0:1a:e5:09:ac:82:7d:04.\n Are you sure you want to continue connecting (yes/no)? \n root@172.16.192.131's password: \n \n Number of key(s) added: 1\n \n Now try logging into the machine, with: \"ssh 'root@172.16.192.131'\"\n and check to make sure that only the key(s) you wanted were added.\n \n .... ....\n \n ```\n \n ---\n\n 3): **测试ansible是否配置成功**\n ```bash\n ansible -m ping host_132\n \n host_132 | SUCCESS =\u003e {\n \"changed\": false,\n \"failed\": false,\n \"ping\": \"pong\"\n }\n ```\n \n ---\n\n 4): **配置mysqltools**\n \n mysqltools的配置文件是**mysqltools/config.yaml** 它是一个yaml格式的文件；配置项中最基本的有**mtls_base_dir、mysql_packages_dir、mysql_package**\n \n 1、**mtls_base_dir用于配置mysqltools的安装路径**：在[下载并解压](#下载并解压)这个步骤中我们把mysqltools解压到了/usr/local/、所以mtls_base_dir的值就应该等于\"/usr/local/mysqltools/\"\n\n ---\n \n 2、**mysql_packages_dir用于配置MySQL二进制安装包保存的位置**：MySQL的安装包有600+MB、出于体量的原因mysqltools并没有直接打包MySQL的二进制安装包、而是留有mysql_packages_dir这个配置项，mysqltools会从这个目录中去找MySQL的二进制安装包。\n\n ---\n \n 3、**mysql_package用于配置MySQL安装包的名字**、有这个变量的因为是为了，可以做到有多个不同的MySQL的版本共存、默认值为 mysql-5.7.22-linux-glibc2.12-x86_64.tar.gz\n \n ---\n\n config.yaml的关键内容大致如下：\n ```yaml\n mtls_base_dir: /usr/local/mysqltools/\n mysql_packages_dir: /usr/local/src/mysql/\n mysql_package: mysql-5.7.21-linux-glibc2.12-x86_64.tar.gz\n ```\n 注意：在mysqltools中所有的目录都是要以'/'号结尾的\n\n ---\n \n **如果你正确的完成了mysqltools相关的配置那么config.yaml看起来就应该是这样的**\n ```\n ---\n #----------------------------------mysqltools全局配置文件---------------------\n # section 1 #mysqltools所在的目录\n mtls_base_dir: /usr/local/mysqltools/\n # #mysqltool自带的各类软件的安装文件所在路径(相对路径)\n mtls_packages: deploy/packages/\n # #mysqltool自带的python脚本、下发到被控主机时所保存的路径\n mtls_client_base_dir: /usr/local/\n \n \n # section 2 #mysqltools自带的各类软件安装文件 的全名、设置这些变量的作用是方便版本共存、mysql不在这里设置是因为\n # #mysql的安装包太大了，mysqltools并没有把它打包进来 \n mtls_apr: apr-1.6.2.tar.gz\n mtls_apr_util: apr-util-1.6.0.tar.gz\n mtls_httpd: httpd-2.4.28.tar.gz\n mtls_php: php-5.6.31.tar.gz\n #mtls_zabbix: zabbix-3.4.3.tar.gz\n mtls_zabbix: zabbix-4.0.0.tar.gz\n mtls_python: python-3.6.2.tar.xz\n mtls_mysql_connector_python: mysql-connector-python-2.1.5.tar.gz\n mtls_mycat: mycat-server-1.6.5-linux.tar.gz\n mtls_mha_node: mhanode.tar.gz\n mtls_mha_manager: mhamanager.tar.gz\n mtls_git: git-2.9.5.tar.gz\n mtls_nginx: nginx-1.13.7.tar.gz\n mtls_sysbench: sysbench-1.1.0.tar.gz\n mtls_meb: meb-4.1.0-linux-glibc2.5-x86-64bit.tar.gz\n mtls_xtrb: percona-xtrabackup-2.4.9-Linux-x86_64.tar.gz\n mtls_mysqlclient: mysqlclient-1.3.12.tar.gz\n mtls_pytz: pytz-2018.4.tar.gz\n mtls_django: django-2.0.4.tar.gz\n mtls_uwsgi: uwsgi-2.0.17.tar.gz\n \n #mysql与php-5.6.x 是否要同时安装在一台主机上、如果是就要把这个设置成yes、以为php导出mysqclient_r.so文件\n mtls_with_php: 1\n #通过ansible在被控机上安装python-3.x的时候，是否自动安装好mysql-connector-python\n mtls_with_mysql_conntor_python: 1\n #是否给mysql用户加密码\n mtls_make_mysql_secure: 1\n #\n mtls_with_mysql_group_replication: 0\n #----------------------------------mysqltools全局配置文件---------------------\n \n \n ####\n #### mysql 相关的配置\n ####\n #mysql 安装包所在的目录\n mysql_packages_dir: /usr/local/src/mysql/\n #mysql 安装包的名字\n #mysql_package: mysql-5.6.40-linux-glibc2.12-x86_64.tar.gz\n #mysql_package: mysql-5.7.23-linux-glibc2.12-x86_64.tar.gz\n mysql_package: mysql-8.0.13-linux-glibc2.12-x86_64.tar.xz\n #linux 系统级别mysql用户相关信息\n mysql_user: mysql\n mysql_group: mysql\n mysql_user_uid: 3306\n mysql_user_gid: 3306\n #mysql 安装目录\n mysql_base_dir: /usr/local/mysql/\n #mysql 真正的datadir就会是mysql_data_dir_base+mysql_port\n mysql_data_dir_base: /database/mysql/data/\n mysql_port: 3306\n mysql_root_password: mtls0352\n mysql_zabbix_password: mtls\n mysql_rple_user: repl\n mysql_rple_password: repl0352\n mysql_mha_user: mha\n mysql_mha_password: mtls0352\n mysql_app_user: appuser\n mysql_app_password: mtls0352\n mysql_monitor_user: monitor\n mysql_monitor_password: monitor0352\n mysql_backup_user: backup\n mysql_backup_password: DX3906\n #mysql 配置文件模版\n mysql_binlog_format: row\n mysql_innodb_log_files_in_group: 16\n mysql_innodb_log_file_size: 256M\n mysql_innodb_log_buffer_size: 64M\n mysql_innodb_open_files: 65535\n mysql_max_connections: 1000\n mysql_thread_cache_size: 256\n mysql_sync_binlog: 1\n mysql_binlog_cache_size: 64K\n mysql_innodb_online_alter_log_max_size: 128M\n mysql_performance_schema: 'on'\n use_write_set: 1\n \n #mysql \n \n ####\n #### zabbix 相关的配置\n #####\n zabbix_server_ip: 172.16.192.131\n \n \n \n ```\n 为了你能更加方便的使用mysqltools我提供了份linux上的标准配置文件**mysqltools/config.yaml-for-linux** 使用时只要把它重命名成**cofnig.yaml**就行了\n \n ---\n\n 5): **下载MySQL**\n \n 根据上面的配置可以知道MySQL的安装包要保存到**/usr/local/src/mysql/**目录下、包的版本为mysql-5.7.22-linux-glibc2.12-x86_64.tar.gz\n \n 下载地址如下：https://dev.mysql.com/get/Downloads/MySQL-5.7/mysql-5.7.22-linux-glibc2.12-x86_64.tar.gz\n \n ```bash\n cd /usr/local/src/mysql/\n wget https://dev.mysql.com/get/Downloads/MySQL-5.7/mysql-5.7.21-linux-glibc2.12-x86_64.tar.gz\n \n \n ```\n\n\n---\n\n## mysqltools快速开始\n **mysqltools的目录组织是根据它的功能来定的**\n 1. ### mysqltools目录介绍\n\n ```\n cd /usr/local/mysqltools/\n tree ./\n ├── README.md // 文档\n ├── config.yaml // 为Mac生成的配置文件(你可能用不着我的开发机是Mac)\n ├── config.yaml-for-linux // 专为linux生成的个配置文件\n ├── deploy // 自动化安装与配置的所有实现\n ├── docs // 文件资源\n ├── trashCan // 垃圾文件\n └── tuning // DBA常用的SQL脚本有助于快速解决问题\n ```\n \n ---\n\n 2. ### 自动化mysql单实例安装\n **mysqltools/deploy/ansible/** 目录下的每一个子目录都对应一类软件环境的自动化安装、我们这次是要安装MySQL所以应该进入到mysql子目录\n\n ---\n \n 1. #### 进入mysql功能目录\n ```\n cd /usr/local/mysqltools/deploy/ansible/mysql/\n ll \n ```\n 输出如下\n ```\n 总用量 24\n drwxr-xr-x. 2 root root 4096 3月 19 15:01 common # 通用组件\n -rw-r--r--. 1 root root 836 3月 19 15:01 install_group_replication.yaml # 自动化安装mysql group replication\n -rw-r--r--. 1 root root 889 3月 19 15:01 install_master_slaves.yaml # 自动化安装mysql 主从复制环境\n -rw-r--r--. 1 root root 924 3月 19 15:01 install_multi_source_replication.yaml # 自动化安装mysql 多源复制\n -rw-r--r--. 1 root root 772 3月 26 13:20 install_single_mysql.yaml # 自动化安装mysql 单机实例\n drwxr-xr-x. 3 root root 203 3月 19 15:01 template # 通用模板 \n -rw-r--r--. 1 root root 892 3月 19 15:01 upgrad_single_mysql.yaml # 自动化升级MySQL(不推荐)\n drwxr-xr-x. 2 root root 99 3月 19 15:01 vars # 自定义变量(重要!)\n -rw-r--r-- 1 root root 99 3月 19 15:01 uninstall.yaml # 专门用来删库的(无力回天的那种删除)\n ```\n 由于/usr/local/mysqltools/deploy/ansible/下的每一个子目录都实现某一类功能，于是我们约定/usr/local/mysqltools/deploy/ansible/下的子目录叫**功能目录**, 功能目录下的会包含若干.yaml文件，每一个文件都实现了特定的功能，如install_single_mysql.yaml实现了自动化安装MySQL的功能。\n \n ---\n \n 2. #### 指定安装的目标主机\n\n install_single_mysql.yaml的前几行大致如下\n\n ```yaml\n ---\n - hosts: cstudio\n remote_user: root\n become_user: yes\n vars_files:\n - ../../../config.yaml\n tasks:\n - name: create user and config file\n import_tasks: common/create_user_and_config_file.yaml\n \n ```\n 其中的**- hosts: cstudio** 这一行中的cstudio 就是用来指定目标主机或主机组的，也就是说它指明了install_single_mysql.yaml将在哪里安装MySQL单实例\n\n 假设我们要在host_132这台主机上安装单机的MySQL所以我们要把cstudio改成host_132、注意这里的host_132引用的是/etc/ansible/hosts文件、修改后的install_single_mysql.yaml文件的前几行大致如下\n\n ```yaml\n ---\n - hosts: host_132\n remote_user: root\n become_user: yes\n vars_files:\n - ../../../config.yaml\n tasks:\n - name: create user and config file\n import_tasks: common/create_user_and_config_file.yaml\n ```\n \n ---\n \n 3. #### 执行自动化安装\n 正如上文所说的mysqltools是借助ansible来完成自动化的、那么我们调用ansible来完成自动化安装MySQL\n ```bash\n ansible-playbook install_single_mysql.yaml \n \n ```\n 输出如下\n\n ```\n \n PLAY [host_132] *********************************************************************************************\n \n TASK [Gathering Facts] **************************************************************************************\n ok: [host_132]\n \n TASK [create mysql user] ************************************************************************************\n changed: [host_132]\n \n TASK [create and config /etc/my.cnf] ************************************************************************\n changed: [host_132]\n \n TASK [install libaio-devel] *********************************************************************************\n ok: [host_132]\n \n TASK [install numactl-devel] ********************************************************************************\n ok: [host_132]\n \n TASK [transfer mysql install package to remote host and unarchive to /usr/local/] ***************************\n changed: [host_132]\n \n TASK [change owner to mysql user] ***************************************************************************\n changed: [host_132]\n \n TASK [make link /usr/local/mysql-xx.yy.zz to /usr/local/mysql] **********************************************\n changed: [host_132]\n \n TASK [export mysql share object (*.os)] *********************************************************************\n ok: [host_132]\n \n TASK [load share object] ************************************************************************************\n changed: [host_132]\n \n TASK [export path env variable] *****************************************************************************\n ok: [host_132]\n \n TASK [export path env to /root/.bashrc] *********************************************************************\n ok: [host_132]\n \n TASK [make link /usr/local/mysql-xx.yy.zz to /usr/local/mysql] **********************************************\n ok: [host_132]\n \n TASK [create libmysqlclient_r.so file for php-5.6] **********************************************************\n changed: [host_132]\n \n TASK [create datadir] ***************************************************************************************\n changed: [host_132]\n \n TASK [initialize-insecure] **********************************************************************************\n changed: [host_132]\n \n TASK [create systemd config file] ***************************************************************************\n changed: [host_132]\n \n TASK [start mysql(sytemctl)] ********************************************************************************\n changed: [host_132]\n \n TASK [config mysql.service start up on boot] ****************************************************************\n ok: [host_132]\n \n TASK [config sysv start script] *****************************************************************************\n skipping: [host_132]\n \n TASK [start mysql(service)] *********************************************************************************\n skipping: [host_132]\n \n TASK [config mysql.service start up on boot] ****************************************************************\n skipping: [host_132]\n \n TASK [transfer sleep script to /tmp/] ***********************************************************************\n changed: [host_132]\n \n TASK [sleep 15 seconds] *************************************************************************************\n changed: [host_132]\n \n TASK [remove /tmp/sleep_15.sh] ******************************************************************************\n changed: [host_132]\n \n TASK [transfer sql statement to remonte] ********************************************************************\n changed: [host_132]\n \n TASK [make mysql secure] ************************************************************************************\n changed: [host_132]\n \n TASK [remove temp file /tmp/make_mysql_secure.sql] **********************************************************\n changed: [host_132]\n \n PLAY RECAP **************************************************************************************************\n host_132 : ok=25 changed=17 unreachable=0 failed=0 \n ```\n\n 测试MySQL是否安装成功\n ```bash\n mysql -uroot -pmtls0352 \n ```\n 输出如下\n ``` \n mysql: [Warning] Using a password on the command line interface can be insecure.\n Welcome to the MySQL monitor. Commands end with ; or \\g.\n Your MySQL connection id is 3\n Server version: 5.7.21-log MySQL Community Server (GPL)\n \n Copyright (c) 2000, 2018, Oracle and/or its affiliates. All rights reserved.\n \n Oracle is a registered trademark of Oracle Corporation and/or its\n affiliates. Other names may be trademarks of their respective\n owners.\n \n Type 'help;' or '\\h' for help. Type '\\c' to clear the current input statement.\n \n mysql\u003e \n ```\n \n ---\n\n 3. ### 谈谈mysqltools如何实现高效 \n 从上面的[自动化mysql单实例安装](#自动化mysql单实例安装)来看自动化安装的过程中我们只执行了**3个命令**\n ```bash\n cd \n vim\n ansible-playbook\n ```\n 其中 1、cd 用于进行mysql功能目录 2、vim用于指定playbook的目标主机或主机组 3、执行playbook\n\n 高效的关键有两点：\n\n **1):自动化安装的方式相对于人肉来说要快很多**\n\n **2):支持批量操作也就是说可以同时在多台主机上执行、只要在hosts: 变量的值是一个主机组就行了**\n\n ---\n\n 4. ### 谈谈mysqltools如何实现高质量\n\n **1、mysqltools是流程化的，如上面的安装MySQL就包含20+的小步骤，尽量做到面面俱到** \n\n ---\n\n **2、mysqltools尽可能的在各个小步骤中都保持高的质量、比如/etc/my.cnf各个参数的配置都会根据主机当前的cpu \u0026 内存进行配置**\n\n ---\n\n 以下是一个host_132的/etc/my.cnf **这些都是由mysqltools针对单机环境动态生成的**\n\n ```cnf\n [mysql]\n auto-rehash\n \n \n [mysqld]\n ####: for global\n user =mysql # mysql\n basedir =/usr/local/mysql/ # /usr/local/mysql/\n datadir =/database/mysql/data/3306 # /usr/local/mysql/data\n server_id =653 # 0\n port =3306 # 3306\n character_set_server =utf8 # latin1\n explicit_defaults_for_timestamp =off # off\n log_timestamps =system # utc\n socket =/tmp/mysql.sock # /tmp/mysql.sock\n read_only =0 # off\n skip_name_resolve =1 # 0\n auto_increment_increment =1 # 1\n auto_increment_offset =1 # 1\n lower_case_table_names =1 # 0\n secure_file_priv = # null\n open_files_limit =65536 # 1024\n max_connections =256 # 151\n thread_cache_size =128 # 9\n table_open_cache =4096 # 2000\n table_definition_cache =2000 # 1400\n table_open_cache_instances =32 # 16\n \n ####: for binlog\n binlog_format =row # row\n log_bin =mysql-bin # off\n binlog_rows_query_log_events =on # off\n log_slave_updates =on # off\n expire_logs_days =7 # 0\n binlog_cache_size =65536 # 65536(64k)\n binlog_checksum =none # CRC32\n sync_binlog =1 # 1\n slave-preserve-commit-order =ON # \n \n ####: for error-log\n log_error =err.log # /usr/local/mysql/data/localhost.localdomain.err\n \n general_log =off # off\n general_log_file =general.log # hostname.log\n \n ####: for slow query log\n slow_query_log =on # off\n slow_query_log_file =slow.log # hostname.log\n log_queries_not_using_indexes =on # off\n long_query_time =2.000000 # 10.000000\n \n ####: for gtid\n gtid_executed_compression_period =1000 # 1000\n gtid_mode =on # off\n enforce_gtid_consistency =on # off\n \n \n ####: for replication\n skip_slave_start =0 # \n master_info_repository =table # file\n relay_log_info_repository =table # file\n slave_parallel_type =logical_clock # database | LOGICAL_CLOCK\n slave_parallel_workers =4 # 0\n rpl_semi_sync_master_enabled =1 # 0\n rpl_semi_sync_slave_enabled =1 # 0\n rpl_semi_sync_master_timeout =1000 # 1000(1 second)\n plugin_load_add =semisync_master.so #\n plugin_load_add =semisync_slave.so #\n binlog_group_commit_sync_delay =500 # 500(0.05%秒)、默认值0\n binlog_group_commit_sync_no_delay_count = 13 # 0\n \n \n ####: for innodb\n default_storage_engine =innodb # innodb\n default_tmp_storage_engine =innodb # innodb\n innodb_data_file_path =ibdata1:64M:autoextend # ibdata1:12M:autoextend\n innodb_temp_data_file_path =ibtmp1:12M:autoextend # ibtmp1:12M:autoextend\n innodb_buffer_pool_filename =ib_buffer_pool # ib_buffer_pool\n innodb_log_group_home_dir =./ # ./\n innodb_log_files_in_group =8 # 2\n innodb_log_file_size =128M # 50331648(48M)\n innodb_file_per_table =on # on\n innodb_online_alter_log_max_size =128M # 134217728(128M)\n innodb_open_files =65535 # 2000\n innodb_page_size =16k # 16384(16k)\n innodb_thread_concurrency =0 # 0\n innodb_read_io_threads =4 # 4\n innodb_write_io_threads =4 # 4\n innodb_purge_threads =4 # 4\n innodb_print_all_deadlocks =on # off\n innodb_deadlock_detect =on # on\n innodb_lock_wait_timeout =50 # 50\n innodb_spin_wait_delay =6 # 6\n innodb_autoinc_lock_mode =2 # 1\n innodb_io_capacity =200 # 200\n innodb_io_capacity_max =2000 # 2000\n #--------Persistent Optimizer Statistics\n innodb_stats_auto_recalc =on # on\n innodb_stats_persistent =on # on\n innodb_stats_persistent_sample_pages =20 # 20\n innodb_buffer_pool_instances =1\n innodb_adaptive_hash_index =on # on\n innodb_change_buffering =all # all\n innodb_change_buffer_max_size =25 # 25\n innodb_flush_neighbors =1 # 1\n #innodb_flush_method = # \n innodb_doublewrite =on # on\n innodb_log_buffer_size =128M # 16777216(16M)\n innodb_flush_log_at_timeout =1 # 1\n innodb_flush_log_at_trx_commit =1 # 1\n innodb_buffer_pool_size =1152M # 134217728(128M)\n autocommit =1 # 1\n #--------innodb scan resistant\n innodb_old_blocks_pct =37 # 37\n innodb_old_blocks_time =1000 # 1000\n #--------innodb read ahead\n innodb_read_ahead_threshold =56 # 56 (0..64)\n innodb_random_read_ahead =OFF # OFF\n #--------innodb buffer pool state\n innodb_buffer_pool_dump_pct =25 # 25 \n innodb_buffer_pool_dump_at_shutdown =ON # ON\n innodb_buffer_pool_load_at_startup =ON # ON\n \n \n \n \n #### for performance_schema\n performance_schema =on # on\n performance_schema_consumer_global_instrumentation =on # on\n performance_schema_consumer_thread_instrumentation =on # on\n performance_schema_consumer_events_stages_current =on # off\n performance_schema_consumer_events_stages_history =on # off\n performance_schema_consumer_events_stages_history_long =off # off\n performance_schema_consumer_statements_digest =on # on\n performance_schema_consumer_events_statements_current =on # on\n performance_schema_consumer_events_statements_history =on # on\n performance_schema_consumer_events_statements_history_long =off # off\n performance_schema_consumer_events_waits_current =on # off\n performance_schema_consumer_events_waits_history =on # off\n performance_schema_consumer_events_waits_history_long =off # off\n performance-schema-instrument ='memory/%=COUNTED'\n ```\n 由上面的内容可以看到mysqltools生成的/etc/my.cnf在各个方面都是比较全面的(配置是在一个2core 2G的vm虚拟机上复制出来的)\n \n ---\n\n## 联系我\n\n **赞助/交流/技术支持**\n\n 如果我分享的知识对你有用、可以打赏任意金额；这些也是我花了许多业余时间才写出来的，也许可以请我喝一杯咖啡。如果你需要一名数据库建模、运维方面的专家也可也联系我、当然技术方面的交流也是欢迎的。\n\n \u003cimg src=\"./docs/imgs/jianglexing_alipay.JPG\" width=\"240px\" /\u003e\n \u003cimg src=\"./docs/imgs/jianglexing_donate_wechart.jpg\" width=\"240px\" /\u003e\n \u003cimg src=\"./docs/imgs/jiangleixng_wechart.jpg\" width=\"240px\" /\u003e\n \n ---\n\n## mysql原生环境安装\n **原生环境安装主要包含 1、MySQL单机 2、主从复制 3、多源复制 4、组复制(mysql-group-replication)**\n 如果这是你第一次使用mysqltools可以从[mysqltools快速开始](#mysqltools快速开始)章节开启你的mysqltools之旅\n\n ---\n\n 1. ### mysql单机\n **1):进入mysql功能目录**\n ```bash\n cd /usr/local/mysqltools/deploy/ansible\n \n cd mysql #进入mysql功能目录\n ```\n\n ---\n\n **2):指定install_single_mysql.yaml中的目标主机**\n\n 假设我要在host_132主机上安装mysql那么install_single_mysql.yaml文件中的hosts变量应该设置为host_132\n ```yaml\n ---\n - hosts: host_132\n ```\n yaml格式对空间是非常敏感的、注意“:”后面是有个空格的\n\n ---\n\n **3):执行自动化安装**\n ```bash\n ansible-playbook install_single_mysql.yaml\n ```\n 输出省略 ... ...\n\n ---\n\n **4):验证mysql是否成功安装**\n ```bash\n mysql -uroot -pmtls0352\n ```\n 输出如下\n ```\n mysql: [Warning] Using a password on the command line interface can be insecure.\n Welcome to the MySQL monitor. Commands end with ; or \\g.\n Your MySQL connection id is 2\n Server version: 5.7.21-log MySQL Community Server (GPL)\n \n Copyright (c) 2000, 2018, Oracle and/or its affiliates. All rights reserved.\n \n Oracle is a registered trademark of Oracle Corporation and/or its\n affiliates. Other names may be trademarks of their respective\n owners.\n \n Type 'help;' or '\\h' for help. Type '\\c' to clear the current input statement.\n \n mysql\u003e\n ```\n root用户的密码是在mysqltools/config.yaml文件中的mysql_root_password配置项指定的、你可以根据你的要求定制\n\n ---\n\n 2. ### mysql主从复制 \n 假设我们要在**10.186.19.15，10.186.19.16，10.186.19.17**三台主机上建设一个一主两从的主从复制环境，其中10.186.19.15为主库\n\n ---\n\n **1):增加主机信息到/etc/ansible/hosts**\n\n 向/etc/ansible/hosts文件中增加如下内容\n ```\n [repl]\n replmaster15 ansible_host=10.186.19.15\n replslave16 ansible_host=10.186.19.16\n replslave17 ansible_host=10.186.19.17\n ```\n\n ---\n\n **2):进入mysql功能目录**\n ```bash\n cd /usr/local/mysqltools/deploy/ansible\n \n cd mysql #进入mysql功能目录\n ```\n\n ---\n\n **3):指定install_master_slaves.yaml中的目标主机**\n\n 假设我要在repl主机组上安装mysql复制环境那么install_master_slaves.yaml文件中的hosts变量应该设置为repl\n ```yaml\n ---\n - hosts: repl\n ```\n 修改vars/master_slaves.yaml 告诉mysqltools那个ip是主那些ip是slave\n ```\n #在创建一主多从环境时会用到的变量\n master_ip: 10.186.19.15\n slave_ips:\n - 10.186.19.16\n - 10.186.19.17\n ```\n\n ---\n\n **4):执行自动化安装**\n ```bash\n ansible-playbook install_master_slaves.yaml \n ```\n 输出如下\n ```\n PLAY [repl] *****************************************************************************************\n \n TASK [Gathering Facts] ******************************************************************************\n ok: [replmaster15]\n ok: [replslave16]\n ok: [replslave17]\n \n ... ... ... ... ... ... 省略 ... ... ... ... ... ... \n \n PLAY RECAP ******************************************************************************************\n replmaster15 : ok=28 changed=18 unreachable=0 failed=0 \n replslave16 : ok=28 changed=19 unreachable=0 failed=0 \n replslave17 : ok=28 changed=19 unreachable=0 failed=0 \n ```\n\n ---\n\n **5):检查主从复制环境是否配置完成**\n ```\n mysql -uroot -pmtls0352\n show slave status \\G\n ```\n 输出如下\n ```\n *************************** 1. row ***************************\n Slave_IO_State: Waiting for master to send event\n Master_Host: 10.186.19.15\n Master_User: rple\n Master_Port: 3306\n Connect_Retry: 60\n Master_Log_File: mysql-bin.000002\n Read_Master_Log_Pos: 595\n Relay_Log_File: actionsky16-relay-bin.000002\n Relay_Log_Pos: 800\n Relay_Master_Log_File: mysql-bin.000002\n Slave_IO_Running: Yes\n Slave_SQL_Running: Yes\n ... ... ... ... ... ... ... ... \n Retrieved_Gtid_Set: 8b5ac555-37ec-11e8-b50e-5a3fdb1cf647:1-2\n Executed_Gtid_Set: 8b5ac555-37ec-11e8-b50e-5a3fdb1cf647:1-2\n Auto_Position: 1\n Replicate_Rewrite_DB: \n Channel_Name: \n Master_TLS_Version:\n ```\n 两个Yes 说明说明主从复制环境成功配置了\n \n ---\n\n 3. ### mysql多源复制\n\n 假设我们要在**10.186.19.15，10.186.19.16，10.186.19.17**这三台机器上搭建两主1从的多源复制环境、其中15，16两机器上的数据向17同步\n\n ---\n\n **1):增加主机信息到/etc/ansible/hosts**\n\n 向/etc/ansible/hosts文件中增加如下内容\n ```\n [repl]\n replmaster15 ansible_host=10.186.19.15\n replslave16 ansible_host=10.186.19.16\n replslave17 ansible_host=10.186.19.17\n ```\n\n ---\n\n **2):进入mysql功能目录**\n ```bash\n cd /usr/local/mysqltools/deploy/ansible\n \n cd mysql #进入mysql功能目录\n ``` \n\n --- \n\n **3):指定install_multi_source_replication.yaml中的目标主机**\n\n 假设我要在repl主机组上安装mysql复制环境那么install_multi_source_replication.yaml文件中的hosts变量应该设置为repl\n ```yaml\n ---\n - hosts: repl\n ```\n 修改vars/multi_source_replication.yaml 告诉mysqltools那些ip是主那个ip是slave\n ```\n #master_ips 定义多个master主机ip组成的列表\n master_ips:\n - '10.186.19.15'\n - '10.186.19.16'\n \n #定义slave的ip\n slave_ip: '10.186.19.17'\n ```\n\n ---\n\n **4):执行自动化安装**\n ```bash\n ansible-playbook install_multi_source_replication.yaml \n ```\n 输出省略\n ```\n ```\n\n ---\n\n **5):检测多源复制环境是否安装成功**\n ```\n mysql -uroot -pmtls0352\n show slave status \\G\n ```\n 输出如下\n ```\n *************************** 1. row ***************************\n Slave_IO_State: Waiting for master to send event\n Master_Host: 10.186.19.15\n Master_User: rple_user\n Master_Port: 3306\n Connect_Retry: 60\n Master_Log_File: mysql-bin.000002\n Read_Master_Log_Pos: 150\n Relay_Log_File: actionsky17-relay-bin-master1.000002\n Relay_Log_Pos: 355\n Relay_Master_Log_File: mysql-bin.000002\n Slave_IO_Running: Yes\n Slave_SQL_Running: Yes\n *************************** 2. row ***************************\n Slave_IO_State: Waiting for master to send event\n Master_Host: 10.186.19.16\n Master_User: rple_user\n Master_Port: 3306\n Connect_Retry: 60\n Master_Log_File: mysql-bin.000002\n Read_Master_Log_Pos: 150\n Relay_Log_File: actionsky17-relay-bin-master2.000002\n Relay_Log_Pos: 355\n Relay_Master_Log_File: mysql-bin.000002\n Slave_IO_Running: Yes\n Slave_SQL_Running: Yes\n ```\n 两个通道都是双Yes、多源监制环境成了哦！\n\n ---\n\n 4. ### mysql组复制\n 假设我们要在**10.186.19.15，10.186.19.16，10.186.19.17**这三台机器上搭建一个group replication 环境\n\n ---\n\n **1):增加主机信息到/etc/ansible/hosts**\n\n 向/etc/ansible/hosts文件中增加如下内容\n ```\n [repl]\n mgr15 ansible_host=10.186.19.15\n mgr16 ansible_host=10.186.19.16\n mgr17 ansible_host=10.186.19.17\n ```\n\n ---\n\n **2):进入mysql功能目录**\n ```bash\n cd /usr/local/mysqltools/deploy/ansible\n \n cd mysql #进入mysql功能目录\n ``` \n\n ---\n\n **3):指定install_group_replication.yaml中的目标主机**\n\n 假设我要在repl主机组上安装mysql复制环境那么install_group_replication.yaml文件中的hosts变量应该设置为repl\n ```yaml\n ---\n - hosts: repl\n ```\n 修改var/group_replication.yaml.yaml 告诉mysqltools那些主机要参与到group replication中、组中成员之前通信的端口是多少、binlog的格式设置成row、开启group replication 复制的开关，这样的话mysqltools就会在生成配置文件的时候写入group replication相关的配置\n ```\n mtls_with_mysql_group_replication: 1\n mysql_binlog_format: row\n mysql_mgr_port: 13306\n mysql_mgr_hosts: \n - '10.186.19.15'\n - '10.186.19.16'\n - '10.186.19.17'\n ```\n\n ---\n\n **4):执行自动化安装**\n ```bash\n ansible-playbook install_group_replication.yaml \n ```\n 输出如下\n ```\n PLAY [repl] *****************************************************************************************\n \n TASK [Gathering Facts] ******************************************************************************\n ok: [mgr15]\n ok: [mgr16]\n ok: [mgr17]\n ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... \n PLAY RECAP ******************************************************************************************\n mgr15 : ok=28 changed=19 unreachable=0 failed=0 \n mgr16 : ok=28 changed=18 unreachable=0 failed=0 \n mgr17 : ok=28 changed=18 unreachable=0 failed=0 \n ```\n\n ---\n \n **5):检查group replication集群是否安装成功**\n ```bash\n mysql -uroot -pmtls0352\n select * from performance_schema.replication_group_members ;\n ```\n 输出如下\n ```\n +---------------------------+--------------------------------------+-------------+-------------+--------------+\n | CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE |\n +---------------------------+--------------------------------------+-------------+-------------+--------------+\n | group_replication_applier | 08de362c-3802-11e8-9e65-5a3fdb1cf647 | actionsky15 | 3306 | ONLINE |\n | group_replication_applier | ef7f3b61-3801-11e8-886d-9a17854b700d | actionsky17 | 3306 | ONLINE |\n | group_replication_applier | f1649143-3801-11e8-8fd3-5a1f0f06c50d | actionsky16 | 3306 | ONLINE |\n +---------------------------+--------------------------------------+-------------+-------------+--------------+\n 3 rows in set (0.00 sec)\n ```\n 三个都是online说明group replication 配置成功了哦\n\n 当mtls_with_mysql_group_replication: 1这个开头打开后mysqltools会自动为/etc/my.cnf增加了group replication 相关的配置项\n ```\n ####: for mysql group replication \n loose-group_replication_recovery_retry_count =10 # 10\n loose-group_replication_recovery_reconnect_interval =60 # 60\n loose-group_replication_allow_local_disjoint_gtids_join=off # off\n loose-group_replication_allow_local_lower_version_join=off # off\n loose-group_replication_ip_whitelist =AUTOMATIC # AUTOMATIC\n loose-transaction_write_set_extraction =XXHASH64 # off\n loose-group_replication_group_name =\"aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa\" #\n loose-group_replication_start_on_boot =off # off\n loose-group_replication_bootstrap_group =off # off\n loose-group_replication_single_primary_mode =on # on\n loose-group_replication_enforce_update_everywhere_checks=off\n loose-group_replication_gtid_assignment_block_size =1000000 # 1000000\n loose-group_replication_poll_spin_loops =0 # 0\n loose-group_replication_compression_threshold =1024 # 1000000\n loose-group_replication_flow_control_mode =QUOTA # QUOTA\n loose-group_replication_local_address =\"10.186.19.15:13306\"\n loose-group_replication_group_seeds =\"10.186.19.15:13306,10.186.19.16:13306,10.186.19.17:13306\"\n ```\n 也就是说目前mysqltools在配置group replication 集群环境时是按单个写结点的方式配置的、**默认mysqltools把第一个ip地址设置为写结点**\n\n ---\n\n 5. ### mysql小版本升级\n **mysqltools为版本升级提供了支持，如果你想使用这个功能，那么你要清楚的知道自己在干什么**\n\n ---\n\n 1、小版本升级并不会去执行`mysql_upgrade`脚本，这个主要是由于mysqltools诞生环境比较特别，单个实例的数据量是以`TB`来衡量的，在这种量级的情况下`mysql_upgrade`可以能执行几天才能完成，所以你懂的\n\n ---\n\n **1):修改upgrad_single_mysql.yaml文件的hosts 变量为你要升级的目标主机**\n\n ```\n ---\n - hosts: sqlstudio\n ```\n\n ---\n\n **2):执行升级程序**\n ```\n ansible-playbook upgrad_single_mysql.yaml\n ```\n 输出如下：\n ```\n PLAY [sqlstudio] **************************************************************************************************************\n \n TASK [Gathering Facts] ********************************************************************************************************\n ok: [sqlstudio]\n \n TASK [stop mysql service] *****************************************************************************************************\n ok: [sqlstudio]\n \n TASK [backup link file] *******************************************************************************************************\n changed: [sqlstudio]\n \n TASK [unarchive new package to /usr/local/] ***********************************************************************************\n changed: [sqlstudio]\n \n TASK [change owner and group] *************************************************************************************************\n changed: [sqlstudio]\n \n TASK [make new link file] *****************************************************************************************************\n changed: [sqlstudio]\n \n TASK [start mysql service] ****************************************************************************************************\n changed: [sqlstudio]\n \n PLAY RECAP ********************************************************************************************************************\n sqlstudio : ok=7 changed=5 unreachable=0 failed=0\n ```\n\n ---\n\n **3):查看升级是否成功**\n \n 1、看/usr/local/下的变化\n\n **升级前**\n ```\n drwxr-xr-x 9 mysql mysql 129 6月 18 14:46 mysql-5.7.21-linux-glibc2.12-x86_64\n lrwxrwxrwx 1 root root 35 6月 18 14:53 mysql -\u003e mysql-5.7.21-linux-glibc2.12-x86_64\n ```\n **升级后**\n ```\n lrwxrwxrwx 1 mysql mysql 46 6月 18 15:30 mysql -\u003e /usr/local/mysql-5.7.22-linux-glibc2.12-x86_64\n drwxr-xr-x 9 mysql mysql 129 6月 18 14:46 mysql-5.7.21-linux-glibc2.12-x86_64\n drwxr-xr-x 9 mysql mysql 129 6月 18 15:30 mysql-5.7.22-linux-glibc2.12-x86_64\n lrwxrwxrwx 1 root root 35 6月 18 14:53 mysql.backup.20180618 -\u003e mysql-5.7.21-linux-glibc2.12-x86_64\n ```\n\n ---\n\n 2、连接进数据库进行检察\n ```\n mysql -uroot -pxxxxxx\n ```\n\n ``` \n mysql: [Warning] Using a password on the command line interface can be insecure.\n Welcome to the MySQL monitor. Commands end with ; or \\g.\n Your MySQL connection id is 3\n Server version: 5.7.22-log MySQL Community Server (GPL)\n \n Copyright (c) 2000, 2018, Oracle and/or its affiliates. All rights reserved.\n \n Oracle is a registered trademark of Oracle Corporation and/or its\n affiliates. Other names may be trademarks of their respective\n owners.\n \n Type 'help;' or '\\h' for help. Type '\\c' to clear the current input statement.\n \n mysql\u003eselect @@version;\n +------------+\n | @@version |\n +------------+\n | 5.7.22-log |\n +------------+\n 1 row in set (0.00 sec)\n ```\n\n ---\n\n \n## 读写分离\n **本来这里是要加入分库分表的功能的、因为用yaml难以表达mycat的相关配置、最终我还是放弃了、所以目前mysqltools还只有自动化安装配置mycat读写分离的功能**\n\n ---\n\n 1. ### mycat读写分离\n 假设我们已经有了套三结点的集群10.186.19.15，10.186.19.16，10.186.19.17，其中15支持读写，16、17只读；为了方面应用程序我们要在10.186.19.14上安装一个mycat用它做读写分离，这样应用只要配置一个连接字符串就可以了(连接mycat)\n\n ---\n\n **1):在数据库中创建用户、mycat会有这个用户连接数据库**\n\n 用户名、密码引用自mysqltools/config.yaml中的mysql_app_user、mysql_app_password\n 在读写库上执行如下代码\n ```sql\n create user appuser@'%' identified by 'mtls0352';\n create database appdb char set utf8;\n grant all on appdb.* to appuser@'%';\n ```\n\n ---\n\n **2):向/etc/ansible/hosts文件增加主机信息**\n 向/etc/ansible/hosts文件增加如下内容\n ```\n mycat ansible_host=10.186.19.14\n ```\n\n ---\n\n **3):更新var/var_mycat.yaml文件**\n\n 告诉mysqltools那个主机是读写库(master_ip)，哪些库是只读库(slave_ips)，导出后端哪些schema(schemas)\n ```\n master_ip: \"10.186.19.15\"\n \n slave_ips:\n - \"10.186.19.16\"\n - \"10.186.19.17\"\n \n schemas:\n - \"appdb\"\n ```\n **可以把master_ip设置为vip、slave_ips设置为所有ip的一个列表，这样就可以不在意高可用环境下的主从切换了**\n\n ---\n\n **4):修改install_mycat.yaml文件中的hosts**\n ```\n ---\n - hosts: mycat\n ```\n\n ---\n\n **5):执行mycat的安装**\n ```\n ansible-playbook install_mycat.yaml\n ```\n 输出如下\n ```\n PLAY [mycat] ****************************************************************************************\n \n TASK [Gathering Facts] ******************************************************************************\n ok: [mycat]\n \n TASK [install java-1.7.0-openjdk] *******************************************************************\n changed: [mycat]\n \n TASK [create mycat user] ****************************************************************************\n changed: [mycat]\n \n TASK [trasfer mycat-server-1.6.5-linux.tar.gz to remonte host] **************************************\n changed: [mycat]\n \n TASK [export MYCAT_HOME env to /etc/profile] ********************************************************\n changed: [mycat]\n \n TASK [config schema.xml] ****************************************************************************\n changed: [mycat]\n \n TASK [config server.xml] ****************************************************************************\n changed: [mycat]\n \n TASK [transfer start_mycat.sh to remonte /tmp/] *****************************************************\n changed: [mycat]\n \n TASK [start mycat] **********************************************************************************\n changed: [mycat]\n \n TASK [remove start_mycat.sh] ************************************************************************\n changed: [mycat]\n \n PLAY RECAP ******************************************************************************************\n mycat : ok=10 changed=9 unreachable=0 failed=0 \n ```\n\n ---\n\n **6):检查mycat是否启动**\n ```\n ps -ef | grep mycat\n ```\n 输出如下\n ```\n root 24415 1 0 09:21 ? 00:00:00 /usr/local/mycat/bin/./wrapper-linux-x86-64 /usr/local/mycat/conf/wrapper.conf wrapper.syslog.ident=mycat wrapper.pidfile=/usr/local/mycat/logs/mycat.pid wrapper.daemonize=TRUE wrapper.lockfile=/var/lock/subsys/mycat\n root 24417 24415 12 09:21 ? 00:00:06 java -DMYCAT_HOME=. -server -XX:MaxPermSize=64M -XX:+AggressiveOpts -XX:MaxDirectMemorySize=2G -Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=1984 -Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.ssl=false -Xmx4G -Xms1G -Djava.library.path=lib -classpath lib/wrapper.jar:conf:lib/zookeeper-3.4.6.jar:lib/jline-0.9.94.jar:lib/ehcache-core-2.6.11.jar:lib/log4j-1.2.17.jar:lib/fastjson-1.2.12.jar:lib/curator-client-2.11.0.jar:lib/joda-time-2.9.3.jar:lib/log4j-slf4j-impl-2.5.jar:lib/libwrapper-linux-x86-32.so:lib/netty-3.7.0.Final.jar:lib/druid-1.0.26.jar:lib/log4j-api-2.5.jar:lib/mapdb-1.0.7.jar:lib/slf4j-api-1.6.1.jar:lib/univocity-parsers-2.2.1.jar:lib/hamcrest-core-1.3.jar:lib/objenesis-1.2.jar:lib/leveldb-api-0.7.jar:lib/hamcrest-library-1.3.jar:lib/wrapper.jar:lib/commons-lang-2.6.jar:lib/reflectasm-1.03.jar:lib/mongo-java-driver-2.11.4.jar:lib/guava-19.0.jar:lib/curator-recipes-2.11.0.jar:lib/curator-framework-2.11.0.jar:lib/libwrapper-linux-ppc-64.so:lib/log4j-core-2.5.jar:lib/mysql-binlog-connector-java-0.6.0.jar:lib/netty-common-4.1.9.Final.jar:lib/leveldb-0.7.jar:lib/sequoiadb-driver-1.12.jar:lib/kryo-2.10.jar:lib/jsr305-2.0.3.jar:lib/commons-collections-3.2.1.jar:lib/mysql-connector-java-5.1.35.jar:lib/disruptor-3.3.4.jar:lib/log4j-1.2-api-2.5.jar:lib/velocity-1.7.jar:lib/Mycat-server-1.6.5-release.jar:lib/libwrapper-linux-x86-64.so:lib/dom4j-1.6.1.jar:lib/minlog-1.2.jar:lib/asm-4.0.jar:lib/netty-buffer-4.1.9.Final.jar -Dwrapper.key=sexYToWnzGO4Glh1 -Dwrapper.port=32000 -Dwrapper.jvm.port.min=31000 -Dwrapper.jvm.port.max=31999 -Dwrapper.pid=24415 -Dwrapper.version=3.2.3 -Dwrapper.native_library=wrapper -Dwrapper.service=TRUE -Dwrapper.cpu.timeout=10 -Dwrapper.jvmid=1 org.tanukisoftware.wrapper.WrapperSimpleApp io.mycat.MycatStartup start\n ```\n\n ---\n\n **7):测试mycat的可用性**\n ```\n mysql -uappuser -pmtls0352 -h10.186.19.14 -P8066\n ```\n ```\n mysql: [Warning] Using a password on the command line interface can be insecure.\n Welcome to the MySQL monitor. Commands end with ; or \\g.\n Your MySQL connection id is 1\n Server version: 5.7.200-mycat-1.6.5-release-20171117203123 MyCat Server (OpenCloundDB)\n \n Copyright (c) 2000, 2018, Oracle and/or its affiliates. All rights reserved.\n \n Oracle is a registered trademark of Oracle Corporation and/or its\n affiliates. Other names may be trademarks of their respective\n owners.\n \n Type 'help;' or '\\h' for help. Type '\\c' to clear the current input statement.\n \n mysql\u003e show databases;\n +----------+\n | DATABASE |\n +----------+\n | appdb |\n +----------+\n 2 rows in set (0.00 sec)\n\n create table person( id int auto_increment primary key,name varchar(16));\n Query OK, 0 rows affected (0.02 sec)\n\n insert into person(name) values('welson');\n Query OK, 1 row affected (0.05 sec)\n\n select * from person;\n +----+--------+\n | id | name |\n +----+--------+\n | 1 | welson |\n +----+--------+\n ```\n mycat 读写分离环境测试成功.\n\n---\n\n\n## 高可用\n **目前mysql高可用的开源解决方案中比较成熟的应该算mha了、相信不久的将来innodb cluster将会成为主流；mysqltools会把两者都包涵进来**\n\n ---\n\n1. ## mha\n **MHA是在mysql主从复制环境的基础上加的一套高可用软件、这套软件逻辑上又可以分成两个组件manager和node；其中manager负责监控master库是否存活，一旦master有问题就开大招做主从切换、切换中的一些脏活累活基本都由node来完；** 相关链接：https://www.cnblogs.com/gomysql/p/3675429.html\n\n mysqltools在安装mha时不需要连接外网，主要是因为它把mha用到的包都打包了进来；不幸的是目前我只打包了centos-7.4版本下的包；由于时间上的成本其它系统版本我就没有打包了\n\n ---\n\n 1. ### 环境规划\n **130、131、132三个实例组成一个mysql复制环境、其中130是master**\n \n |主机名 | ip地址 | 操作系统版本 | mysql角色| mha角色| vip |\n |---------|----------------|------------|---------|--------|------------------|\n |mhamaster| 192.168.29.130 | centos-7.4 | master | node |192.168.29.100 |\n |mhaslave1| 192.168.29.131 | centos-7.4 | slave | node | |\n |mhaslave2| 192.168.29.132 | centos-7.4 | slave | manager| |\n\n ---\n\n 2. ### 配置mha的一些前置备件\n **1、已经规范的安装配置了mysql复制环境、可以参考[mysql主从复制](#mysql主从复制)**\n\n **2、各个主机之间都已经完成了ssh信任(mha会采用scp的方式从宕机的master主机上采集binlog日志、所以你事先要把ssh信任关系做好)**\n\n **3、在安装配置的过程中有一些组件要用到gcc编译、mysqltools会去安装gcc gcc-c++ ，所以你要事先在这些主机上配置好yum**\n\n ---\n\n 3. ### 配置ansible的hosts文件\n **把要配置mha的几台主机作为一个组配置到/etc/ansible/hosts文件中去**\n ```\n [mhacluster]\n mhamaster ansible_host=192.168.29.130 ansible_user=root\n mhaslave1 ansible_host=192.168.29.131 ansible_user=root\n mhaslave2 ansible_host=192.168.29.132 ansible_user=root\n ```\n ---\n\n 4. ### 配置mysqltools中mha相关的配置项\n **mysqltools/config.yaml配置文件中有两项、是用来指明安装mha时所用的安装包的、默认值如下；用默认值就行一定不要改、除非你知道自己在做什么。**\n ```\n mtls_mha_node: mhanode.tar.gz\n mtls_mha_manager: mhamanager.tar.gz\n ```\n ---\n\n 5. ### 配置mha的相关信息\n **针对单个mha的详细配置都记录在了mysqltools/deploy/ansible/mha/vars/var_mha.yaml这个配置文件中了**\n ```\n master_ip: \"192.168.29.130\"\n slave_ips:\n - \"192.168.29.131\"\n - \"192.168.29.132\"\n \n manager_ip: \"192.168.29.132\"\n \n net_work_interface: \"ens33\"\n vip: \"192.168.29.100\"\n \n os_release: '7.4'\n\n ```\n 1、master_ip 表示mysql master库所在主机的ip地址\n\n 2、slave_ips 表示mysql slave库的ip地址列表\n\n 3、manager_ip 表示mha manager要安装在主机ip、mysqltools希望你用最后一个slave的ip作为manger要安装的地方、mysqltools还会把第一个slave作为新的master\n\n 4、net_work_inferface 表示vip要绑定到的网卡\n\n 5、vip 表示vip地址\n\n 6、os_release 表示操作系统的版本、目前还只支持\"7.4\"这一个rhel版本\n\n ---\n\n 6. ### 准备安装配置mha\n **mysqltools/deploy/ansible/mha/install_mha.yaml文件中包涵了所有的步骤、修改文件中的host项以告诉ansible把这些步骤应用到哪些主机、根据上面的规划我要把hosts: 修改成如下内容**\n ```\n ---\n - hosts: mhacluster\n ```\n 也mhacluster这个组内的机器应用install_mha.yaml中定义的步骤\n\n ---\n\n 7. ### 安装配置mha\n ```\n cd mysqltools/deploy/ansible/mha\n ansible-playbook install_mha.yaml\n ```\n 输出如下：\n ```\n PLAY [mhacluster] ********************************************************************\n TASK [Gathering Facts] ***************************************************************\n ok: [slave2]\n ok: [slave1]\n ok: [master]\n TASK [install gcc] ********************************************************************\n changed: [slave1]\n changed: [slave2]\n changed: [master]\n TASK [install gcc-c++] ****************************************************************\n changed: [slave2]\n changed: [slave1]\n changed: [master]\n TASK [transfer mhanode.tar.gz to remote host and unarchive to /tmp/] ******************\n changed: [master]\n changed: [slave1]\n changed: [slave2]\n TASK [install mha node] ***************************************************************\n changed: [master]\n changed: [slave2]\n changed: [slave1]\n TASK [export path env to /root/.bashrc] ***********************************************\n changed: [master]\n changed: [slave2]\n changed: [slave1]\n TASK [stransfer create_mha_user.sql to master] ****************************************\n skipping: [slave1]\n skipping: [slave2]\n changed: [master]\n TASK [create mha user in mysql(master)] ***********************************************\n skipping: [slave1]\n skipping: [slave2]\n changed: [master]\n TASK [copy bind_vip.sh to /tmp/] ******************************************************\n skipping: [slave1]\n skipping: [slave2]\n changed: [master]\n TASK [bind vip] ***********************************************************************\n skipping: [slave1]\n skipping: [slave2]\n changed: [master]\n TASK [transfer mhamanager.tar.gz to remote host and unarchive to /tmp/] ***************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [install mha manager] ************************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [create directory(/etc/masterha/)] ***********************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [create directory(/var/log/masterha)] *********************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [config mha app.cnf] ***************************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [config master_ip_failover] ********************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [config master_ip_online_change] ***************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [copy start_mha.sh to /usr/local/] **************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n TASK [start mha manager] *****************************************************************\n skipping: [master]\n skipping: [slave1]\n changed: [slave2]\n PLAY RECAP ********************************************************************************\n master : ok=10 changed=9 unreachable=0 failed=0 \n slave1 : ok=6 changed=5 unreachable=0 failed=0 \n slave2 : ok=15 changed=14 unreachable=0 failed=0 \n ```\n\n ---\n\n 8. ### 验证是否成功完成\n **mysqltools在安装配置mha后并没有并没有把相关安装包直接删除、而是保留在了/tmp/mhanode /tmp/mhamanager 这两个目录下；目录下还包含用于验证结果的的脚本**\n\n ---\n\n **1、验证master主机是否成功的绑定了vip**\n ```\n ifconfig\n ```\n 输出如下：\n ```\n ens33: flags=4163\u003cUP,BROADCAST,RUNNING,MULTICAST\u003e mtu 1500\n inet 192.168.29.130 netmask 255.255.255.0 broadcast 192.168.29.255\n inet6 fe80::413c:fcac:858:64bc prefixlen 64 scopeid 0x20\u003clink\u003e\n ether 00:0c:29:e3:07:d3 txqueuelen 1000 (Ethernet)\n RX packets 10816 bytes 13581709 (12.9 MiB)\n RX errors 0 dropped 0 overruns 0 frame 0\n TX packets 2535 bytes 310584 (303.3 KiB)\n TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0\n \n ens33:0: flags=4163\u003cUP,BROADCAST,RUNNING,MULTICAST\u003e mtu 1500\n inet 172.16.192.100 netmask 255.255.0.0 broadcast 172.16.255.255\n ether 00:0c:29:e3:07:d3 txqueuelen 1000 (Ethernet)\n \n lo: flags=73\u003cUP,LOOPBACK,RUNNING\u003e mtu 65536\n inet 127.0.0.1 netmask 255.0.0.0\n inet6 ::1 prefixlen 128 scopeid 0x10\u003chost\u003e\n loop txqueuelen 1 (Local Loopback)\n RX packets 203 bytes 29718 (29.0 KiB)\n RX errors 0 dropped 0 overruns 0 frame 0\n TX packets 203 bytes 29718 (29.0 KiB)\n TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0\n ```\n ens33:0 已经绑上了192.168.29.100 这个vip、说明vip绑定成功了\n\n ---\n\n **2、验证mha-manager是否正常启动**\n ```\n ps -ef | grep man\n ```\n 输出如下：\n ```\n root 3549 1 1 11:38 ? 00:00:00 perl /usr/local/bin//masterha_manager --conf=/etc/masterha/app.cnf --ignore_last_failover\n ```\n masterha_manager 这个进程存在说明manager已经成功启动(默认情况下日志保存在/var/log/masterha/manager.log你可以在这个日志看到更多的详细信息)\n\n ---\n\n **事实上完成上面的两项如果都成功了、那么你的mha就算配置成功了、但是为了排错的方便我还是在/tmp/mhamanager目录下留下了一些用于check脚本**\n\n ---\n\n **3、检查ssh互信是否配置正确**\n ```\n cd /tmp/mhamanager/\n ./check_ssh.sh \n ```\n 输出如下：\n ```\n Thu May 17 11:33:49 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.\n Thu May 17 11:33:49 2018 - [info] Reading application default configuration from /etc/masterha/app.cnf..\n Thu May 17 11:33:49 2018 - [info] Reading server configuration from /etc/masterha/app.cnf..\n Thu May 17 11:33:49 2018 - [info] Starting SSH connection tests..\n Thu May 17 11:33:50 2018 - [debug] \n Thu May 17 11:33:49 2018 - [debug] Connecting via SSH from root@192.168.29.130(192.168.29.130:22) to root@192.168.29.131(192.168.29.131:22) ..\n Thu May 17 11:33:49 2018 - [debug] ok.\n Thu May 17 11:33:49 2018 - [debug] Connecting via SSH from root@192.168.29.130(192.168.29.130:22) to root@192.168.29.132(192.168.29.132:22) ..\n Thu May 17 11:33:50 2018 - [debug] ok.\n Thu May 17 11:33:50 2018 - [debug] \n Thu May 17 11:33:49 2018 - [debug] Connecting via SSH from root@192.168.29.131(192.168.29.131:22) to root@192.168.29.130(192.168.29.130:22) ..\n Thu May 17 11:33:50 2018 - [debug] ok.\n Thu May 17 11:33:50 2018 - [debug] Connecting via SSH from root@192.168.29.131(192.168.29.131:22) to root@192.168.29.132(192.168.29.132:22) ..\n Thu May 17 11:33:50 2018 - [debug] ok.\n Thu May 17 11:33:51 2018 - [debug] \n Thu May 17 11:33:50 2018 - [debug] Connecting via SSH from root@192.168.29.132(192.168.29.132:22) to root@192.168.29.130(192.168.29.130:22) ..\n Thu May 17 11:33:50 2018 - [debug] ok.\n Thu May 17 11:33:50 2018 - [debug] Connecting via SSH from root@192.168.29.132(192.168.29.132:22) to root@192.168.29.131(192.168.29.131:22) ..\n Thu May 17 11:33:51 2018 - [debug] ok.\n Thu May 17 11:33:51 2018 - [info] All SSH connection tests passed successfully.\n ```\n 最后一行`All SSH connection tests passed successfully.`说明ssh信任是配置好了的\n\n ---\n\n **4、检查mysql 复制是否配置正确**\n ```\n cd /tmp/mhamanager/\n ./check_repl.sh \n ```\n 输出如下：\n ```\n Thu May 17 11:33:24 2018 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping.\n Thu May 17 11:33:24 2018 - [info] Reading application default configuration from /etc/masterha/app.cnf..\n Thu May 17 11:33:24 2018 - [info] Reading server configuration from /etc/masterha/app.cnf..\n Thu May 17 11:33:24 2018 - [info] MHA::MasterMonitor version 0.57.\n Thu May 17 11:33:25 2018 - [info] GTID failover mode = 1\n Thu May 17 11:33:25 2018 - [info] Dead Servers:\n Thu May 17 11:33:25 2018 - [info] Alive Servers:\n Thu May 17 11:33:25 2018 - [info] 192.168.29.130(192.168.29.130:3306)\n Thu May 17 11:33:25 2018 - [info] 192.168.29.131(192.168.29.131:3306)\n Thu May 17 11:33:25 2018 - [info] 192.168.29.132(192.168.29.132:3306)\n Thu May 17 11:33:25 2018 - [info] Alive Slaves:\n Thu May 17 11:33:25 2018 - [info] 192.168.29.131(192.168.29.131:3306) Version=5.7.22-log (oldest major version between slaves) log-bin:enabled\n Thu May 17 11:33:25 2018 - [info] GTID ON\n Thu May 17 11:33:25 2018 - [info] Replicating from 192.168.29.130(192.168.29.130:3306)\n Thu May 17 11:33:25 2018 - [info] Primary candidate for the new Master (candidate_master is set)\n Thu May 17 11:33:25 2018 - [info] 192.168.29.132(192.168.29.132:3306) Version=5.7.22-log (oldest major version between slaves) log-bin:enabled\n Thu May 17 11:33:25 2018 - [info] GTID ON\n Thu May 17 11:33:25 2018 - [info] Replicating from 192.168.29.130(192.168.29.130:3306)\n Thu May 17 11:33:25 2018 - [info] Current Alive Master: 192.168.29.130(192.168.29.130:3306)\n Thu May 17 11:33:25 2018 - [info] Checking slave configurations..\n Thu May 17 11:33:25 2018 - [info] read_only=1 is not set on slave 192.168.29.131(192.168.29.131:3306).\n Thu May 17 11:33:25 2018 - [info] read_only=1 is not set on slave 192.168.29.132(192.168.29.132:3306).\n Thu May 17 11:33:25 2018 - [info] Checking replication filtering settings..\n Thu May 17 11:33:25 2018 - [info] binlog_do_db= , binlog_ignore_db= \n Thu May 17 11:33:25 2018 - [info] Replication filtering check ok.\n Thu May 17 11:33:25 2018 - [info] GTID (with auto-pos) is supported. Skipping all SSH and Node package checking.\n Thu May 17 11:33:25 2018 - [info] Checking SSH publickey authentication settings on the current master..\n Thu May 17 11:33:26 2018 - [info] HealthCheck: SSH to 192.168.29.130 is reachable.\n Thu May 17 11:33:26 2018 - [info] \n 192.168.29.130(192.168.29.130:3306) (current master)\n +--192.168.29.131(192.168.29.131:3306)\n +--192.168.29.132(192.168.29.132:3306)\n \n Thu May 17 11:33:26 2018 - [info] Checking replication health on 192.168.29.131..\n Thu May 17 11:33:26 2018 - [info] ok.\n Thu May 17 11:33:26 2018 - [info] Checking replication health on 192.168.29.132..\n Thu May 17 11:33:26 2018 - [info] ok.\n Thu May 17 11:33:26 2018 - [info] Checking master_ip_failover_script status:\n Thu May 17 11:33:26 2018 - [info] /usr/local/bin/master_ip_failover --command=status --ssh_user=root --orig_master_host=192.168.29.130 --orig_master_ip=192.168.29.130 --orig_master_port=3306 \n \n \n IN SCRIPT TEST====/sbin/ifconfig ens33:0 down==/sbin/ifconfig ens33:0 172.16.192.100===\n \n Checking the Status of the script.. OK \n Thu May 17 11:33:26 2018 - [info] OK.\n Thu May 17 11:33:26 2018 - [warning] shutdown_script is not defined.\n Thu May 17 11:33:26 2018 - [info] Got exit code 0 (Not master dead).\n \n MySQL Replication Health is OK.\n ```\n\n 最后一行`MySQL Replication Health is OK.`说明mysql复制是正常的\n \n ---\n \n## 备份\n **备份的作用在此不表。单单从备份工具来看就有mysqldup,xtrabackup,mysqlbackup三种可选的工具；备份工具是只完成备份计划的手段，作为一个dba我更加关心备份策略，比如说周日做全备，其它几天每天一个差异备份；就这样一个备份计划而言我选择xtrabackup,mysqlbackup都是可以的。但是为了方便使用mysqltools把备份实现的细节通过一个python程序包装起来，dba只要告诉mysqltools他要什么时候做全备，什么时候做差异备份就行了。隐去了实现的细节dba可以更加的专注问题的核心**\n\n **由于备份是一个比较通用的功能所以我把它打包进了mysqltools-python这个软件包中可以通过pip直接安装，如果你用的mysqltools安装的python默认我会把这个包也安装上去**\n\n 你可以通过如下地址了解更多`mysqltools-python`相关的内容\n\n **1):Pypi:** https://pypi.org/project/mysqltools-python/\n \n \u003cimg src=\"docs/imgs/mysqltoools-python.png\"\u003e\n\n ---\n\n **2):Github:** https://github.com/Neeky/mysqltools-python\n\n \u003cimg src=\"docs/imgs/mysqltools-python-github.png\"\u003e\n\n ---\n\n 安装完`mysqltools-python`后你在python的bin目录下会看到如下可执行文件\n ```\n ll /usr/local/python/bin/ | grep mtls \n -rwxr-xr-x. 1 root root 11770 9月 17 16:47 mtlsbackup # mtlsbackup用于备份 \n -rwxr-xr-x. 1 root root 11211 9月 17 16:47 mtlsmonitor # mtlsbackup用于监控\n ```\n 如果你是用mysqltools自带的安装脚本安装的python-3.x.x那么mysqltools-python会被自动安装上，当然你也可以自己手工安装它，像这样\n ```\n pip3 install mysqltools-python\n ```\n ---\n\n 1. ### mtlsbackup备份相关的实现细节\n **1): mtlsbackup 备份脚本**\n\n 这是一个由python编写的程序、它的目的有两个 1):隔离mysqldump,xtrabackup,mysqlbackup这三个备份工具的差异，用户在做备份时只要调用`mtlsbackup`就行了。2):它还要完成完整的备份决策(根据dba的配置来决定什么时候做全备，什么时候做增备；它也会根据实际情况对决策进行调整，如果dba配置的是每周日一个全备，其它几天做差异备份。假设dba配置这个备份计划的时候是周3，根据配置要求周3是要做差异备份的，由于现在(周三)还没有全备呢，mtlsbackup会这次的备份执行情况做调整，把它从差异备份调整为全备)\n\n ---\n\n **2): /etc/mtlsbackup.cnf 定义备份策略**\n\n 为了方便使用`mtlsbackup.py`尽可能的避免命令行参数，而是采用从配置文件直接读取备份计划的方式来完成备份，以下是一份完成的备份计划的样例\n ```\n [global]\n backup_tool=xtrabackup #备份工具xtrabackup,mysqldump,meb 之一(现在只直接xtrabackup)\n user=backup #备份用户(mysql级别)\n password=DX3906 #密码\n host=127.0.0.1 #主机\n port=3306 #端口\n full_backup_days=6 #指定哪些天做全备 6 --\u003e 周日 \n diff_backup_days=0,1,2,3,4,5 #指定哪些天做差异备 0 --\u003e 周一 ，1 --\u003e 周2 ... ...\n backup_data_dir=/database/backups/3306/data/ #备份保存的路径\n backup_log_dir=/database/backups/3306/log/ #使用xtrackup备份时check_point文件的目录\n backup_temp_dir=/database/backups/3306/temp/ #xtrabackup的工作目录\n \n [xtrabackup]\n # xtrabackup 备份时对应的命令模板\n full_backup_script=/usr/local/xtrabackup/bin/xtrabackup --defaults-file=/etc/my.cnf --host={self.host} --port={self.port} --user={self.user} --password={self.password} --no-version-check --compress --compress-threads=4 --use-memory=200M --stream=xbstream --parallel=8 --backup --extra-lsndir={self.lsndir} --target-dir={self.backup_temp_dir} \u003e {self.full_backup_file} 2\u003e{self.full_backup_log_file} \u0026\n diff_backup_script=/usr/local/xtrabackup/bin/xtrabackup --defaults-file=/etc/my.cnf --host={self.host} --port={self.port} --user={self.user} --password={self.password} --no-version-check --compress --compress-threads=4 --use-memory=200M --stream=xbstream --parallel=8 --backup --extra-lsndir={self.lsndir} --target-dir={self.backup_temp_dir} --incremental --incremental-lsn={self.tolsn} \u003e {self.diff_backup_file} 2\u003e{self.diff_backup_log_file} \u0026\n ```\n 一直以提高生产力为目标的mysqltools是不会让你手工编写这个配置文件它，它会根据机器的配置自动生成\n\n ---\n\n **3): /database/backups/这个目录是用来保存备份文件的地方**\n ```\n tree /database/backups/\n ```\n 输出如下：\n ```\n /database/backups/\n ├── 3306\n │   ├── data\n │   │   └── 2018-07-28T13:38:01\n │   │   ├── 2018-07-28T13:38:01-full.log\n │   │   ├── 2018-07-28T13:38:01-full.xbstream\n │   │   ├── 2018-07-28T13:40:05-diff.log\n │   │   ├── 2018-07-28T13:40:05-diff.xbstream\n │   │   ├── 2018-07-28T13:42:04-diff.log\n │   │   ├── 2018-07-28T13:42:04-diff.xbstream\n │   ├── log\n │   │   ├── 2018-07-28T13:38:01\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   ├── 2018-07-28T13:40:05\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   ├── 2018-07-28T13:42:04\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n ```\n\n ---\n\n 2. ### 实施备份计划的前期准备\n **1): 如前面所说的mtlsbackup是一个python写的脚本它的运行依赖于python3你要在目标主机上安装python3，见[安装python](#安装python)**\n\n ---\n\n **2): 配置MySQL数据库的备份计划**\n \n 配置文件`mysqltools/deploy/ansible/backup/template/mtlsbackup.cnf`中的`full_backup_days`代表着哪些天执行全备，`diff_backup_days`代表着哪些天执行差异备份\n ```\n full_backup_days=6 #指定哪些天做全备 6--\u003e周日 5--\u003e周六 4--\u003e周五... ...\n diff_backup_days=0,1,2,3,4,5 #指定哪些天做差异备 6--\u003e周日 4--\u003e周六 4--\u003e周五... ...\n ```\n **建议不要改保持默认值**\n\n ---\n\n **3): 配置crontab执行备份的时机**\n\n 配置文件`mysqltools/deploy/ansible/backup/vars/mtlsbackup.yaml`中`backup_minute`对应linux crontab 中的minute,`backup_hour`对应linux crontab 中的hour,`backup_user`对应linux crontab \b中的user\n ```\n ---\n backup_minute: \"0\" \n backup_hour: \"2\"\n backup_user: \"mysql\"\n \n #backup_minute 对应linux crontab 中的minute\n #backup_hour 对应linux crontab 中的hour\n #backup_user 对应linux crontab \b中的user\n \n # 上面的默认配置表示每天的02:00:00时对数据库进行备份\n ```\n **建议不要改保持默认值**\n\n ---\n\n 3. ### 配置备份\n **1): 配置备份的工作目录在mysqltools/deploy/ansible/backup**\n\n 修改config_backup.yaml文件中的hosts为你的目标主机,这里以sqlstudio这个主机为例，所以config_backup.yaml的内容如下\n ```\n ---\n - hosts: sqlstudio\n ```\n ---\n\n **2):实施备份计划**\n ```\n ansible-playbook config_backup.yaml\n ```\n 输出如下：\n ```\n PLAY [sqlstudio] **************************************************************************************************************\n \n TASK [Gathering Facts] ********************************************************************************************************\n ok: [sqlstudio]\n \n TASK [transfer qperss to remonte host(rhel-7.x)] ******************************************************************************\n changed: [sqlstudio]\n \n TASK [install qpress(rhel-7.x)] ***********************************************************************************************\n changed: [sqlstudio]\n \n TASK [remove qpress install package(rhel-7.x)] ********************************************************************************\n changed: [sqlstudio]\n \n TASK [transfer extrabackup install package to remonte host] *******************************************************************\n ok: [sqlstudio]\n \n TASK [make link file fore percona-xtrabackup-2.4.9-Linux-x86_64] **************************************************************\n ok: [sqlstudio]\n \n TASK [export path env variable] ***********************************************************************************************\n ok: [sqlstudio]\n \n TASK [export path env to /root/.bashrc] ***************************************************************************************\n ok: [sqlstudio]\n \n TASK [transfer create_backup_user.sql file to remote host] ********************************************************************\n skipping: [sqlstudio]\n \n TASK [execute create_backup_user.sql] *****************************************************************************************\n skipping: [sqlstudio]\n \n TASK [remove /tmp/create_backup_user.sql] *************************************************************************************\n skipping: [sqlstudio]\n \n TASK [config /etc/mtlsbackup.cnf] *********************************************************************************************\n ok: [sqlstudio]\n \n TASK [config crontab] *********************************************************************************************************\n ok: [sqlstudio]\n \n PLAY RECAP ********************************************************************************************************************\n sqlstudio : ok=10 changed=3 unreachable=0 failed=0 \n ```\n ---\n\n 4. ### 检查配置是否成功的几个点\n **1): 检查crond是否正确配置备份任务**\n ```\n sudo -umysql crontab -l \n ```\n 输出如下：\n ``` \n #Ansible: mtlsbackup\n 0 2 * * * /usr/local/python/bin/python3 /usr/local/mysqltoolsclient/mtlsbackup.py 2\u003e\u003e/database/backups/mtlsbackup.log\n ```\n 由上面的配置可以看出crontab已经配置好了\n\n ---\n\n **2): 检查mysql用户是否能成功执行备份任务***\n\n 手动调用crontab中的备份命令\n ```\n su mysql\n /usr/local/python/bin/mtlsbackup \n ```\n 输出如下：\n ```\n [2018-09-21 14:38:46,682] [mtlsbackup] [INFO] read config file /etc/mtlsbackup.cnf\n [2018-09-21 14:38:46,683] [mtlsbackup] [INFO] 开始检查 /database/backups/3306/data/ \n [2018-09-21 14:38:46,684] [mtlsbackup] [INFO] 开始检查 /database/backups/3306/log/ \n [2018-09-21 14:38:46,684] [mtlsbackup] [INFO] 开始检查 /database/backups/3306/temp/ \n [2018-09-21 14:38:46,684] [mtlsbackup] [INFO] 今天星期 4 根据配置文件中的备份计划，决定进行差异备份\n [2018-09-21 14:38:46,684] [mtlsbackup] [INFO] 进入差异备份流程\n [2018-09-21 14:38:46,684] [mtlsbackup] [INFO] 准备检查最近一次的全备是否成功...\n [2018-09-21 14:38:46,685] [mtlsbackup] [INFO] 检查最后一个备份集2018-09-12T02:16:49的可用性\n [2018-09-21 14:38:46,685] [mtlsbackup] [INFO] 检查/database/backups/3306/data/2018-09-12T02:16:49/2018-09-12T02:16:49-full.log\n [2018-09-21 14:38:46,686] [mtlsbackup] [WARNING] 检查到最后一个全备 备份成功\n [2018-09-21 14:38:46,686] [mtlsbackup] [INFO] 从xtrabackup_checkpoints文件中读到tolsn=2589231\n [2018-09-21 14:38:46,687] [mtlsbackup] [INFO] 使用如下命令对MySQL数据库进行差异备 /usr/local/xtrabackup/bin/xtrabackup --defaults-file=/etc/my.cnf --host=127.0.0.1 --port=3306 --user=backup --password=DX3906 --no-version-check --compress --compress-threads=4 --use-memory=200M --stream=xbstream --parallel=8 --backup --extra-lsndir=/database/backups/3306/log/2018-09-21T14:38:46 --target-dir=/database/backups/3306/temp/ --incremental --incremental-lsn=2589231 \u003e /database/backups/3306/data/2018-09-12T02:16:49/2018-09-21T14:38:46-diff.xbstream 2\u003e/database/backups/3306/data/2018-09-12T02:16:49/2018-09-21T14:38:46-diff.log \u0026\n ```\n 根据上面的输出可以知道mtlsbackup已经成功执行了，不过备份有没有成功这个两是要看一下xtrabckup的日志才行，从mtlsbackup.py的日志可以看到xtrabackup把日志保存到了/database/backups/3306/data/2018-09-12T02:16:49/2018-09-21T14:38:46-diff.log \n ```\n tail -2 /database/backups/3306/data/2018-09-12T02:16:49/2018-09-21T14:38:46-diff.log\n ```\n 输出如下：\n ```\n xtrabackup: Transaction log of lsn (2663437) to (2663446) was copied.\n 180921 14:38:48 completed OK!\n ```\n 说明备份成功了！\n\n ---\n\n **3): 查看备份文件**\n ```\n tree /database/backups/\n\n ```\n 输出如下：\n ```\n /database/backups/\n ├── 3306\n │   ├── data\n │   │   └── 2018-09-12T02:16:49\n │   │   ├── 2018-09-12T02:16:49-full.log\n │   │   ├── 2018-09-12T02:16:49-full.xbstream\n │   │   ├── 2018-09-21T02:12:33-diff.log\n │   │   ├── 2018-09-21T02:12:33-diff.xbstream\n │   │   ├── 2018-09-21T13:19:23-diff.log\n │   │   ├── 2018-09-21T13:19:23-diff.xbstream\n │   │   ├── 2018-09-21T14:38:32-diff.log\n │   │   ├── 2018-09-21T14:38:32-diff.xbstream\n │   │   ├── 2018-09-21T14:38:46-diff.log\n │   │   └── 2018-09-21T14:38:46-diff.xbstream\n │   ├── log\n │   │   ├── 2018-09-12T02:16:49\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   ├── 2018-09-21T02:12:33\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   ├── 2018-09-21T13:19:23\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   ├── 2018-09-21T14:38:32\n │   │   │   ├── xtrabackup_checkpoints\n │   │   │   └── xtrabackup_info\n │   │   └── 2018-09-21T14:38:46\n │   │   ├── xtrabackup_checkpoints\n │   │   └── xtrabackup_info\n │   └── temp\n └── mtlsbackup.log\n\n ```\n \n ---\n\n 5. ### 注意事项\n **1): mtlsbakup还在开发中目前只包装了xtrabackup,以后有时候会把meb,mysqldump都会包进去**\n\n ---\n\n **2): 如果你使用mysqltools-2.18.07.28之前的版本创建了mysql数据库，它默认是不会创建backup用户的，所以你应该为数据库手工创建backup用户，相关SQL在mysqltools/deploy/ansible/backup/template/create_backup_user.sql文件中用。如果你不想手工创建backup用户而是让mysqltools去创建，那么你要把config_backup.yaml中的create_user变量设置为1**\n\n---\n \n\n\n\n\n\n## 监控\n **mysqltools的出发点是以提升生产力为目标的、但凡能用电解决的事、就不要动用人力;我们的目标是认机器检测到问题后尽可能的自动解决掉它、解决完成后发个通知就行。这一切的基础是要有一套完善的监控系统，mysqltools在这方面使用的是zabbix这个开源解决方案。zabbix是一套通过的监控框架，要实现对MySQL我们还要自己写程序用来收集各个监控指标，好在我已经把程序写好了见** \u003ca href=\"https://github.com/Neeky/mysqltools-python\"\u003emysqltools-python \u003c/a\u003e\n\n ---\n \n 1. ### 目前已经实现的监控项\n 目前我已经用python实现了200+个监控项，可以把监控项大致分成四大块 1): variable 类型的监控项、这类监控项的作用是为了可以追踪参数变更后对MySQL各个方面的影响；\n 2): status 类型的监控项、这类监控项主要用于对实例实时信息的收集可用于提前发现问题、解决问题；3):ps 类型的监控项 这类监控项主要会对MySQL某一特定维度的监控、如\n 半同步复制是否正常、mysql group replication 集群是否正常等等 4):自动发现\n\n *监控项名* | *简介* | *采集方式* \n ----------------------------------|----------------------------------- |----------------------------------------------\n |`mysql配置(variable)相关的监控项列表`|如果人为修改了mysql参数(variable)并引起了问题、那么对关键参数的监控就能方便的定位问题\n |`-- ServerID` | 对应server_id | variable |\n |`-- BaseDir` | 对应basedir | variable |\n |`-- DataDir` | 对应datadir | variable |\n |`-- Port` | 对应port | variable |\n |`-- CharacterSetServer` | 对应character_set_server | variable |\n |`-- Socket` | 对应socket | variable |\n |`-- ReadOnly` | 对应readonly | variable |\n |`-- SkipNameResolve` | 对应skip_name_resolve | variable |\n |`-- LowerCaseTableNames` | 对应lower_case_table_names | variable |\n |`-- ThreadCacheSize` | 对应thread_cache_size 、线程池的大小、如果池有空闲的线程、那么新的连接就不单独创建新的线程了 |variable|\n |`-- TableOpenCache` | 对应table_open_cache | variable |\n |`-- TableDefinitionCache` | 对应table_definition_cache | variable |\n |`-- TableOpenCacheInstances` | 对应table_open_cache_instance | variable |\n |`-- MaxConnections` | 对应max_connections | variable |\n |`-- BinlogFormat` | 对应binlog_format | variable |\n |`-- LogBin` | 对应log_bin | variable |\n |`-- BinlogRowsQueryLogEvents` | 对应binlog_rows_query_log_events | variable |\n |`-- LogSlaveUpdates` | 对应log_slave_updates | variable |\n |`-- ExpireLogsDays` | 对应expire_logs_days | variable |\n |`-- BinlogCacheSize` | 对应binlog_cache_size | variable |\n |`-- SyncBinlog` | 对应sync_binlog | variable |\n |`-- ErrorLog` | 对应error_log | variable |\n |`-- GtidMode` | 对应gtid_mode | variable |\n |`-- EnforceGtidConsistency` | 对应enforce_gtid_consistency | variable |\n |`-- MasterInfoRepository` | 对应master_info_repository | variable |\n |`-- RelayLogInfoRepository` | 对应relay_log_info_repository | variable |\n |`-- SlaveParallelType` | 对应slave_parallel_type | variable |\n |`-- SlaveParallelWorkers` | 对应slave_parallel_workers | variable |\n |`-- InnodbDataFilePath` | 对应innodb_data_file_path | variable |\n |`-- InnodbTempDataFilePath` | 对应innodb_temp_data_file_path | variable |\n |`-- InnodbBufferPoolFilename` | 对应innodb_buffer_pool_filename | variable |\n |`-- InnodbLogGroupHomeDir` | 对应innodb_log_group_home_dir | variable |\n |`-- InnodbLogFilesInGroup` | 对应innodb_log_file_in_group | variable |\n |`-- InnodbLogFileSize` | 对应innodb_log_file_size | variable |\n |`-- InnodbFileformat` | 对应innodb_fileformat | variable |\n |`-- InnodbFilePerTable` | 对应innodb_file_per_table | variable |\n |`-- InnodbOnlineAlterLogMaxSize` | 对应innodb_online_Alter_log_max_size |variable |\n |`-- InnodbOpenFiles` | 对应innodb_open_files | variable |\n |`-- InnodbPageSize` | 对应innodb_page_size | variable |\n |`-- InnodbThreadConcurrency` | 对应innodb_thread_concurrency | variable |\n |`-- InnodbReadIoThreads` | 对应innodb_read_io_threads | variable |\n |`-- InnodbWriteIoThreads` | 对应innodb_write_io_threads | variable |\n |`-- InnodbPurgeThreads'` | 对应innodb_purge_threads | variable |\n |`-- InnodbLockWaitTimeout` | 对应innodb_lock_wait_timeout | variable |\n |`-- InnodbSpinWaitDelay` | 对应innodb_spin_wait_delay | variable |\n |`-- InnodbAutoincLockMode` | 对应innodb_autoinc_lock_mode | variable |\n |`-- InnodbStatsAutoRecalc` | 对应innodb_stats_auto_recalc | variable |\n |`-- InnodbStatsPersistent` | 对应innodb_stats_persistent | variable |\n |`-- InnodbStatsPersistentSamplePages` |对应innodb_stats_persistent_sample_pages | variable |\n |`-- InnodbBufferPoolInstances` | 对应innodb_buffer_pool_instances | variable |\n |`-- InnodbAdaptiveHashIndex` | 对应innodb_adaptive_hash_index | variable |\n |`-- InnodbChangeBuffering` | 对应innodb_change_buffering | variable |\n |`-- InnodbChangeBufferMaxSize` | 对应innodb_change_buffer_max_size | variable |\n |`-- InnodbFlushNeighbors` | 对应innodb_flush_neighbors | variable |\n |`-- InnodbFlushMethod` | 对应innodb_flush_method | variable |\n |`-- InnodbDoublewrite` | 对应innodb_doublewrite | variable |\n |`-- InnodbLogBufferSize` | 对应innodb_log_buffer_size | variable |\n |`-- InnodbFlushLogAtTimeout` | 对应innodb_flushLog_at_timeout | variable |\n |`-- InnodbFlushLogAtTrxCommit` | 对应innodb_flushLog_at_trx_commit | variable |\n |`-- InnodbBufferPoolSize` | 对应innodb_buffer_pool_size | variable |\n |`-- Autocommit` | 对应autocommit | variable |\n |`-- InnodbOldBlocksPct` | 对应innodb_lld_blocks_pct | variable |\n |`-- InnodbOldBlocksTime` | 对应innodb_old_blocks_time | variable |\n |`-- InnodbReadAheadThreshold` | 对应innodb_read_ahead_threshold | variable |\n |`-- InnodbRandomReadAhead` | 对应innodb_random_read_ahead | variable |\n |`-- InnodbBufferPoolDumpPct` | 对应innodb_buffer_pool_dump_pct | variable |\n |`-- InnodbBufferPoolDumpAtShutdown` |对应innodb_buffer_pool_dump_at_shutdown | variable |\n |*********************************| | |\n |`mysql状态(status)相关监控` | 通过对status进行监控可得知mysql当前的性能表现\n |`-- AbortedClients` | 对应aborted_clients 、client异常退出使得连接没有被正常关闭的次数 | status | \n |`-- AbortedConnects` | 对应borted_connects 、没有成功连接到server端的次数 | status |\n |`-- BinlogCacheDiskUse` | 对应binlog_cache_disk_use 、使用临时文件存储事务语句的次数 | status |\n |`-- BinlogCacheUse` | 对应binlog_cache_user 、使用binlog_cache存储事务语句的次数 | status |\n |`-- BinlogStmtCacheDiskUse` | 对应binlog_stmt_cache_disk_use 、非事务语句使用临时文件存储的次数 | status |\n |`-- BinlogStmtCacheUse` | 对应binlog_stmt_cache_use 、非事务语句使用binlog_cache存储的次数 | status |\n |`-- BytesReceived` | 对应bytes_received、从客户端收到的字节数 | status |\n |`-- BytesSent` | 对应bytes_sent、发送给客户端的字节数 | status |\n |`-- ComBegin` | 对应com_begin、 语句执行的次数 | status |\n |`-- ComCallProcedure` | 对应com_call_procedure、语句执行的次数 | status |\n |`-- ComChangeMaster` | 对应com_change_master、 语句执行的次数 | status |\n |`-- ComCommit` | 对应com_commit、 语句执行的次数 | status |\n |`-- ComDelete` | 对应com_delete、 语句执行的次数 | status |\n |`-- ComDeleteMulti` | 对应com_delete_multi、 语句执行的次数 | status |\n |`-- ComInsert` | 对应com_insert、 语句执行的次数 | status |\n |`-- ComInsertSelect` | 对应com_insert_select、 语句执行的次数 | status |\n |`-- ComSelect` | 对应com_select、 语句执行的次数 | status |\n |`-- ComUpdate` | 对应com_update、 语句执行的次数 | status |\n |`-- ComUpdateMulti` | 对应com_update_multi、 语句执行的次数 | status |\n |`-- Connections` | 对应connections、尝试连接的次数 | status |\n |`-- CreatedTmpDiskTable` | 对应created_tmp_disk_table、创建磁盘临时表的次数 | status |\n |`-- CreatedTmpFiles` | 对应created_tmp_files、创建临时文件的次数 | status |\n |`-- CreatedTmpTables` | 对应created_tmp_tables、创建临时表的次数 | status |\n |`-- InnodbBufferPoolDumpStatus` | 对应innodb_buffer_pool_dump_status innodb_xx_dump的进度 | status |\n |`-- InnodbBufferPoolLoadStatus` | 对应innodb_buffer_pool_load_status innodb_xx_load的进度 | status |\n |`-- InnodbBufferPoolResizeStatus`| 对应innodb_buffer_pool_resize_status 进度 | status |\n |`-- InnodbBufferPoolBytesData` | 对应innodb_buffer_pool_bytes_data buffer_pool中的数据量(单位字节) | status |\n |`-- InnodbBufferPoolPagesData` | 对应innodb_buffer_pool_pages_data buffer_pool中数据页面数 | status |\n |`-- InnodbBufferPoolPagesDirty` | 对应innodb_buffer_pool_pages_dirty buffer_pool中脏页数量 | status |\n |`-- InnodbBufferPoolBytesDirty` | 对应innodb_buffer_pool_bytes_dirty buffer_pool中脏数据量(单位字节) | status |\n |`-- InnodbBufferPoolPagesFlushed`| 对应innodb_buffer_pool_pages_flushed 请求刷新出buffer_pool的页面数 | status |\n |`-- InnodbBufferPoolPagesFree` | 对应innodb_buffer_pool_pages_free buffer_pool中空闲页面数 | status |\n |`-- InnodbBufferPoolPagesMisc` | 对应innodb_buffer_pool_pages_misc buffer_pool total_page -(free + data) | status |\n |`-- InnodbBufferPoolPagesTotal` | 对应innodb_buffer_pool_pages_total buffer_pool 总项目数 | status |\n |`-- InnodbBufferPoolReadAhead` | 对应innodb_buffer_pool_read_ahead 由read-ahead机制读入的页面数 | status |\n |`-- InnodbBufferPoolReadAheadEvicted` | 对应innodb_buffer_pool_read_ahead_evicted 由raed-ahead机制读入的页面中、由于读入后没有被访问而淘汰的页 面\n |`-- InnodbBufferPoolReadRequests`| 对应innodb_buffer_pool_read_requests 逻辑读的次数(读buffer_pool) | status |\n |`-- InnodbBufferPoolReads` | 对应innodb_buffer_pool_reads 物理读的次数(读磁盘) | status |\n |`-- InnodbBufferPoolWaitFree` | 对应innodb_buffer_pool_wait_free 等待有可用页面的次数 | status |\n |`-- InnodbBufferPoolWriteRequests`|对应innodb_buffer_pool_write_requests 请求写buffer_pool的次数 | status |\n |`-- InnodbDataFsyncs` | 对应innodb_data_fsyncs fsyncs()函数调用的次数 | status |\n |`-- InnodbDataPendingFsyncs` | 对应innodb_data_pending_fsyncs 当前挂起的fsyncs操作 | status |\n |`-- InnodbDataPendingReads` | 对应innodb_data_pending_reads 当前挂起的读操作 | status |\n |`-- InnodbDataPendingWrites` | 对应innodb_data_pending_writes 当前挂起的写操作 | status |\n |`-- InnodbDataRead` | 对应innodb_data_read 自启动后读了多少数据进buffer_pool | status |\n |`-- InnodbDataReads` | 对应innodb_data_reads 自启动后读了多少次数据进buffer_pool | status |\n |`-- InnodbDataWrites` | 对应innodb_data_writes 自启动后写了多少次数据到buffer_pool | status |\n |`-- InnodbDataWritten` | 对应innodb_data_written 自启动后写了多少数据到buffer_pool | status |\n |`-- InnodbDblwrPagesWritten` | 对应innodb_dblwr_pages_written double_write写入到磁盘的页面数量 | status |\n |`-- InnodbDblwrWrites` | 对应innodb_dblwr_writes double_write 执行的次数 | status |\n |`-- InnodbLogWaits` | 对应innodb_log_waits 写日志时的等待次数 | status |\n |`-- InnodbLogWriteRequests` | 对应innodb_log_write_requests 写请求次数 | status |\n |`-- InnodbLogWrites` | 对应innodb_log_writes 写磁盘的次数 | status |\n |`-- InnodbOsLogFsyncs` | 对应innodb_os_log_fsyncs fsync()函数调用的次数(针对redo log file) | status |\n |`-- InnodbOsLogPendingFsyncs` | 对应innodb_os_log_pending_fsyncs 挂起的fsync操作数量 | status |\n |`-- InnodbOsLogPendingWrites` | 对应innodb_os_log_pending_writes 挂起的write操作数量 | status |\n |`-- InnodbOsLogWritten` | 对应innodb_os_log_written 写入的字节数量 | status |\n |`-- InnodbPagesCreated` | 对应innodb_pages_created 创建的页面数量 | status |\n |`-- InnodbPagesRead` | 对应innodb_pages_read 从buffer_pool中读出的页面数量 | status |\n |`-- InnodbPagesWritten` | 对应innodb_pages_written 向buffer_pool写入的页面数量 | status |\n |`-- InnodbRowLockCurrentWaits` | 对应innodb_row_lock_current_waits 当前的行锁等待数量 | status |\n |`-- InnodbRowLockTime` | 对应innodb_row_lock_time 花费在获取行锁上的总时间 | status |\n |`-- InnodbRowLockTimeAvg` | 对应innodb_row_lock_time_avg 花费在获取行锁上的平均时间 | status |\n |`-- InnodbRowLockTimeMax` | 对应innodb_row_lock_time_max 花费在获取行锁上的最大时间 | status |\n |`-- InnodbRowLockWaits` | 对应innodb_row_lock_waits 行锁等待的总次数 | status |\n |`-- InnodbRowsDeleted` | 对应innodb_rows_deleted 删除的行数 | status |\n |`-- InnodbRowsInserted` | 对应innodb_rows_inserted 插入的行数 | status |\n |`-- InnodbRowsRead` | 对应innodb_rows_read 读取的行数 | status |\n |`-- InnodbRowsUpdated` | 对应innodb_rows_updated 更新的行数 | status |\n |`-- OpenTableDefinitions` | 对应open_table_definitions 缓存中的.frm文件数量 | status |\n |`-- OpenTables` | 对应open_tables 当前打开的表的数量 | status |\n |`-- OpenedTableDefinitions` | 对应opened_table_definitions 曾经缓存过的.frm文件数量 | status |\n |`-- OpenedTables` | 对应opened_tables 曾经打开过的表 | status |\n |`-- TableOpenCacheOverflows` | 对应table_open_cache_overflows 表打开又关闭的次数 | status |\n |`-- ThreadsCached` | 对应threads_cached 当前线程池中线程的数量 | status |\n |`-- ThreadsConnected` | 对应threads_connected 当前打开的连接 | status |\n |`-- ThreadsCreated` | 对应threads_created 为了处理连接所创建的线程总数 | status |\n |`-- ThreadsRunning` | 对应threads_running 非sleep状态下的线程数 | status |\n |`-- Uptime` | 对应uptime 从启动开始到现在已经运行了多少秒 | status | \n |`-- MgrTotalMemberCount` | mgr集群中成员的数量 | p_s |\n |`-- MgrOnLineMemberCount` | mgr集群中online状态下的成员数量 | p_s |\n |`-- MgrMemberState` | 当前mgr成员的状态 | p_s | \n |`-- MgrCountTransactionsInQueue` | 当前mgr成员上等待进行冲突检查的事务数量 | p_s |\n |`-- MgrCountTransactionsChecked` | 当前mgr成员上已经完成冲突检测的事务数量 | p_s |\n |`-- MgrCountConflictsDetected` | 当前mgr成员上没能通过冲突检测的事务数量 | p_s |\n |`-- MgrTransactionsCommittedAllMembers`|当前mgr成员上已经应用的事务总数量 | p_s |\n |`-- RplSemiSyncMasterClients` | 当前master端处理半同步状态的slave数量 | status |\n |`-- RplSemiSyncMasterStatus` | master的半同步状态 | status |\n |`-- RplSemiSyncMasterNoTx` | 没有收到半同步slave确认的事务数量 | status |\n |`-- RplSemiSyncMasterYesTx` | 有收到半同步slave确认的事务数量 | status |\n |`-- RplSemiSyncSlaveStatus` | slave的半同步状态 | status |\n |`-- SlaveIORunning` | IO线程的状态(-1:说明当前实例是master,0:非Yes,1:Yes) | show slave status |\n |`-- SlaveSQLRunning` | SQL线程的状态(-1:说明当前实例是master,0:非Yes,1:Yes) | show slave status |\n |`-- SecondsBehindMaster` | Seconds behind master | show slave status |\n\n ---\n\n\n 2. ### 监控项的人肉使用方法\n 前面说到监控项的采集程序**mysqltools-python**已经写好了的，直接用就行；这只是为了让你过一下手瘾事实上你是用不着这样的，监控是全自动的\n\n ```\n mtlsmonitor --user=monitor --password=monitor0352 --host=127.0.0.1 --port=3306 BinlogCacheDiskUse\n ```\n 输出如下\n ```\n 0\n ```\n\n ---\n\n 3. ### zabbix监控环境介绍\n **zabbix采用的是server/agent的架构设计，agent负责采集数据、数据经由可选的proxy(也就是说可以不经过proxy)上报到server；数据最终是保存到数据库中的、这里后台的数据库我选择用mysql；为了方便使用zabbix还自带了一个用php写的网站，这样用户就可以通过web界面来配置监控系统了，体验上来说要好不少。网站不直接与server交互它直接操作数据。**\n\n **从软件组件上看一套zabbix监控环境涉及到多少组件**\n \n 1): **linux** 主机 2): **apache(httpd)** 3): **mysql** 4): **php** 5):**zabbix-server** 6):**zabbix-agent** 7): **zabbix-proxy** \n \n 从涉及到的组件上看一个监控环境是比较复杂的；mysqltools自动化安装的默认行为是在zabbix-server主机上同时安装上mysql、apache(httpd)、php 当然为了有监控zabbix-server所在主机的能力mysqltools还会在zabbix-server主机上也安装上zabbix-agent\n\n ---\n\n 4. ### zabbix监控环境安装规划\n\n **安装规划**\n\n 主机名 | ip地址 | 角色\n ----------|----------------|-----\n sqlstudio | 172.16.192.101 | zabbix-server\n mysqldb | 172.16.192.128 | zabbix-agent\n\n ---\n\n 5. ### 安装用于保存监控数据的mysql数据库与pyton-3.6.2\n **mysql-8.0.xx 还刚出来；我在使用mysql-8.0.11这个版本时编译php出错了；最好还在这里用mysql-5.7.x版本的mysql**\n 见([mysql单机](#mysql单机))\n\n 见([用ansible安装python](#用ansible安装python)) 如果你是用mysqltools来安装python-3.6.2的话它默认为把mysqltools-python这个包也给你安装上，mysqltool-python的主要功能之一就是mysql监控数据采集，所以这个轮子是你的zabbix监控系统所必要的。\n \n ---\n\n 6. ### 改配置文件中zabbix_server_ip这个配置项\n **修改mysqltools/config.yaml配置文件中的zabbix_server_ip配置项的值为zabbix-server主机的ip地址**\n ```\n zabbix_server_ip: 172.16.192.101\n ```\n\n ---\n\n 7. ### 安装httpd\n **1):进入httpd的安装目录**\n ```\n cd mysqltools/deploy/ansible/httpd\n ```\n **2):修改install_httpd.yaml文件中的目标主机为zabbixstudio**\n ```\n ---\n - hosts: zabbixstudio\n vars_files:\n ```\n **3):安装httpd**\n ```\n ansible-playbook install_httpd.yaml\n ```\n 输出如下：\n ```\n PLAY [zabbixstudio] ***********************************************************************************************************\n \n TASK [Gathering Facts] ********************************************************************************************************\n ok: [zabbixstudio]\n \n TASK [install gcc] ************************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [install gcc-c++] ********************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [install pcre-devel] *****************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [openssl-devel] **********************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [expat-devel] ************************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [perl] *******************************************************************************************************************\n ok: [zabbixstudio]\n \n TASK [transfer apr-1.6.2.tar.gz to remote host] *******************************************************************************\n changed: [zabbixstudio]\n \n TASK [copy install script to remote] ******************************************************************************************\n changed: [zabbixstudio]\n \n TASK [install apr] ************************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/install_apr.sh] *********************************************************************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/apr-1.6.2] **************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [transfer apr-util-1.6.0.tar.gz to remote host] **************************************************************************\n changed: [zabbixstudio]\n \n TASK [copy install script to remote] ******************************************************************************************\n changed: [zabbixstudio]\n \n TASK [install apr_util] *******************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [clear /tmp/ directory] **************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [clear /tmp/ directory] **************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [copy httpd-2.4.28.tar.gz to remonte host] *******************************************************************************\n changed: [zabbixstudio]\n \n TASK [copy install scripts to remonte host] ***********************************************************************************\n changed: [zabbixstudio]\n \n TASK [install httpd] **********************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/install_httpd.sh] *******************************************************************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/httpd-2.4.28.tar.gz] ****************************************************************************************\n changed: [zabbixstudio]\n \n TASK [config httpd.service] ***************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [start httpd.service] ****************************************************************************************************\n changed: [zabbixstudio]\n \n TASK [enable httpd.service] ***************************************************************************************************\n changed: [zabbixstudio]\n \n PLAY RECAP ********************************************************************************************************************\n zabbixstudio : ok=25 changed=23 unreachable=0 failed=0\n ```\n **4):查看目标主机上的httpd进行种是否已经启动**\n ```\n ps -ef | grep httpd \n root 38860 1 0 15:11 ? 00:00:00 /usr/local/httpd/bin/httpd -DFOREGROUND \n daemon 38861 38860 0 15:11 ? 00:00:00 /usr/local/httpd/bin/httpd -DFOREGROUND \n daemon 38862 38860 0 15:11 ? 00:00:00 /usr/local/httpd/bin/httpd -DFOREGROUND \n daemon 38863 38860 0 15:11 ? 00:00:00 /usr/local/httpd/bin/httpd -DFOREGROUND \n daemon 43716 38860 0 15:15 ? 00:00:00 /usr/local/httpd/bin/httpd -DFOREGROUND \n ```\n **5):通过浏览器查看效果**\n \u003cimg src=\"./docs/imgs/httpd-0001.png\"/\u003e\n\n 8. ### 安装php\n **1):进入php的安装目录**\n ```\n cd mysqltools/deploy/ansible/php\n ```\n ---\n\n **2):修改install_php.yaml文件中的目标主机为zabbixstudio**\n ```\n ---\n - hosts: zabbixstudio\n vars_files:\n ```\n\n ---\n **3):安装php**\n ```\n ansible-playbook install_php.yaml \n ```\n 输出如下：\n ```\n PLAY [zabbixstudio] ************************************************************\n \n TASK [Gathering Facts] *********************************************************\n ok: [zabbixstudio]\n \n TASK [install gcc] *************************************************************\n ok: [zabbixstudio]\n \n TASK [install gcc-c++] *********************************************************\n ok: [zabbixstudio]\n \n TASK [install bzip2-devel] *****************************************************\n changed: [zabbixstudio]\n \n TASK [install libjpeg-devel] ***************************************************\n changed: [zabbixstudio]\n \n TASK [install libpng-devel] ****************************************************\n changed: [zabbixstudio]\n \n TASK [install freetype-devel] **************************************************\n changed: [zabbixstudio]\n \n TASK [install freetype-devel] **************************************************\n changed: [zabbixstudio]\n \n TASK [stop httpd.service] ******************************************************\n changed: [zabbixstudio]\n \n TASK [unarchive php-5.6.31.tar.gz to remonte host] *****************************\n changed: [zabbixstudio]\n \n TASK [copy install_php.sh to remonte host] *************************************\n changed: [zabbixstudio]\n \n TASK [install php] *************************************************************\n changed: [zabbixstudio]\n \n TASK [copy php.ini to remote] **************************************************\n changed: [zabbixstudio]\n \n TASK [remove /usr/local/httpd/htdocs/index.html] *******************************\n changed: [zabbixstudio]\n \n TASK [copy index.php to remote] ************************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/install_php.sh] **********************************************\n changed: [zabbixstudio]\n \n TASK [remove /tmp/php-5.6.31] **************************************************\n changed: [zabbixstudio]\n \n TASK [config httpd] ************************************************************\n changed: [zabbixstudio]\n \n TASK [start httpd.service] *****************************************************\n changed: [zabbixstudio]\n \n PLAY RECAP *********************************************************************\n zabbixstudio : ok=19 changed=16 unreachable=0 failed=0 \n ```\n ---\n\n **4):通过web页面查看是否安装成功**\n \u003cimg src=\"./docs/imgs/php-0001.png\"/\u003e\n\n ---\n\n 9. ### 安装zabbix服务端\n **1):配置mysqltools/config.yaml**\n\n mysqltools配置文件中的`zabbix_server_ip: xxx.xxx.xxx.xxx`用来指定zabbix_server主机的ip地址，这个地址会在zabbix_agent的配置文件中用到，由于zabbix_server所在的主机也是要监控的，所以在zabbix_server主机上也要安装上zabbix_agent。此外，为了方安装mysqltools会自动在zabbix_server所在的主机上安装上zabbix_agent。\n 根据上面规划提到的zabbix_server在172.16.192.101这个主机上，所以config.yaml的内容应该如下\n ```\n zabbix_server_ip: 172.16.192.101\n ```\n ---\n\n **2):修改install_zabbix_server.yaml文件中的目标主机为zabbixstudio**\n ```\n ---\n - hosts: zabbixstudio\n vars_files:\n ```\n\n ---\n\n **3):安装zabbix_server**\n \n ```\n cd mysqltools/deploy/ansible/zabbix\n ansible-playbook install_zabbix_server.yaml\n ```\n 输出如下：\n ```\n PLAY [zabbixstudio] ***********************************************************************************************************\n \n TASK [Gathering Facts] ********************************************************************************************************\n ok: [zabbixstudio]\n \n TASK [add zabbix user to system] **********************************************************************************************\n changed: [zabbixstudio]\n ....\n \n TASK [config zabbix_server start up on boot] **********************************************************************************\n changed: [zabbixstudio]\n \n PLAY RECAP ********************************************************************************************************************\n zabbixstudio : ok=32 changed=26 unreachable=0 failed=0 \n ```\n\n ---\n\n **4):通过web配置zabbix**\n\n 1、访问zabbix-web界面 \n\n \u003cimg src=\"./docs/imgs/zabbix-0001.png\"/\u003e\n\n ---\n\n 2、下一步\n\n \u003cimg src=\"./docs/imgs/zabbix-0002.png\"/\u003e\n\n ---\n\n 3、配置zabbix管理界面连接后台数据库的方式\n\n \u003cimg src=\"./docs/imgs/zabbix-0003.png\"/\u003e\n 这里的password指的是数据库端zabbix用户的密码、这个东西引用的是config.yaml中的`mysql_zabbix_password`这个配置项\n\n 然后就一直点“next step ”\n\n ---\n\n 4、登录到zabbix-web\n\n \u003cimg src=\"./docs/imgs/zabbix-0004.png\"/\u003e\n\n 这里的账号密码是死的，账号名：`admin` 密码：`zabbix`\n \n ---\n\n 5、zabbix-web 主页\n\n \u003cimg src=\"./docs/imgs/zabbix-0005.png\"/\u003e\n\n ---\n\n 10. ### 安装zabbix客户端\n **1): 和安装服务端一样只不过把yaml文件换成install_zabbix_agent.yaml**\n ```\n ansible-playbook install_zabbix_agent.yaml \n ```\n 输出如下：\n ```\n PLAY [sqlstudio] *******************************************************************************\n TASK [Gathering Facts] *************************************************************************\n ok: [sqlstudio]\n TASK [transfer zabbix install package to remote host and unarchive to /tmp/] *******************\n changed: [sqlstudio]\n TASK [transfer install script to remonte host] *************************************************\n changed: [sqlstudio]\n TASK [install zabbix_agent_node] ***************************************************************\n changed: [sqlstudio]\n TASK [change owner to zabbix user] *************************************************************\n changed: [sqlstudio]\n TASK [make link] *******************************************************************************\n changed: [sqlstudio]\n TASK [transfer zabbix config file to remonte host] *********************************************\n changed: [sqlstudio]\n TASK [remove /tmp/install_zabbix_agent.sh] *****************************************************\n changed: [sqlstudio]\n TASK [remove /tmp/zabbix-3.4.3] ****************************************************************\n changed: [sqlstudio]\n TASK [export path env variable] ****************************************************************\n ok: [sqlstudio]\n TASK [export path env to /root/.bashrc] ********************************************************\n ok: [sqlstudio]\n TASK [transfer monitor script to remonte host] *************************************************\n changed: [sqlstudio]\n TASK [config file mode] ************************************************************************\n changed: [sqlstudio]\n TASK [transfer mtls.conf to remonte] ***********************************************************\n changed: [sqlstudio]\n TASK [config mtls.conf's mode] *****************************************************************\n ok: [sqlstudio]\n TASK [config zabbix_agent systemd] *************************************************************\n ok: [sqlstudio]\n TASK [start zabbix_agent] **********************************************************************\n changed: [sqlstudio]\n TASK [config zabbix_agent start up on boot] ****************************************************\n ok: [sqlstudio]\n PLAY RECAP *************************************************************************************\n sqlstudio : ok=28 changed=12 unreachable=0 failed=0 \n ```\n \n 11. ### 通过mysqltools中给出的模板来监控mysql\n **mysqltools提供了对MySQL单机的监控模板、把模板导入再关联到你的MySQL主机就能完成监控项的收集与画图了**\n\n \u003cimg src=\"./docs/imgs/mysql_basic_screan-0001.png\"/\u003e\n\n --- \n \n 12. ### mysqltools中定义的MySQL的监控模板\n **1): zbx_export_mysql_basic_templates.xml** 针对MySQL单实例的各个维度进行监控\n\n **2): zbx_export_mysql_replication_templates.xml** 继承了zbx_export_mysql_basic_templates.xml中定义的监控项并扩展了在replication方面的监控\n ---\n \n\n\n\n## lnmp\n **lnmp指的是:** linux + nginx + mysql + python 可以用这些组件来搭建django框架写成的网站；(我们将在一台机器上集齐所有组件)\n\n **主机名** | **ip地址** | **系统版本** | 软件 |\n -------------:|:-------------------|--------------|-------------|\n uwsgiweb | 172.16.192.133 |centos-7.4 | linux + nginx + mysql + python3.6.x + uwsgi + django |\n\n\n\n 1. ### 安装mysql单机\n [mysql单机](#mysql单机)\n 2. ### 用ansible安装python\n **mysqltools在自动化安装python的时候还会自动安装下些python包 django2.0.x、mysqlclient、uwsgi**\n\n **1):修改deploy/ansible/python/install_python.yaml文件中的hosts**\n ```\n ---\n - hosts: uwsgiweb\n ```\n\n **2):安装python**\n ```\n ansible-playbook install_python.yaml\n ```\n 输出如下\n ```\n PLAY [uwsgiweb] *********************************************************************************\n \n TASK [Gathering Facts] **************************************************************************\n ok: [uwsgiweb]\n \n ... ... ... ... ... ... \n\n TASK [install python-3.6.2] *********************************************************************\n changed: [uwsgiweb]\n \n TASK [install mysqlclient and mysql-connector-python] *******************************************\n changed: [uwsgiweb]\n \n TASK [install django-2.0.4] *********************************************************************\n changed: [uwsgiweb]\n \n TASK [install uwsgi] ****************************************************************************\n changed: [uwsgiweb]\n \n TASK [install bs4] ******************************************************************************\n changed: [uwsgiweb]\n \n TASK [install requests-2.18.4] ******************************************************************\n changed: [uwsgiweb]\n \n PLAY RECAP **************************************************************************************\n uwsgiweb : ok=39 changed=22 unreachable=0 failed=0 \n ```\n 我们要建设的是lnmp环境，所以mysqltools还会把django、uwsgi 这两个包安装上去、同时运行起uwsgi\n\n ```\n ps -ef | grep uwsgi\n ```\n ```\n uwsgi 40729 1 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40731 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40732 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40733 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40734 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40735 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40736 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40737 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n uwsgi 40738 40729 0 13:04 ? 00:00:00 /usr/local/python/bin/uwsgi --ini=/etc/uwsgi.cnf \n ```\n 由上面可以看出uwsgi成功启动了\n\n 3. ### 安装nginx\n nginx的安装与配置也都是自动化的\n\n **1):修改deploy/ansible/nginx/install_nginx.yaml的hosts 变量**\n ```\n ---\n - hosts: uwsgiweb\n ```\n **2):安装nginx**\n ```\n ansible-playbook install_nginx.yaml \n\n ```\n 输出如下：\n ```\n PLAY [uwsgiweb] *********************************************************************************\n \n TASK [Gathering Facts] **************************************************************************\n ok: [uwsgiweb]\n ... ... ... ... ... ... ... ... ... ...\n \n TASK [install nginx] ****************************************************************************\n changed: [uwsgiweb]\n \n TASK [config nginx] *****************************************************************************\n changed: [uwsgiweb]\n \n TASK [create systemd config file for nginx] *****************************************************\n ok: [uwsgiweb]\n \n TASK [start nginx(sytemctl)] ********************************************************************\n changed: [uwsgiweb]\n \n TASK [config nginx.service start up on boot] ****************************************************\n ok: [uwsgiweb]\n \n PLAY RECAP **************************************************************************************\n uwsgiweb : ok=17 changed=8 unreachable=0 failed=0 \n ```\n **3):检查nginx是否运行**\n ```\n ps -ef | grep nginx\n ```\n ``` \n root 45536 1 0 13:25 ? 00:00:00 nginx: master process /usr/local/nginx/sbin/nginx \n nginx 45537 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45538 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45539 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45540 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45541 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45542 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45543 45536 0 13:25 ? 00:00:00 nginx: worker process \n nginx 45544 45536 0 13:25 ? 00:00:00 nginx: worker process \n ```\n\n 4. ### 查看效果\n 查看环境是否正确配置\n \u003cimg src=\"./docs/imgs/lnmp-0001.png\"/\u003e\n\n\n\n\n\n\n\n\n\n\n\n\n","funding_links":[],"categories":["zabbix","Python"],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FNeeky%2Fmysqltools","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FNeeky%2Fmysqltools","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FNeeky%2Fmysqltools/lists"}