腾讯TDSQL数据库集群环境安装文档

作者：mobiledu2502877697 | 来源：互联网 | 2023-06-06 13:09

**腾讯TDSQL数据库集群环境安装文档**#使用4台虚拟机搭建了TDSQL数据库集群环境，在搭建过程中，遇到了一些问题，在文章中对遇到的这些问题的处理方式进行了详解的介绍，希

一、环境介绍

在4台服务器上的/etc/hosts上配置上相应的ip地址和主机名的映射

配置互信

配置互信，保证tdsql4可以将安装包无密码推送到其它服务器上

[root@tdsql4 ~]# ssh-keygen -t rsa [root@tdsql4 ~]# ssh-copy-id tdsql1 [root@tdsql4 ~]# ssh-copy-id tdsql2 [root@tdsql4 ~]# ssh-copy-id tdsql3 [root@tdsql4 ~]# ssh-copy-id tdsql4

规划磁盘

在每个服务器上配置/data,/data1两个磁盘，每个磁盘至少100G

[root@tdsql4 ~]# df -h Filesystem Size Used Avail Use% Mounted on /dev/mapper/centos-root 20G 6.9G 14G 35% / devtmpfs 7.9G 0 7.9G 0% /dev tmpfs 7.9G 0 7.9G 0% /dev/shm tmpfs 7.9G 8.9M 7.9G 1% /run tmpfs 7.9G 0 7.9G 0% /sys/fs/cgroup /dev/sr0 4.3G 4.3G 0 100% /mnt/cdrom /dev/sda1 5.0G 132M 4.9G 3% /boot /dev/mapper/vg-data1 100G 33M 100G 1% /data1 /dev/mapper/vg-data 100G 586M 100G 1% /data tmpfs 1.6G 0 1.6G 0% /run/user/0

安装包

安装TDSQL数据库集群环境，主要用到以下3个安装包

[root@tdsql4 ~]# ls ansible_install tdsql_install tdsql_packet
二、配置安装前的环境

[root@tdsql4 script]# ls -l ansible_install/script/ total 20 -rw-r--r--. 1 root root 2390 Jun 10 2020 init_os_for_apt_install.sh -rw-r--r--. 1 root root 1513 Jun 10 2020 init_os_for_yum_install.sh -rw-r--r--. 1 root root 1058 Jun 10 2020 install_ansible.sh -rw-r--r--. 1 root root 266 Jun 10 2020 install_python3_for_arm.sh -rw-r--r--. 1 root root 253 Jun 10 2020 install_python3_for_x86.sh

安装必要的系统依赖包

将系统盘挂载到挂载点后，配置好yum环境，执行以下脚本来安装TDSQL环境必要的系统包

[root@tdsql4 script]# sh init_os_for_yum_install.sh

安装python3工具，并使其生效

[root@tdsql4 script]# sh install_python3_for_x86.sh [root@tdsql4 script]# source /etc/profile [root@tdsql4 script]# python3 --version Python 3.8.2

安装ansible工具

[root@tdsql4 script]# sh install_ansible.sh
三、开始TDSQL的安装

TDSQL的主要配置文件

[root@tdsql4 ~]# ls /root/tdsql_install group_files group_vars playbooks roles tdsql_hosts tdsql_update_hosts

TDSQL的主要配置文件只有2个，一个是tdsql_hosts，另一个是group_vars/all文件。

配置tdsql_hosts文件

备注：此文件为安装TDSQL环境的主要文件之一

[root@tdsql4 tdsql_install]# cat /root/tdsql_install/tdsql_hosts [tdsql_allmacforcheck] tdsql_tdsql1 ansible_ssh_host=90.90.90.221 tdsql_tdsql2 ansible_ssh_host=90.90.90.222 tdsql_tdsql3 ansible_ssh_host=90.90.90.223 tdsql_tdsql4 ansible_ssh_host=90.90.90.224 [tdsql_zk] tdsql_zk1 ansible_ssh_host=90.90.90.221 tdsql_zk2 ansible_ssh_host=90.90.90.222 tdsql_zk3 ansible_ssh_host=90.90.90.223 [tdsql_scheduler] tdsql_scheduler1 ansible_ssh_host=90.90.90.222 tdsql_scheduler2 ansible_ssh_host=90.90.90.223 [tdsql_oss] tdsql_oss1 ansible_ssh_host=90.90.90.222 tdsql_oss2 ansible_ssh_host=90.90.90.223 [tdsql_chitu] tdsql_chitu1 ansible_ssh_host=90.90.90.222 tdsql_chitu2 ansible_ssh_host=90.90.90.223 [tdsql_monitor] tdsql_monitor1 ansible_ssh_host=90.90.90.222 tdsql_monitor2 ansible_ssh_host=90.90.90.223 [tdsql_db] tdsql_db1 ansible_ssh_host=90.90.90.221 tdsql_db2 ansible_ssh_host=90.90.90.222 tdsql_db3 ansible_ssh_host=90.90.90.223 [tdsql_proxy] tdsql_proxy1 ansible_ssh_host=90.90.90.221 tdsql_proxy2 ansible_ssh_host=90.90.90.222 tdsql_proxy3 ansible_ssh_host=90.90.90.223 [tdsql_hdfs] tdsql_hdfs1 ansible_ssh_host=90.90.90.224 [tdsql_lvs] tdsql_lvs1 ansible_ssh_host=90.90.90.222 tdsql_lvs2 ansible_ssh_host=90.90.90.223 [tdsql_kafka] tdsql_kafka1 ansible_ssh_host=90.90.90.221 tdsql_kafka2 ansible_ssh_host=90.90.90.222 tdsql_kafka3 ansible_ssh_host=90.90.90.223 [tdsql_consumer] tdsql_consumer1 ansible_ssh_host=90.90.90.224 [tdsql_es] tdsql_es1 ansible_ssh_host=90.90.90.224 [tdsql_intercity] tdsql_intercity1 ansible_ssh_host=90.90.90.224 [tdsql_newdb] tdsql_newdb1 ansible_ssh_host=1.1.1.1 tdsql_newdb2 ansible_ssh_host=2.2.2.2 tdsql_newdb3 ansible_ssh_host=3.3.3.3 [tdsql_ansible_test] tdsql_ansible_test1 ansible_ssh_host=1.1.1.1 tdsql_ansible_test2 ansible_ssh_host=2.2.2.2 tdsql_ansible_test3 ansible_ssh_host=3.3.3.3

配置group_vars/all文件

备注：此文件为安装TDSQL环境的主要文件之一

[root@tdsql4 ]# cat /root/tdsql_install/group_vars/all --- tdsql_env_cpu: x86 tdsql_env_os: yum_install tdsql_zk_num: 3 tdsql_sche_netif: eth0 //修改为当前IP地址所在的网卡名称 tdsql_os_pass: a+complex+password tdsql_zk_rootdir: /tdsqlzk tdsql_metadb_ip: 10.235.49.132 tdsql_metadb_port: 15001 tdsql_metadb_ip_bak: 10.240.179.3 tdsql_metadb_port_bak: 15001 tdsql_metadb_user: hanlon tdsql_metadb_password: hanlon tdsql_hdfs_num: 1 //hdfs节点，根据前期规划，配置个数为1 tdsql_hdfs_ssh: 22 //hdfs的ssh端口，设置为22 tdsql_hdfs_datadir: /data2/hdfs,/data3/hdfs,/data4/hdfs tdsql_hdfsdatadir_count: 3 tdsql_kafka_logdir: /data2/kafka,/data3/kafka,/data4/kafka tdsql_es_mem: 2 //es节点的内存大小，配置为2GB tdsql_es_log_days: 7 tdsql_es_base_path: /data/application/es-install/es tdsql_intercity_zkrootdir: /tdsqlcross tdsql_intercity_netif: eth0 tdsql_zk_clientport: 2118 tdsql_zk_serverport1: 2338 tdsql_zk_serverport2: 2558 tdsql_oc_server_pass: K2JatUv5llBbMrske/k2YbqC tdsql_oc_client_pass: LGhVs0v5nVxcOLQie/k9bb2I tdsql_clouddba_metadb_pass: h5Wyg2Xy tdsql_oss_metadb_pass: q9Cjn0Wl

对配置文件中的口令进行加密

此步骤必须执行，虽然上面的配置文件没有修改口令，但是在每次安装组件前建议执行下口令加密脚本

[root@tdsql4 tdsql_install]# sh playbooks/tdsql_encrypt.sh

安装组件DB,chitu,oss,proxy,scheduler,tdsql,zk

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_part1_site.yml PLAY [checkenv server] *********************************************************************************************************** ....................................................... ....................................................... ................................................ PLAY RECAP *********************************************************************************************************************** tdsql_chitu1 : ok=28 changed=18 unreachable=0 failed=0 skipped=2 rescued=0 ignored=0 tdsql_chitu2 : ok=27 changed=17 unreachable=0 failed=0 skipped=2 rescued=0 ignored=0 tdsql_db1 : ok=10 changed=10 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_db2 : ok=10 changed=10 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_db3 : ok=10 changed=10 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_oss1 : ok=12 changed=11 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_oss2 : ok=5 changed=4 unreachable=0 failed=1 skipped=0 rescued=0 ignored=0 tdsql_proxy1 : ok=10 changed=9 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_proxy2 : ok=10 changed=9 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_proxy3 : ok=10 changed=9 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_scheduler1 : ok=18 changed=18 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_scheduler2 : ok=14 changed=14 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_tdsql1 : ok=41 changed=39 unreachable=0 failed=0 skipped=5 rescued=0 ignored=0 tdsql_tdsql2 : ok=37 changed=35 unreachable=0 failed=0 skipped=5 rescued=0 ignored=0 tdsql_tdsql3 : ok=37 changed=35 unreachable=0 failed=0 skipped=5 rescued=0 ignored=0 tdsql_tdsql4 : ok=37 changed=33 unreachable=0 failed=0 skipped=5 rescued=0 ignored=0 tdsql_zk1 : ok=11 changed=11 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_zk2 : ok=10 changed=10 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_zk3 : ok=10 changed=10 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

备注：此安装过程持续的时间比较长，大概是10分钟左右，中途不需要人为干预，安装完成后，保证每一个组件的failed=0.

我在安装的过程中，出现了tdsql_oss组件failed=1的情况，这个的原因是由于我使用了vmware平台上的虚拟主机环境，环境存在一些差异，其它人使用腾讯的云主机CVM环境，均未出现过failed=1的情况。

解决办法：

{

当tdsql_oss组件的failed=1时，所以此组件安装失败了。登录到失败节点上，通用ps -ef | grep oss发现并未有oss的进程，此时需要手工修改oss的配置文件，然后手工启动OSS

此处存在failed=1的oss节点未tdsql_oss2，在前面的规划中此节点位于90.90.90.223服务器上

[tdsql@tdsql3 ]$ cat /data/application/oss/conf/scheduler.xml [root@tdsql3 conf]# more scheduler.xml

然后手工修改为当前节点的IP地址，如下

[root@tdsql3 conf]# more scheduler.xml

然后手工启动OSS服务

[tdsql@tdsql3 ]$ /data/application/oss/bin/oss_server

}

四、初始化赤兔管理平台

登录http://90.90.90.222/tdsqlpcloud界面，进行初始化

环境检测

保证每个组件的检查结果为通过，然后点击下一步

集群接入

在集群命名处填写自定义的集群名称

OSS服务列表处填写前期规划的OSS地址和端口

OSS服务的端口可以从OSS的配置文件中获取

[root@tdsql2 conf]# cat /data/application/oss/conf/scheduler.xml "1" />

然后点击测试服务连接，则会自动检测出Zookeeper列表信息和 Zookeeper节点信息。确认无误后点击下一步。

集群初始化

配置IDC信息

添加IDC，机型规格，设备资源，网关组信息

配置机型规格

备注：机型的名字要全部大写，DB机型的资源配置尽量高一些，PROXY机型的资源可以小一些

配置设备资源

之后点击设备资源的"上报DB资源"依次添加3个DB资源

之后点击设备资源的"上报网关资源"依次添加3个网关资源

配置资源池(网关组)

配置资源池，由于前面配置了3个IDC，每个IDC里1个DB服务资源，所以在配置资源池的时候，选在"从3个IDC中，取3台机器"

创建赤兔的数据库实例

备注：此过程大概会持续4-5分钟的时间

待数据库实例创建成功后，需要对实例进行初始化

配置数据库

输入密码tdsqlcloud，后点击测试数据库连接，提示"数据库测试成功"后点击下一步[开始安装]

至此，TDSQL运营平台，赤兔就配置好了，初始用户名为admin，初始密码为123456

使用mysql客户端测试下是初始化的数据库实例是否正常，如果数据库正常的话，则就可以配置组件信息了。

连接的地址为proxy的地址，端口为proxy的端口15001，此端口可通过proxy的配置文件/data/tdsql_run/15001/gateway/conf/instance_15001.cnf获取

[root@tdsql2 conf]# more instance_15001.cnf cl="0" />

连接测试

[root@tdsql4 playbooks]# mysql -h90.90.90.221 -P15001 -utdsqlpcloud -ptdsqlpcloud Welcome to the MariaDB monitor. Commands end with ; or \g. Your MySQL connection id is 4393 Server version: 5.7.30-33-V2.0R630D001-v17-20200522-1443-log Source distribution Copyright (c) 2000, 2018, Oracle, MariaDB Corporation Ab and others. Type 'help;' or '\h' for help. Type '\c' to clear the current input statement. MySQL [(none)]> select 1; +---+ | 1 | +---+ | 1 | +---+ 1 row in set (0.00 sec) [root@tdsql4 playbooks]# mysql -h90.90.90.222 -P15001 -utdsqlpcloud -ptdsqlpcloud Welcome to the MariaDB monitor. Commands end with ; or \g. Your MySQL connection id is 4441 Server version: 5.7.30-33-V2.0R630D001-v17-20200522-1443-log Source distribution Copyright (c) 2000, 2018, Oracle, MariaDB Corporation Ab and others. Type 'help;' or '\h' for help. Type '\c' to clear the current input statement. MySQL [(none)]> select 1; +---+ | 1 | +---+ | 1 | +---+ 1 row in set (0.00 sec) [root@tdsql4 playbooks]# mysql -h90.90.90.223 -P15001 -utdsqlpcloud -ptdsqlpcloud Welcome to the MariaDB monitor. Commands end with ; or \g. Your MySQL connection id is 4455 Server version: 5.7.30-33-V2.0R630D001-v17-20200522-1443-log Source distribution Copyright (c) 2000, 2018, Oracle, MariaDB Corporation Ab and others. Type 'help;' or '\h' for help. Type '\c' to clear the current input statement. MySQL [(none)]> select 1; +---+ | 1 | +---+ | 1 | +---+ 1 row in set (0.00 sec)
五、配置赤兔上的各个组件

修改group_vars/all配置文件

修改group_vars/all配置文件，更新以下内容

tdsql_metadb_port: 15001 #赤兔监控库的proxy 主port tdsql_metadb_ip_bak: 90.90.90.222 #赤兔监控库的proxy 备IP tdsql_metadb_port_bak: 15001 #赤兔监控库的proxy 备port tdsql_metadb_user: tdsqlpcloud #赤兔监控库的用户名 tdsql_metadb_password: tdsqlpcloud #赤兔监控库的密码

对配置文件中的口令进行加密

此步骤必须执行，对上面的配置文件中的口令进行加密

[root@tdsql4 tdsql_install]# sh playbooks/tdsql_encrypt.sh

配置DB,chitu,oss,proxy,scheduler,tdsql,zk的信息

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_part2_site.yml PLAY [install allmac beginning] ************************************************************************************************** TASK [tdsql_beginning : create the directory] ************************************************************************************ [WARNING]: Consider using the file module with state=directory rather than running 'mkdir'. If you need to use command because file is insufficient you can add 'warn: false' to this command task or set 'command_warnings=False' in ansible.cfg to get rid of this message. changed: [tdsql_tdsql1] changed: [tdsql_tdsql3] changed: [tdsql_tdsql2] changed: [tdsql_tdsql4] PLAY RECAP *********************************************************************************************************************** tdsql_chitu1 : ok=18 changed=16 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_chitu2 : ok=5 changed=4 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_monitor1 : ok=13 changed=13 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_monitor2 : ok=12 changed=12 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_oss1 : ok=12 changed=9 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_oss2 : ok=5 changed=3 unreachable=0 failed=1 skipped=0 rescued=0 ignored=0 tdsql_scheduler1 : ok=30 changed=29 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_scheduler2 : ok=15 changed=15 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_tdsql1 : ok=31 changed=19 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_tdsql2 : ok=28 changed=16 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_tdsql3 : ok=28 changed=16 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_tdsql4 : ok=28 changed=14 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0

确保每个组件的配置信息failed=0,如上图出现的tdsql_oss2的failed=1,则处理方法详见上面的提示.

安装配置hdfs组件

[root@tdsql4 ]#cat /root/tdsql_install/tdsql_hosts .................... [tdsql_hdfs] tdsql_hdfs1 ansible_ssh_host=90.90.90.224 ................... [root@tdsql4 ]#cat /root/tdsql_install/group_vars/all ...................... tdsql_hdfs_num: 1 tdsql_hdfs_ssh: 22 tdsql_hdfs_datadir: /data2/hdfs,/data3/hdfs,/data4/hdfs tdsql_hdfsdatadir_count: 3 .....................

创建必要的目录(如果目录不手工创建的话，在安装的时候也会自动创建的)

[root@tdsql4 ~]# mkdir -p /data2/hdfs /data3/hdfs /data4/hdfs [root@tdsql4 ~]# chown tdsql:users -R /data2/hdfs /data3/hdfs /data4/hdfs

安装hdfs组件

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_hdfs_single.yml PLAY [install hdfs beginning] ***************************************************************************************************************************************** ....................................... PLAY RECAP ************************************************************************************************************************************************************ tdsql_db1 : ok=42 changed=27 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_db2 : ok=39 changed=25 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_db3 : ok=39 changed=25 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_hdfs1 : ok=49 changed=34 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0

格式化hdfs

[root@tdsql4 ~]# su - tdsql [tdsql@tdsql4 ~]$ hdfs namenode -format WARNING: /data/hadoop/pids does not exist. Creating. WARNING: /data/home/tdsql/hadoop/logs does not exist. Creating. Formatting using clusterid: CID-9f486d22-bbac-46ff-9191-754ef7508e33

启动hdfs服务

[tdsql@tdsql4 ~]$ hdfs --daemon start namenode [tdsql@tdsql4 ~]$ hdfs --daemon start datanode

测试hdfs服务

Found 1 items drwxr-xr-x - tdsql supergroup 0 2021-02-22 18:38 /tdsqlbackup

安装配置lvs

备注：lvs 机器建议单独部署，不建议和别人复用（最多只能和 zk 复用）

[root@tdsql4 ]#cat /root/tdsql_install/tdsql_hosts ......................... [tdsql_lvs] tdsql_lvs1 ansible_ssh_host=90.90.90.222 tdsql_lvs2 ansible_ssh_host=90.90.90.223 ......................

安装lvs组件

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_lvs.yml PLAY [install lvs beginning] ***************************************************************************************************** TASK [tdsql_beginning : create the directory] ************************************************************************************ [WARNING]: Consider using the file module with state=directory rather than running 'mkdir'. If you need to use command because file is insufficient you can add 'warn: false' to this command task or set 'command_warnings=False' in ansible.cfg to get rid of this message. changed: [tdsql_lvs1] changed: [tdsql_lvs2] .............................. .............................. .............................. PLAY RECAP *********************************************************************************************************************** tdsql_lvs1 : ok=41 changed=26 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_lvs2 : ok=34 changed=19 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0

安装配置kafka

[root@tdsql4 ]#cat /root/tdsql_install/tdsql_hosts ......................... [tdsql_kafka] tdsql_kafka1 ansible_ssh_host=90.90.90.221 tdsql_kafka2 ansible_ssh_host=90.90.90.222 tdsql_kafka3 ansible_ssh_host=90.90.90.223 ...................... [root@tdsql4 ]#cat /root/tdsql_install/group_vars/all ...................... tdsql_kafka_logdir: /data2/kafka,/data3/kafka,/data4/kafka .....................

创建必要的目录(如果目录不手工创建的话，在安装的时候也会自动创建的)

[root@tdsql4 ~]# mkdir -p /data2/kafka /data3/kafka /data4/kafka [root@tdsql4 ~]# chown tdsql:users -R /data2/kafka /data3/kafka /data4/kafka

安装kafka组件

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_kafka.yml PLAY [install kafka beginning] **************************************************************************************************************************************** ........................ ....................... ...................... TASK [tdsql_kafka : start kafka service] ****************************************************************************************************************************** changed: [tdsql_kafka2] PLAY RECAP ************************************************************************************************************************************************************ tdsql_kafka1 : ok=41 changed=26 unreachable=0 failed=1 skipped=1 rescued=0 ignored=0 tdsql_kafka2 : ok=40 changed=25 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0 tdsql_kafka3 : ok=35 changed=20 unreachable=0 failed=1 skipped=1 rescued=0 ignored=0

备注：安装完成后，保证每一个组件的failed=0.

我在安装的过程中，出现了tdsql_kafka1和tdsql_kafka3组件failed=1的情况，这个的原因是由于我使用了vmware平台上的虚拟主机环境，环境存在一些差异，其它人使用腾讯的云主机CVM环境，均未出现过failed=1的情况。

解决办法：

{

此时需要手工修改kafka的配置文件，然后手工启动kafka

登录到failed=1的kafka节点上，修改配置文件/data/application/kafka/config/server.properties

kafka节点1：

[tdsql@tdsql1 ]$ cat /data/application/kafka/config/server.properties ................................. # A comma seperated list of directories under which to store log files #log.dirs=/tmp/kafka-logs log.dirs=/data2/kafka,/data3/kafka,/data4/kafka # The port the socket server listens on #port=9092 # Hostname the broker will bind to. If not set, the server will bind to all interfaces host.name=wait_for_change_hostname # Hostname the broker will advertise to producers and consumers. If not set, it uses the # value for "host.name" if configured. Otherwise, it will use the value returned from # java.net.InetAddress.getCanonicalHostName(). advertised.host.name=wait_for_change_hostname ..............................

更新为

# A comma seperated list of directories under which to store log files #log.dirs=/tmp/kafka-logs log.dirs=/data2/kafka,/data3/kafka,/data4/kafka # The port the socket server listens on #port=9092 # Hostname the broker will bind to. If not set, the server will bind to all interfaces host.name=90.90.90.221 # Hostname the broker will advertise to producers and consumers. If not set, it uses the # value for "host.name" if configured. Otherwise, it will use the value returned from # java.net.InetAddress.getCanonicalHostName(). advertised.host.name=90.90.90.221

然后对kafka目录设置属组，并启动kafka服务

[root@tdsql1 ~]# chown -R tdsql:users /data/application/kafka [tdsql@tdsql1 bin]$ cd /data/application/kafka/bin [tdsql@tdsql1 bin]$ ./kafka-server-start.sh -daemon ../config/server.properties

kafka节点3：

[tdsql@tdsql3 ]$ cat /data/application/kafka/config/server.properties ................................. # A comma seperated list of directories under which to store log files #log.dirs=/tmp/kafka-logs log.dirs=/data2/kafka,/data3/kafka,/data4/kafka # The port the socket server listens on #port=9092 # Hostname the broker will bind to. If not set, the server will bind to all interfaces host.name=wait_for_change_hostname # Hostname the broker will advertise to producers and consumers. If not set, it uses the # value for "host.name" if configured. Otherwise, it will use the value returned from # java.net.InetAddress.getCanonicalHostName(). advertised.host.name=wait_for_change_hostname ..............................

更新为

# A comma seperated list of directories under which to store log files #log.dirs=/tmp/kafka-logs log.dirs=/data2/kafka,/data3/kafka,/data4/kafka # The port the socket server listens on #port=9092 # Hostname the broker will bind to. If not set, the server will bind to all interfaces host.name=90.90.90.223 # Hostname the broker will advertise to producers and consumers. If not set, it uses the # value for "host.name" if configured. Otherwise, it will use the value returned from # java.net.InetAddress.getCanonicalHostName(). advertised.host.name=90.90.90.223

然后对kafka目录设置属组，并启动kafka服务

[root@tdsql3 ~]# chown -R tdsql:users /data/application/kafka [tdsql@tdsql3 bin]$ cd /data/application/kafka/bin [tdsql@tdsql3 bin]$ ./kafka-server-start.sh -daemon ../config/server.properties

}

安装配置consumer

[root@tdsql4 ~]#cat /root/tdsql_install/tdsql_hosts ................... [tdsql_consumer] tdsql_consumer1 ansible_ssh_host=90.90.90.224 ..................

安装consumer组件

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_consumer.yml PLAY [install consumer beginning] ************************************************************************************************************************************* ........................ ........................ ........................ TASK [tdsql_consumer : write /etc/crontab] **************************************************************************************************************************** changed: [tdsql_consumer1] PLAY RECAP ************************************************************************************************************************************************************ tdsql_consumer1 : ok=40 changed=27 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0

启动consumer服务

[root@tdsql4 ]# su - tdsql [tdsql@tdsql4 ]$cd /data/application/consumer/bin [tdsql@tdsql4 bin]$ ./binlogconsumermgn --zklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --zkrootpath /tdsqlzk --kafkazklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --kafkazkrootpath /kafka --dev eth0 localip:90.90.90.224 [tdsql@tdsql4 bin]$ arglist: [tdsql@tdsql4 bin]$ ps -ef | grep consum tdsql 48445 1 0 19:36 ? 00:00:00 ./binlogconsumermgn --zklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --zkrootpath /tdsqlzk --kafkazklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --kafkazkrootpath /kafka --dev eth0 tdsql 48446 48445 0 19:36 ? 00:00:00 ./binlogconsumermgn --zklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --zkrootpath /tdsqlzk --kafkazklist 90.90.90.221:2118,90.90.90.222:2118,90.90.90.223:2118 --kafkazkrootpath /kafka --dev eth0 tdsql 48487 48396 0 19:36 pts/0 00:00:00 grep --color=auto consum

安装配置ES

[root@tdsql4 ~]#cat /root/tdsql_install/tdsql_hosts ......................... [tdsql_es] tdsql_es1 ansible_ssh_host=90.90.90.224 ....................... [root@tdsql4 ]#cat /root/tdsql_install/group_vars/all ....................... tdsql_es_mem: 2 tdsql_es_log_days: 7 tdsql_es_base_path: /data/application/es-install/es .......................

安装ES组件

[root@tdsql4 tdsql_install]# ansible-playbook -i tdsql_hosts playbooks/tdsql_es_single.yml PLAY [install es beginning] ******************************************************************************************************************************************* ...................... ...................... ...................... PLAY [update conf on clouddba] **************************************************************************************************************************************** TASK [tdsql_update_clouddba : replace diagnosis.conf esip] ************************************************************************************************************ changed: [tdsql_chitu1] TASK [tdsql_update_clouddba : start clouddba service] ***************************************************************************************************************** changed: [tdsql_chitu1] PLAY RECAP ************************************************************************************************************************************************************ tdsql_chitu1 : ok=2 changed=2 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 tdsql_es1 : ok=39 changed=29 unreachable=0 failed=0 skipped=2 rescued=0 ignored=0

启动elasticsearch服务

[tdsql@tdsql4 ]$ cd /data/application/es-install/es/master/tools/ [tdsql@tdsql4 ]$ nohup ./start-elasticsearch.sh & [tdsql@tdsql4 ]$ cd /data/application/es-install/es/data [tdsql@tdsql4 ]$ nohup ./start-elasticsearch.sh & [tdsql@tdsql4 ]$ ps -ef |grep elasticsearch | grep -v 'grep' |wc -l 4

启动logstash服务

[tdsql@tdsql4 tools]$ cd /data/application/es-install/logstash/tools/ [tdsql@tdsql4 tools]$ nohup ./start-logstash.sh & [tdsql@tdsql4 es-cleaner]$ ps -ef |grep logstash | grep -v 'grep' |wc -l 2

启动es-cleaner服务

[tdsql@tdsql4 tools]$ cd /data/application/es-install/es-cleaner/ nohup ./start-es-cleaner.sh & 1

启动kibana服务

[tdsql@tdsql4 es-cleaner]$ cd /data/application/es-install/kibana-5.6.4-linux-x86_64/ nohup ./kibana & [tdsql@tdsql4 es-cleaner]$ ps -ef |grep 'node/bin/node' | grep -v 'grep' |wc -l 1

至此，TDSQL集群环境就安装成功了。

接下来就可以使用赤兔管理平台对TDSQL数据库集群环境进行管理和维护了。