Home > Database > Mysql Tutorial > body text

Greenplum测试环境部署

WBOY
Release: 2016-06-07 16:00:26
Original
1188 people have browsed it

本实例是部署实验环境,采用的是Citrix的虚拟化环境,分配了3台RHEL6.4的主机。

1.准备3台主机

本实例是部署实验环境,采用的是Citrix的虚拟化环境,分配了3台RHEL6.4的主机。

Master 创建模板后,额外添加20G一块磁盘/dev/xvdb,额外添加2块网卡eth1,eth2

Standby 创建模板后,额外添加20G一块磁盘/dev/xvdb,额外添加2块网卡eth1,eth2

Segment01 创建模板后,额外添加50G一块磁盘/dev/xvdb,额外添加2块网卡eth1,eth2

网络规划

 eth0(外部IP)eth1eth2

Master 192.168.9.123 172.16.10.101 172.16.11.101

Standby 192.168.9.124 172.16.10.102 172.16.11.102

Segment01 192.168.9.125(可选) 172.16.10.1 172.16.11.1

实验环境资源有限暂时配置3个节点,后续可能会根据需求添加Segment02,Segment03...

修改主机名

将Master,Standby,Segment01的三台主机名分别设置为mdw, smdw, sdw1

主机名修改方法:

hostname 主机名 vi /etc/sysconfig/network 修改hostname

Options:配置脚本,前期为了方便同步节点间的配置,可选。

export NODE_LIST='MDW SMDW SDW1'

vi /etc/hosts 临时配置

192.168.9.123 mdw 192.168.9.124 smdw 192.168.9.125 sdw1

配置第一个节点到自身和其他机器的无密码登录

ssh-keygen -t rsa ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.9.123 ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.9.124 ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.9.125 cluster_run_all_nodes "hostname ; date"

磁盘规划

gp建议使用xfs文件系统,所有节点需要安装依赖包
# rpm -ivh xfsprogs-3.1.1-10.el6.x86_64.rpm

所有节点建立/data文件夹,用来挂载xfs的文件系统

mkdir /data

mkfs.xfs /dev/xvdb

[root@smdb Packages], agsize=1310720 blks = , , imaxpct=25 = blks internal log , version=2 = blks, none , rtextents=0

vi /etc/fstab 添加下面一行

/dev/xvdb /data xfs rw,noatime,inode64,allocsize=16m1 1 2.关闭iptables和selinux cluster_run_all_nodes "hostname; service iptables stop" cluster_run_all_nodes "hostname; chkconfig iptables off" cluster_run_all_nodes "hostname; chkconfig ip6tables off" cluster_run_all_nodes "hostname; chkconfig libvirtd off" cluster_run_all_nodes "hostname; setenforce 0" cluster_run_all_nodes "hostname; sestatus" vi /etc/selinux/config cluster_copy_all_nodes /etc/selinux/config /etc/selinux/

注:所有节点都要统一设定,我这里先配置了信任,用脚本实现的同步,如果没有配置,是需要每台依次设定的。

3.设定建议的系统参数

vi /etc/sysctl.conf

kernel.shmmax = 500000000 kernel.shmmni = 4096 kernel.shmall = 4000000000 kernelkernel.sysrq = 1 kernel.core_uses_pid = 1 kernel.msgmnb = 65536 kernel.msgmax = 65536 net.ipv4.tcp_syncookies = 1 net.ipv4.ip_forward = 0 net= 0 net.ipv4.tcp_tw_recycle = 1 net.ipv4.tcp_max_syn_backlog = 4096 net= 1 net= 1 net.core.netdev_max_backlog = 10000 vm.overcommit_memory = 2 kernel.msgmni = 2048 net

vi /etc/security/limits.conf

* soft nofile 65536 * hard nofile 65536 * soft nproc 131072 * hard nproc 131072

同步到各个节点:

cluster_copy_all_nodes /etc/sysctl.conf /etc/sysctl.conf cluster_copy_all_nodes /etc/security/limits.conf /etc/security/limits.conf

磁盘预读参数及 deadline算法

在/etc/rc.d/rc.local 添加

blockdev --setra 16385 /dev/xvdb echo deadline > /sys/block/xvdb/queue/scheduler cluster_copy_all_nodes /etc/rc.d/rc.local /etc/rc.d/rc.local

注:重启后 blockdev --getra /dev/xvdb 验证是否生效

验证所有节点的字符集

cluster_run_all_nodes "hostname; echo $LANG"

重启所有节点,验证修改是否生效:

blockdev --getra /dev/xvdb more /sys/block/xvdb/queue/scheduler cluster_run_all_nodes "hostname; service iptables status" 4.在Master上安装 mkdir -p /data/soft 上传greenplum-db-4.3.4.2-build-1-RHEL5-x86_64.zip到Master unzip greenplum-db-4.3.4.2-build-1-RHEL5-x86_64.zip /bin/bash greenplum-db-4.3.4.2-build-1-RHEL5-x86_64.bin 5.在所有的节点上安装配置Greenplum

配置/etc/hosts

192.168.9.123 mdw 172.16.10.101 mdw-1 172.16.11.101 mdw-2 192.168.9.124 smdw 172.16.10.102 smdw-1 172.16.11.102 smdw-2 192.168.9.125 sdw1 172.16.10.1 sdw1-1 172.16.11.1 sdw1-2

同步/etc/hosts配置

cluster_copy_all_nodes /etc/hosts /etc/hosts

配置gp需要的互信

Related labels:
source:php.cn
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template
About us Disclaimer Sitemap
php.cn:Public welfare online PHP training,Help PHP learners grow quickly!