PBS+maui安装文档

PBS安装文档

centOS 装机

系统选择 compute node

development tools 选上

语言支持选上汉语

密码 wjC井01

机器名字（本次装机忘做设置，后来改的）

1. 修改机器名字与/etc/hosts #

hostnamectl set-hostname igc-head

cat /etc/hosts

127.0.0.1 localhost localhost.localdomain

10.1.10.1 igc-head

10.1.10.100 is1

10.1.10.11 i1

10.1.10.12 i2

10.1.10.13 i3

cat /etc/hostname

igc-head

2. 节点之间无密码登陆

ssh-keygen

ssh-copy-id root@is1

ssh-copy-id root@i1

ssh-copy-id root@i2

ssh-copy-id root@i3

3. 安装库,安装torque

yum update

yum install libtool openssl-devel libxml2-devel boost-devel gcc gcc-c++

yum install git

mkdir –p software/torque

git clone https://github.com/adaptivecomputing/torque.git -b 5.1.1 5.1.1

cd 5.1.1

./autogen.sh

./configure

make

make install

4.Configure the trqauthd daemon to start automatically at system boot.

cp contrib/init.d/trqauthd /etc/init.d/

chkconfig --add trqauthd

echo /usr/local/lib > /etc/ld.so.conf.d/torque.conf

ldconfig

service trqauthd start

export PATH=/usr/local/bin/:/usr/local/sbin/:$PATH

./torque.setup root

5. 编辑 /var/spool/torque/server_priv/nodes , 曾加节点信息 #

node-name[:ts] [np=] [gpus=] [properties]

[np=] 节点核数

[gpus=] 节点GPU个数

[properties] 节点名字

cat /var/spool/torque/server_priv/nodes

i1 np=16 normal all

i2 np=16 normal all

i3 np=16 bigmem all

6. make packages

会生成一些.sh文件 , 将torque-package-clients-linux-x86_64.sh torque-package-mom-linux-x86_64.sh 拷贝到各个节点

scp torque-package-clients-linux-x86_64.sh torque-package-mom-linux-x86_64.sh 192.168.0.151:/software/torque/5.1.1

同时, 将pbs_mom拷贝到各节点

scp contrib/init.d/pbs_mom 192.168.72.25:/software/torque

7.每一个节点做如下配置

more /var/spool/torque/mom_priv/config

$pbsserver ih # hostname running pbs server

$logevent 225 # bitmap of which events to log

8. 自动启动

管理节点配置

cp contrib/init.d/pbs_server /etc/init.d

chkconfig --add pbs_server

service pbs_server restart

其他自动启动

./torque-package-mom-linux-x86_64.sh --install

./torque-package-clients-linux-x86_64.sh --install

cp /software/torque/pbs_mom /etc/init.d

/usr/local/sbin/pbs_mom -c /var/spool/torque/mom_priv/config

chkconfig --add pbs_mom

service pbs_mom start

设置队列

qmgr -c "create queue all.q"

qmgr -c "set queue all.q queue_type = Execution"

qmgr -c "set queue all.q Priority = 100"

qmgr -c "set queue all.q resources_default.neednodes = all"

qmgr -c "set queue all.q resources_default.nodes = 1"

qmgr -c "set queue all.q enabled = True"

qmgr -c "set queue all.q started = True"

qmgr -c "create queue normal.q"

qmgr -c "set queue normal.q queue_type = Execution"

qmgr -c "set queue normal.q Priority = 100"

qmgr -c "set queue normal.q resources_default.neednodes = normal"

qmgr -c "set queue normal.q resources_default.nodes = 1"

qmgr -c "set queue normal.q enabled = True"

qmgr -c "set queue normal.q started = True"

qmgr -c "create queue bigmem.q"

qmgr -c "set queue bigmem.q queue_type = Execution"

qmgr -c "set queue bigmem.q Priority = 100"

qmgr -c "set queue bigmem.q resources_default.neednodes = bigmem"

qmgr -c "set queue bigmem.q resources_default.nodes = 1"

qmgr -c "set queue bigmem.q enabled = True"

qmgr -c "set queue bigmem.q started = True"

qmgr -c "set server default_queue = all.q"

maui 安装与配置

1. 官网下载，需先注册 http://www.adaptivecomputing.com/products/open-source/maui/

http://www.adaptivecomputing.com/support/download-center/maui-cluster-scheduler/

2 安装

./configure --with-pbs=/usr/local

make

make install

3. 开机启动

cp /software/torque/maui-3.3.1/etc/maui.d /etc/init.d/

vi /etc/init.d/maui.d # "MAUI_PREFIX=/opt/maui" 改为 "MAUI_PREFIX=/usr/local/maui"

chkconfig maui.d on

启动maui:

/etc/init.d/maui.d start

防火墙端口

所有计算节点和头节点都需要做

firewall-cmd --zone=public --add-port=15001-15004/tcp --permanent

firewall-cmd --zone=public --add-port=15001-15004/udp --permanent

firewall-cmd --reload

9. 运行与测试

[root@igc-head ~]# qnodes

state = free

power_state = Running

np = 16

properties = normal,all

ntype = cluster

status = rectime=1437594868,macaddr=44:a8:42:27:6a:3d,cpuclock=Fixed,varattr=,jobs=,state=free,netload=42405454151,gres=,loadave=0.00,ncpus=32,physmem=131752888kb,availmem=133165940kb,totmem=135947188kb,idletime=113308,nusers=1,nsessions=1,sessions=1275,uname=Linux i1 3.10.0-229.el7.x86_64 #1 SMP Fri Mar 6 11:36:42 UTC 2015 x86_64,opsys=linux

mom_service_port = 15002

mom_manager_port = 15003

state = free

power_state = Running

np = 16

properties = normal,all

ntype = cluster

status = rectime=1437594868,macaddr=44:a8:42:27:6a:0d,cpuclock=Fixed,varattr=,jobs=,state=free,netload=36553902893,gres=,loadave=0.00,ncpus=32,physmem=131752888kb,availmem=133604660kb,totmem=135947188kb,idletime=113156,nusers=1,nsessions=1,sessions=1256,uname=Linux i2 3.10.0-229.el7.x86_64 #1 SMP Fri Mar 6 11:36:42 UTC 2015 x86_64,opsys=linux

mom_service_port = 15002

mom_manager_port = 15003

state = free

power_state = Running

np = 16

properties = bigmem,all

ntype = cluster

status = rectime=1437594867,macaddr=44:a8:42:27:84:5b,cpuclock=Fixed,varattr=,jobs=,state=free,netload=37740503968,gres=,loadave=0.00,ncpus=32,physmem=263865204kb,availmem=264675684kb,totmem=268059504kb,idletime=113035,nusers=1,nsessions=1,sessions=1271,uname=Linux i3 3.10.0-229.el7.x86_64 #1 SMP Fri Mar 6 11:36:42 UTC 2015 x86_64,opsys=linux

mom_service_port = 15002

mom_manager_port = 15003

测试

su - igc

echo "sleep 30; hostname" |qsub

单节点特殊配置

完成安装后，需要注意如下几点：

hostname不能是localhost，需要修改为其他，例如

sudo hostname new_name

sudo vi /etc/hostname

127.0.0.1 localhost

x.x.x.x new_name

对应修改pbs_server目录下server_name, server_priv/nodes, 还有pbs_mom下面config里的server name

qnodes测试应通过，然后进行

PBS+maui安装文档相关推荐

ELK安装文档及相关优化
前言:随着硬件成本的不断低廉,我们可以存储更多数据内容,也会对各数据加以利用,其中一项很重要的数据内容便是日志文件,无论是访问日志还是系统日志或是应用日志,都显得十分重要,而怎么加以利用一直是一个难题 ...
mysql数据库21_Mysql5.7.21安装文档
Mysql5.7.21安装文档发布时间:2020-06-27 13:59:56 来源:51CTO 阅读:946 作者:PengChonggui --1.创建用户组mysql,创建用户mysql并将其 ...
Subversion（SVN）安装文档
Subversion(SVN)安装文档 SVN中文网址(操作手册) http://www.iusesvn.com/ 软件版本(当前版本): subversion-1.6.17.tar.gz su ...
GitLab安装文档
GitLab安装文档 GitLab 是一个用于仓库管理系统的开源项目,使用Git作为代码管理工具,并在此基础上搭建起来的web服务. GitLab与GitHub的功能相似,通常企业使用GitLab在局 ...
Oracle 11g 单实例安装文档
这里介绍在Red Hat Enterprise Linux Server release 5.7 (Tikanga)下安装ORACLE 11.2.0.1.0的过程,本文仅仅是为了写这样安装指导文档而整 ...
KVM 标准化安装文档
KVM 标准化安装文档转载于:https://blog.51cto.com/wangwei007/600111
hadoop2.6.5安装文档及解决root用户无法ssh localhost的问题
hadoop2.6.5官方安装文档解决root用户无法ssh localhost的问题:
转自 David dai linux平台 oracle 数据库安装文档
Oracle 官方的安装文档 http://download.oracle.com/docs/cd/B28359_01/install.111/b32002/toc.htm 一. 安装相关包 1.1 ...
php oracle817,Oracle 817 For Linux/Unix安装文档-数据库专栏,ORACLE
oracle 817安装文档 1．创建dba组和oracle用户 root@hz_cmode # groupadd dba root@hz_cmode # useradd -m -d /o ...
ORACLE LINUX 6.3 + ORACLE 11.2.0.3 RAC + VBOX安装文档
ORACLE LINUX 6.3 + ORACLE 11.2.0.3 RAC + VBOX安装文档 2015-10-21 12:51 525人阅读评论(0) 收藏举报分类: Oracle RA ...

PBS+maui安装文档

PBS+maui安装文档相关推荐

最新文章

热门文章