Centos 7, Torque 单节点部署
1.准备工作
安装Torque必须首先配置linux主机名称，服务器主机名称大多默认localhost,不建议直接使用localhost。
linux主机名称修改地址：http://www.cnblogs.com/smbin/p/8488909.html
linux系统：Centos 7
主机名称：master
系统用户：root
Torque官网下载地址：http://www.adaptivecomputing.com/support/download-center/torque-download/
作者下载的版本：http://wpfilebase.s3.amazonaws.com/torque/torque-6.1.2.tar.gz
2.安装和配置Torque
首先在/opt下创建文件夹torque，在此文件夹中下载压缩包，并解压下载并解压Torque文件
[root@mastar ]# cd /opt [root@mastar ]# mkdir torque [root@mastar ]# cd torque [root@mastar torque]# wget http://wpfilebase.s3.amazonaws.com/torque/torque-6.1.2.tar.gz ......省略下载过程 [root@mastar torque]# tar -zxvf torque-6.1.2.tar.gz ......省略解压过程 [root@mastar torque]#cd torque-6.1.2/ [root@mastar torque-6.1.2]#
加载、安装和master配置。master配置就是主机和PBS之间的配置，master就是主机名
[root@master torque-6.1.2]# yum install libxml2-devel openssl-devel gcc gcc-c++ boost-devel libtool-y Loaded plugins: fastestmirror, langpacks base | 3.6 kB 00:00:00 extras | 3.4 kB 00:00:00 mysql-connectors-community | 2.5 kB 00:00:00 mysql-tools-community | 2.5 kB 00:00:00 mysql56-community | 2.5 kB 00:00:00 updates | 3.4 kB 00:00:00 Determining fastest mirrors * base: mirrors.cn99.com * extras: mirrors.tuna.tsinghua.edu.cn * updates: mirrors.tuna.tsinghua.edu.cn Package libxml2-devel-2.9.1-6.el7_2.3.x86_64 already installed and latest version Package 1:openssl-devel-1.0.2k-8.el7.x86_64 already installed and latest version Package gcc-4.8.5-16.el7_4.1.x86_64 already installed and latest version Package gcc-c++-4.8.5-16.el7_4.1.x86_64 already installed and latest version Package boost-devel-1.53.0-27.el7.x86_64 already installed and latest version No package libtool-y available. Nothing to do [root@master torque-6.1.2]# ./configure --prefix=/usr/local/torque --with-scp--with-default-server=master ......省略加载过程 Building components: server=yes mom=yes clients=yes gui=no drmaa=no pam=no PBS Machine type : linux Remote copy : /bin/scp -rpB PBS home : /var/spool/torque Default server : master Unix Domain sockets : Linux cpusets : no Tcl : disabled Tk : disabled Authentication : trqauthd configure: WARNING: This compilation has strict compiler options enabled that cause the build to fail if any compiler warnings are emitted. If this build fails because of a harmless warning, please report the problem to torqueusers@supercluster.org and run configure again without --enable-gcc-warnings. Ready for 'make'. [root@master torque-6.1.2]# make ......省略加载过程 [root@master torque-6.1.2]# make install ......省略加载过程 [root@master torque-6.1.2]# make packages
[root@master torque-6.1.2]# make packages
Building packages from /opt/torque/torque-6.1.2/tpackages
rm -rf /opt/torque/torque-6.1.2/tpackages
mkdir /opt/torque/torque-6.1.2/tpackages
Building ./torque-package-server-linux-x86_64.sh ...
libtool: install: warning: remember to run `libtool --finish /usr/local/torque/lib' //需要去执行命令：libtool --finish /usr/local/torque/lib
Building ./torque-package-mom-linux-x86_64.sh ...
libtool: install: warning: remember to run `libtool --finish /usr/local/torque/lib'
Building ./torque-package-clients-linux-x86_64.sh ...
libtool: install: warning: remember to run `libtool --finish /usr/local/torque/lib'
Building ./torque-package-devel-linux-x86_64.sh ...
libtool: install: warning: remember to run `libtool --finish /usr/local/torque/lib'
Building ./torque-package-doc-linux-x86_64.sh ...
Done.
The package files are self-extracting packages that can be copied
and executed on your production machines. Use --help for options.
[root@master torque-6.1.2]# libtool --finish /usr/local/torque/lib
libtool: finish: PATH="/usr/lib/jvm/java-1.7.0-openjdk/bin:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/root/bin:/usr/local/torque/bin:/usr/local/torque/sbin:/root/bin:/sbin" ldconfig -n /usr/l ocal/torque/lib
----------------------------------------------------------------------
Libraries have been installed in:
/usr/local/torque/lib
If you ever happen to want to link against installed libraries
in a given directory, LIBDIR, you must either use libtool, and
specify the full pathname of the library, or use the `-LLIBDIR'
flag during linking and do at least one of the following:
- add LIBDIR to the `LD_LIBRARY_PATH' environment variable
during execution
- add LIBDIR to the `LD_RUN_PATH' environment variable
during linking
- use the `-Wl,-rpath -Wl,LIBDIR' linker flag
- have your system administrator add LIBDIR to `/etc/ld.so.conf'
See any operating system documentation about shared libraries for
more information, such as the ld(1) and ld.so(8) manual pages.
配置服务：pbs_server PBS，pbs_sched，pbs_mom，trqauthd
[root@master torque-6.1.2]# cp contrib/init.d/{pbs_{server,sched,mom},trqauthd} /etc/init.d/ [root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do chkconfig --add $i; chkconfig $i on; done //遇见y/n选择y回车继续
设置Torque环境变量
[root@master torque-6.1.2]# TORQUE=/usr/local/torque [root@master torque-6.1.2]# echo "TORQUE=$TORQUE" >> /etc/profile [root@master torque-6.1.2]# echo "export PATH=\$PATH:$TORQUE/bin:$TORQUE/sbin" >> /etc/profile [root@master torque-6.1.2]# source /etc/profile
以root用户启动，报错服务指向的主机名和现有主机名不一致，安装过程中暂时没有找到解决方案！安装完毕后有解决方案，在本文最下方！！！
[root@master torque-6.1.2]# ./torque.setup root //尝试以root启动，报错：服务“pbs_server”已经启动 initializing TORQUE (admin: root) pbs_server already running... run 'qterm' to stop pbs_server and rerun //运行sterm关闭服务 [root@master torque-6.1.2]# qterm //发现服务指向的主机名称和正常显示的主机名称不一致，命令qterm无法关闭 Can not resolve name for server mastar. (rc = -2 - ) Cannot resolve specified server host 'mastar'. qterm: could not connect to server '' (15010) Access from host not allowed, or unknown host [root@master mom_priv]# ps -e | grep pbs //查询服务，尝试以kill -9命令关闭服务 30505 ? 00:00:00 pbs_server [root@master mom_priv]# kill -9 30505 [root@master mom_priv]# ps -e | grep pbs [root@master torque-6.1.2]# ./torque.setup root //发现服务关闭后仍无法启动，服务指向的主机名和现有主机名不一致！经确认上边配置的时候没有配置错误：
//‘./configure --prefix=/usr/local/torque --with-scp--with-default-server=master’ configure没有错误，未找到解决方案，怀疑是系统缓存的问题。 initializing TORQUE (admin: root) //暂时只能修改/etc/hosts文件的内容 You have selected to start pbs_server in create mode. If the server database exists it will be overwritten. do you wish to continue y/(n)?y Can not resolve name for server mastar. (rc = -2 - ) Cannot resolve specified server host 'mastar'. qmgr: cannot connect to server (errno=15010) Access from host not allowed, or unknown host ERROR: cannot set root@master in operators list Can not resolve name for server mastar. (rc = -2 - ) Cannot resolve specified server host 'mastar'. qterm: could not connect to server '' (15010) Access from host not allowed, or unknown host [root@master torque-6.1.2]# vi /etc/hosts //修改/etc/hosts文件 10.131.101.142 master 10.131.101.142 mastar //添加这一行的内容 27.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
[root@master torque-6.1.2]# ./torque.setup root //此时执行成功
initializing TORQUE (admin: root)
You have selected to start pbs_server in create mode.
If the server database exists it will be overwritten.
do you wish to continue y/(n)?y //输入y
开始pbs_server，pbs_sched服务，pbs_mom和trqauthd
[root@master torque-6.1.2]# qterm //关闭服务 [root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do service $i start; done Starting pbs_server (via systemctl): [ OK ] Starting pbs_sched (via systemctl): [ OK ] Starting pbs_mom (via systemctl): [ OK ] Starting trqauthd (via systemctl): [ OK ]
指定计算节点
添加计算节点”master”，设置CPU的数量
检查CPU的数量通过使用命令“lscpu”或“nproc”
[root@master torque-6.1.2]# vi /var/spool/torque/server_priv/nodes master np=8 //添加本行信息，注意等号前后不要有空格 master是主机名 [root@master torque-6.1.2]# vi /var/spool/torque/mom_priv/config pbsserver master //添加这两行信息 master是主机名 logevent 255
检查PBS的信息
[root@master torque-6.1.2]# ps -e | grep pbs 11188 ? 00:00:00 pbs_sched 11215 ? 00:00:00 pbs_mom 29683 ? 00:00:00 pbs_server [root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do service $i restart; done Restarting pbs_server (via systemctl): [ OK ] Restarting pbs_sched (via systemctl): [ OK ] Restarting pbs_mom (via systemctl): [ OK ] Restarting trqauthd (via systemctl): [ OK ]
创建队列的默认信息
[root@master torque-6.1.2]# qmgr -c 'create queue master' [root@master torque-6.1.2]# qmgr -c 'set queue master queue_type= execution' [root@master torque-6.1.2]# qmgr -c 'set queue master started= true' [root@master torque-6.1.2]# qmgr -c 'set queue master enabled= true' [root@master torque-6.1.2]# qmgr -c 'set queue master resources_default.walltime= 240:00:00' [root@master torque-6.1.2]# qmgr -c 'set queue master resources_default.nodes= 1' [root@master torque-6.1.2]# qmgr -c 'set server default_queue= master'
提交任务测试：
[root@master torque-6.1.2]# qnodes //查询计算节点的状态 master state = free power_state = Running np = 8 ntype = cluster status = opsys=linux,uname=Linux master 3.10.0-514.el7.x86_64 #1 SMP Tue Nov 22 16:42:41 UTC 2016 x86_64,sessions=3154 3489 41105 41699,nsessions=4,nusers=3,idletime=3198,
totmem=94868512kb,availmem=92195284kb,physmem=32367652kb,ncpus=56,loadave=0.85,gres=,netload=4005925534,state=free,varattr= ,cpuclock=Fixed,macaddr=68:cc:6e:c3:cf:87,version=6.1.2,rectime=1519980694,jobs= mom_service_port = 15002 mom_manager_port = 15003 [root@master torque-6.1.2]# su master //切换用户：此master不是主机名，而是一个用户的名字 [master@master torque-6.1.2]$ echo sleep 10 | qsub 0.master [master@master torque-6.1.2]$ qstat //查询任务状态 Job ID Name User Time Use S Queue ------------------------- ---------------- --------------- -------- - ----- 0.master STDIN master 0 R master [master@master torque-6.1.2]$ qstat -a -n //查询任务状态和每个任务占用cpu核数 master: Req'd Req'd Elap Job ID Username Queue Jobname SessID NDS TSK Memory Time S Time ----------------------- ----------- -------- ---------------- ------ ----- ------ --------- --------- - --------- 0.master master master STDIN 12470 1 1 -- 240:00:00 C -- master/0 [master@master torque-6.1.2]$
主机名和现有主机名不一致的问题解决方案：
这个问题一直没有找到出现的原因，但是怀疑是之前的Torque删除时没有删除干净，在“创建队列的默认信息”这一步的缓存依然存在。
在Torque安装成功后，停止Torque
[root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do service $i stop; done //停止服务T，start改为stop Stopping pbs_server (via systemctl): [ OK ] Stopping pbs_sched (via systemctl): [ OK ] Stopping pbs_mom (via systemctl): [ OK ] Stopping trqauthd (via systemctl): [ OK ] [root@master torque-6.1.2]# ./torque.setup root //重新运行这一步 hostname: master Currently no servers active. Default server will be listed as active server. Error 15133 Active server name: master pbs_server port is: 15001 trqauthd daemonized - port /tmp/trqauthd-unix trqauthd successfully started initializing TORQUE (admin: root) You have selected to start pbs_server in create mode. If the server database exists it will be overwritten. do you wish to continue y/(n)?y //输入y [root@master torque-6.1.2]# vi /var/spool/torque/server_priv/nodes master np=8 //=前后不要带空格
[root@master torque-6.1.2]# qterm //关闭pbs_server、 pbs_sched、 pbs_mom、 trqauthd服务 [root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do service $i start; done //重启服务 Starting pbs_server (via systemctl): [ OK ] Starting pbs_sched (via systemctl): [ OK ] Starting pbs_mom (via systemctl): [ OK ] Starting trqauthd (via systemctl): [ OK ]
[root@master torque-6.1.2]# qnodes //查询状态，报错服务trqauthd没有启动
socket_connect_unix failed: 15137
qnodes: cannot connect to server master, error=15137 (could not connect to trqauthd)
[root@master torque-6.1.2]# for i in pbs_server pbs_sched pbs_mom trqauthd; do service $i restart; done //重新启动服务
Restarting pbs_server (via systemctl): [ OK ]
Restarting pbs_sched (via systemctl): [ OK ]
Restarting pbs_mom (via systemctl): [ OK ]
Restarting trqauthd (via systemctl): [ OK ]
[root@master torque-6.1.2]# qnodes //查询状态，成功 master state = free power_state = Running np = 8 ntype = cluster status = opsys=linux,uname=Linux master 3.10.0-514.el7.x86_64 #1 SMP Tue Nov 22 16:42:41 UTC 2016 x86_64,sessions=3154 3489 10903 41105 41699,nsessions=5,nusers=4,idletime=5287,totmem=94868512kb,
availmem=92236268kb,physmem=32367652kb,ncpus=56,loadave=0.01,gres=,netload=8920006882,state=free,varattr= ,cpuclock=Fixed,macaddr=68:cc:6e:c3:cf:87,version=6.1.2,rectime=1519982783,jobs= mom_service_port = 15002 mom_manager_port = 15003
