Installing TORQUE

Build the distribution on the machine that will act as the TORQUE server - the machine which monitors and controls all compute nodes by running the pbs_server daemon.

The built distribution package works only on compute nodes of a similar architecture. Nodes with different architecture must have the installation package built on them individually.

The following software is required to run TORQUE 5.0.0:

Important: If you intend to use TORQUE 5.0.0 with Moab, you must run Moab version 8.0.0 or later. TORQUE 5.0.0 will not work with versions earlier than Moab 8.0.0.

To install TORQUE

  1. Install the gcc, libssl-devel, and libxml2-devel packages to build 5.0.0. The package names may vary.

    Use these commands to install the packages on the following operating systems:

    RHEL, CentOS, and Scientific Linux:

    [root]# yum update

    [root]# yum install libtool-devel libxml2-devel openssl-devel gcc gcc-c++ boost-devel

    SLES (You must have a licensed installation of SuSE and have installed the SuSE Linux Enterprise Software Development Kit and added the ISO to the repository):

    [root]# zypper update

    [root]# zypper install libtool-devel libxml2-devel libopenssl-devel gcc gcc-c++ boost-devel

  2. Verify that the following ports are open for essential communication:
    • For client communication to pbs_server, all privileged ports must be open (ports under 1024).
    • For pbs_server communication to pbs_mom, the default port is 15003.
    • For pbs_mom to pbs_server, the default port is 15001.

    For more information on configuring ports, see Configuring ports.

  3. Download the latest TORQUE build (torque-4.5.0.tar.gz, for example) from the Adaptive Computing website. It can also be downloaded via command line.

    1. Clone source from github.

      [root]# git clone https://github.com/adaptivecomputing/torque.git -b 5.0.0 5.0.0
      [root]# cd 5.0.0

      After downloading a copy of the repository, you can list the current branches by typing git branch -a from within the directory of the branch you cloned.

      If you're checking source out from git, read the README.building-40 file in the repository.

    2. Get tarball source distribution.

      [root]# wget http://www.adaptivecomputing.com/download/torque/torque-5.0.0.tar.gz
      [root]# tar -xzvf torque-5.0.0.tar.gz
      [root]# cd torque-5.0.0/

    To download a different version, replace each 5.0.0 with the desired version.

  4. After TORQUE is downloaded, do the following to configure the package:
    1. Run ./configure. TORQUE has many options you can specify to configure your installation. For more information, see Customizing the install.
    2. By default, the TORQUE directory is /var/spool/torque. If you want to change the location, use the --with-server-home configuration option.
    3. By default, make install installs all files in /usr/local/bin, /usr/local/lib, /usr/local/sbin, /usr/local/include, and /usr/local/man . You can specify an installation prefix other than /usr/local by using --prefix as an argument to ./configure. Note that TORQUE cannot be installed into a directory path that contains a space.

      If you decide to use --prefix to specify a custom directory and that directory does not already exist, you must create it before running ./configure --prefix=....

      [root]# ./configure

  5. Run make and make install.
  6. TORQUE must be installed by a root user. If running sudo fails, switch to root with su -.

    [root]# make

    [root]# make install

    OSX 10.4 users need to change #define __TDARWIN in src/include/pbs_config.h to #define __TDARWIN_8. Note that Mac OSX is not officially supported.

  7. The trqauthd daemon must be started to complete the remaining installation steps. For the appropriate distribution, do the following:

    * If Debian distribution, do the following * 

    [root]# cp contrib/init.d/debian.trqauthd /etc/init.d/trqauthd

     

    * If SLES distribution, do the following *

    [root]# cp contrib/init.d/suse.trqauthd /etc/init.d/trqauthd

     

    * If RHEL distribution, do the following *

    [root]# cp contrib/init.d/trqauthd /etc/init.d/

     

    * For all aforementioned distributions, do the following:

     

    [root]# chkconfig --add trqauthd

    [root]# echo /usr/local/lib > /etc/ld.so.conf.d/torque.conf

    [root]# ldconfig

    [root]# service trqauthd start

  8. Start the trqauthd daemon manually.

    [root]# trqauthd

    If needed, you can also stop trqauthd manually by running trqauthd -d. (For additional information, see Configuring trqauthd for client commands).

  9. The make packages command can be used to create self-extracting packages that can be copied and executed on your nodes. For information on creating packages and deploying them, see Specifying compute nodes.

    You will also want to scp the init.d scripts to the compute nodes and install them there.

    The TORQUE initialization scripts are provided in the /init.d directory as a courtesy and may be modified at your discretion to work on your system.

  10. Verify that the /var/spool/torque/server_name file exists and contains the correct name of the server.

    [root]# echo <pbs_server's_hostname> > /var/spool/torque/server_name

  11. After installation, verify that you have the PATH environment variable configured to include /usr/local/bin/ and /usr/local/sbin/ for both the installation user and the root user.
  12. By default, make install creates a directory at /var/spool/torque. This directory is referred to as TORQUE_HOME. TORQUE_HOME has several sub-directories, including server_priv/, server_logs/, mom_priv/, mom_logs/, and other directories used in the configuration and running of TORQUE.

  13. Verify you have environment variables configured so your system can find the shared libraries and binary files for TORQUE. This step is not necessary if the shared libraries are in their default locations.

  14. To set the library path, add the directory where the TORQUE libraries are installed. For example, if your TORQUE libraries are installed in /usr/local/lib (if you changed the default library path at configure time, use that path instead.), execute the following:

    [root]# echo '/usr/local/lib' > /etc/ld.so.conf.d/torque.conf

    [root]# ldconfig

  15. Initialize serverdb by executing the torque.setup script.

    [root]# ./torque.setup root

    If you are upgrading from TORQUE 2.5.9, run pbs_server -u before running torque.setup.

    [root]# pbs_server -u

  16. Add nodes to the /var/spool/torque/server_priv/nodes file. For information on syntax and options for specifying compute nodes, see Specifying compute nodes.
  17. Configure the MOMs if necessary. See Configuring TORQUE on compute nodes.
  18. Configure pbs_server and pbs_mom to start automatically at system boot, then start their daemons. See Enabling TORQUE as a service.

If you have a multi-homed host, or if your host name resolves to the loopback address, you may encounter an "unauthorized request" error when running qmgr. The resolution is to as a manager manually add the host name of the specified interface to the serverdb file with the attributes tag.

<managers>[email protected]</managers>

Related topics 

© 2014 Adaptive Computing