.. _admin-install: .. warning:: **THIS DOCUMENTATION IS OUTDATED AND CORRESPONDS TO OAR 2.5.X NOT 3.0.0** Installation ============ .. image:: https://img.shields.io/badge/docs-outdated-yellow :target: http://oar.imag.fr/oar_3 Overview -------- There are currently 2 methods to install OAR: - from the Debian packages - from sources Before going further, please have in mind OAR's architecture. A common OAR installation is composed of: - a **server** which will hold all of OAR "smartness". That host will run the OAR server daemon; - one or more **frontends**, which users will have to login to, in order to reserve computing nodes (oarsub, oarstat, oarnodes, ...); - **computing nodes** (or basically *nodes*), where the jobs will execute; - optionally a **visualisation server** which will host the visualisation webapps (monika, drawgantt, ...); - optionally an **API server**, which will host OAR restful API service. Many OAR data are stored and archived in a a postgresql database. Beside this documentation, please have a look at **OAR website**: http://oar.imag.fr, which also provides a lot of information, especially in the **Download** and **Contribs** sections. Server ------ Installation from the packages ______________________________ **Instructions** *Dependencies for Debian like system*:: # Build dependencies apt-get install gcc make tar python-docutils # Common dependencies apt-get install perl perl-base openssh-client openssh-server libdbi-perl libsort-versions-perl libterm-ui-perl # PostgreSQL dependencies apt-get install postgresql postgresql-client libdbd-pg-perl # Python3 dependencies apt-get install python3 \ python3-alembic python3-click python3-flask \ python3-passlib python3-psutil python3-requests \ python3-simplejson python3-sqlalchemy-utils \ python3-tabulate python3-toml python3-yaml \ python3-zmq python3-psycopg2 python3-fastapi \ python3-escapism python3-clustershell python3-rich # To execute admin commands on noces apt-get install taktuk # Install procset wget http://ftp.de.debian.org/debian/pool/main/p/python-procset/python3-procset_1.0-2_all.deb dpkg -i python3-procset_*.deb # Install sqlalchemy 2 wget http://ftp.de.debian.org/debian/pool/main/s/sqlalchemy/python3-sqlalchemy_2.0.19+ds1-1_all.deb dpkg -i python3-sqlalchemy_2*.deb # Then install oar node package dpkg -i python3-oar_*.deb oar-common_*.deb oar-server_*.deb # Due to usage of experimental packages, you might need to run apt --fix-broken install Installation from the tarball _____________________________ **Requirements** dependencies: - make - perl (pod2man) - gcc - python3 (with distutils package) - postgres - python3-dev (to build python extension) - `poetry `_ - pip3 - libpq package - taktuk **On Debian you can use the following command**:: apt update && apt install make perl python3 python3-dev python3-pip curl libpq-dev **Instructions** Get the sources:: export OAR_VERSION=3.0.0.dev7 wget -O - https://github.com/oar-team/oar3/archive/refs/tags/${OAR_VERSION}.tar.gz | tar xzvf - cd oar3-${OAR_VERSION} Install the python sources:: poetry build && pip install dist/*.whl Build/Install/Setup the OAR server:: # build make PREFIX=/usr/local server-build # install make PREFIX=/usr/local server-install # setup make PREFIX=/usr/local server-setup Configuration _____________ The oar database ~~~~~~~~~~~~~~~~ **Install and create the database (postgresql) on debian you can install postgres with the following commands**:: apt-get update && \ apt-get install -y postgresql postgresql-contrib libjson-perl && \ apt-get clean # Configure the database postgresql_main=$(find /etc/postgresql -name "main") \ && sed -i "s/#listen_addresses = 'localhost'/listen_addresses = '*'/" ${postgresql_main}/postgresql.conf \ && echo "host all all 0.0.0.0/0 md5" >> ${postgresql_main}/pg_hba.conf systemctl enable postgresql # Reboot or use systemctl start postgresql **Install the perl dependencies for the oar database management tool**:: # For debian system use apt-get install libdbi-perl perl perl-base libsort-versions-perl libdbd-pg-perl libdatetime-perl libterm-ui-perl Define the database configuration in /etc/oar/oar.conf. You need to set the variables ``DB_HOSTNAME, DB_PORT, DB_BASE_NAME, DB_BASE_LOGIN, DB_BASE_PASSWD, DB_BASE_LOGIN_RO, DB_BASE_PASSWD_RO``:: vi /etc/oar/oar.conf Create the database and the database users:: # General case oar-database --create --db-admin-user --db-admin-pass # OR, for PostgreSQL, in case the database is installed locally oar-database --create --db-is-local Init.d scripts ~~~~~~~~~~~~~~ If you have installed OAR from sources, you need to become root user and install manually the init.d/default/sysconfig scripts present in the folders:: $PREFIX/share/doc/oar-server/examples/scripts/{init.d,default,sysconfig} Then use the script ``/etc/init.d/oar-server`` to start the OAR server daemon. Adding resources to the system ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ To **automatically** initialize resources for your cluster, you can run the ``oar_resources_init`` command. It will detect the resources from nodes set in a file and give the OAR commands to initialize the database with the appropriate values for the memory and the cpuset properties. Another tool is also available to create resources beforehand: that tool does not require nodes to be up and accessible by SSH. See ``oar_resources_add``. *Otherwise:* To add resources to your system, you can use (as root) the ``oarnodesetting`` command. For a complete understanding of what that command does, see the manual page. For a basic usage, the main options are **-a** (means add a resource) and **-h** (defines the resource hostname or ip adress). For instance, to add a computing resource for node to your setup, type:: oarnodesetting -a -h This adds a resource with as host IP address (network_address property). You can modify resources properties with **-p** option, for instance:: oarnodesetting -r 1 -p "besteffort=YES" This allows the resource #1 to accept jobs of type *besteffort* (an admission rule forces besteffort jobs to execute on resources with the property "besteffort=YES"). Notes _____ Security issues ~~~~~~~~~~~~~~~ For security reasons it is hardly **recommended** to configure a read only account for the OAR database (like the above example). Thus you will be able to add it in DB_BASE_LOGIN_RO and DB_BASE_PASSWD_RO in *oar.conf*. PostgreSQL: autovacuum ~~~~~~~~~~~~~~~~~~~~~~~ Be sure to activate the "autovacuum" feature in the "postgresql.conf" file (OAR creates and deletes a lot of records and this setting cleans the postgres database from unneeded records). PostgreSQL: authentication ~~~~~~~~~~~~~~~~~~~~~~~~~~~ In case you've installed a PostgreSQL database remotely, if your PostgreSQL installation doesn't authorize the local connections by default, you need to enable the connections to this database for the oar users. Assuming the OAR server has the address , you can add the following lines in the ``pg_hba.conf`` file:: # in /etc/postgresql/8.1/main/pg_hba.conf or /var/lib/pgsql/data/pg_hba.conf host oar oar_ro /32 md5 host oar oar /32 md5 Using Taktuk ~~~~~~~~~~~~ .. note:: Taktuk is currently the recommended way to run admin commands on nodes. Is should be possible to use other programs such as `clustershell `_. Reach us on `github `_ for questions. OAR3 uses taktuk for remote administration operations; you have to install it. You can find information about taktuk from its website: http://taktuk.gforge.inria.fr. Then, you have to edit your oar configuration file and fill in the related parameters: *Install taktuk*:: apt-get install taktuk Then update your `configuration`: - ``TAKTUK_CMD`` (the path to the taktuk command) - ``PINGCHECKER_TAKTUK_ARG_COMMAND`` (the command used to check resources states) - ``SCHEDULER_NODE_MANAGER_SLEEP_CMD`` (command used for halting nodes) CPUSET feature ~~~~~~~~~~~~~~ .. warning:: Currently OAR3 is not compatible with cgroupv2. Make sure your nodes have cgroupv1 enabled. On debian you can set the kernel parameter `systemd.unified_cgroup_hierarchy=0`. OAR uses the CPUSET features provided by the Linux kernel >= 2.6. This enables to restrict user processes to reserved processors only and provides a powerful clean-up mechanism at the end of the jobs. For more information, have a look at the CPUSET file. Energy saving ~~~~~~~~~~~~~ Starting with version 2.4.3, OAR provides a module responsible of advanced management of wake-up/shut-down of nodes when they are not used. To activate this feature, you have to: - provide 2 commands or scripts which will be executed on the oar server to shutdown (or set into standby) some nodes and to wake-up some nodes (configure the path of those commands into the ``ENERGY_SAVING_NODE_MANAGER_WAKE_UP_CMD`` and ``ENERGY_SAVING_NODE_MANAGER_SHUT_DOWN_CMD`` variables in oar.conf) Thes 2 commands are executed by the oar user. - configure the ``available_upto`` property of all your nodes: - ``available_upto=0`` : to disable the wake-up and halt - ``available_upto=1`` : to disable the wake-up (but not the halt) - ``available_upto=2147483647`` : to disable the halt (but not the wake-up) - ``available_upto=2147483646`` : to enable wake-up/halt forever - ``available_upto=`` : to enable the halt, and the wake-up until the date given by Ex: to enable the feature on every nodes forever: :: oarnodesetting --sql true -p available_upto=2147483646 - activate the energy saving module by setting ``ENERGY_SAVING_INTERNAL="yes"`` and configuring the ``ENERGY_*`` variables into oar.conf - configure the metascheduler time values into ``SCHEDULER_NODE_MANAGER_IDLE_TIME``, ``SCHEDULER_NODE_MANAGER_SLEEP_TIME`` and ``SCHEDULER_NODE_MANAGER_WAKEUP_TIME`` variables of the oar.conf file. - restart the oar server (you should see an "Almighty" process more). You need to restart OAR each time you change an ``ENERGY_*`` variable. More informations are available inside the oar.conf file itself. For more details about the mechanism, take a look at the "Greta" module documentation. Disabling SELinux ~~~~~~~~~~~~~~~~~ On some distributions, SELinux is enabled by default. There is currently no OAR support for SELinux. So, you need to disable SELinux, if enabled. Cpuset id issue ~~~~~~~~~~~~~~~ On some rare servers, the core ids are not persistent across reboot. So you need to update the cpuset ids in the resource database at startup for each computing node. You can do this by using the ``/etc/oar/update_cpuset_id.sh`` script. The following page give more informations on how configuring it: http://oar.imag.fr/wiki:old:customization_tips#start_stop_of_nodes_using_ssh_keys Frontends --------- Installation from the packages ______________________________ **Instructions** *For the Debian like systems*:: # Install dependencies apt-get update && \ apt-get install -y python3 perl \ python3-alembic \ python3-click python3-flask \ python3-passlib python3-psutil python3-requests \ python3-simplejson python3-sqlalchemy-utils \ python3-tabulate python3-toml python3-yaml \ python3-zmq python3-psycopg2 python3-fastapi \ python3-escapism python3-clustershell python3-rich # Install procset wget http://ftp.de.debian.org/debian/pool/main/p/python-procset/python3-procset_1.0-2_all.deb dpkg -i python3-procset_*.deb # Install sqlalchemy 2 wget http://ftp.de.debian.org/debian/pool/main/s/sqlalchemy/python3-sqlalchemy_2.0.19+ds1-1_all.deb dpkg -i python3-sqlalchemy_2*.deb # Then install oar node package dpkg -i python3-oar_*.deb oar-common_*.deb oar-user_*.deb # Due to usage of experimental packages, you might need to run apt --fix-broken install Installation from the tarball _____________________________ **Requirements** - python3, pip, poetry and libpq-dev (to build psycopg2) **On Debian you can use the following command**:: apt update && apt install make perl python3 python3-dev python3-pip curl libpq-dev *For Debian like system*:: # Build dependencies apt-get install gcc make tar curl python3 # Common dependencies apt-get install perl perl-base openssh-client openssh-server libdbi-perl # PostgreSQL dependencies apt-get install postgresql-client libdbd-pg-perl **Instructions** Get the sources:: export OAR_VERSION=3.0.0.dev7 wget -O - https://github.com/oar-team/oar3/archive/refs/tags/${OAR_VERSION}.tar.gz | tar xzvf - cd oar3-${OAR_VERSION} Install the python sources:: poetry build && pip install dist/*.whl Build/Install/setup:: # build make user-build # install make user-install # setup make user-setup Configuration _____________ SSH setup ~~~~~~~~~ OAR uses SSH to connect from machine to machine (e.g. from server or frontend to nodes or from nodes to nodes), using a dedicated SSH daemon usually running on port 6667. Upon installation of the OAR server on the server machine, a SSH key pair along with an authorized_keys file is created for the oar user in ``/var/lib/oar/.ssh``. You need to copy that directory from the oar server to the frontend (if not the same machine). Please note that public key in the authorized_keys file must be prefixed with ``environment="OAR_KEY=1"``, e.g.:: environment="OAR_KEY=1" ssh-rsa AAAAB3NzaC1yc2[...]6mIcqvcwG1K7V6CHLQKHKWo/ root@server Also please make sure that the ``/var/lib/oar/.ssh`` directory and contained files have the right ownership (oar.oar) and permissions for SSH to function. Coherent configuration files between server node and user nodes ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ You need to have a coherent oar configuration between the server node and the user nodes. So you can just copy the /etc/oar/oar.conf directory from to server node to the user nodes. About X11 usage in OAR ~~~~~~~~~~~~~~~~~~~~~~ The easiest and scalable way to use X11 application on cluster nodes is to open X11 ports and set the right DISPLAY environment variable by hand. Otherwise users can use X11 forwarding via SSH to access cluster frontends. You must configure the SSH server on the frontends nodes with:: X11Forwarding yes X11UseLocalhost no With this configuration, users can launch X11 applications after a 'oarsub -I' on the given node or "oarsh -X node12". Admission rules ~~~~~~~~~~~~~~~ In OAR3 the admission rules can be written as plain files located in `/etc/oar/admission_rules.d` in the **frontend**. The files must start with an integer to be read by OAR, to disable a rule simply rename it with a non-integer prefix (for instance `OFF_`). *Update the oar configuration*:: # Add the following line to the configuration ADMISSION_RULES_IN_FILES="yes" The rules are not yet automatically copied during the installation. This is the list of the default admission rules available in `OAR3 repository `_. *Default admission rules*:: etc/oar/admission_rules.d ├── 01_default_queue.py ├── 02_prevent_root_oar_toSubmit.py ├── 03_avoid_jobs_on_resources_in_drain_mode.py ├── 04_submit_in_admin_queue.py ├── 05_filter_bad_resources.py ├── 06_besteffort_formatting.py ├── 07_besteffort_advance_reservation.py ├── 08_deploy_formatting.py ├── 09_prevent_deploy_on_non-entire_nodes.py ├── 11_advance_reservation_limitation.py ├── 13_default_walltime.py ├── 14_interactive_max_walltime.py ├── 15_check_types.py ├── OFF_12_perform_action_for_user.py ├── OFF_16_default_resource_property.py ├── OFF_20_job_properties_cpu.py ├── OFF_20_job_properties_cputype.py └── OFF_21_add_sequentiel_constraint.py Computing nodes --------------- Installation from the packages ______________________________ **Instructions** *For the Debian like systems* .. note:: OAR3 is not shipped in the official distribution. The packages can be found at : https://github.com/oar-team/oar3/releases/latest. Also download the debian package for `ProcSet `_ (which is an OAR3 dependency). *First install OAR3 dependencies*:: apt-get update && \ apt-get install -y python3 perl python3-alembic \ python3-click python3-flask \ python3-passlib python3-psutil python3-requests \ python3-simplejson python3-sqlalchemy-utils \ python3-tabulate python3-toml python3-yaml \ python3-zmq python3-psycopg2 python3-fastapi \ python3-escapism python3-clustershell python3-rich # Install procset wget http://ftp.de.debian.org/debian/pool/main/p/python-procset/python3-procset_1.0-2_all.deb dpkg -i python3-procset_*.deb # Install sqlalchemy 2 wget http://ftp.de.debian.org/debian/pool/main/s/sqlalchemy/python3-sqlalchemy_2.0.19+ds1-1_all.deb dpkg -i python3-sqlalchemy_2*.deb # Then install oar node package dpkg -i python3-oar_*.deb oar-common_*.deb oar-node*.deb # Due to usage of experimental packages, you might need to run apt --fix-broken install Installation from the tarball (sources) _______________________________________ **Instructions** dependencies: - make - perl (pod2man) - gcc Get the sources:: export OAR_VERSION=3.0.0.dev7 wget -O - https://github.com/oar-team/oar3/archive/refs/tags/${OAR_VERSION}.tar.gz | tar xzvf - cd oar3-${OAR_VERSION} build/install/setup:: # build make PREFIX=/usr/local node-build # install make PREFIX=/usr/local node-install # setup make PREFIX=/usr/local node-setup Configuration _____________ Init.d scripts ~~~~~~~~~~~~~~ If you have installed OAR from sources, you need to become root user and install manually the {init.d,default,sysconfig} scripts present in the folders:: $PREFIX/share/oar/oar-node/{init.d,default,sysconfig}/oar-node Then you just need to use the script ``/etc/init.d/oar-node`` to start the SSH daemon dedicated to oar-node. SSH setup ~~~~~~~~~ OAR uses SSH to connect from machine to machine (e.g. from server or frontend to nodes or from nodes to nodes), using a dedicated SSH daemon usually running on port 6667. Upon installation of the OAR server on the server machine, a SSH key pair along with an authorized_keys file is created for the oar user in ``/var/lib/oar/.ssh``. You need to copy that directory from the oar server to the nodes. Please note that public key in the authorized_keys file must be prefixed with ``environment="OAR_KEY=1"``, e.g.:: environment="OAR_KEY=1" ssh-rsa AAAAB3NzaC1yc2[...]6mIcqvcwG1K7V6CHLQKHKWo/ root@server Also please make sure that the ``/var/lib/oar/.ssh`` directory and contained files have the right ownership (oar.oar) and permissions for SSH to function. API server ---------- Description ___________ Since the version 2.5.3, OAR offers an API for users and admins interactions. This api must be installed on a frontend node (with the user module installed). Installation from the packages ______________________________ **Instructions** *For the Debian like systems*:: apt-get update && \ apt-get install -y python3 perl \ python3-sqlalchemy python3-alembic \ python3-click python3-flask \ python3-passlib python3-psutil python3-requests \ python3-simplejson python3-sqlalchemy-utils \ python3-tabulate python3-toml python3-yaml \ python3-zmq python3-psycopg2 python3-fastapi # Install procset dpkg -i .deb # Then install oar node package dpkg -i python3-oar_*.deb oar-common_*.deb oar-restful-api_*.deb Installation from the tarball _____________________________ **Requirements** *For Debian like system*:: # Build dependencies apt-get install gcc make tar python-docutils # Common dependencies apt-get install perl perl-base libdbi-perl libjson-perl libyaml-perl libwww-perl apache2 libcgi-fast-perl # Install apache FastCGI and Suexec modules (optional but highly recommended) # MySQL dependencies apt-get install libdbd-mysql-perl # PostgreSQL dependencies apt-get install libdbd-pg-perl **Instructions** Get the sources:: OAR_VERSION=2.5.4 wget -O - http://oar-ftp.imag.fr/oar/2.5/sources/stable/oar-${OAR_VERSION}.tar.gz | tar xzvf - cd oar-${OAR_VERSION}/ build/install/setup:: # build make api-build # install make api-install # setup make api-setup Configuration _____________ *Configuring OAR* For the moment, the API needs the user tools to be installed on the same host ('``make user-install``' or oar-user packages). A suitable ``/etc/oar/oar.conf`` should be present. For the API to work, you should have the oarstat/oarnodes/oarsub commands to work (on the same host you installed the API) *Configuring Apache* The api provides a default configuration file (``/etc/oar/apache-api.conf``) that is using an identd user identification enabled only from localhost. Edit the ``/etc/oar/apache-api.conf`` file and customize it to reflect the authentication mechanism you want to use. For ident, you may have to install a "identd" daemon on your distrib. The steps may be: - Install and run an identd daemon on your server (like *pidentd*). - Activate the ident auth mechanism into apache (``a2enmod ident``). - Activate the headers apache module (``a2enmod headers``). - Activate the rewrite apache module (``a2enmod rewrite``). - Customize apache-api.conf to allow the hosts you trust for ident. *YAML, JSON, XML* You need at least one of the YAML or JSON perl module to be installed on the host running the API. *Test* You may test the API with a simple wget:: wget -O - http://localhost/oarapi/resources.html It should give you the list of resources in the yaml format but enclosed in an html page. To test if the authentication works, you need to post a new job. See the example.txt file that gives you example queries with a ruby rest client. Visualization server -------------------- .. warning:: This part has been written for OAR2, and hasn't been updated yet. Description ___________ OAR provides two webapp tools for visualizing the resources utilization:: - monika which displays the current state of resources as well as all running and waiting jobs - drawgantt-svg which displays gantt chart of nodes and jobs for the past and future. Installation from the packages ______________________________ **Instructions** *For RedHat like systems*:: # OAR provides a Yum repository. # For more information see: http://oar.imag.fr/download#rpms # Install OAR web status package yum --enablerepo=OAR install oar-web-status *For the Debian like systems*:: # OAR is shipped as part of Debian official distributions (newer versions can be available in backports) # For more info see: http://oar.imag.fr/download#debian # Install OAR web status package apt-get install oar-web-status Installation from the tarball _____________________________ **Requirements** *For RedHat like systems*:: # Build dependencies yum install gcc make tar python-docutils # Common dependencies yum install perl perl-base perl-DBI ruby-GD ruby-DBI perl-Tie-IxHash perl-Sort-Naturally perl-AppConfig php # MySQL dependencies yum install mysql perl-DBD-MySQL ruby-mysql php-mysql # PostgreSQL dependencies yum install postgresql perl-DBD-Pg ruby-pg php-pgsql *For Debian like system*:: # Build dependencies apt-get install gcc make tar python-docutils # Common dependencies apt-get install perl perl-base ruby libgd-ruby1.8 libdbi-perl libtie-ixhash-perl libappconfig-perl libsort-naturally-perl libapache2-mod-php5 # MySQL dependencies apt-get install libdbd-mysql-perl libdbd-mysql-ruby php5-mysql # PostgreSQL dependencies apt-get install libdbd-pg-perl libdbd-pg-ruby php5-pgsql **Instructions** Get the sources:: OAR_VERSION=2.5.4 wget -O - http://oar-ftp.imag.fr/oar/2.5/sources/stable/oar-${OAR_VERSION}.tar.gz | tar xzvf - cd oar-${OAR_VERSION}/ build/install/setup:: # build make monika-build drawgantt-build drawgantt-svg-build www-conf-build # install make monika-install drawgantt-install drawgantt-svg-install www-conf-install # setup make monika-setup drawgantt-setup drawgantt-svg-setup www-conf-setup Configuration _____________ **Monika configuration** - Edit ``/etc/oar/monika.conf`` to fit your configuration. **Drawgantt-SVG configuration** - Edit ``/etc/oar/drawgantt-config.inc.php`` to fit your configuration. **httpd configuration** - You need to edit ``/etc/oar/apache.conf`` to fit your needs and verify that you http server configured.