Initial Release - copy from internal repo.

.gitignore

@@ -0,0 +1,6 @@
+# ignore the file that should contain the mysql_slurm_password
+group_vars/*/mysql
+
+# Backup files
+*~
+*.swp

README.md

@@ -0,0 +1,93 @@
+# Creates a slurm cluster in pouta
+
+Tested with slurm versions:
+ - 14.11.0
+ - 14.11.3
+ - 15.08.0
+
+## How-To
+
+### Launch the Openstack instances:
+
+ - ansible-playbook launcopenstackinstance.yml # launches the VMs. source the openstack-rc script before running this playbook. Also update the playbook to include the names of your key and tenant.
+  - see the group_vars/all/all file for default variables used for launching an OS instance
+
+### Initial configuration of the instances and your workstation:
+
+ - First yum -y install nc on the bastion host.
+ - Then setup SSH config so you don't have to have a public IP on each instance. Change the Hostname in "Host bastion" to the service node.
+
+Put this in ~/.ssh/config :
+
+<pre>
+# http://edgeofsanity.net/article/2012/10/15/ssh-leap-frog.html
+# This applies to all hosts in your ssh config.
+ControlMaster auto
+ControlPath ~/.ssh/ssh_control_%h_%p_%r
+
+# Always ssh as cloud-user and don't save hostkeys
+Host bastion
+  User cloud-user
+  StrictHostKeyChecking no
+  UserKnownHostsFile /dev/null
+  Hostname 86.50.168.39
+
+Host slurm* 
+   User cloud-user
+   StrictHostKeyChecking no
+   UserKnownHostsFile /dev/null
+   ForwardAgent yes
+   ProxyCommand ssh bastion nc %h %p
+</pre>
+
+ - Second update the /etc/hosts on the service node (playbook should update the others):
+
+</pre>
+# For second cluster
+192.168.36.126 slurm2-compute1
+192.168.36.125 slurm2-service
+192.168.36.124 slurm2-login
+
+# For first cluster
+192.168.36.129 slurm-compute1
+192.168.36.128 slurm-service
+192.168.36.127 slurm-login
+</pre>
+
+### Then we can finally run the slurm configuration playbooks:
+
+\_ Update the files in group_vars/ to your settings
+
+You also need to add a mysql_slurm_password: "PASSWORD" string somewhere. This will be used to set a password for the slurm mysql user.
+
+
+#### Description of the playbooks:
+
+ - site.yml - calls the slurm*.yml playbooks
+ - slurm_*.yml # The playbooks that configure the servers
+  - set slurm_version - this is used to determine which version to download from schedmd.com
+
+#### Run them in this order:
+
+ - Update stage to have the right IP addresses and hostnames.
+ - configuring 1st slurm: ansible-playbook site.yml
+
+ - Update stage to have the right IP addresses and hostnames.
+ - configuring 2st slurm: ansible-playbook site.yml
+
+#### Add cloud-user to slurm
+
+<pre>
+sacctmgr create account name=csc
+sacctmgr create user name=cloud-user account=csc
+</pre>
+
+### Make changes to slurm.conf and distribute it to nodes and restart/reconfigure:
+
+ - Would be nice with a role /tag where one could just run ansible-playbook site.yml --tag new-slur-config and it pushes new config and restarts/reconfigs as necessary.
+
+# Authors:
+
+ - Marco Passerini (original author)
+ - Johan Guldmyr (updates done as part of FGCI work)
+

group_vars/all/all

@@ -0,0 +1,9 @@
+---
+
+tenant_name: "tenant_name"
+key_name: "key_name"
+image: "CentOS-7.0"
+flavor: "small"
+security_groups: "default,slurm"
+network_name: "default"
+auth_url: "https://pouta.csc.fi:5000/v2.0/"

launcopenstackinstance.yml

@@ -0,0 +1,27 @@
+---
+    - name: launch an instance
+      hosts: localhost
+      gather_facts: False
+
+      tasks:
+        - name: launch the Slurm servers
+          nova_compute:
+            state: present
+            login_username: "{{ lookup('env','OS_USERNAME') }}"
+            login_password: "{{ lookup('env','OS_PASSWORD') }}"
+            login_tenant_name: "{{ tenant_name }}"
+            auth_url: "{{ auth_url }}"
+            name: "{{ item }}"
+            image: "{{ image }}"
+            key_name: "{{ key_name }}"
+            wait_for: 200
+            flavor: "{{ flavor }}"
+            security_groups: "{{ security_groups }}"
+            nics:
+              - net-id: "{{ network_name }}"
+            meta:
+              hostname: slurms
+          with_items:
+            - slurm2-login
+            - slurm2-service
+            - slurm2-compute1

roles/common/files/iptables

@@ -0,0 +1,15 @@
+# Firewall configuration written by system-config-firewall
+# Manual customization of this file is not recommended.
+*filter
+:INPUT ACCEPT [0:0]
+:FORWARD ACCEPT [0:0]
+:OUTPUT ACCEPT [0:0]
+-A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
+-A INPUT -p icmp -j ACCEPT
+-A INPUT -i lo -j ACCEPT
+-A INPUT -m state --state NEW -m tcp -p tcp --dport 22 -j ACCEPT
+-A INPUT -i eth0 -j ACCEPT
+-A INPUT -j REJECT --reject-with icmp-host-prohibited
+-A FORWARD -j REJECT --reject-with icmp-host-prohibited
+COMMIT
+

roles/common/tasks/main.yml

@@ -0,0 +1,65 @@
+---
+    - name: update all packages first
+      yum: pkg=* state=latest
+
+    - name: Remove duplicate CentOS repos
+      file: name=/etc/yum.repos.d/centos6-latest.repo state=absent
+
+    - name: install EPEL6
+      yum: name='http://www.nic.funet.fi/pub/mirrors/fedora.redhat.com/pub/epel/6/i386/epel-release-6-8.noarch.rpm' state=present
+      when: major_relase is "6"
+
+    - name: install software
+      yum: name="{{item}}" state=present
+      with_items:
+        - "xterm"
+        - "sssd"
+        - "gcc" 
+        - "make"
+        - "gcc-c++"
+        - "wget"
+        - "vim"
+        - "man"
+        - "rpm-build"
+        - "pam"
+        - "pam-devel"
+        - "hwloc"
+        - "munge"
+        - "munge-devel"
+        - "munge-libs"
+        - "readline-devel"
+        - "openssl-devel"
+        - "perl-ExtUtils-MakeMaker"
+        - "lua"
+        - "lua-devel"
+        - "lua-posix"
+        - "lua-filesystem"
+
+    - debug: msg="{{groups['all']}}"
+      tags: debug
+
+
+
+#    - name: create /etc/hosts
+#      lineinfile: dest=/etc/hosts regexp='.*{{ item }}$' line="{{ hostvars[item]['ansible_default_ipv4']['address'] }} {{item}}" state=present
+#      when: hostvars[item]['ansible_default_ipv4']['address'] is defined
+#      with_items: groups['all']
+#      tags: debug
+
+    - name: Add cluster hosts to local /etc/hosts
+      sudo: yes
+      action: lineinfile
+        state=present
+        dest=/etc/hosts
+        line="{{ hostvars[item]['ssh_host'] }} {{ item }}"
+      when: hostvars[item]['ssh_host'] is defined
+      with_items: groups.all
+      tags: debug
+
+    - name: copy iptables settings
+      copy: src=iptables
+            dest=/etc/sysconfig/iptables owner=root mode=600
+
+
+    - name: restart iptables
+      service: name=iptables state=restarted

roles/slurm_common/files/slurm

@@ -0,0 +1,3 @@
+        auth    required        pam_localuser.so
+        account required        pam_unix.so
+        session required        pam_limits.so

roles/slurm_common/tasks/main.yml

@@ -0,0 +1,91 @@
+---
+
+#  - name: Copy slurm files
+#    synchronize: mode=pull src={{ item }} dest=/root/rpmbuild/RPMS/x86_64/
+#    delegate_to: slurm-service
+#    with_items:
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-plugins-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-perlapi-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-slurmdb-direct-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-sql-{{ slurm_version }}-1.el6.x86_64.rpm"
+         #- "roles/slurm_common/files/slurm-lua-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-pam_slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-sjstat-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-slurmdbd-{{ slurm_version }}-1.el6.x86_64.rpm"
+         #- "roles/slurm_common/files/slurm-spank-x11-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-torque-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-devel-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-munge-{{ slurm_version }}-1.el6.x86_64.rpm"
+#         - "/root/rpmbuild/RPMS/x86_64/slurm-sjobexit-{{ slurm_version }}-1.el6.x86_64.rpm"
+
+
+
+  - name:  distribute the slurm RPMs to the nodes
+    copy: src={{ item }}
+           dest=/root/rpmbuild/RPMS/x86_64/ 
+    with_items:
+         - "roles/slurm_common/files/slurm-plugins-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-perlapi-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-slurmdb-direct-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-sql-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-lua-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-devel-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-pam_slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-sjstat-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-slurmdbd-{{ slurm_version }}-1.el6.x86_64.rpm"
+        #- "roles/slurm_common/files/slurm-spank-x11-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-torque-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-munge-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "roles/slurm_common/files/slurm-sjobexit-{{ slurm_version }}-1.el6.x86_64.rpm"
+
+
+  - name: install Slurm
+    #yum: name="/root/rpmbuild/RPMS/x86_64/{{ item }}-{{ slurm_version }}-1.el6.x86_64.rpm" state=present
+    yum: name={{ item }} state=present
+    tags: slurm
+    with_items:
+
+         - "/root/rpmbuild/RPMS/x86_64/slurm-plugins-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-perlapi-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-slurmdb-direct-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-sql-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-lua-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-devel-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-pam_slurm-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-sjstat-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-slurmdbd-{{ slurm_version }}-1.el6.x86_64.rpm"
+         #- "/root/rpmbuild/RPMS/x86_64/slurm-spank-x11-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-torque-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-munge-{{ slurm_version }}-1.el6.x86_64.rpm"
+         - "/root/rpmbuild/RPMS/x86_64/slurm-sjobexit-{{ slurm_version }}-1.el6.x86_64.rpm"
+
+
+  - name: pam.d/slurm
+    copy: src=slurm dest=/etc/pam.d/slurm owner=root mode="644"
+    tags: slurm
+
+
+  - name: slurm.conf
+    template: src=slurm.conf.j2 dest=/etc/slurm/slurm.conf owner=root mode="644"
+    tags: slurm
+
+  - name: add slurm user
+    user: name=slurm shell=/sbin/nologin createhome=no home=/nonexixtent system=yes append=yes
+    tags: slurm
+
+  - name: add slurm log dir
+    file: path="/var/log/slurm" state=directory owner=slurm group=slurm mode=750
+    tags: slurm
+
+  - name: add slurm tmp dir
+    file: path="/tmp/slurmd" state=directory owner=slurm group=slurm mode=750
+    tags: slurm
+
+  - name: add slurm tmp dir
+    file: path="/tmp/slurmstate" state=directory owner=slurm group=slurm mode=750
+    tags: slurm
+
+