Merge branch 'master' of https://github.com/monash-merc/ansible_cluster_in_a_box

roles/OpenVPN-Client/tasks/main.yml

@@ -3,6 +3,6 @@
  include: installOpenVPN.yml
 
 - name: "Start OpenVPN"
-  service: name=openvpn state=started
+  service: name=openvpn state=started enabled=yes
   sudo: true
 

roles/OpenVPN-Server/tasks/main.yml

@@ -3,5 +3,5 @@
  include: installOpenVPN.yml
 
 - name: "Start OpenVPN"
-  service: name=openvpn state=started
+  service: name=openvpn state=started enabled=yes
   sudo: true

roles/apache2/tasks/apacheDebian.yml

@@ -29,6 +29,6 @@
 
 -
  name: "Starting Apache2"
- service: name=apache2 state=started
+ service: name=apache2 state=started enabled=yes
  sudo: true
 

roles/calculateSlurmConf/templates/slurm.conf.j2

@@ -30,7 +30,7 @@ SwitchType=switch/none
 MpiDefault=pmi2
 SlurmctldPidFile={{ slurmpiddir }}/slurmctld.pid
 SlurmdPidFile={{ slurmpiddir }}/slurmd.pid
-ProctrackType=proctrack/linuxproc
+ProctrackType=proctrack/cgroup
 #PluginDir=
 CacheGroups=0
 #FirstJobId=

roles/ldapclient/tasks/configLdapClient.yml

@@ -40,5 +40,5 @@
   notify: restart sssd
 
 - name: "start sssd"
-  service: name=sssd state=started
+  service: name=sssd state=started enabled=yes
   sudo: true

roles/lustre-client/tasks/main.yml

 ---
 - include_vars: "{{ ansible_distribution }}_{{ ansible_distribution_major_version }}.yaml"
 
-- name: copy rpms
+- name: copy rpms/debs
   copy: dest=/tmp/ src=lustre-install/{{ item }}
   with_items:
     "{{ lustre_pkgs }}"
 
+
 #- name: install rpms
 #  yum: name="/tmp/{{ item }}"
 #  sudo: true
@@ -14,13 +15,38 @@
 - name: install rpms
   yum: name=/tmp/lustre-client-modules-2.7.0-3.10.0_229.14.1.el7.x86_64.x86_64.rpm
   sudo: true
+  when: ansible_os_family == "RedHat"
 
 - name: install rpms
   yum: name=/tmp/lustre-client-2.7.0-3.10.0_229.14.1.el7.x86_64.x86_64.rpm
   sudo: true
+  when: ansible_os_family == "RedHat"
+
+# instructions to build these debs:
+# Instantiate an Ubuntu 14.04 instance
+# git clone git://git.hpdd.intel.com/fs/lustre-release.git
+# cd lustre-release
+# optionally git checkout 0754bc8f2623bea184111af216f7567608db35b6 <- I know this commit works on Ubuntu, but I had a lot of trouble with other branches
+# sh autogen.sh
+# ./configure --enable-dist --disable-doc  --disable-server --disable-dependency-tracking --with-o2ib=/var/lib/dkms/mlnx-ofed-kernel/3.1/build/
+# mkdir BUILD
+# cd BUILD
+# ln -s ../lustre-2.7.62.tar.gz lustre-2.7.62.orig.tar.gz
+# tar zxvf ../lustre-2.7.62.tar.gz
+# cd lustre-2.7.62
+# ./configure --disable-doc  --disable-server --disable-dependency-tracking --with-o2ib=/var/lib/dkms/mlnx-ofed-kernel/3.1/build/
+# vi debian/changelog (the version number on the first line is incorrect)
+# make debs
+#
+- name: install debs
+  apt: name="/tmp/{{ item }}"
+  sudo: true
+  with_items: "{{ lustre_pkgs }}"
+  when: ansible_distribution == "Ubuntu" and ansible_distribution_major_version == "14"
 
-  
 - name: "Mount lustre filesystems"
   mount: name="{{ item.mntpt }}" src="{{ item.servers }}"/"{{ item.src }}" state="mounted" fstype="lustre" opts="_netdev,flock" 
   sudo: true
   with_items: "{{ mntlist }}"
+
+

roles/lustre-client/vars/Ubuntu_14.yaml

+---
+lustre_pkgs:
+  - linux-patch-lustre_2.7.62-1_all.deb
+  - lustre-client-modules-3.13.0-58-generic_2.7.62-1_amd64.deb
+  - lustre-utils_2.7.62-1_amd64.deb

roles/mellanox_drivers/tasks/main.yml

 ---
+- include_vars: mellanoxVars.yml
+
 
 - name: yum install dependencies
   yum:  name=perl,pciutils,gtk2,atk,cairo,gcc-gfortran,libxml2-python,tcsh,libnl,lsof,tcl,tk
   sudo: true
   ignore_errors: true
+  when: ansible_os_family == "RedHat"
   
 - name: yum update to upgrade kernel
   shell: "yum update -y"
   sudo: true
-  when: ansible_os_family=="RedHat" and ansible_distribution_major_version == "7"
+  ignore_errors: true
+  when: ansible_os_family == "RedHat"
   
 #
 # A REBOOT IS NEEDED AFTER a KERNEL UPDATE
@@ -19,6 +23,7 @@
   poll: 0
   ignore_errors: true
   sudo: true
+  when: ansible_os_family == "RedHat"
 
 - name: waiting for server to come back
   #local_action: wait_for host={{ ansible_host }} state=started port=22 delay=10 search_regex=OpenSSH
@@ -36,26 +41,78 @@
   ignore_errors: true
 
 - name: copy driver source
-  unarchive: copy=yes src=files/MLNX_OFED_LINUX-3.1-1.0.3-rhel7.1-x86_64-ext.tgz dest=/tmp
+  #make this a variable
+  unarchive: copy=yes src="files/{{ MELLANOX_DRIVER_SRC }}.tgz" dest=/tmp 
   sudo: true
-  when: drivers_installed|failed and ansible_os_family=="RedHat" and ansible_distribution_major_version == "7"
+  #when: drivers_installed|failed and ansible_os_family=="RedHat" and ansible_distribution_major_version == "7"
+  when: drivers_installed|failed 
 
 - name: install drivers
   shell: ./mlnxofedinstall -q
   args:
-    chdir: /tmp/MLNX_OFED_LINUX-3.1-1.0.3-rhel7.1-x86_64-ext
+    #more changes
+    chdir: "/tmp/{{ MELLANOX_DRIVER_SRC }}"
   sudo: true
-  when: drivers_installed|failed and ansible_distribution_major_version == "7"
+  when: drivers_installed|failed 
 
 #
+# get IP address before reboot
+#
+- name: get IP address
+  local_action: command ./scripts/map_ib_ip.pl {{ inventory_hostname }}
+  register: ip_address
+  sudo: false
+
+  
+- name: template IP address
+  template: dest=/etc/sysconfig/network-scripts/ifcfg-ens6 src=ifcfg-ens6.j2 owner=root group=root
+  sudo: true
+  when:  ansible_os_family=="RedHat" and  ansible_distribution_major_version == "7"
+#ubuntu equivalent of previous command
+- name: Ubuntu network interfaces - line 1
+  lineinfile:
+  args: 
+      dest: /etc/network/interfaces
+      line: auto {{ MELLANOX_DEVICE_NAME }}
+      state: present
+  sudo: true
+  when: ansible_os_family=="Debian"
+- name: Ubuntu network interfaces - line 2
+  lineinfile:
+  args: 
+      dest: /etc/network/interfaces
+      line: iface {{ MELLANOX_DEVICE_NAME }} inet static
+      state: present
+      insertafter: "auto {{ MELLANOX_DEVICE_NAME }}"
+  sudo: true
+  when: ansible_os_family=="Debian"
+- name: Ubuntu network interfaces - line 3
+  lineinfile:
+  args: 
+      dest: /etc/network/interfaces
+      line: address {{ ip_address.stdout }}
+      state: present
+      insertafter: "iface {{ MELLANOX_DEVICE_NAME }} inet static"
+  sudo: true
+  when: ansible_os_family=="Debian"
+#
 # A REBOOT IS NEEDED AFTER SUCCESSFUL INSTALL
 #
 - name: restart machine
-  command: sleep 5; sudo shutdown -r now "Ansible updates triggered"
+  command: "sleep 5; sudo shutdown -r now"
   async: 2
   poll: 0
   ignore_errors: true
   sudo: true
+  when: ansible_os_family=="Centos"
+
+- name: restart machine for Ubuntu -cos it is 'special'
+  command: "sudo shutdown -r now"
+  async: 2
+  poll: 0
+  ignore_errors: true
+  sudo: true
+  when: ansible_os_family=="Debian"
 
 - name: waiting for server to come back
   local_action: wait_for host={{ ansible_ssh_host }} state=started port=22 delay=10 search_regex=OpenSSH
@@ -63,23 +120,11 @@
 
 - name: waiting for server to come back 2
   local_action: wait_for host={{ ansible_ssh_host }} state=started port=22 delay=10 search_regex=OpenSSH
-  sudo: false
 
-- name: get IP address
-  local_action: command ./scripts/map_ib_ip.pl {{ inventory_hostname }}
-  register: ip_address
-  
-- name: template IP address
-  template: dest=/etc/sysconfig/network-scripts/ifcfg-ens6 src=ifcfg-ens6.j2 owner=root group=root
-  sudo: true
-  when: ansible_distribution_major_version == "7"
- 
 - name: bring up interface
-  command: ifup ens6
+  #variable=eth0 or ens6
+  command: ifup {{ MELLANOX_DEVICE_NAME }} 
   sudo: true
   when: ansible_distribution_major_version == "7"
    
   
-
-
-

roles/mellanox_drivers/vars/mellanoxVars.yml

+---
+ #note. do not add '.tgz' to driver src. done in playbook
+ MELLANOX_DRIVER_SRC: "{% if ansible_os_family == 'RedHat'  %}MLNX_OFED_LINUX-3.1-1.0.3-rhel7.1-x86_64-ext{% elif ansible_os_family == 'Debian' %}MLNX_OFED_LINUX-3.1-1.0.3-ubuntu14.04-x86_64{% endif %}"
+ MELLANOX_DEVICE_NAME: "{% if ansible_os_family == 'RedHat'  %}ens6{% elif  ansible_os_family == 'Debian' %}eth1{% endif %}"

roles/nfs-common/handlers/main.yml

 --- 
 - 
   name: "Run rpcbind service"
-  service: "name=rpcbind state=started"
+  service: "name=rpcbind state=started enabled=yes"

roles/nfs-server/tasks/startServer.yml

@@ -5,16 +5,16 @@
   with_items: exportList
 
 - name: "Starting rpcbind"
-  service: "name=rpcbind state=started"
+  service: "name=rpcbind state=started enabled=true"
   sudo: true
   when: ansible_os_family == "RedHat"
 
 - name: "Start the Server"
-  service: "name=nfs state=started"
+  service: "name=nfs state=started enabled=true"
   sudo: true
   when: ansible_os_family == "RedHat"
 
 - name: "Start the Server"
-  service: "name=nfs-kernel-server state=started"
+  service: "name=nfs-kernel-server state=started enabled=true"
   sudo: true
   when: ansible_os_family == "Debian"

roles/shibboleth-sp/tasks/shibbolethConfig.yml

@@ -140,4 +140,4 @@
 -
  name: "Starting shibboleth"
  sudo: true
- service: name=shibd state=started
+ service: name=shibd state=started enabled=yes

roles/slurm-start/tasks/main.yml

@@ -36,7 +36,7 @@
   when: use_systemd is defined
 
 - name: start munge
-  service: name=munge state=started
+  service: name=munge state=started enabled=yes
   sudo: true
 
 - name: start slurmdbd