• manage-users
• manage-nfs
• manage-lvm

# vim /home/user/test_manifest.pp

group { 'user_test':
  ensure => 'present',
  gid    => '13000',
     }

user { 'user_test':
  ensure     => present,
  uid        => '13000',
  gid        => '13000',
  shell      => '/bin/bash',
  home       => '/home/user_test',
  password         => 'password',
  password_max_age => '99999',
  password_min_age => '0',
}

file { '/Puppet_test/directory':
    ensure => 'directory',
    owner  => 'user_test',
    group  => 'user_test',
    mode   => '0750',
  }

file { "/Puppet_test/directory/file":
        ensure => "present",
        owner => "user_test",
        group => "user_test",
        mode => "0644",
        content => "This is a test, puppet manifest"
}

#!/bin/bash

DIRECTORY=prueba
#No arguments
if [ $# -eq 0 ];
then
 echo ""
 echo "Option -$OPTARG needs an argument." >&2
 echo ""
 exit 1
fi

while getopts ":c:*:" option; do
 case $option in
 c)
 if [[ $2 == "upper" ]];then
 for file in $(ls $DIRECTORY); do mv -- "$DIRECTORY/$file" "$DIRECTORY/${file^^}"; done
 echo "UPPER"
 elif [[ $2 == "lower" ]];then
 for file in $(ls $DIRECTORY); do mv -- "$DIRECTORY/$file" "$DIRECTORY/${file,,}"; done
 echo "LOWER"
 else
 echo "Invalid option"
 fi
 ;;

 \?)
 echo "Invalid option: -$OPTARG" >&2
 exit 1
 ;;
 
 echo "Option -$OPTARG requires an argument." >&2
 exit 1
 ;;

 esac
done

*.- Mount a CD/DVD iso image:

# mount -t iso9660 -o loop /dev/cdrom /mnt/

*.- Create a local repo with the CD/DVD, to install all the packages require:

# vim /etc/yum.repos.d/local.repo
[local-repo]
name=Local Repo
baseurl=file:///local-repo
enabled=1
gpgcheck=0

*.- The following packages are required for a X11 server:

 
# yum install -y xorg-x11-font* xorg-x11-utils xorg-x11-xauth xeyes

*.- Then you will be able to export X’s from that server, connecting through ssh with flags -XY:

*.- Boot into rescue mode (with a USB image).
*.- Continue

# loadkeys es
# chroot /mnt/sysimage
# ifup bond0

##Reinstalling GRUB 2 on BIOS-Based Machines

# grub2-install /dev/sda 
# grub2-mkconfig -o /boot/grub2/grub.cfg 

##Reinstalling GRUB 2 on UEFI-Based Machines

# yum reinstall grub2-efi shim
# grub2-mkconfig -o /boot/efi/EFI/centos/grub.cfg

# nmcli con show
# nmcli con add type bond con-name bond0 ifname bond0 mode balance-rr
# nmcli con add type bond-slave con-name bond0-slave-ens3 ifname ens3 master  bond0
# nmcli con add type bond-slave con-name bond0-slave-ens7 ifname ens3 master bond0
# nmcli con mod bond0 ipv4.addresses 192.168.1.10/24
# nmcli con mod bond0 ipv4.gateway 192.168.1.1
# nmcli con mod bond0 ipv4.method manual
# nmcli con up bond0

###Teaming

# nmcli con show
# nmcli con add type team con-name team0 ifname team0 config '{"runner": {"name" : "loadbalancer"}}'
# nmcli con add type team-slave con-name slave-ens3 ifname ens3 master team0
# nmcli con add type team-slave con-name slave-ens7 ifname ens7 master team0
# nmcli con mod team0 ipv4.addresses 192.168.1.10/24
# nmcli con mod team0 ipv4.gateway 192.168.1.1
# nmcli con mod team0 ipv4.method manual
# nmcli con up team0

Examples:

# man nmcli-examples

*.- Mount a CD/DVD iso image:

# mount -t iso9660 -o loop /dev/cdrom /mnt/

*.- Create a local repo with the CD/DVD, to install all the packages require:

# vim /etc/yum.repos.d/local.repo
[local-repo]
name=Local Repo
baseurl=file:///local-repo
enabled=1
gpgcheck=0

*.- Create a directory to put all CD/DVD data

# mkdir /local-repo
# cp -pr /mnt/* /local-repo

*.- CREATEREPO

# yum install -y createrepo
# createrepo /local-repo

*.- FTP SERVER

# yum install -y vsftp
# systemctl enable vsftp
# systemctl start vsftp
# firewall-cmd --add-service=ftp --permanent
# firewall-cmd --reload
# mkdir /var/ftp/pub/repo
# echo "/local-repo /var/ftp/pub/repo	none	defaults,bind	0 0" /etc/fstab
# mount /var/ftp/pub/repo

*.- Check the SElinux context and apply it to the new dir:

# ls -sZ /var/ftp/pub/
total 4
drwxr-xr-x. root root unconfined_u:object_r:public_content_t:s0 repo
# chcon -R -t public_content_t /var/ftp/pub/repo

***REMOTE (client)****

# vim /etc/yum.repos.d/remote.repo
[remote-repo]
name=Remote Repo
baseurl=ftp://server-local-repo/pub/repo
enabled=1
gpgcheck=0

# yum clean all
# yum repolist all

# yum provides /*sepolicy
# yum install -y policycoreutils-devel
# man -k
# sepolicy manpage -a -p /usr/share/man/man8
# mandb 
# man -k _selinux | grep nfs
# man nfsd_selinux

# yum install -y nfs-utils policycoreutils-python
# echo -e "/srv/nfsexport	*(rw)" >> /etc/exports
# semanage fcontext -a -t nfs_t "/srv/nfsexport/(/.*)?"
# restorecon -R -v /srv/nfsexport
# systemctl start nfs-server;systemctl enable nfs-server
# firewall-cmd --get-services
# firewall-cmd --add-service={nfs,mountd,rpc-bind} --permanent
# firewall-cmd --reload

###Client

# showmount -e 192.168.122.201
# echo -e "192.168.122.201:/srv/nfsexport /export-client  nfs   defaults,_netdev   0 0" >> /etc/fstab 

5.1.1.- VMWare:
</clusternodes>
    <clusternode name="node-clu" nodeid="1">
        <fence>
            <method name="vmwarefence">
		<device name="vmwarefence" port="node1" ssl="on" uuid="4229f0ef-1fb7-6adb-4c30-da3d92f2e386"/>
            </method>
        </fence>
    </clusternode>
</clusternodes>

5.1.2.- KVM:

<clusternode name="NAME_PRIVATE_NW" nodeid="NODE_ID">
  <fence>
    <method name="NAME_fence_Method">
	<device identity_file="/path/to/pub_key" name="fence_agent_name" port="VM_UUID" secure="Though_SSH"/>
    </method>
  </fence>
</clusternodes>

<clusternodes>
   <clusternode name="node1" nodeid="1">
      <fence>
	<method name="virsh_fence_mtd">
	   <device identity_file="/root/.ssh/id_rsa" name="virsh_fence" port="59d34fee-688d-f2e1-3569-f3ce3b900463" secure="on"/>
	</method>
     </fence>
   </clusternode>
</clusternodes>

5.2.- : As I said before, you may use as many as fence devices you have available, then you can stablish a priority to trigger the one you trust more. In this section it will define things such as an fence_agent, IP address, username and password that can be used to login to and control the fence device.

5.2.2.- KVM:

<fencedevices>
	<fencedevice agent="fence_virsh"  ipaddr="192.168.24.24" login="kvm_user" name="virsh_fence" passwd="p4ssw0rd" verbose="true" debug="/var/log/fence_kvm.log"/>
</fencedevices>

5.2.2.- VMWare:

<fencedevices>
	<fencedevice agent="fence_vmware_soap" ipaddr="vcententer.test.lan" login="vmware_user" name="vmware_fence" passwd="p4ssw0rd"/>
</fencedevices>

5.3.- Fencing check:

# fence_check
fence_check run at Sun Jul 19 17:07:29 CEST 2015 pid: 23476
Testing node1 method 1: success
Testing node2 method 1: success
Testing node3 method 1: success

5.3.1.- Crasing a node to test the fence, on the node will crash launch:

# sync; echo s > /proc/sysrq-trigger
# echo c > /proc/sysrq-trigger

Checking the logs:

# tail -f /var/log/messages
Jan 23 18:50:03 node1 corosync[2073]:   [TOTEM ] A processor joined or left the membership and a new membership was formed.
Jan 23 18:50:03 node1 corosync[2073]:   [QUORUM] Members[7]: 1 2 3 4 5 6 7
Jan 23 18:50:03 node1 corosync[2073]:   [QUORUM] Members[7]: 1 2 3 4 5 6 7
Jan 23 18:50:03 node1 corosync[2073]:   [CPG   ] chosen downlist: sender r(0) ip(193.2.123.10) ; members(old:6 left:0)
Jan 23 18:50:03 node1 corosync[2073]:   [MAIN  ] Completed service synchronization, ready to provide service.
Jan 23 18:50:14 node1 kernel: dlm: got connection from 3
Jan 23 19:21:40 node1 ricci[32280]: Executing '/usr/bin/virsh nodeinfo'
Jan 23 19:21:40 node1 ricci[32282]: Executing '/usr/libexec/ricci/ricci-worker -f /var/lib/ricci/queue/1906940672'
Jan 23 19:21:40 node1 modcluster: Updating cluster.conf
Jan 23 19:21:41 node1 corosync[2073]:   [QUORUM] Members[7]: 1 2 3 4 5 6 7
Jan 23 19:23:57 node1 corosync[2073]:   [QUORUM] Members[6]: 1 3 4 5 6 7
Jan 23 19:23:57 node1 corosync[2073]:   [TOTEM ] A processor joined or left the membership and a new membership was formed.
Jan 23 19:23:57 node1 kernel: dlm: closing connection to node 2
Jan 23 19:23:57 node1 corosync[2073]:   [CPG   ] chosen downlist: sender r(0) ip(193.2.123.10) ; members(old:7 left:1)
Jan 23 19:23:57 node1 corosync[2073]:   [MAIN  ] Completed service synchronization, ready to provide service.
Jan 23 19:24:28 node1 corosync[2073]:   [TOTEM ] A processor joined or left the membership and a new membership was formed.
Jan 23 19:24:28 node1 corosync[2073]:   [QUORUM] Members[7]: 1 2 3 4 5 6 7
Jan 23 19:24:28 node1 corosync[2073]:   [QUORUM] Members[7]: 1 2 3 4 5 6 7
Jan 23 19:24:28 node1 corosync[2073]:   [CPG   ] chosen downlist: sender r(0) ip(193.2.123.10) ; members(old:6 left:0)
Jan 23 19:24:28 node1 corosync[2073]:   [MAIN  ] Completed service synchronization, ready to provide service.
Jan 23 19:24:39 node1 kernel: dlm: got connection from 2

5.4.- Failover Domains: Defines a logical grouping of nodes which can fail over to each other:

		<failoverdomains>
			<failoverdomain name="node1_domain" nofailback="1" ordered="0" restricted="1">
				<failoverdomainnode name="node1"/>
			</failoverdomain>
			<failoverdomain name="node2_domain" nofailback="1" ordered="0" restricted="1">
				<failoverdomainnode name="node2"/>
			</failoverdomain>
			<failoverdomain name="node3_domain" nofailback="1" ordered="0" restricted="1">
				<failoverdomainnode name="node3"/>
			</failoverdomain>
			<failoverdomain name="All_nodes" nofailback="1" ordered="0" restricted="0">
				<failoverdomainnode name="node1"/>
				<failoverdomainnode name="node2"/>
				<failoverdomainnode name="node3"/>
			</failoverdomain>
		</failoverdomains>

5.5.- Shared Resources: A resource used by a cluster service, such GFS, a shared filesystem, ip address, NFS resource, script, or Samba service.

<resources>
			<ip address="192.168.5.241" monitor_link="1"/>
			<script file="/etc/init.d/clvmd" name="clvmd"/>
			<script file="/etc/init.d/vsftpd" name="vsftpd"/>
			<clusterfs device="/dev/mapper/VG_shareable-lv_ftp" force_unmount="1" fsid="ec13017f-55e5-c8ba-d546-a4b5ab0dc59c" fstype="gfs2" mountpoint="/ClusterFS/FTP/" name="lv_ftp" options="defaults,noatime,nodiratime,acl"/>
			<clusterfs device="/dev/mapper/VG_shareable-lv_shareable" force_unmount="1" fsid="379d0a7a-f14c-9150-dbfe-b37f39428a82" fstype="gfs2" mountpoint="/ClusterFS/Shareable/" name="lv_shareable" options="defaults,noatime,nodiratime,acl"/>
</resources>

5.6.- Services: An HA service provided by the cluster, which combines together shared resources within a failover domain utilizing one or more nodes and their associated fence device.

<service autostart="1" domain="All_nodes" exclusive="0" name="float_ip" recovery="relocate">
			<ip ref="192.168.5.241"/>
		</service>
		<service autostart="1" domain="All_nodes" exclusive="0" name="float_vsftp" recovery="relocate">
			<script ref="vsftp"/>
		</service>
		<service autostart="1" domain="node1_domain" exclusive="0" name="storage_node1" recovery="restart">
			<script ref="clvmd">
				<clusterfs ref="lv_ftp"/>
				<clusterfs ref="lv_shareable"/>
			</script>
		</service>
		<service autostart="1" domain="node2_domain" exclusive="0" name="storage_node2" recovery="restart">
			<script ref="clvmd">
				<clusterfs ref="lv_ftp"/>
				<clusterfs ref="lv_shareable"/>
			</script>
		</service>
		<service autostart="1" domain="node3_domain" exclusive="0" name="storage_node3" recovery="restart">
			<script ref="clvmd">
				<clusterfs ref="lv_ftp"/>
				<clusterfs ref="lv_shareable"/>
			</script>
		</service>