PACEMAKER

Installation

[ALL] Initial setup

Install required packages:

sudo apt-get install pacemaker cman resource-agents fence-agents gfs2-utils gfs2-cluster ocfs2-tools-cman openais drbd8-utils

Make sure each host can resolve all other hosts. Best way to achive this is by adding their IPs and hostnames to /etc/hosts on all nodes. In this example, that would be:

eth0
192.168.244.161 fix
192.168.244.162 foxy
eth1
10.168.244.161   fix-ha
10.168.244.162   foxy-ha

Disable o2cb from starting:

update-rc.d -f o2cb remove

[ALL] Create /etc/cluster/cluster.conf

Paste this into /etc/cluster/cluster.conf :

<?xml version="1.0"?>
<cluster config_version="1" name="pacemaker">
<cman two_node="1" expected_votes="1"> </cman>
   <fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"/>
   <clusternodes>
           <clusternode name="fix" nodeid="1" votes="1">
               <fence>
                       <method name="pcmk-redirect">
                               <device name="pcmk" port="fix"/>
                       </method>
               </fence>
           </clusternode>
           <clusternode name="foxy" nodeid="2" votes="1">
               <fence>
                       <method name="pcmk-redirect">
                               <device name="pcmk" port="foxy"/>
                       </method>
               </fence>
           </clusternode>
   </clusternodes>
 <fencedevices>
   <fencedevice name="pcmk" agent="fence_pcmk"/>
 </fencedevices>
</cluster>

Validate the config

# ccs_config_validate 
Configuration validates

[ALL] Edit /etc/corosync/corosync.conf

Find pacemaker service in /etc/corosync/corosync.conf and bump version to 1:

service {
        # Load the Pacemaker Cluster Resource Manager
        ver:       1
        name:      pacemaker
}

Replace bindnetaddr with the IP of your network. For example:

                bindnetaddr: 10.168.244.0

'0' is not a typo.

[ALL] Enable pacemaker init scripts

update-rc.d -f pacemaker remove
update-rc.d pacemaker start 50 1 2 3 4 5 . stop 01 0 6 .

[ALL] Start cman service and then pacemaker service

deactivat quorum

# echo CMAN_QUORUM_TIMEOUT=0 >> /etc/default/cman

start cman

# service cman start
Starting cluster: 
   Checking if cluster has been disabled at boot... [  OK  ]
   Checking Network Manager... [  OK  ]
   Global setup... [  OK  ]
   Loading kernel modules... [  OK  ]
   Mounting configfs... [  OK  ]
   Starting cman... [  OK  ]
   Waiting for quorum... [  OK  ]
   Starting fenced... [  OK  ]
   Starting dlm_controld... [  OK  ]
   Unfencing self... [  OK  ]
   Joining fence domain... [  OK  ]

# cman_tool nodes
Node  Sts   Inc   Joined               Name
   1   M      4   2012-09-11 19:14:58  fix
   2   M      8   2012-09-11 19:18:35  foxy

[ALL] Enable pacemaker init scripts

update-rc.d -f pacemaker remove
update-rc.d pacemaker start 50 1 2 3 4 5 . stop 01 0 6 .

service pacemaker start
Starting Pacemaker Cluster Manager: [  OK  ]

[ONE] Setup resources

Wait for a minute until pacemaker declares all nodes online:
# crm status
============
Last updated: Fri Sep  7 21:18:12 2012
Last change: Fri Sep  7 21:17:17 2012 via crmd on fix
Stack: cman
Current DC: fix - partition with quorum
Version: 1.1.6-9971ebba4494012a93c03b40a2c58ec0eb60f50c
2 Nodes configured, unknown expected votes
0 Resources configured.
============ 

Online: [ fix foxy ]

[ALL] install services that will fail over between servers

In this example, I'm installing apache2

sudo apt-get install apache2

Disable their init scripts:

update-rc.d -f apache2 remove

In this example, I'll create failover for the apache2. I'll also add one additional and tie apache2.

sudo crm configure edit
primitive resAPACHE ocf:heartbeat:apache \
       params configfile="/etc/apache2/apache2.conf" httpd="/usr/sbin/apache2" \
       op monitor interval="5s"
primitive resIP-APACHE ocf:heartbeat:IPaddr2 \
       params ip="192.168.244.171" cidr_netmask="21" nic="eth0
group groupAPACHE resIP-APACHE resAPACHE
order apache_after_ip inf: resIP-APACHE:start resAPACHE:start

Links

• http://www.linux-magazin.de/Online-Artikel/GFS2-und-OCFS2-zwei-Cluster-Dateisysteme-im-Linux-Kernel
• https://wiki.ubuntu.com/ClusterStack/Precise
• https://wiki.ubuntu.com/ClusterStack/LucidTesting
• http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Clusters_from_Scratch
• http://www.drbd.org/users-guide/
• http://wiki.techstories.de/display/IT/Pacemaker+HA-Cluster+Schulung+GFU+Koeln
• http://theclusterguy.clusterlabs.org/post/907043024/introducing-the-pacemaker-master-control-process-for
• http://www.hastexo.com/resources/hints-and-kinks/fencing-libvirtkvm-virtualized-cluster-nodes
• http://linux.die.net/man/5/cluster.conf
• http://burning-midnight.blogspot.de/2012/05/cluster-building-ubuntu-1204.html
• http://burning-midnight.blogspot.de/2012/07/cluster-building-ubuntu-1204-revised.html
• http://www.hastexo.com/resources/hints-and-kinks/ocfs2-pacemaker-debianubuntu
• http://www.gossamer-threads.com/lists/drbd/users/23267
• http://blog.simon-meggle.de/tutorials/score-berechnung-im-pacemaker-cluster-teil-1/
• http://www.clusterlabs.org/doc/en-US/Pacemaker/1.0/html/Pacemaker_Explained/s-cluster-options.html
• http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Pacemaker_Explained/s-ocf-return-codes.html
• https://aseith.com/plugins/viewsource/viewpagesrc.action?p
• http://publications.jbfavre.org/virtualisation/cluster-xen-corosync-pacemaker-drbd-ocfs2.enageId=9601113
• http://www.youtube.com/watch?v=3GoT36cK6os&feature=youtu.be
• http://www.debian-administration.org/articles/578
• http://blog.datentraeger.li/?cat=31
• http://publications.jbfavre.org/virtualisation/cluster-xen-corosync-pacemaker-drbd-ocfs2.en