Tonneson DokuWiki

• Enable Clustering subscription within Red Hat Network
• Install clustering components

yum install ricci system-config-cluster
chkconfig ricci on; service ricci start

• Go to Luci and create a new cluster (or add a node to an existing cluster)

• Login to Luci interface
• Cluster tab, Create a New Cluster
  • Enter a Cluster Name
  • Enter hostnames and root passwords
  • Check “Use locally installed packages”
  • Un-check “Enable Shared Storage Support”
  • Click Sumbit
• Provided you completed the Installation instructions fully, your cluster should form successfully (without rebooting the nodes).

• First, the Xen/KVM hosts must be clustered
• Create a shared key for all to use

# dd if=/dev/urandom of=/etc/cluster/fence_xvm.key bs=4096 count=1

• Ownership and mode on shared key must be root:root & 600

# chown root:root /etc/cluster/fence_xvm.key; chmod 600 /etc/cluster/fence_xvm.key

This step does not need to be done for existing Xen/KVM hosts. The file is already located in /etc/cluster.

• Copy key to all hosts and virtual instances that are clustered
• Add a fence device to each clustered virtual instance
  • Name = xvm_(hostname)
  • Domain = (hostname as listed via a xm list or virsh list)

If you have a host that you definitely know is down (or you do not want to be fenced) and fenced is throwing errors in syslog like:

fenced{some_pid}: fencing node "some_node.domain.com"
fenced{some_pid}: fence "some_node.domain.com" failed

Do this, where some_node matches node name in /etc/cluster/cluster.conf

# fence_ack_manual -n some_node -e

In Red Hat Cluster Suite, there are two methods for configuring LVM on shared storage devices, CLVMD and HA_LVM:

clvmd

An extension to the lvm2 package called lvm2-cluster exists in the Cluster Storage channels for Red Hat Enterprise Linux 4 and 5 on Red Hat Network (RHN). This package provides the clvmd daemon which is responsible for managing clustered volume groups and communicating metadata changes for them back to other cluster members. With this option, the same volume group and logical volumes can be activated and mounted on multiple nodes as long as clvmd is running and the cluster is quorate. This option is better suited for clusters utilizing GFS or GFS2 filesystems, since they are often mounted on multiple nodes at the same time (and thus require that the logical volume be active on each).

NOTE: Although clvmd allows for the same volume group to be activated on multiple systems concurrently, this does not make it possible to mount a non-clustered filesystem such as ext2 or ext3 on multiple nodes at the same time. Doing so will likely cause corruption of the filesystem and data. Proper precautions should be taken to ensure no two nodes mount the same non-clustered filesystem at once.

To configure clvmd, first ensure the lvm2-cluster package is installed. Next the lvm configuration file /etc/lvm/lvm.conf must be modified to tell it to allow for the usage of clustered volume groups. Change the locking_type option to 3 as follows: locking_type=3

Alternatively, you can enable clustered LVM by using the lvmconf command:

# lvmconf --enable-cluster

Now the daemon can be configured to start at boot time:

# chkconfig clvmd on

And started (provided the cluster services are running and the cluster is quorate):

# service clvmd start

Any volume groups with the clustered flag set will be activated when clvmd starts. To determine if a volume group has the clustered flag set, use vgdisplay:

# vgdisplay myVG | grep Clustered
  Clustered             yes

To set this flag while creating a volume group:

# vgcreate -cy <volume group name> <device path>

Or to add the clustered flag to an existing volume group:

# vgchange -cy <volume group name>

NOTE: Any volume group with the clustered flag set will not be available until the clvmd daemon is started and the cluster is quorate.

All nodes in the cluster must have the same view of all storage devices contained within clustered volume groups. For instance if a new device is presented to one node and added to a volume group, errors may be thrown if that same device is not presented to the other nodes. Or if a device is partitioned on one node and then added to a volume group, errors may be thrown if the other nodes have not re-read the partition table by running # partprobe

• Login to the Red Hat Network and alter each server's subscription, adding Cluster Storage access
• Install the package that provides “clvmd”, the Cluster Logical Volume Manager Daemon.
• yum install lvm2-cluster
• chkconfig clvmd on

• Edit /etc/lvm/lvm.conf, uncomment and set the following: locking_type = 3
• service clvmd start