Chapter 5. Manual Configuration of a Cluster | ||||
|---|---|---|---|---|
 | 4.2. Migrating a Cluster Resource | 5.2. Configuring Constraints |  | |
Chapter 5. Manual Configuration of a Cluster | ||||
|---|---|---|---|---|
| 4.2. Migrating a Cluster Resource | 5.2. Configuring Constraints | | |
Manual configuration of a Heartbeat cluster is often the most effective way of creating a reliable cluster that meets specific needs. Because of the extensive configurability of Heartbeat and the range of needs it can meet, it is not possible to document every possible scenario. To introduce several concepts of the Heartbeat configuration and demonstrate basic procedures, consider a real world example of an NFS file server. The goal is to create an NFS server that can be built with very low-cost parts and is as redundant as possible. For this, set up the following cluster:
Two machines that have redundant hardware
Data is mirrored on the disks of those machines with drbd
Only one machine at a time accesses and exports the data
Assign a special IP address to the computer for exporting the file system
Before starting with the cluster configuration, set up two nodes as described in Chapter 2, Installation and Setup. In addition to the system installation, both should have a data partition of the same size to setup drbd.
The configuration splits into two main parts. First, all the
resources must be configured. After this, create a set
of constraints that define the starting rules for
those resources.
All the configuration data is written in XML. For convenience, the example relies on snippets that may be loaded into the cluster configuration individually.
There are three types of RAs (Resource Agents) available with Heartbeat. First, there are legacy Heartbeat 1 scripts. Heartbeat can make use of LSB initialization scripts. Finally, Heartbeat has its own set of OCF (Open Cluster Framework) agents. This documentation concentrates on LSB scripts and OCF agents.
All LSB scripts are commonly found in the directory
/etc/init.d. They must have several actions
implemented, which are at least start,
stop, restart,
reload, force-reload, and
status as explained in http://www.linux-foundation.org/spec/refspecs/LSB_1.3.0/gLSB/gLSB/iniscrptact.html.
The configuration of those services is not standardized. If you
intend to use an LSB script with Heartbeat, make sure that you
understand how the respective script is configured.
Often you can find some documentation to this in the
documentation of the respective package in
/usr/share/doc/packages/<package_name>.
When used by Heartbeat, the service should not be touched by other means. This means that it should not be started or stopped on boot or reboot and it also should not be started or stopped manually. However, if you want to check if the service is configured properly, you may start it manually, but make sure that it is stopped again before Heartbeat takes over.
Before using an LSB resource, make sure that the configuration of this resource is present and identical on all cluster nodes. The configuration is not managed by Heartbeat. You must take care of that yourself.
All OCF agents are located in
/usr/lib/ocf/resource.d/. These are small
programs that have a functionality similar to that of LSB scripts.
However, the configuration is always done with environment
variables. All OCF Resource Agents are required to have at least the
actions start, stop,
status, monitor, and
meta-data. The meta-data
action is retrieves information about how to
configure the agent. For example, if you want to know more about
the IPaddr agent, use the command:
/usr/lib/ocf/resource.d/heartbeat/IPaddr meta-data
The output is lengthy information in a simple XML format. Unfortunately, there is not yet an automation that converts this to some print quality documentation, but it may still be read in this format by the experienced user. Basically this XML format has three sections—first several common descriptions, second all the available parameters, and last the available actions for this agent.
A typical parameter of a OCF RA as shown with the
meta-data command looks like this:
<parameter name="ip" unique="1" required="1"><longdesc lang="en">
The IPv4 address to be configured in dotted quad notation, for example "192.168.1.1". </longdesc> <shortdesc lang="en">IPv4 address</shortdesc> <content type="string" default="" />
</parameter>
This is part of the IPaddr RA. The information
about how to configure the parameter of this RA can be read as
follows:
| |
The name of the |
| | The description of the parameter is available in a long and a short description tag. |
| | The content of the value of this parameter is a string. There is no default value available for this resource. |
Find a configuration example for this RA at Chapter 3, Setting Up a Simple Resource.
To set up the NFS server, three resources are needed: a
file system resource, a drbd resource, and a group of an NFS server and an
IP address. You can write each of
the resource configurations to a separate file then load them to
the cluster with cibadmin -C -o resources -x
resource_configuration_file.
The filesystem resource is configured as an
OCF primitive resource. It has the task to mount and unmount a device
to a directory on start and stop requests. In this case, the
device is /dev/drbd0 and the directory to use as
mount point is /srv/failover.
The file system used is reiserfs.
The configuration for this resource looks like the following:
<primitive id="filesystem_resource" class="ocf" provider="heartbeat" type="Filesystem">
<instance_attributes id="ia-filesystem_1">
<attributes>
<nvpair id="filesystem-nv-1" name="device" value="/dev/drbd0"/>
<nvpair id="filesystem-nv-2" name="directory" value="/srv/failover"/>
<nvpair id="filesystem-nv-3" name="fstype" value="reiserfs"/>
</attributes>
</instance_attributes>
</primitive>
Before starting with the drbd Heartbeat configuration, set up a
drbd device manually. Basically this is configuring
drbd in /etc/drbd.conf and letting it
synchronize. The exact procedure for configuring drbd is
described in the Storage Administration Guide. For now, assume that you
configured a resource r0 that may be accessed
at the device /dev/drbd0 on both of your
cluster nodes.
The drbd resource is an OCF master slave resource.
This can be found in the description of the metadata of the drbd
RA. However, more important is that there are the actions
promote and demote in the
actions section of the metadata. These are
mandatory for master slave resources and commonly not available
to other resources.
For Heartbeat, master slave resources may have multiple masters on
different nodes. It is even possible to have a master and slave on
the same node. Therefore, configure this resource in a
way that there is exactly one master and one slave, each running on
different nodes. Do this with the meta
attributes of the master_slave resource.
Master slave resources are a special kind of clone resources in
Heartbeat. Every master and every slave counts as a clone.
<master_slave id="drbd_resource" ordered="false"><nvpair id="drbd-nv-4" name="master_node_max" value="1"/>
<nvpair id="drbd-nv-5" name="notify" value="yes"/>
</attributes> </meta_attributes> <primitive id="drbd_r0" class="ocf" provider="heartbeat" type="drbd">
<instance_attributes id="ia-drbd_1"> <attributes> <nvpair id="drbd-nv-5" name="drbd_resource" value="r0"/>
</attributes> </instance_attributes> </primitive> </master_slave>
| |
The master element of this resource is
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
The actually working RA inside this |
| | The most important parameter this resource needs to know about is the name of the drbd resource to handle. |
To make the NFS server always available at the same IP address, use an additional IP address as well as the ones the machines use for their normal operation. This IP address is then assigned to the active NFS server in addition to the system's IP address.
The NFS server and the IP address of the NFS server should always be active on the same machine. In this case, the start sequence is not very important. They may even be started at the same time. These are the typical requirements for a group resource.
Before starting the Heartbeat RA configuration, configure
the NFS server with YaST. Do not let the system start the NFS
server. Just set up the configuration file. If you want to
do that manually, see the manual page exports(5) (man 5
exports. The
configuration file is /etc/exports.
The NFS server is configured as an LSB resource.
Configure the IP address completely with the Heartbeat RA configuration. No additional modification is necessary in the system. The IP address RA is an OCF RA.
<group id="nfs_group">
| | In a group resource, there may be several other resources. It must have an ID set. |
| |
The |
| |
The |
| |
There is only one mandatory instance attribute in the
|
