Primary Server
Complete the following steps on the primary server to configure the cluster:
- Create TCP Communication (Comm) Path(s)
- Verify the Communication (Comm) Path(s)
Before you begin, SIOS recommends at least two TCP communications paths between each server within the cluster to each remote server for heartbeat redundancy.
Access the LifeKeeper GUI
The LifeKeeper Graphical User Interface (GUI) is a Java based application that can be run as a native Linux application, or as an applet within your Java-Enabled Web Browser.
The LifeKeeper GUI is based on Java RMI with callbacks. Hostnames must be resolvable or you may receive a Java 115 or 116 errors.
- Verify that both short and fully qualified hostnames of all cluster nodes resolve to the proper locations
# ping LinuxPrimary
# ping LinuxPrimary.domain.com
# ping LinuxSecondary
# ping LinuxSecondary.domain.com
- To start the LifeKeeper Linux GUI Application:
a. /opt/LifeKeeper/bin/lkGUIapp &
- To Connect to the LifeKeeper GUI Applet from a Web Browser, go to:
a. http://<hostname>:81
- Enter the name of the server you wish to connect to (this field will be populated with the name of the server you are on, if you are running the GUI from a server with LifeKeeper installed) along with your root credentials and click OK.
Create Communication (Comm) Paths
- Within the LifeKeeper GUI, from the File menu, select Connect. Enter the name of your Secondary server, login and password when the Cluster Connect window displays.
- Within the LifeKeeper GUI, click the Create Comm Path button on the toolbar. You can also right click one of the servers and click Create Comm Path from the pop-up menu as well.
- Select your Local and Remote Server(s) from the list box. If a server does not appear in the list box, you may enter it by typing its name and clicking the Add Server button. When using the Add Server procedure, you must make sure that the computer names for both network interfaces on the servers respond correctly when you ping them (from all of the partner server(s)) using the ping –a IP ADDRESS syntax. If they do not, this must be corrected prior to continuing. Click Next.
- Select TCP for Device Type and Click Next.
- Provide all the required information and click Next for the following series of dialog boxes. For each field in the dialog box you can click Help for further information or refer to the table below for an explanation or recommendation
- After entering data in all the required fields, select Create. A message will display indicating the network communication path is successfully created. Click Next.
If you selected multiple Local IP Addresses or multiple Remote Servers and the Device Type was set to TCP, then the procedure will return you to the setup wizard the next Comm Path.
- Select Done in the last dialog box.
Repeat this process until you have defined all the communication paths you plan to use. SIOS strongly recommends that you define at least two communication paths for redundancy.
Verify the Communications Paths
- Verify that the communications paths are configured properly by viewing the Server Properties dialog box. From the LK GUI, select Edit, Server, Properties and then the Comm Paths tab.
- Note the State displayed is ALIVE. You can also check the server icon in the right, main pane of the GUI. If only one comm path has been created, the server icon shows a yellow warning icon on the server icon, indicating that one comm. path is ALIVE, but there is no redundant comm path. The server icon will display a green heartbeat checkmark when there are at least two comm paths configured and ALIVE.
Create the LifeKeeper Hierarchy
Create a Mirror and Begin Data Replication
In this section we will setup and configure the Data Replication resource, which be used to synchronize our MySQL’s data between cluster nodes. The data we will replicate resides in the /var/lib/mysql partition on our Primary cluster node
Please note:
- The source volume to be replicated must be mounted on the Primary server
- The target volume, which will received replicated data, must NOT be mounted on the Secondary server.
- The target volume’s size must equal to or larger than the size of its source volume.
- From the LifeKeeper GUI toolbar, click Create Resource Hierarchy.
The Create Resource Wizard dialog box will appear with a drop down list box displaying all recognized Recovery Kits installed within the cluster.
- Select Data Replication and click Next.
- Follow the Data Replication wizard, and enter the following values:
Switchback Type Server |
Intelligent LinuxPrimary (Primary Cluster Node, i.e. Mirror Source) |
Hierarchy Type Existing Mount Point |
Select: “Replicate Existing Filesystem” At this step you will select the mounted partition to replicate. In our example, select “/var/lib/mysql” |
Data Replication Resource Tag File System Resource Tag Bitmap File |
Leave as default Leave as default Leave as default (Note: if using high speed SSD storage you will want to create a small partition and use it for bitmap placement, i.e. /bitmaps) |
Enable Asynchronous Replication | Leave as default (Yes) |
- Click Next to begin creation of the Data Replication resource hierarchy. Status will be displayed in the GUI as follows:
- Click Next to begin the process to Extend the Data Replication Resource. Select all default settings. When it asks for the target disk, select a free partition on your Target server which is the same size (or greater) than the Source Volume we are replicating. This partition should NOT be mounted on the Target system.
- Continue through the wizard, and you will be prompted to select the network you would like replication to take place over. In general, it’s a best practice to separate your user/application and your replication traffic. In our example setup we will replicate over our backend network, 192.168.198.X
- Click Next and continue through the wizard. Once completed, your resource hierarchy will look as follows
Disable Automatic Failover
In this section we will review the procedure for disabling automatic failover to the standby server.
- From the LifeKeeper GUI, right click on one of the cluster nodes and select Properties.
- Select the Source server from the “Server:” drop down at the top of the window
- Once the Server Properties window loads, check all boxes at the bottom of the page. This will prevent any automatic failovers from happening.
- Click Apply
- Repeat steps 2-4, this time selecting the Target server from the “Server:” drop down
- Next, edit /etc/default/LifeKeeper on both nodes
a. Set CONFIRMSODEF=1 (change from 0 to 1)
Post your comment on this topic.