A.2 Installation Procedure

1. Create a separate partition and specify the partition path as the iSCSI shared storage partition.

2. Install the iSCSI target packages.

1. Create a block device on the target server.

2. Type the yast2 disk command in terminal.

3. Create a new Linux partition, and select Do not format.

4. Select Do not mount the partition.

5. Specify the partition size.

6. Type the yast2 iscsi-server command in terminal.

7. Click the Service tab, then select When Booting in Service Start.

8. In the Targets tab, click Add to enter the partition path (as created during the SLES installation).

9. Click Finish.

10. Run the cat /proc/net/iet/volume command in the terminal to verify if the iSCSI target is installed

1. Install the iSCSI initiator packages.

2. Run the yast2 iscsi-client in terminal.

3. Click the Service tab and select When Booting in Service Start.

4. Click the Connected Targets tab, and click Add to enter the IP address of the iSCSI target server.

5. Select No Authentication.

6. Click Next, then click Connect.

7. Click Toggle Start-up to change the start-up option from manual to automatic, then click Next.

8. Click Next, then click OK.

9. To check the status of the connected initiator on the target server, run the cat /proc/net/iet/session command on the target server. The list of initiators that are connected to iSCSI server are displayed.

1. Run the yast2 disk command in terminal.

2. In the Expert Partitioner dialog box, select the shared volume. In our example, select sdb from the Expert Partitioner dialog box.

3. Click Add, select Primary partition option, and click Next.

4. Select Custom size, and click Next. In our example, the custom size is 10 MB.

5. Under Formatting options, select Do not format partition. In our example, the File system ID is 0x83 Linux.

6. Under Mounting options, select Do not mount partition, then click Finish.

7. Click Add, then select Primary partition.

8. Click Next, then select Maximum Size, and click Next.

9. In Formatting options, select Do not format partition. In our example, specify the File system ID as 0x83 Linux.

10. In Mounting options, select Do not mount partition, then click Finish.

1. Go to the NetIQ Downloads website.

2. In the Product or Technology menu, select SUSE Linux Enterprise HAExtension, then click Search.

   NOTE:Select and install the appropriate HA extension ISO file based on your system architecture.

3. Download the ISO file on each server.

4. Open YaST Control Center dialog box, click Add-on products > Add.

5. Click Browse and select the local ISO image, then click Next.

6. In the Software selection and system tasks dialog box, select High Availability. Repeat this step on the other server.

Configure the unicast IP addresses for Heartbeat:

1. Configure the other interface on both the nodes with the static IP addresses, which will be used for node communication (Heartbeat). In our example, the IP addresses are 10.10.10.13 and 10.10.10.14 on Node1 and Node2, respectively.

2. Ping the two servers using their host names to test the connectivity between the two servers.

   IMPORTANT:If the machines are unable to ping each other, edit the local /etc/hosts file and add the host names of the other nodes and their IP addresses. In our example, the /etc/hosts file contains the following:

   • 10.10.10.13 sles11sp2-idm1

   • 10.10.10.14 sles11sp2-idm2

3. On Node1, run the yast2 cluster command in the terminal.

4. In the Cluster - Communication Channels dialog box, specify the following details:

   1. Set the Transport protocol to UDPU.

   2. Specify the Bind Network Address, which is the network address of the unicast IP addresses. In our example, the bind network address is 10.10.10.0.

   3. Specify the Multicast port. In our example, the Multicast port is 5405.

   4. Click Add to enter the IP address for each node at the member address. In our example, the IP addresses are 10.10.10.13 and 10.10.10.14 on Node1 and Node2, respectively.

   5. Select Auto generate Note ID, then click Next.

5. In the Cluster -Security dialog box, select the Enable Security Auth, set Threads to 1, then click Generate Auth Key File.

   This creates an authentication key to allow other nodes to join your cluster. The key is stored in the /etc/corosync/authkey location. Copy this file to the other node.

6. In the Cluster - Service dialog box, select On--Start openais at booting, then click Start openais Now.

7. Select Start Management as well to allow the cluster to be managed by crm_gui. For more information, see Section A.2.2, Configuring the iSCSI initiator on all Nodes.

8. In the Sync Host panel, perform the following actions:

   1. Click Add to add hostnames of the cluster nodes.

   2. Click Generate Pre-Shared-Keys to synchronize the configuration file between nodes, then copy it to the other node. The key file is stored in /etc/csync2/key_hagroup.

   3. In the Sync File pane, click Add Suggested Files to automatically generate a list of common files to synchronize between nodes.

   4. Click Turn csync2 ON, then click Next.

   5. Click Next, then click Finish.

9. Run the passwd hacluster command to set the hacluster user password on all nodes.

   NOTE:Set the same password for hacluster user on nodes.

10. Run the following commands to copy the configuration files and authentication keys to the other node:

    • # scp /etc/csync2/csync2.cfg node2:/etc/csync2/

    • # scp /etc/csync2/key_hagroup node2:/etc/csync2/

    • # scp /etc/corosync/authkey node2:/etc/corosync/

    • # scp /etc/corosync/corosync.conf node2:/etc/corosync/

11. Reboot all the nodes after the configuration files are copied to Node2.

12. Run the csync2 -xv command.

13. Create the mkdir -p /share directory to mount the shared storage.

14. On Node2, do the following:

    1. Run the yast2 cluster command in the terminal.

       NOTE:The wizard window does not appear, because the configuration file is already copied over.

    2. In the Service tab, select Check On -- Start openais at booting, then click Start openais Now.

    3. In the Configure Csync2 tab, click Turn csync2 ON, then click Finish.

    4. Create the mkdir -p /share directory to mount the shared storage.

       The cluster should be up and running.

15. Run the crm_mon command in the terminal to verify the status. Following is a sample output:

    ============
    Last updated: Fri Aug 5 16:38:36 2011
    Stack: openais
    Current DC: node1 - partition with quorum
    Version: 1.1.2-2e096a41a5f9e184a1c1537c82c6da1093698eb5
    2 Nodes configured, 2 expected votes
    0 Resources configured.
    ============
    Online: [node1 node2]

1. Run the crm_gui command in the terminal.

2. Click Connection menu > Login. Log in using the IP address of either of the nodes.

3. Click the CRM Config tab, then change Default Resource Stickiness to a positive value.

   This is to ensure that the resources in the cluster remain in the current location. In our example, the value is 1.

4. Change No Quorum Policy to ignore.

   This ensures that the cluster services are up and running even if one of the nodes is down.

5. Click Apply.

1. To install eDirectory on cluster nodes:

   Install eDirectory a supported version of eDirectory. For step-by-step instructions to configure eDirectory on HA clusters, see “Deploying eDirectory on High Availability Clusters” in the eDirectory 8.8 Installation Guide.

   IMPORTANT:Ensure that the virtual IP is configured on the Node1 before you install eDirectory on Node1.

2. Install Identity Manager on Node 1 using the Metadirectory Server option.

3. Install Identity Manager engine on Node 2 Server using the DCLUSTER_INSTALL option.

   Run the ./install.bin -DCLUSTER_INSTALL="true" command in the terminal.

   The installer installs the Identity Manager files are installed without any interaction with eDirectory.