Node agent and application server

Node agent and application server

A node agent process will run on each node that is part of the cell. The node agent will look after application server processes running on that node.

Installation

To install a node agent and create a profile, do the following:

1. Start HACMP services on the first LPAR to mount the shared file system from the VIO Server.

2. Install WAS Network Deployment V6.0 on the first LPAR. The installation path should be on the shared file system.

3. If necessary, install Refresh Packs or Fix Packs. WebSphere Install Factory could (and should) be used to streamline this process.

4. Create a custom profile on the first LPAR. The profile directory should be a subdirectory of the WebSphere <install_root>, which is on the shared file system. If we select a different profile directory, then ensure that it is on the shared file system. When creating the profile, use the virtual host name as the Host name. Do not use the physical host name or physical IP address.

5. Federate the node to the Deployment Manager while creating the custom profile or later using the addNode command.

6. Create application servers on the node as desired.

Configuration

To configure HACMP to run the node agents and application servers, add the node agent stop and start commands as well as the appropriate start and stop commands for each application server in the failover unit to the HACMP stop and start scripts. Follow these steps:

1. Add the node agent and application server stop commands to the HACMP stop script (ha.stop) on both LPARs. Be aware that the application server names are case sensitive, and make sure that you specify the correct profile directory.

For example, if wer node had two clones called svr01 and svr02, then in the profile called svr, the script might contain the following:
/opt/WAS70/AppServer/profiles/svr/bin/stopServer.sh svr01
/opt/WAS70/AppServer/profiles/svr/bin/stopServer.sh svr02
/opt/WAS70/AppServer/profiles/svr/bin/stopNode.sh
2. Add the node agent and application server start commands to the HACMP start script (ha.start) on both LPARs. It is usually prudent to add a sleep command after the node agent start command, because the application servers depend on an active node agent to be able to start successfully. The length of the sleep will need to be adjusted for your environment, depending on the time the node agent takes to start.

Using the preceding example with a sleep length of 60 seconds, the start script might contain the following:
/opt/WAS70/AppServer/profiles/svr/bin/startNode.sh
sleep 60
/opt/WAS70/AppServer/profiles/svr/bin/startServer.sh svr01
/opt/WAS70/AppServer/profiles/svr/bin/startServer.sh svr02
3. Verify the HACMP services are active on the second LPAR.

Testing

During the failover process, the application servers hosted on the HACMP node cannot serve any client requests. At a minimum, the following two failover scenarios should be tested.
Conduct a graceful failover test using the HACMP takeover option. Watch the SystemOut.log on the node agent and each application server for a successful start. Check the node agent to make sure it can synchronize with the Deployment Manager before and after failover. There might be failed synchronization attempts during the failover.
Fall back to the first LPAR and conduct a shutdown failover test using reboot -q on this LPAR. Watch the SystemOut.log on the node agent and each application server for a successful start.