Errors setting up multiserver environments

After creating and starting a cluster, the cluster does not start, and logs show that servers in the cluster are not found

This error can occur when the configuration is not synchronized from the deployment manager to a node. If auto synchronization is enabled, wait until the synchronization has had a chance to run. If you are using manual synchronization, explicitly request a sync to each node on the cluster.

To determine whether synchronization has taken place, look at the configuration on the node machines using the administrative console and verify that the new cluster members are defined on each node.

One or more nodes do not show up in the administrative console

This can occur when there is a basic connectivity problem between the deployment manager server and other servers in the topology. To determine whether this is the problem, look for the fileserverindex.xml in the deployment manager directory structure.

If the problem node does not appear in the list, review the steps for adding a node to the cluster.
If the problem node does appear in the list:
- From the deployment manager server, ping the server name as it appears in the list. If the ping command shows no communication, verify that the host name is correct in the list, and correct it if necessary, then restart the deployment manager.
- If the name that appears in the list is the short name, ping the fully qualified network name. If the corrected name works, update the list and restart the deployment manager.
- If the problem server uses DHCP, try replacing the logical host name with the IP address and restart the deployment manager. If this resolves the problem, be aware that change serverindex.xml each time the problem server address changes, potentially each time the problem machine is rebooted. To avoid this problem, consider assigning a static IP address to the server.

The addNode command fails

This error can occur when the deployment manager DNS configuration is set up improperly. The default installation on Linux uses the loopback address (127.0.0.1) as the default host address. To verify that this is the problem, query the host name of the suspect machine. If it returns localhost 127.0.0.1, or if file transfer traces at the node show the node trying to upload files to a URL that includes 127.0.0.1, the node has an incorrect DNS configuration.

To correct this problem, update the /etc/hosts file or the name service configuration file, /etc/nsswitch.conf, to query the Domain Name Server or Network Information Server (NIS) before searching hosts.

Application files are not present on all nodes

In the WAS Network Deployment environment, application binary files are transferred to the individual nodes where applications are supported as part of the node sync operation. During node sync, application files are only propagated if their deployment descriptors specify enableDistribution=true. This flag is specified as part of the application installation procedure in the administrative console, and is stored as a property in...

install_root/config/cells/cell/applications/appname/deployment.xml

To confirm that this is the cause, check to see whether the enableDistribution flag is set. If it is already set to true, ensure that the target node is configured to run auto file synchronization.

If both of these settings are correct and the problem persists, manually perform an explicit synchronization. If the application files still do not appear in the installation directory, use the EARExpander tool (located in install_root/bin) to expand the EAR file from the repository to the installation destination. On remote nodes, the repository should appear in...

config/cells/cell/applications/appname.ear/

After downloading the Network Deployment plug-in to my system, my server does not start

If you experience this situation, the most likely cause is that the transport paths in the plug-in must be modified to work in your environment.

In a clustered environment, a server with debug mode enabled does not start

This problem occurs when the following three conditions exist:

Multiple server processes are configured to run on the same node
More than one server has Debug Mode enabled
The debug arguments for more than one of the servers have been left at the default values, so that more than one server in the node is trying to use the same debug port (port number 7777).

The server will not start because multiple servers processes running on the same physical host machine with debug enabled cannot use the same debug port.

To correct this problem, for each server:

On the Administrative Console select...
Server | Application servers | servername | Java and Process Management | Process Definition | Java Virtual Machine
Update the Debug argument so that the address of the debug port (address=port number) is unique for each server process.

If none of these problem solution descriptions fixes your problem

Browse the JVM logs of the problem deployment manager and application servers:
1. Look up any error messages by selecting the Reference view of the information center navigation and expanding Messages in the navigation tree.
2. Use the Log Analyzer to browse and analyze the service log (activity.log) of the deployment manager and any nodes encountering problems. View the activity.log files in both...
  NetworkDeployment_install_root/logs
  
  ...and...
  ApplicationServer_install_root/logs
3. If Java exceptions appear in the log files, try to determine the actual subcomponent directly involved in the problem by examining the trace stack and looking for a WAS-related class near the top of the stack (names beginning with com.ibm.websphere or com.ibm.ws) that threw the exception. Review the steps for troubleshooting the appropriate subcomponent in the Troubleshooting by component: what is not working? topic.
  For example, if the exception appears to have been thrown by a class in the com.ibm.websphere.naming package, review the Naming Services Component troubleshooting tips topic.
Ensure that all the machines in your configuration have TCP/IP connectivity to each other by running the ping command:
1. From each physical server to the Deployment Manager
2. From the Deployment Manager to each physical server
Although the problem is happening in a clustered environment, the actual cause might be only indirectly related, or unrelated, to clustering. Investigate all relevant possibilities:
1. If an enterprise bean on one or more servers is not serving requests, review the Cannot access an enterprise bean from a servlet, JSP, stand-alone program, or other client and Cannot access an object hosted by WAS from a servlet, JSP file, or other client topics.
2. If problems seem to appear after enabling security, review the Errors or access problems after enabling security topic.
3. If an application server stops responding to requests, or spontaneously dies (its process closes), review the Web module or application server dies or hangs topic.
4. If SOAP requests are not being served by some or all servers, review the Errors returned to client trying to send a SOAP request topic.
5. If you have problems installing or deploying an application on servers on one or more nodes, review the Errors or problems deploying, installing, or promoting applications topic.
If your topology consists of a Windows-based Deployment Manager with UNIX-based servers, browse any recently-updated .xml and .policy files on the UNIX-based platform using vi to ensure that Control-M characters are not present in the files. Edit these files using vi on the UNIX-based platform, to avoid inserting these characters.
Check the steps for troubleshooting the Workload Management component..
Check to see if the problem is identified and documented by looking at available online support (hints and tips, technotes, and fixes).