Node agent and DMGR discovery problem

Node agent and DMGR discovery problem is one of the common task I face while working with customer where the Network team are very rigid in opening the ports between the systems. When you have a clustered architecture where we federate the node to deployment manager, the connection between system takes place through ports and the common problem will be where the Nodeagent and Deployment manager couldn’t communicate as a result of port block issue.

As mentioned in my previous blog entry, at the maximum there could be a problem with ports.

There is a IBM Technote that mentions on how you start troubleshooting the issue,

Troubleshooting: http://www-01.ibm.com/support/docview.wss?uid=swg21201625

Collecting Data: http://www-01.ibm.com/support/docview.wss?rs=180&uid=swg21196220

Analyzing: http://www-01.ibm.com/support/docview.wss?rs=180&uid=swg21201623

As part of the technote, Steps to help resolve Node agent and Deployment Manager discovery problems

1. If you have multiple network interface cards (NIC) on the nodeagent or Deployment Manager node, and you do not want to use the primary NIC, you must apply Interim Fix PQ71669 or use Fix Pack 1 or higher for version V5.0.

2. Are both processes running? (Message ADMD0004E can be seen in an FFDC log if either process is not running.)

3. Is there a firewall between the processes?

4. Is the discovery port listening?

5. Has discovery occurred successfully? Check the SystemOut.log file for the message ADMD0023I indicating successful discovery.