Having trouble resuming, completing, or aborting the workflow

Products

• CA Process Automation Base
• Process Automation Manager

Issue/Introduction

When trying to resume, complete, or abort the workflow, it does not always work.  It all depends on which PAM server was processing the workflow when it stopped responding. You have to log in to each PAM server and work the workflow from the server that shows it as not responding. Before the upgrade to PAM 4.4, we were able to work on workflows using the main URL without this behavior. 

Environment

Process Automation - v4.4 and up
This applies when the Orchestrator servers use 2 NIC cards

Resolution

If one of the NIC cards on the Orchestrator server is disabled, the problem disappears.  However, if multiple NIC cards are required for the Orchestrator server for other reasons, this could not be implemented as the solution.  

These steps resolved the problem:

Actions for the network team:

1. The environment should be updated so that whenever the host with FQDN (fully qualified domain name) is pinged, it should return the results with Primary IP instead of backup IP
   
   
2. Whenever running the ipconfig command, only the primary IP should show as Primary instead of both IPs as Primary.

Actions for the ITPAM Administrator:

Once the above-mentioned issue is resolved, configure PAM ("OasisConfig.properties" and "c2osvcw.conf") to use FQDN (fully qualified domain name) instead of IP address

• in OasisConfig.properties
     oasis.jxta.host=<fqdn>
     oasis.local.hostname=<fqdn>
     pam.mq.tcp.url=tcp://<fqdn>:61618
  
  
• in c2osvcw.conf
     wrapper.app.parameter.6=0.0.0.0  <=== change to the IP address of primary NIC card