We have a clustered environment. we have 2 servers running on JVM's and 4 nodes. the BASE URL in the server.confg file was set as the URL of the Primary Server but then we changed it to the dns name for i.e ppm..com on both the servers and on all four nodes. The reason to change was that we were getting an error [Connection not established. Please close this window, refresh your dashboard, and try again. (KNTA-10976)] after changing the BASE URL to the DNS name we are still getting this error.
The workbench URL is set with the RMI_URL directive in server.conf. A unique value must be set for each instance in the @node section for that instance. The BASE URL value is not used by the workbench.
The BASE URL value should be used by users when connecting to PPM. With a cluster, it would be the URL of the load balancer. It will be used to build all links for log files, attachments, etc, and the user will have to log in again if they happen to be routed to the wrong instance by the load leveler.
You would have to post your server.conf file for us to examine. I don't know if you would want to do that.
I am a bit confused by your problem statement. You indicated you made the change to the load balancer's DNS name because connections failed when you had BASE URL set to the primary server, and that the problem remained after the change.Is this correct?
yes it is correct that we changed the base url from Primary Server URL to DNS name it was told by a support engineer. We first tried it in our QA environment and then later on in PROD. In QA we never had this problem even if base url was not the DNS name.
Base URL affects only the Dashboard and any web page links like log files, etc. The Workbench address is defined uniquely for each instance with the RMI URL directive. It is defined in the @node section for the instance and should look something like:
Did you able to resolve the problem? We have the same problem. When we are adding DNS name instead of url we are facing the problem in opening workbench. Please let us know if you found a solution then we'll try the same. It seems there is some relation between base url and rmi url. Workbench is not directly related to rmi url.
You are absolutely right. It is showing in the workbench header when I was able to open the workbench with base url as :.(i.e. com.kintana.core.server.BASE_URL=http://126.96.36.199:2222/itg/) But when I changed the base_url to dns name (i.e com.kintana.core.server.BASE_URL=https://uappm.ual.com/itg/) I can't even open the workbench. So it looks like there is some relation of workbench with base_url in PPM 8.0. I am not getting the reason behind it. I didn't change anything in the RMI_url.