This is something new that I have not yet come across in an Insight (Cognos) deployment. When I was attempting to either run a report or load a Framework Manager model package, I got the following errors:
This was an intermittent issue, so sometimes I could run a report successfully and seconds later, I could not. I observed the system processes and notice that at times, the “BiBusTKServerMain.exe” process would spawn, terminate, respawn and terminate again when trying to either run a report or load an FM package.
I reviewed the cogserver.log, and found these errors/failures:
“DPR-ERR-2074 Failed to receive a timely response from an external process with a PID XXXX. Ensure that the process with a process ID XXXX is started. If the cause of problem cannot be determined, increase the logging level for the target service in the IBM Cognos Administration tool and reproduce the conditions that caused the error.”
“Failure RSV-SRV-0040 The report server encountered an internal error. Check additional information associated with this error message. If cause of problem cannot be ascertained, increase the logging level in the IBM Cognos administration tool and reproduce the conditions that caused the error.”
“Failure CCL-SRV-0501 The BIBusTKServer process caught an unexpected exception.”
There were no DMP files in /cognos/bin or /cognos/bin64 directories. I turned on additional logging using “ipfBiBUSclientconfig.xml” however this did not really give me any additional clues to the resolution.
I ended up chatting with one of my colleagues and after much investigation, we found the issue. In my Cognos Configuration > Environment entries, I had all the URIs using the “localhost” reference. Its always a best practice to change that to a FQDN. I had not done this in this particular environment because it was a temporary test environment that I was using for a short period of time.
I replaced all “localhost” references in every URI property to be the machine hostname (which was mapped in the hosts file), restarted the Cognos BI Report server and the issue was resolved. This appeared to be an issue where the server could [intermittently] not resolve “localhost”.
I had a chat with another colleague and we discussed that this also may have had something to do with the IP version (IPv4 versus IPv6) of the particular environment I was using . There is a setting in Cognos Configuration that regulates this and by default is set to and IPv4 address. I did not go back to see if it would do the trick because I already had resolved the issue and needed to proceed, but just wanted to mention it as this likely could have helped.
Well, I hope that this helps you all save a few hours time of troubleshooting. It was definitely a good learning experience for myself as a seasoned Cognos veteran.