I am using JCA adapter mechanism in Java to talk to CICS program running on a mainframe. We use Java-J2EE technology...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
with application deployed on IBM Websphere. I have been getting following errors consistently for large transaction in java.
CTG9631E: Error occurred during interaction with CICS. Error Code=ECI_ERR_SYSTEM_ERROR javax.resource.spi.ResourceAdapterInternalException: CTG9631E: Error occurred during interaction with CICS. Error Code=ECI_ERR_SYSTEM_ERROR
This happens when Websphere application tries to get the connection from CICS Transaction server. I get this error quite inconsistently depending on length of transaction in application and/or number of calls to CICS in one transaction. I will very appreciate if you can share your experties or comments on this error and suggest any specific are to look into.
It's difficult to understand the environment being used here: and these sorts of questions are VERY release and environment dependant.
However, I'm assuming that you have MVS WebSphere using a CTG on MVS talking to CICS via EXCI - but I don't know and advice is VERY release dependant.
As you do not mention any circumstances in CICS - this is an indication of problems in the CTG or Java or Websphere.
If you are getting this CTG initiated error, then you should turn on the CTG diagnostics to see what message/trace accompanies the error. Unfortunately, this type of error is somewhat generic and could arise with the CTG classes executing in Webspere, or the functions in the CTG or something happening within CICS.
As the error is intermittant and only seems to happen in a big transaction (big in what sense: long running, largeCPU, vast Recoverable Resources being updated/held??), this could be an indication of some sort of Storage problem (in the CTG maybe) . There seem to be a lot of these errors clustered around CodePage Conversion, so check that something deep in the large transaction does not have anything odd in this area.
Therefore, the only thing I can suggest is bumping up Region Sizes and running with Trace to trap the error. In the end, I suspect, you will be raising a PMR so you will need all Diagnostic info collected from the three places (SebSphere, CTG, CICS).
Dig Deeper on IBM system z and mainframe systems
Related Q&A from Robert Crawford
For better mainframe capacity planning, how do I convert CPU hours to MIPS? And is there a way to calculate the relationship between MIPS and MSUs?continue reading
I have two years of experience in mainframe technology, currently working as a mainframe developer. I want to change to Java technology.continue reading
I want to replicate DB2 from the mainframe to an AIX box since it's cheaper and the copy can be used for testing. Is this possible?continue reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.