Tuesday, 15 April 2014

Interesting exception seen whilst starting IBM Business Process Manager 8.5.0.1

So I saw this exception: -

[15/04/14 19:25:40:071 BST] 00000001 WsServerImpl  E   WSVR0009E: Error occurred during startup
com.ibm.ws.exception.RuntimeError: Exception while verifying the datasource version: failed to get stardard DB version by jndi jdbc/PerformanceDB

( Yes, that's correct - it did say stardard DB version !! )

This occurred whilst I was starting the Support cluster of an IBM BPM 8.5.0.1 Advanced environment.

From a bit of digging about, it looks like I failed to completely create the required databases schemata, tables etc.

It looks like I had at least one missing table - LSW_SYSTEM_SCHEMA.

Once I ran the appropriate scripts to create the Performance Data Warehouse (PDW) DB schemas: -

db2 -tvf createSchema_Advanced.sql 

I was able to validate the existence and contents of the missing table: -

DB2 "SELECT PROPVALUE FROM LSW_SYSTEM_SCHEMA WHERE PROPNAME = 'DatabaseSchemaVersion' "

which now returns: -

PROPVALUE                                                                                                                                                                                              
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
8.5.0                                                                                                                                                                                                  

  1 record(s) selected.


Now the cluster starts nicely, which is nice.

Memo to self - remember to create all the database objects, not just the containers themselves :-)

2 comments:

Arturo Arenas said...

I came here because I got this issue when installing BPM 856. My mistake was a commit... yes, a commit.
I have my Oracle database, and After BPMConfig.sh command execution, I took the dbscripts and run them. No errors where shown during scripts execution. But still got this error.
I realized my mistake when closing sqlDeveloper, when a pop up ask me if I wanted to commit my changes... Hope this help anyone. Best Regards

Dave Hay said...

Arturo, yes, I know what you mean - this is also the case when using SQL Plus in Oracle ( it's caught me out before ).

Thanks for the feedback, Dave

Note to self - use kubectl to query images in a pod or deployment

In both cases, we use JSON ... For a deployment, we can do this: - kubectl get deployment foobar --namespace snafu --output jsonpath="{...