I am just confused because of the following scenario.

* I have turned on the new 8.2 registry variables skipdeleted,
skipinserted, evaluncommitted to ON
* I gave a select * from <>, it was waiting for a lock, it needed an
intent share lock on the particular table..
the lock that it was waiting on was EXclusive row lock that another
process(process-2) was having, process-2 was updating the table 500
rows at a time before
committing
* I was hoping this will not result because of the new registry
variables!!!!

I tried db2pd to get the isolation levels of the applications, but
with no joy...
How can I get the isolation level?

Hi, Arun.

It's possible that one of the updated rows from process-2 satisfies
predicates of the query. I believe that DB2_EVALUNCOMMITTED can only
avoid lock wait where the modified row does not satisfy the
predicates.

There are also some limitations to when the feature can play in, and
even issues like lock escalation to consider.

Here's a ref: http://www.ibm.com/developerworks/db...m-0509schuetz/

Regards,
- Steve P.
--
Steve Pearson, DB2 for Linux, UNIX, and Windows, IBM Software Group
"Portland" Development Team, IBM Beaverton Lab, Beaverton, OR, USA