I have an update trigger that fails (it inserts an audit table record)
in some circumstances. This is causing the triggering transaction to
fail and roll back. Is there any way to prevent this? If the trigger
fails I still want the triggering transaction to continue.

Why does the execution plan have a nested loop join for a simple select
with an UDF in the where clause?
Here is the query:
select * from test_plan where vCol = my_udf('test')

Is there any way to invoke and perform a given subroutine within an SQL
procedure several different times, doing other work in between invocations
of the subroutine?
For instance, let's say that I want my SQL procedure, named A, to call

The following Update fails with SQL0407N on Colum "Name", although the
source-table t_Addresses_2005 is defined with NOT NULL on that column:
UPDATE
Addresses.t_Paddresses Old

I have a large table (about 50G) which stores data for over 7 years. I
decided to split this table up into a yearly basis and in order to
allow minimum changes to the applications which access this table, I
created a union all view over the 7 yearly tables.

Company is planning a Sybase to UDB migration that is projected to take
1 year.  During that long period of time any Code Freeze put into place
would impact our users (we normally release code changes once a month).
Any suggestions on how to manage a partial code freeze while under

I'm trying to locate the downloadable versions of the DB2 manuals for DB2 V8
for Windows/Linux/Unix. HTML or PDF versions would both be acceptable.
All I can find is the Information Center which, by its nature, can't be
downloaded.

Helpful folks,
Platform: DB2 UDB v7.2, Win2000, KEEPDARI=YES
I am trying to reduce memory usage of our SP's. We only have 80 or so
SP's but in the Task Manager/Processes display, there are over 200

For DB2 V8 on Windows/Linux/Unix, the idea of having the Information Center
online and accessible from IBM via the Internet is probably the best
approach for most users. It keeps the information current and centralized so
that everyone is seeing the "latest and greatest" versions of ...

I am having a problem with a nested SQL Procedure on DB2 V8.2.1 on Windows.
This simple-looking proc is giving me this error message when I try to build
it in the Development Center:
RHINO.CALLED01: 18: [IBM][CLI Driver][DB2/NT] SQL0104N An unexpected token

who could tell me if there is a SYSIBM table that would give me the
current db2 version accessed (on MVS) - i.e. Oracle offers a v$version
system table that gives that information.
Thanks

We have seen this a lot, but have just experienced the opposite to what
we have always seen previously, so this has prompted me to ask a high
level - why do we get this behaviour?
If we re-write queries in the following pattern (simplified)

I'm writing a program that interacts with a db2 database through jdbc.
I frequently need to run a quite query q1 that involves several full
outer join of a table with itself.  Once q1 is executed I need to
postprocess the result.  This postprocessing involves several queries on

For example - very large table with a lot indexes and updates work very
slow. I'd like found unused or rarely used indexes. How? Yes, for
static I found all sql in packages and, after analyse, drop/recreate
some indexes. But for dynamic I have no idea. Only drop all and monitor