I'm using DB2 8.1.
Suppose table foo has columns name and lname:
create table foo
(name as varchar(200),

I'm using DB2 UDB 7.2 on a Linux server and I've installed the DB2
client on my Windows 2000 machine. I've defined on my client two
instances, one for test databases and one for production databases
and I've cataloged the same database (of course with the same name)

I've worked with DB2 for a long time, but I'm new to the AS400
environment.
I'm having a problem with an SQL statement that runs perfectly when I
use STRSQL, but when I batch it and call RUNSQLSTM, the exact same

Informix recommend that symbolic links be used for chunks. These links
are then used in the dbspace definition.
Does this advice hold for DB2? The equivalent would be to create
symbolic links for containers and use the link name when defining

I am running this query at home PC with DB2 EXPRESS 9.5C, but at work on
mainframe i don't
have option VALUES.
What is the best way to UNPIVOT without using VALUES?

My DBA says that a column defined, for instance, as DECIMAL(11,2) and
containing a value of 1.00 takes up no more space on the database disk than
a column defined as DECIMAL(7,2) and containing a value of 1.00.
I am honestly having a hard time believing this, but I don't know where ...

When collecting snapshot, there is seemingly discrepency between query
running time and some metrics collected.
For example, on a DPF with 8 nodes, I have   8863201 seconds for
"Total buffer pool read time (milliseconds)" from running  "db2 get

This question is related to iSeries V5R4 and db2.
I want to implement an AFTER DELETE trigger to save the deleted rows
to an archive table, I initially defined it as a FOR EACH STATEMENT
trigger that would insert all the deleted rows in one operation like

The following query has a high cost (over 4 million), uses a nested
loop join and runs much slower than I would like. Does anyone know if
is possible to rewrite this query more efficiently (probably without
the correlated subquery)? If so, could you please show me how.

Here is the input table:
   MPID     MGRID  EMPNAME    SALARY  
----------- ----------- -------------------------       ---------
        14           9 Didi                        1500.00

Table definition:
CREATE TABLE "SCHEMA1  "."X2"  (
                 "C1" CHAR(20) NOT NULL ,
                 "C2" CHAR(10) NOT NULL ,

I have a strange problem with federated access between two DB2 databases,
both on AIX 5, both instances V9.1 (FP02).
For the sake of this post, the two DB2 databases are named:
1. "live"

DB2 v8 FP15 luw.
Serge´s SQL on Fire has the following code for granular commits on
delete statements:
loop: LOOP

is it possible to define a CHECK CONSTRAINT on a table, whereas the
CHECK-OPTION will be a SELECT.
With this CHECK CONSTRAINT i try to ensure, that there will allways a
state only set once within several rows. If possible, does someone can

we have an application which uses online backups, so archive logging
is a requirement. We have a small/moderate size database (a few
hundred megabytes isn't much by today's standards) with relatively low
insert/update activity (a few thousand per day).