Hi,

I am new to DB2, need help.

I have an application using Static Embedded SQL with C. I need to
change the db schema to satisfy new business requirements by adding an
extra column to an existing table. This new column will be used in new
executables only. However, the changed table IS used in existing
ESQLs.

My question is: Do I need to rebuild (recompile/rebind, etc) all the
existing ESQL C programs that use this table? Is there anyway to
avoid this?

regards,

k.

we are using MS 2000, with version 7.2 DB2.

I also noticed the following statement from DB2 online doc:

"Adding a column to a table will result in invalidation of all
packages with insert usage on the altered table."

If I have insert statements in the package like:

EXEC SQL
INSERT INTO anExistingTableBeenAltered
(anExistingColumn)
VALUES
(:value)

which does not have anything to do with the added column. Will this
package become invalid because of the insert?

appreciate any input.

k.


"Rhino" <rhino1 (AT) NOSPAM (DOT) sympatico.ca> wrote

In article <7632f111.0308271218.2a325c67 (AT) posting (DOT) google.com>,
wctdy2003 (AT) yahoo (DOT) com says...
The package will become invalid but you can use the command
'db2 rebind package <packagename>' to rebind it. If you do nothing
DB2 will automatically try to rebind the package when it is first
executed but I've seen severe locking problems because of that.

Gert van der Kooij <gk-ibm-db2 (AT) xs4all (DOT) nl> wrote

Thanks for the help.

What is the real reason behind this (package becoming invalid)? The
insert statement does not reference the new column. What really
happens behind the scene during bind/rebind for this SQL? Isn't just
the table statistics info needed?