ibm_db Python driver/wrapper update 0.2.7 now available

[abhigyan...@in.ibm.com]

It is my pleasure to inform you that ibm_db Python driver/Wrapper
update 0.2.7 is release and now available on google code and pypi.
http://code.google.com/p/ibm-db/
http://pypi.python.org/pypi/ibm_db

This release fixes bug #151164 where we found that ibm_db_dbi.execute
method was raising System error when connection handle is invalid.
Please continue flowing in your valuable suggestions and comments.
Together we can refine and improve this project.

[Jaimy Azle]

On Friday, March 7, 2008, 12:33:39 AM, abhigyan...@in.ibm.com wrote:

> This release fixes bug #151164 where we found that
> ibm_db_dbi.execute method was raising System error when connection
> handle is invalid.

Hi,

Is this fix include bug that causing python to crash? I tried this
driver with SQLAlchemy to access DB2-Express C, but any execution
after long idle seems causing runtime error in Python interpreter
itself. I can not log what kind of execution that causing error, but
it seems that the DB2 itself will disconnect any client connections
after n-seconds idle. Anyhow, is there any suggestion how long
(n-seconds) should I use in SQLAlchemy connection pool configuration to
recycle DB2 database connection? or in what configuration should I
change in DB2 that define something like max-iddle-time? I am new to
DB2, the RDBMS itself was installed on Windows using default
configuration right now.

--
Salam,

-Jaimy Azle

[abhigyan...@in.ibm.com]

Hi Jaimy,
No this fix doesn't include the bug you mentioned. This
looks more like configuration problem.
Please check your CONNECTIONIDLETIME value which may be the problem.
Also, check QueryTimeoutInterval, which if set to zero, will result in
SQLSTATE 01S02 if application tries to set SQL_ATTR_QUERY_TIMEOUT.
Also, have a look at following link -
http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/index.jsp?topic=/com.ibm.db2.luw.admin.gui.doc/doc/c0021917.html

If this doesn't solve the problem, please provide more details like -
the code you are trying to execute, current SQLAlchemy connection pool
configuration and time duration for this long idle. Usually, the
default values are fine. Also, which version of DB2 are you using?

Thanks,
Abhigyan

On Mar 8, 8:14 am, Jaimy Azle <jaimy.a...@gmail.com> wrote:

[Jaimy Azle]

On Sunday, March 9, 2008, 5:39:14 AM, abhigyan...@in.ibm.com wrote:

> Please check your CONNECTIONIDLETIME value which may be the problem.
> Also, check QueryTimeoutInterval, which if set to zero, will result in
> SQLSTATE 01S02 if application tries to set SQL_ATTR_QUERY_TIMEOUT.

Thanks, I will check on it.

> If this doesn't solve the problem, please provide more details like -
> the code you are trying to execute, current SQLAlchemy connection pool
> configuration and time duration for this long idle. Usually, the
> default values are fine. Also, which version of DB2 are you using?

DB2 version being used is 9.5, and python were implemented as an
embedded business process executor on a middleware written with
Delphi. This system were previously runs well with SQL Server, problem
only occurs when i test it using DB2 as backend. But after applying
the most up-to-date db2 wrapper, it seems the previous problem I
describe does not occur anymore. However, I will try to activate
deep-logging meanwhile to make sure.

--
Salam,

-Jaimy Azle

"Great new ideas usually come from very small teams...
don't give up so easily."
-- John Kaster