Bug #52196 SQL_COLUMN_IGNORE not checked in SQLBulkOperations AGAIN

• Submitted: 19 Mar 2010 0:42
• Modified: 28 May 2015 11:26
• Reporter: Ivan Peev
• Status: Can't repeat
• Category: Connector / ODBC
• Severity: S3 (Non-critical)
• Version: 5.1.6
• OS: Windows
• CPU Architecture: Any
• Tags: Bulk-inserts, regression

[19 Mar 2010 0:42] Ivan Peev

Description:
The bug reported here:

http://bugs.mysql.com/bug.php?id=3499

has been reintroduced.

When trying to add data to a table using SQLBulkOperations and SQL_COLUMN_IGNORE is used in SQLBindCol, the application fails with memory corruption/allocation errors. I suspect the issue has been reintroduced with this fix:

http://bugs.mysql.com/file.php?id=6011

How to repeat:
1. Create a table with columns having default values. Use date, timestamp columns.
2. Use the routine described here: http://msdn.microsoft.com/en-us/library/ms712471%28VS.85%29.aspx

to setup ODBC bulk-load process.

3. Execute the process with 100,000 records.

[19 Mar 2010 13:14] Tonci Grgin

Hi Ivan and thanks for your report.

Is it possible for you to attach small but complete test case demonstrating this behavior? I really do not have time to invent test cases just to find out we have different DNS configuration or something trivial like that.

[19 Apr 2010 23:00] Bugs System

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".

[19 Apr 2010 23:02] Ivan Peev

I don't have time to prepare test case. But the bug is 100% there.

[28 May 2015 11:26] Chiranjeevi Battula

Hello Ivan Peev,

Thank you for your feedback.
I could not repeat this issue at my end with the provided information. 
Could you please try the newer versions and feel free to open the bug if it is not working with the complete reproduce test cases.

Thanks,
Chiranjeevi.