Bugs item #2938065, was opened at 2010-01-23 17:53
Message generated for change (Settings changed) made by nielsnes
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2938065&group_id=56967

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: SQL/Core
Group: SQL "candidate"
>Status: Closed
Resolution: Fixed
Priority: 5
Private: No
Submitted By: zeuner (zeuner)
Assigned to: Niels Nes (nielsnes)
Summary: corrupted values after multiple INSERT/DELETEs

Initial Comment:
The attached SQL statements lead to wrong output on MonetDB-SQL Feb2010. The 
final SELECT statement should result in the string 'modified' in the column 
"value", but it doesn't. Most likely it is a memory-related bug as most of the 
time, I get an empty string there, but during isolation of a reasonably small 
test case, I also got arbitrary binary data, and on occasion a string length 
overflow assertion breaking inside str.mx.

The test system runs CentOS 5.4 on x86_64, having MonetDB compiled with 64-bit 
OIDs. The compiler was "gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-46)".

----------------------------------------------------------------------

Comment By: Niels Nes (nielsnes)
Date: 2010-01-24 15:41

Message:
fixed in the candidate feb2010 branch. Sometimes the bats with strings were
incorrectly initialized.
Also added the test to src/test/BugTracker-2010

----------------------------------------------------------------------

Comment By: zeuner (zeuner)
Date: 2010-01-24 09:16

Message:
I first became aware of the issue on a 2010-01-22 checkout. After tracking
down how to trigger it, I tried again using a 2010-01-23 checkout, which
made no difference.

----------------------------------------------------------------------

Comment By: Martin Kersten (mlkersten)
Date: 2010-01-24 07:29

Message:
to be sure, you are using a check out of the candidate code for the Feb2010
release of what date?
There are major changes in that area for which any indication of an error
is relevant information.

----------------------------------------------------------------------

Comment By: zeuner (zeuner)
Date: 2010-01-24 04:33

Message:
I encountered a similar data corruption on other queries. Interestingly,
the corrupted column was also initialized from a max() aggregate. I also
had max() aggregates over integer columns in the same query, but these were
processed correctly, so it seems to apply only to string columns.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2938065&group_id=56967

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs

Reply via email to