Bugs item #2779462, was opened at 2009-04-23 14:30
Message generated for change (Comment added) made by stmane
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2779462&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: MonetDB5 CVS Head
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Wouter Alink (vzzzbx)
Assigned to: Milena Ivanova (mivanova)
Summary: SQL: table leftovers
Initial Comment:
Using todays CVS head.
The following sql statements give wrong results (both SELECT yield 1 row with
number '10'). Note that without setting the optimizers, it works fine.
start transaction;
set
optimizer='inline,remap,evaluate,costModel,coercions,emptySet,aliases,mergetable,deadcode,constants,commonTerms,reorder,joinPath,deadcode,recycle,reduce,garbageCollector,dataflow,history,multiplex';
CREATE TABLE y (x int);
INSERT INTO y VALUES (10);
SELECT * FROM y;
rollback;
start transaction;
CREATE TABLE y (x int);
INSERT INTO y VALUES (13);
SELECT * FROM y;
rollback;
----------------------------------------------------------------------
>Comment By: Stefan Manegold (stmane)
Date: 2009-04-27 11:36
Message:
I can reproduce this one, both with the Feb2009-SP2 code base and with this
mornings CVS HEAD:
It works by default, i.e., with the following optimizers activated:
Feb2009-SP2:
sql>select optimizer;
0 tuples
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| single_value
|
+===============================================================================================================================================================================+
|
inline,remap,evaluate,costModel,coercions,emptySet,aliases,mergetable,constants,commonTerms,joinPath,accumulators,deadcode,reduce,garbageCollector,dataflow,history,multiplex
|
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
CVS HEAD:
sql>select optimizer;
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| single_value
|
+===================================================================================================================================================================================+
|
inline,remap,evaluate,costModel,coercions,emptySet,mitosis,aliases,mergetable,deadcode,constants,commonTerms,joinPath,deadcode,reduce,garbageCollector,dataflow,history,multiplex
|
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
but fails as reported when recycle is activated (between deadcode &
reduce), i.e.,
Feb2009-SP2:
sql>set
optimizer='inline,remap,evaluate,costModel,coercions,emptySet,aliases,mergetable,constants,commonTerms,joinPath,accumulators,deadcode,recycle,reduce,garbageCollector,dataflow,history,multiplex';
CVS HEAD:
sql>set
optimizer='inline,remap,evaluate,costModel,coercions,emptySet,mitosis,aliases,mergetable,deadcode,constants,commonTerms,joinPath,deadcode,recycle,reduce,garbageCollector,dataflow,history,multiplex';
This looks as if recycle is too eager caching/recycling update operations
(even) across transaction boundaries...
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2009-04-23 14:57
Message:
re-assigned to Mrs. "recycler" --- Milena ;-)
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2009-04-23 14:47
Message:
the 'recycle' optimizer caused the problems.
This optimizer has been switched off a while ago apparently. I don't know
whether it will be enabled again any time soon, but if so, i assume this
bug should stay open. If not, this bug could be postponed or closed.
Martin/Milena, could you comment?
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2779462&group_id=56967
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensign option that enables unlimited
royalty-free distribution of the report engine for externally facing
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
