Bugs item #2186337, was opened at 2008-10-22 10:16
Message generated for change (Comment added) made by mlkersten
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2186337&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 CVS Head
Status: Closed
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Stefan Manegold (stmane)
Assigned to: Martin Kersten (mlkersten)
Summary: mserver5 segfaults on mclient -lsql connection
Initial Comment:
After yesterdays late(st) changes (see below), mserver5 segfaults as soon as
mclient -lsql tried to connect:
0x00002aaabef92a97 in create_col (tr=0x104ace8, c=0x104d3e8) at
/ufs/manegold/_/scratch0/Monet/Testing/Current/source/sql/src/storage/bat/bat_storage.mx:531
531 int type = c->type.type->localtype;
(gdb) print c
$1 = (sql_column *) 0x104d3e8
(gdb) print *c
$2 = {base = {wtime = 0, rtime = 0, flag = 1, id = 1061, name = 0x104d458
"id"}, type = {type = 0x0, digits = 0, scale = 0, comp_type = 0x0}, colnr = 0,
null = 1 '\001', def = 0x0, unique = 0 '\0', t = 0x104af98, data = 0x0}
(gdb) print c->type
$3 = {type = 0x0, digits = 0, scale = 0, comp_type = 0x0}
(see attached GDB log for details)
Before these changes, it did work fine:
===================================================================
2008/10/21 - mlkersten: MonetDB5/src/modules/mal/cluster.mx,1.4
MonetDB5/src/modules/mal/Tests/cluster00.mal,1.4
Make a guess about the number of (memory) buckets to use for clustering.
===================================================================
2008/10/21 - sjoerd: MonetDB/src/common/stream.mx,1.173
Added function to clear stream error so that we can continue after an
error.
This is needed for mclient.
===================================================================
2008/10/21 - sjoerd: clients/src/mapiclient/MapiClient.mx,1.124
Report write error and clear the stream so that we can continue.
Unfortunately the reason for the write error is lost, so the best we
can do now is to report that there was an error.
Convert command line query according to specified or implied encoding.
===================================================================
2008/10/22 - mlkersten: sql/src/common/sql_keyword.mx,1.13
sql/src/common/sql_string.mx,1.19 sql/src/server/rel_select.mx,1.89
sql/src/server/sql_select.mx,1.249 sql/src/server/sql_semantic.mx,1.189
Added toUpper and avoided name clash. Added all upper case versions
to the symbol table. This accomodates most use cases, if we fail
to find then we re-cast. Savings around 1% on micro-test of Fabian.
===================================================================
----------------------------------------------------------------------
>Comment By: Martin Kersten (mlkersten)
Date: 2008-10-22 13:19
Message:
The changes made to sql_semantics remain, because
mkUpper had confusing semantics compared to mkLower
defined in common/sql_strings.
This has been rectified by keeping mkUpperCopy
and also mkUpper.
----------------------------------------------------------------------
Comment By: Martin Kersten (mlkersten)
Date: 2008-10-22 12:46
Message:
Reverted part of the changes. Tests run as expected again.
Sorry for the inconvenience.
----------------------------------------------------------------------
Comment By: Martin Kersten (mlkersten)
Date: 2008-10-22 12:24
Message:
The find_keyword casted everything to lower before searching.
This was done in place which creates a side-effect of a find.
The keyword_exist call is recasted.
Problem was not detected after a make;make install
over an existing database test run. Testing in progress.
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2008-10-22 12:01
Message:
I see a discrepancy in the code:
in sql/src/common/sql_types.mx the type system is initialized using
capitalized names;
in sql/src/storage/store.mx columns are created using lower case type
names.
The function that is called to do the comparison (sql_find_subtype) does
not do case mapping so doesn't find the type when called from
bootstrap_create_column, called from store_init.
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-10-22 10:18
Message:
overruling automatic assignment, since Niels is/was not actively involved
(Martin and Sjoerd are/were);
Niels "may" of course still have a look at it ... ;-)
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2186337&group_id=56967
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
