BUG #1790: coredump in postgres

From: "Dmitry Karasik" <dmitry(at)karasik(dot)eu(dot)org>
To: pgsql-bugs(at)postgresql(dot)org
Subject: BUG #1790: coredump in postgres
Date: 2005-07-27 10:31:42
Message-ID: 20050727103142.38A7DF0B0B@svr2.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs


The following bug has been logged online:

Bug reference: 1790
Logged by: Dmitry Karasik
Email address: dmitry(at)karasik(dot)eu(dot)org
PostgreSQL version: 8.1-devel
Operating system: freebsd 5.4
Description: coredump in postgres
Details:

pgsql is 8.1-devel, cvsup-latest

configured with:

/configure --enable-debug --with-perl --with-libraries=/usr/local/lib
--with-includes=/usr/local/include
--with-docdir=/usr/local/share/doc/postgresql --with-openssl --enable-nls
--prefix=/usr/local i386-portbld-freebsd5.4 'LDFLAGS=
-rpath=/usr/lib:/usr/local/lib' 'CFLAGS=-O -pipe -O3 -funroll-loops'
host_alias=i386-portbld-freebsd5.4 build_alias=i386-portbld-freebsd5.4
target_alias=i386-portbld-freebsd5.4 CC=cc

postgres.conf has all entries commented out

# gdb postgres --core=/usr/local/pgsql/data/postgres.core
GNU gdb 6.1.1 [FreeBSD]
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i386-marcel-freebsd"...
Core was generated by `postgres'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/local/lib/libintl.so.6...done.
Loaded symbols for /usr/local/lib/libintl.so.6
Reading symbols from /usr/lib/libssl.so.3...done.
Loaded symbols for /usr/lib/libssl.so.3
Reading symbols from /lib/libcrypto.so.3...done.
Loaded symbols for /lib/libcrypto.so.3
Reading symbols from /lib/libz.so.2...done.
Loaded symbols for /lib/libz.so.2
Reading symbols from /usr/local/lib/libreadline.so.5...done.
Loaded symbols for /usr/local/lib/libreadline.so.5
Reading symbols from /lib/libcrypt.so.2...done.
Loaded symbols for /lib/libcrypt.so.2
Reading symbols from /lib/libm.so.2...done.
Loaded symbols for /lib/libm.so.2
Reading symbols from /lib/libutil.so.4...done.
Loaded symbols for /lib/libutil.so.4
Reading symbols from /lib/libc.so.5...done.
Loaded symbols for /lib/libc.so.5
Reading symbols from /usr/local/lib/libiconv.so.3...done.
Loaded symbols for /usr/local/lib/libiconv.so.3
Reading symbols from /lib/libncurses.so.5...done.
Loaded symbols for /lib/libncurses.so.5
Reading symbols from /libexec/ld-elf.so.1...done.
Loaded symbols for /libexec/ld-elf.so.1
#0 create_or_index_quals (root=0x848bcc0, rel=0x84bb2b8) at pg_list.h:81
81 pg_list.h: No such file or directory.
in pg_list.h
(gdb) bt
#0 create_or_index_quals (root=0x848bcc0, rel=0x84bb2b8) at pg_list.h:81
#1 0x0817fff5 in set_plain_rel_pathlist (root=0x848bcc0, rel=0x84bb2b8,
rte=0x84a2538) at allpaths.c:204
#2 0x08180a09 in make_one_rel (root=0x848bcc0) at allpaths.c:177
#3 0x0818ecf3 in query_planner (root=0x848bcc0, tlist=0x84af558,
tuple_fraction=0,
cheapest_path=0xbfbfd21c, sorted_path=0xbfbfd220) at planmain.c:166
#4 0x0818f387 in grouping_planner (root=0x848bcc0, tuple_fraction=0) at
planner.c:849
#5 0x08190854 in subquery_planner (parse=0x83b0e58, tuple_fraction=0,
subquery_pathkeys=0x0)
at planner.c:376
#6 0x0819092c in planner (parse=0x83b0e58, isCursor=0 '\0',
cursorOptions=0, boundParams=0x0)
at planner.c:135
#7 0x081db7a3 in pg_plan_queries (querytrees=0x0, boundParams=0x0,
needSnapshot=0 '\0')
at postgres.c:726
#8 0x081dc6bd in exec_simple_query (
query_string=0x83af228 "SELECT i.guid AS instanceid, i.parent_id AS
checkdef, \n CASE\n", ' ' <repeats 12 times>, "WHEN i.lastchecked IS
NOT NULL THEN i.lastchecked + i.checkintvl\n", ' ' <repeats 12 times>, "ELSE
now()\n END AS schedtime\n FROM"...) at postgres.c:959
#9 0x081ded42 in PostgresMain (argc=4, argv=0x8385528, username=0x8385508
"pgsql") at postgres.c:3165
#10 0x081a7f99 in ServerLoop () at postmaster.c:2860
#11 0x081a971a in PostmasterMain (argc=3, argv=0xbfbfe99c) at
postmaster.c:940
#12 0x0816494d in main (argc=3, argv=0xbfbfe99c) at main.c:268

I added prints to backend/optimizer/path/orindxpath.c that say that
newrinfos is 0x0, and bestpath is 0x 84b8aa0 after this line:

newrinfos = make_restrictinfo_from_bitmapqual((Path *) bestpath, true);

and then

(gdb) p *(BitmapOrPath*)0x84b8aa0
$1 = {path = {type = T_BitmapOrPath, pathtype = T_BitmapOr, parent =
0x84b73a0,
startup_cost = 3.0029999999999997, total_cost = 3.0029999999999997,
pathkeys = 0x0},
bitmapquals = 0x84ba658, bitmapselectivity = 0.014999999999999999}

digging further into make_restrictinfo_from_bitmapqual I found that this
code emits NIL on bestpath:

else if (IsA(bitmapqual, IndexPath))
{
IndexPath *ipath = (IndexPath *) bitmapqual;
result = list_copy(ipath->indexclauses);
}

and in my case ipath->indexclauses was NIL:

(gdb) p ((BitmapOrPath*)0x84b8aa0)->bitmapquals->head->data.ptr_value
$8 = (void *) 0x84a2960

(gdb) p *((IndexPath*)0x84a2960)
$16 = {path = {type = T_IndexPath, pathtype = T_IndexScan, parent =
0x849f8d8, startup_cost = 0,
total_cost = 11.465782608695651, pathkeys = 0x0}, indexinfo = 0x849fca0,
indexclauses = 0x0,
indexquals = 0x0, isjoininner = 0 '\0', indexscandir =
ForwardScanDirection,
indextotalcost = 1.0009999999999999, indexselectivity =
0.0050000000000000001, rows = 1}

That seemed to me like a problem in the planner where I don't have
competence, so I stopped here.

The full query was:
SELECT i.guid AS instanceid, i.parent_id AS checkdef, CASE WHEN
i.lastchecked IS NOT NULL THEN i.lastchecked + i.checkintvl ELSE now() END
AS schedtime FROM v_rt_active_checks_with_status i WHERE ((i.lastchecked +
i.checkintvl) < now() OR i.lastchecked IS NULL OR i.alertlevel > 49 AND
(i.lastchecked + i.recheckintvl) < now()) AND (i.alertlevel < 30 OR
i.alertlevel > 31 OR i.alertlevel IS NULL);

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message uniware 2005-07-27 12:19:27 BUG #1791: Typo
Previous Message Richard Huxton 2005-07-27 07:40:38 Re: BUG #1788: charset of sorting bug