vacuum problem - Mailing list pgsql-hackers

From Tatsuo Ishii
Subject vacuum problem
Date
Msg-id 199808250959.SAA15161@srapc451.sra.co.jp
Whole thread Raw
List pgsql-hackers
vacuum still doesn't work for me.

I did a regresion test and found the sanity_check test failed. The
result shows that the backend died while doing vacuum. I destroyed and
re-created the regression database. Vacuum for a fresh database seems
working.

Next I ran each test one by one manually to see when vacuum began to
fail.

Vacuuming after the boolean test was ok but after the char test the
backend died.

Here is a backtrace of the backend.

gdb ../../backend/postgres ../../../../data/base/regression/postgres.core
GDB is free software and you are welcome to 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.
GDB 4.16 (i386-unknown-freebsd),
Copyright 1996 Free Software Foundation, Inc...
Core was generated by `postgres'.
Program terminated with signal 6, Abort trap.
Cannot access memory at address 0x20105080.
#0  0x201c4b61 in ?? ()
(gdb) where
#0  0x201c4b61 in ?? ()
#1  0x201c43d3 in ?? ()
#2  0xcee08 in ExcAbort (excP=0xf9800, detail=0, data=0x0,
    message=0xd20ce "!(0 < (size) && (size) <= (0xfffffff))") at excabort.c:26
#3  0xced67 in ExcUnCaught (excP=0xf9800, detail=0, data=0x0,
    message=0xd20ce "!(0 < (size) && (size) <= (0xfffffff))") at exc.c:173
#4  0xcedba in ExcRaise (excP=0xf9800, detail=0, data=0x0,
    message=0xd20ce "!(0 < (size) && (size) <= (0xfffffff))") at exc.c:190
#5  0xce5ef in ExceptionalCondition (
    conditionName=0xd20ce "!(0 < (size) && (size) <= (0xfffffff))",
    exceptionP=0xf9800, detail=0x11d85c "size=0 [0x0]",
    fileName=0xd20b8 "mcxt.c", lineNumber=228) at assert.c:73
#6  0xd2140 in MemoryContextAlloc (context=0x1320ac, size=0) at mcxt.c:227
#7  0xd23b2 in palloc (size=0) at palloc.c:69
#8  0xde02 in GetIndexValue (tuple=0x210850d8, hTupDesc=0x12d250, attOff=0,
    attrNums=0x21054bbc, fInfo=0x13ae10, attNull=0xefbfaa8f "")
    at indexam.c:386
#9  0x222e8 in FormIndexDatum (numberOfAttributes=1,
    attributeNumber=0x21054bbc, heapTuple=0x210850d8, heapDescriptor=0x12d250,
    datum=0x19fe90, nullv=0x1a16d0 "\220#\032", fInfo=0x13ae10) at index.c:1284
#10 0x2f45d in vc_rpfheap (vacrelstats=0x12e850, onerel=0x132b10,
    vacuum_pages=0xefbfac44, fraged_pages=0xefbfac38, nindices=3,
    Irel=0x1a16b0) at vacuum.c:1146
#11 0x2e1b7 in vc_vacone (relid=1249, analyze=0, va_cols=0x0) at vacuum.c:552
#12 0x2d911 in vc_vacuum (VacRelP=0x0, analyze=0 '\000', va_cols=0x0)
    at vacuum.c:256
#13 0x2d72e in vacuum (vacrel=0x0, verbose=0, analyze=0 '\000', va_spec=0x0)
    at vacuum.c:159
#14 0xa6337 in ProcessUtility (parsetree=0x12d630, dest=Remote)
    at utility.c:634
#15 0xa3442 in pg_exec_query_dest (query_string=0xefbfad6c "vacuum;",
    dest=Remote, aclOverride=0) at postgres.c:706
#16 0xa3334 in pg_exec_query (query_string=0xefbfad6c "vacuum;")
    at postgres.c:644
#17 0xa496b in PostgresMain (argc=6, argv=0xefbfcde4, real_argc=4,
    real_argv=0xefbfd6b8) at postgres.c:1481
#18 0x8c0f4 in DoBackend (port=0x131000) at postmaster.c:1412
#19 0x8bb56 in BackendStartup (port=0x131000) at postmaster.c:1191
#20 0x8b046 in ServerLoop () at postmaster.c:725
#21 0x8a92b in PostmasterMain (argc=4, argv=0xefbfd6b8) at postmaster.c:534
#22 0x4bfc7 in main (argc=4, argv=0xefbfd6b8) at main.c:93



pgsql-hackers by date:

Previous
From: Sferacarta Software
Date:
Subject: Re[2]: [GENERAL] DELETE statement DOESN'T WORK ANYMORE
Next
From: Michael Meskes
Date:
Subject: Re: [HACKERS] initdb problem