Re: BUG #2129: dblink problem

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Akio Iwaasa <iwaasa(at)mxs(dot)nes(dot)nec(dot)co(dot)jp>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: BUG #2129: dblink problem
Date: 2005-12-28 03:46:28
Message-ID: 200512280346.jBS3kSE11753@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs pgsql-patches


We have significantly improved dblink since 7.4. Would you try 8.1 and
see if you can reproduce the problem?

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

Akio Iwaasa wrote:
>
> The following bug has been logged online:
>
> Bug reference: 2129
> Logged by: Akio Iwaasa
> Email address: iwaasa(at)mxs(dot)nes(dot)nec(dot)co(dot)jp
> PostgreSQL version: 7.4.10
> Operating system: Redhat EL ES 3.0
> Description: dblink problem
> Details:
>
> I'm very sorry for my poor English.
>
> "postgres" process terminated with "signal 11"
> because of my wrong SQL statement using "dblink".
>
> --- SQL statement(Select statement a function) ---
> select into RET *
> from dblink(''select C1,C2,C3 from TABLE01 where ... '') <<<<< 3 column
> as LINK_TABLE01(LC1 varchar(5),LC2 varchar(5),
> LC3 varchar(5),LC4 varchar(5)) ; <<<<< 4 column
> ---------------------------------------------------
>
> Backtrace is below.
>
> ---
> (gdb) core /usr/local/pgsql74a/data/base/10218530/core.20823
> Core was generated by `postgres: postgres nwops [local] CO'.
> Program terminated with signal 11, Segmentation fault.
> :
> (gdb) bt
> #0 0x00575ffb in strlen () from /lib/tls/libc.so.6
> #1 0x081f222a in varcharin (fcinfo=0xbfffbf70) at
> varchar.c:368
> #2 0x0821b86e in FunctionCall3 (flinfo=0x89f5b60,
> arg1=29795, arg2=0, arg3=64) at fmgr.c:1016
> #3 0x08120647 in BuildTupleFromCStrings (attinmeta=
> 0x89f5b00, values=0x8a2e2f0) at execTuples.c:730
> #4 0x00b5173d in dblink_record (fcinfo=0xbfffc160)
> at dblink.c:699
> #5 0x0811c8c8 in ExecMakeTableFunctionResult
> (funcexpr=0x89f5058, econtext=0x89f4c70,
> expectedDesc=0x89f4e08, returnDesc=0xbfffc244)
> at execQual.c:1057
> #6 0x0812882c in FunctionNext (node=0x89f4be8) at
> nodeFunctionscan.c:78
> #7 0x0811fba6 in ExecScan (node=0x89f4be8,
> accessMtd=0x81287dc <FunctionNext>) at
> execScan.c:98
> #8 0x08128906 in ExecFunctionScan (node=0x89f4be8) at
> nodeFunctionscan.c:128
> #9 0x0811aa0c in ExecProcNode (node=0x89f4be8) at
> execProcnode.c:322
> #10 0x081190b8 in ExecutePlan (estate=0x89f4ad8,
> planstate=0x89f4be8, operation=CMD_SELECT,
> numberTuples=1, direction=ForwardScanDirection,
> dest=0x82b4734) at execMain.c:1103
> #11 0x081182f5 in ExecutorRun (queryDesc=0x89e66b0,
> direction=ForwardScanDirection, count=1) at
> execMain.c:249
> #12 0x0812fb65 in _SPI_pquery (queryDesc=0x89e66b0,
> runit=1 '\001', useCurrentSnapshot=0 '\0',
> tcount=1) at spi.c:1254
> #13 0x0812fa5c in _SPI_execute_plan (plan=0x8a1b728,
> ValuNulls=0x8a19398 " 1111112", useCurrentSnapshot=0
> '\0', tcount=1) at spi.c:1201
> #14 0x0812da75 in SPI_execp (plan=0x8a1b728, Values=
> 0x8a193e0, Nulls=0x8a19398 " 1111112", tcount=1) at
> spi.c:241
> #15 0x00ed97a3 in exec_run_select (estate=0xbfffc620,
> expr=0x89e9170, maxtuples=1, portalP=0x0) at
> pl_exec.c:3223
> #16 0x00ed6abe in exec_stmt_select (estate=0xbfffc620,
> stmt=0x89e9250) at pl_exec.c:1519
> #17 0x00ed5eb6 in exec_stmt (estate=0xbfffc620,
> stmt=0x89e9250) at pl_exec.c:967
> #18 0x00ed5d3e in exec_stmts (estate=0xbfffc620,
> stmts=0x89e8ed0) at pl_exec.c:903
> #19 0x00ed5c28 in exec_stmt_block (estate=0xbfffc620,
> block=0x89ebfd8) at pl_exec.c:859
> #20 0x00ed56f0 in plpgsql_exec_trigger (func=0x89e86a8,
> trigdata=0xbfffc830) at pl_exec.c:645
> #21 0x00ed149f in plpgsql_call_handler
> (fcinfo=0xbfffc700) at pl_handler.c:121
> #22 0x0810442b in ExecCallTriggerFunc
> (trigdata=0xbfffc830, finfo=0x89e26d8,
> per_tuple_context=0x89e0500) at trigger.c:1150
> #23 0x0810562c in DeferredTriggerExecute
> (event=0x89eeb10, itemno=0, rel=0xb5549300,
> trigdesc=0x89e2370, finfo=0x89e26c0,
> per_tuple_context=0x89e0500) at trigger.c:1867
> #24 0x0810589a in deferredTriggerInvokeEvents (
> immediate_only=1 '\001') at trigger.c:2008
> #25 0x08105a38 in DeferredTriggerEndQuery ()
> at trigger.c:2143
> #26 0x0819d61a in finish_xact_command () at
> postgres.c:1757
> #27 0x0819c427 in exec_simple_query
> (query_string=0x89dcfb8 "COPY user_info_tbl FROM
> STDIN ;") at postgres.c:946
> #28 0x0819eb0e in PostgresMain (argc=4, argv=0x8990430,
> username=0x89903a0 "postgres") at postgres.c:2918
> #29 0x081728b8 in BackendFork (port=0x899cd70) at
> postmaster.c:2564
> #30 0x08171fe2 in BackendStartup (port=0x899cd70) at
> postmaster.c:2207
> #31 0x08170661 in ServerLoop () at postmaster.c:1119
> #32 0x08170100 in PostmasterMain (argc=1, argv=0x898f538)
> at postmaster.c:897
> #33 0x08137ccb in main (argc=1, argv=0xbfffd9e4)
> at main.c:214
> (gdb) up
> #1 0x081f222a in varcharin (fcinfo=0xbfffbf70)
> at varchar.c:368
> (gdb) p s
> $10 = 0x7463 <Address 0x7463 out of bounds>
> (gdb) up
> #2 0x0821b86e in FunctionCall3 (flinfo=0x89f5b60,
> arg1=29795, arg2=0, arg3=64) at fmgr.c:1016
> (gdb) up
> #3 0x08120647 in BuildTupleFromCStrings (attinmeta=
> 0x89f5b00, values=0x8a2e2f0) at execTuples.c:730
> (gdb) p i
> $12 = 3
> (gdb) p values[0]
> $13 = 0x8a2066c "11111112"
> (gdb) p values[1]
> $16 = 0x8a20675 "11111112"
> (gdb) p values[2]
> $14 = 0x8a2067e "1"
> (gdb) p values[3]
> $15 = 0x7463 <Address 0x7463 out of bounds>
> (gdb) up
> #4 0x00b5173d in dblink_record (fcinfo=0xbfffc160)
> at dblink.c:699
> (gdb) p *((PGresult *)funcctx->user_fctx)
> $9 = {
> ntups = 1,
> numAttributes = 3,
> attDescs = 0x8a205dc,
> tuples = 0x8a21470,
> tupArrSize = 128,
> resultStatus = PGRES_TUPLES_OK,
> cmdStatus = "SELECT", '\0' <repeats 33 times>,
> binary = 0,
> noticeHooks = {
> noticeRec = 0xac0635 <defaultNoticeReceiver>,
> noticeRecArg = 0x0,
> noticeProc = 0xac0679 <defaultNoticeProcessor>,
> noticeProcArg = 0x0
> },
> client_encoding = 1,
> errMsg = 0x0,
> errFields = 0x0,
> null_field = "",
> curBlock = 0x8a205d8,
> curOffset = 168,
> spaceLeft = 1880
> }
> (gdb) p *(funcctx->attinmeta->tupdesc)
> $18 = {
> natts = 4,
> attrs = 0x89f58a0,
> constr = 0x0,
> tdhasoid = 0 '\0'
> }
> (gdb)
> ---
>
> Of course this trouble occurred because of my
> mistake, but I think that it is better if "dblink"
> returns an error for a wrong SQL statement.
>
> If "numAttributes"*1 was smaller than "natts"*2
> before "dblink_record" calls "BuildTupleFromCStrings",
> dblink can not return error ?
>
> *1 funcctx->user_fctx->numAttributes
> *2 funcctx->attinmeta->tupdesc->natts
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: Don't 'kill -9' the postmaster
>

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Amit 2005-12-28 08:47:09 BUG #2134: Broken Pipe Error
Previous Message Bruce Momjian 2005-12-28 03:27:57 Re: [BUGS] BUG #2114: (patch) COPY FROM ... end of

Browse pgsql-patches by date

  From Date Subject
Next Message Jeremy Drake 2005-12-28 04:30:15 Re: sending mail from Postgres
Previous Message Bruce Momjian 2005-12-28 03:43:06 Re: [BUGS] BUG #2114: (patch) COPY FROM ... end of