Re: Psql crashes with Segmentation fault on copy from

From: Gregory Stark <stark(at)enterprisedb(dot)com>
To: <lists(at)stringsutils(dot)com>
Cc: "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, <pgsql-general(at)postgresql(dot)org>
Subject: Re: Psql crashes with Segmentation fault on copy from
Date: 2008-05-28 19:09:42
Message-ID: 87lk1up7nt.fsf@oxford.xeocode.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


"Francisco Reyes" <lists(at)stringsutils(dot)com> writes:

> #1 0x0000002a955820ae in pqPutMsgBytes (buf=0x2a9860a010, len=2147483647,

So it's trying to execute an sql query that's MAXINT bytes long which is
pretty off-course.

> #5 0x0000000000406e95 in SendQuery (
> query=0x2a9860a010 "yjbjK8WKLRHKavptOqlPD4QeI24=\t200803\t2008-03-11\t0\t10.47\t08620667742MDF1\t\tMCDONALD'S F7742\t5814\t1003\t1187954\nyjbjK8WKLRHKavptOqlPD4QeI24=\t200803\t2008-03-11\t0\t16.99\t00614111869AWG1\t\tPIGGLY WIGGLY#52\t5411"...)
> at common.c:878
> #6 0x0000000000409dfa in MainLoop (source=0x5414f0) at mainloop.c:340
> #7 0x000000000040408d in process_file (filename=0x541d70
> "data/usb_t_60M.sql", single_txn=0 '\0') at command.c:1394
> #8 0x00000000004057cd in exec_command (cmd=0x541bc0 "i",
> scan_state=0x549b30, query_buf=0x549890) at command.c:546

Could you post the first couple lines of data/usb_t_60M.sql ? Does it really
have a COPY command at the beginning? Are you really doing \i
data/usb_t_60M.sql or were you trying to do a copy from this file?

> #9 0x0000000000406286 in HandleSlashCmds (scan_state=0x549b30,
> query_buf=0x549890) at command.c:92
> #10 0x0000000000409f66 in MainLoop (source=0x541070) at mainloop.c:259
> #11 0x000000000040408d in process_file (filename=0x560750
> "bench-index-after.sql", single_txn=0 '\0') at command.c:1394
> #12 0x00000000004057cd in exec_command (
> cmd=0x569360 "Hxnh9aOwMviVVP9I=\t200803\t2008-03-10\t0\t15.54\t434240509886\t\tNU-YALE CLEANERS (BARD\t4225\t0\t11636644\n3MM9gwBGMHiyPGmUrqIEOOha1Ao=\t200803\t2008-03-11\t0\t739.95\t554172100100007\t\tUSAA P&C PREMIUM\t6300\t0\t183718"...,
> scan_state=0x549c20, query_buf=0x54a480) at command.c:546

I'm a bit puzzled by this. The command being executed here seems to have been
overwritten by data from the later command -- before it was overwritten it was
apparently "\i bench-index-after.sql". It seems something with the
PsqlScanState is not being sufficient to make the lexer completely reentrant
here.

What version of flex is this built with?

--
Gregory Stark
EnterpriseDB http://www.enterprisedb.com
Ask me about EnterpriseDB's 24x7 Postgres support!

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Gregory Stark 2008-05-28 19:18:51 Re: Bottom Posting
Previous Message Richard Huxton 2008-05-28 18:59:11 Revealing the "postgres" user service-account in Windows XP Home