Обсуждение: BUG #3773: psql segfault on exit
The following bug has been logged online:
Bug reference: 3773
Logged by: guillaume 'ioguix' de Rorthais
Email address: ioguix@free.fr
PostgreSQL version: 8.3 beta3
Operating system: macosx 10.4.10
Description: psql segfault on exit
Details:
I just maked/make installed pg8.3b3 from cvs.
Here is a copy of the segfault with the psql and backend versions :
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~)-(Thu 22/Nov/07 18:30)
(ioguixigx@igix:p3:%)- /sw/opt/pg83//bin/psql --version
psql (PostgreSQL) 8.3beta3
contains support for command-line editing
(~)-(Thu 22/Nov/07 18:30)
(ioguixigx@igix:p3:%)- /sw/opt/pg83//bin/psql -U postgres -p 5433 pagila
Password for user postgres:
Welcome to psql 8.3beta3, the PostgreSQL interactive terminal.
Type: \copyright for distribution terms
\h for help with SQL commands
\? for help with psql commands
\g or terminate with semicolon to execute query
\q to quit
pagila=# select version();
version
----------------------------------------------------------------------------
----------------------------------------------------------------
PostgreSQL 8.3beta3 on i386-apple-darwin8.10.1, compiled by GCC
i686-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 5367)
(1 row)
pagila=# \q
psql(26486) malloc: *** error for object 0x1806200: incorrect checksum for
freed object - object was probably modified after being freed, break at
szone_error to debug
psql(26486) malloc: *** set a breakpoint in szone_error to debug
zsh: segmentation fault /sw/opt/pg83//bin/psql -U postgres -p 5433 pagila
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I don't know which other information I could add. Feel free to ask me for
some test/info.
guillaume 'ioguix' de Rorthais wrote: > PostgreSQL version: 8.3 beta3 > Operating system: macosx 10.4.10 > Description: psql segfault on exit > Details: > > I just maked/make installed pg8.3b3 from cvs. > > pagila=# \q > psql(26486) malloc: *** error for object 0x1806200: incorrect checksum for > freed object - object was probably modified after being freed, break at > szone_error to debug > psql(26486) malloc: *** set a breakpoint in szone_error to debug > zsh: segmentation fault /sw/opt/pg83//bin/psql -U postgres -p 5433 pagila > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I bet this is the same libedit bug that's been reported by various people, on OS X. For example: http://archives.postgresql.org/pgsql-bugs/2007-11/msg00128.php Since quite a few people seem to bump into this, I wonder if we could have configure to test for this? -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com
"guillaume 'ioguix' de Rorthais" <ioguix@free.fr> writes: > Operating system: macosx 10.4.10 > Description: psql segfault on exit If you have any lines longer than 255 bytes in your ~/.psql_history, this is a known bug in Apple's libedit. (They seem to have fixed it in Leopard, btw.) Recommended workaround is to install libreadline from source (the version in /usr/lib isn't really readline) and recompile with that instead. http://archives.postgresql.org/pgsql-hackers/2006-12/msg01222.php regards, tom lane
Heikki Linnakangas <heikki@enterprisedb.com> writes: > Since quite a few people seem to bump into this, I wonder if we could > have configure to test for this? Probably not worth the trouble, seeing that Apple has finally fixed the problem in 10.5 --- not tested by me, but a look at their source code on-line shows it's fixed: http://www.opensource.apple.com/darwinsource/10.5/libedit-11/libedit/src/history.c regards, tom lane
Tom Lane a écrit : > "guillaume 'ioguix' de Rorthais" <ioguix@free.fr> writes: > >> Operating system: macosx 10.4.10 >> Description: psql segfault on exit >> > > If you have any lines longer than 255 bytes in your ~/.psql_history, > this is a known bug in Apple's libedit. (They seem to have fixed it > in Leopard, btw.) Recommended workaround is to install libreadline from > source (the version in /usr/lib isn't really readline) and recompile > with that instead. > > http://archives.postgresql.org/pgsql-hackers/2006-12/msg01222.php > > regards, tom lane > Ok, thanks guys.