BUG #1678: pw_shadow BUS ERROR

Haluk GUNCER wrote:

fixing permissions on existing directory /usr/local/pgsql/data ... ok
creating directory /usr/local/pgsql/data/global ... ok
creating directory /usr/local/pgsql/data/pg_xlog ... ok
creating directory /usr/local/pgsql/data/pg_xlog/archive_status ... ok
creating directory /usr/local/pgsql/data/pg_clog ... ok
creating directory /usr/local/pgsql/data/pg_subtrans ... ok
creating directory /usr/local/pgsql/data/base ... ok
creating directory /usr/local/pgsql/data/base/1 ... ok
creating directory /usr/local/pgsql/data/pg_tblspc ... ok
selecting default max_connections ... 100
selecting default shared_buffers ... 1000
creating configuration files ... ok
creating template1 database in /usr/local/pgsql/data/base/1 ... ok
initializing pg_shadow ... sh: 2419 Bus error(coredump)
child process exited with exit code 138

Can you get a stack trace from the core file? (Depending on where the
core file gets created, you may need to use the '-n' option to initdb to
stop the partially-created data directory from being removed.)

-Neil

I've seen this before, and that was due to unaligned memory access,
which is not supported on some platforms.

I thought we didn't do unaligned memory access??

... John

Haluk G�N�ER wrote:

I did it with -n option. The generated core file is attached. I hope it
will help.

The core file itself is not much use -- you need to get a stack trace
from the core file (which I can't do, since I don't have the binaries
you built -- or an HPUX system for that matter). You can get a stack
trace via gdb using "gdb /path/to/binary /path/to/core" followed by
"bt"; there may be a different / better way to do things on HPUX (HP
support should be able to tell you, at any rate).

-Neil

Hello Neil,

Does this output mean anything to you?

$ gdb /usr/local/pgsql/bin/postgres /usr/local/pgsql/data/base/1/core bt
.
Core was generated by `postgres'.
Program terminated with signal 10, Bus error.

warning: The shared libraries were not privately mapped; setting a
breakpoint in a shared library will not work until you rerun the program.

#0 0x23cf64 in CreatePortal+0x1f4 ()
(gdb) bt
#0 0x23cf64 in CreatePortal+0x1f4 ()
#1 0x23ce28 in CreatePortal+0xb8 ()
#2 0x19c094 in exec_simple_query+0x194 ()
#3 0x19daa8 in PostgresMain+0x1520 ()
#4 0x12f918 in main+0x1e0 ()

thanks.
Haluk

----- Original Message -----
From: "Neil Conway" <neilc@samurai.com>
To: "Haluk G�N�ER" <halukguncer@hotmail.com>
Cc: <pgsql-bugs@postgresql.org>
Sent: Thursday, May 26, 2005 4:02 AM
Subject: Re: [BUGS] BUG #1678: pw_shadow BUS ERROR

On Mon, 2005-05-30 at 19:36 +0300, Haluk G�N�ER wrote:

#0 0x23cf64 in CreatePortal+0x1f4 ()
(gdb) bt
#0 0x23cf64 in CreatePortal+0x1f4 ()
#1 0x23ce28 in CreatePortal+0xb8 ()
#2 0x19c094 in exec_simple_query+0x194 ()
#3 0x19daa8 in PostgresMain+0x1520 ()
#4 0x12f918 in main+0x1e0 ()

Weird -- it's difficult to see how there could be a problem in
CreatePortal(), as that is both simple and widely used code. (The
recursive call to CreatePortal() also makes no sense, but perhaps that
is just weirdness in the backtrace.)

Would it be possible to get a stacktrace with a copy of Postgres that
has been compiled with --enable-debug?

-Neil

Neil Conway <neilc@samurai.com> writes:

Weird -- it's difficult to see how there could be a problem in
CreatePortal(), as that is both simple and widely used code.

The other interesting point is that we've certainly tested on HPUX
recently --- I know I've checked both 11.11 and 11.23 using HP's
testdrive machines. So the question here is not so much "why doesn't
PG work on HPUX" as "what weird configuration are you using?"

regards, tom lane

I have already rebuilt it with -g option.

Is it possible to get the binaries for HPUX from you? I wonder if the
problem persists with yours.
I would also like to get for solaris8 afterwards as well.

thanks a lot.
Haluk.

----- Original Message -----
From: "Tom Lane" <tgl@sss.pgh.pa.us>
To: "Neil Conway" <neilc@samurai.com>
Cc: "Haluk G�N�ER" <halukguncer@hotmail.com>; <pgsql-bugs@postgresql.org>
Sent: Tuesday, May 31, 2005 9:13 AM
Subject: Re: [BUGS] BUG #1678: pw_shadow BUS ERROR

Hello Tom,

You wrote that you ran postgresql on HPUX11.11.
Our Server type: V2500 12CPU 16GB RAM.

I would like to get the patch list installed on the machine you compiled
postgresQL FROM YOU , so that I can compare with ours. It takes about 12kB.

Command for patch list
# swlist
..
PHSS_30478 1.0 X11 Font Library Patch
PHSS_30791 1.0 XClients Periodic Patch
PHSS_30970 1.0 ld(1) and linker tools
cumulative patch
..
#
# swlist > patchlist.txt
#

Regards.
Haluk.

----- Original Message -----
From: "Haluk G�N�ER" <halukguncer@hotmail.com>
To: "Neil Conway" <neilc@samurai.com>; "Tom Lane" <tgl@sss.pgh.pa.us>
Cc: <pgsql-bugs@postgresql.org>
Sent: Thursday, June 02, 2005 6:17 PM
Subject: Re: [BUGS] BUG #1678: pw_shadow BUS ERROR

=?iso-8859-9?Q?Haluk_G=DCN=C7ER?= <halukguncer@hotmail.com> writes:

I would like to get the patch list installed on the machine you compiled
postgresQL FROM YOU , so that I can compare with ours. It takes about 12kB.

I used one of the PA-RISC servers provided by HP's testdrive program
http://www.testdrive.hp.com ... I don't remember exactly which one,
but hopefully they are all configured alike. (Of course, they may have
updated them since I last tried it, which was early January or so.)
You can sign up for an account and check all the details you like...

regards, tom lane

Tom,

My problem is solved! I compiled postgresql 8.0.3 on HP test drive machine.
Then it worked succesfully.
I tar'ed and copied the generated code to our server , and it worked.

I suggest to put binaries for each vendor to this website for downloading
so that we wont waste out time for compiling.

Thanks a lot to you all.
Haluk

----- Original Message -----
From: "Tom Lane" <tgl@sss.pgh.pa.us>
To: "Haluk G�N�ER" <halukguncer@hotmail.com>
Cc: <pgsql-bugs@postgresql.org>
Sent: Friday, June 03, 2005 5:29 PM
Subject: Re: [BUGS] BUG #1678: pw_shadow BUS ERROR