Call for platforms

* Thomas Lockhart <lockhart@alumni.caltech.edu> [010320 20:04]:

OK, here is my current platform list taken from the -hackers list and
from Vince's web page. I'm sure I've missed at least a few reports, but
please confirm that platforms are actually running and passing
regression tests with recent betas or the latest release candidate.

If a platform you are running on is not listed, make sure it gets
included! Platforms with reports for 7.0 risk being demoted to the "used
to be supported list", and platforms with reports for only 6.5 are on a
deathwatch, so be sure to speak up! Also, I've included names below to
remind us who helped last time, but feel free to report even if your
name is not already listed.

FreeBSD 4.3-BETA (will be -RELEASE by the time we release) works too.

I reported FreeBSD 4.[23].

LER

I've separated out recent reports and put them at the end of the list.
Thanks in advance.

- Thomas

AIX 4.3.2 RS6000 7.0 2000-04-05, Andreas Zeugswetter
Compaq Tru64 5.0 Alpha 7.0 2000-04-11, Andrew McMurry
IRIX 6.5.6f MIPS 6.5.3 2000-02-18, Kevin Wheatley
Linux 2.2.x armv4l 7.0 2000-04-17, Mark Knox
Linux 2.0.x MIPS 7.0 2000-04-13, Tatsuo Ishii
mklinux PPC750 7.0 2000-04-13, Tatsuo Ishii
NetBSD 1.4 arm32 7.0 2000-04-08, Patrick Welche
NetBSD 1.4U x86 7.0 2000-03-26, Patrick Welche
NetBSD m68k 7.0 2000-04-10, Henry B. Hotz
NetBSD Sparc 7.0 2000-04-13, Tom I. Helbekkmo
QNX 4.25 x86 7.0 2000-04-01, Dr. Andreas Kardos
SCO OpenServer 5 x86 6.5 1999-05-25, Andrew Merrill
Solaris x86 7.0 2000-04-12, Marc Fournier
Solaris 2.5.1-2.7 Sparc 7.0 2000-04-12, Peter Eisentraut
SunOS 4.1.4 Sparc 7.0 2000-04-13, Tatsuo Ishii
Windows/Win32 x86 7.0 2000-04-02, Magnus Hagander (clients only)
WinNT/Cygwin x86 7.0 2000-03-30, Daniel Horak

BeOS 5.0.3 x86 7.1 2000-12-18, Cyril Velter
BSDI 4.01 x86 7.1 2001-03-19, Bruce Momjian
FreeBSD 4.2 x86 7.1 2001-03-19, Vince Vielhaber
HPUX 10.20 PA-RISC 7.1 2001-03-19, Tom Lane
IBM S/390 7.1 2000-11-17, Neale Ferguson
Linux 2.2.x Alpha 7.1 2001-01-23, Ryan Kirkpatrick
Linux 2.2.16 x86 7.1 2001-03-19, Thomas Lockhart
Linux 2.2.15 Sparc 7.1 2001-01-30, Ryan Kirkpatrick
LinuxPPC G3 7.1 2001-03-19, Tom Lane
SCO UnixWare 7.1.1 x86 7.1 2001-03-19, Larry Rosenman
MacOS-X Darwin PowerPC 7.1 2000-12-11, Peter Bierman

---------------------------(end of broadcast)---------------------------
TIP 5: Have you checked our extensive FAQ?

http://www.postgresql.org/users-lounge/docs/faq.html

--
Larry Rosenman http://www.lerctr.org/~ler
Phone: +1 972-414-9812 E-Mail: ler@lerctr.org
US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749

SCO OpenServer 5 x86...

OK, I see that Billy Allie recently updated FAQ_SCO to indicate
demonstrated (?) support for OpenServer. I will reflect that in the
platform support info.

- Thomas

mklinux PPC750 7.0 2000-04-13, Tatsuo Ishii

I got core dump while running the parallel regression test of beta6.
Will look at...
--
Tatsuo Ishii

Compaq Tru64 5.0 Alpha 7.0 2000-04-11, Andrew McMurry

We've got 7.0.3 and 7.1b4 running on

Compaq Tru64 4.0G Alpha

Will do the regression test once RC1 is out.

Adriaan

mklinux PPC750 7.0 2000-04-13, Tatsuo Ishii

I got core dump while running the parallel regression test of beta6.
Will look at...
--
Tatsuo Ishii

VACUUM;
! FATAL 2: ZeroFill(logfile 0 seg 1) failed: No such file or directory
! pqReadData() -- backend closed the channel unexpectedly.

maybe a bug related to Tom recently fixed?
If so, I will try RC1...
--
Tatsuo Ishii

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

! FATAL 2: ZeroFill(logfile 0 seg 1) failed: No such file or directory
! pqReadData() -- backend closed the channel unexpectedly.

Is it possible you ran out of disk space?

regards, tom lane

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

! FATAL 2: ZeroFill(logfile 0 seg 1) failed: No such file or directory
! pqReadData() -- backend closed the channel unexpectedly.

Is it possible you ran out of disk space?

Probably not.
--
Tatsuo Ishii

Hi,

I reported Linux RedHat 6.2 - 2.2.14-5.0smp #1 SMP Tue Mar 7 21:01:40 EST
2000 i686
2 cpu - 1Go RAM

Gilles DAROLD

Thomas Lockhart writes:

SCO OpenServer 5 x86...

OK, I see that Billy Allie recently updated FAQ_SCO to indicate
demonstrated (?) support for OpenServer. I will reflect that in the
platform support info.

The last FAQ_SCO update was by me, and it was rather the consequence of
some implementational developments and not a good indicator of any
actually working platform. (I do have access to a Unixware box, but that
was already reported.)

--
Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/

Hi,

I am currently testing beta6 on AIX 4.3.3 on a RS6000 H80 with 4 cpu and 4
Go RAM
I use :

./configure
--with-CC=/usr/local/bin/gcc
--with-includes=/usr/local/include
--with-libraries=/usr/local/lib

All seem to be ok, There just the geometry failure in regression test
(following the AIX FAQ
it's normal ?)

But when I configure with --with-perl I have the following error :

make[4]: cc : Command not found

Any idea ?

Gilles DAROLD

Gilles DAROLD wrote:

Hi,

I am currently testing beta6 on AIX 4.3.3 on a RS6000 H80 with 4 cpu and 4
Go RAM
I use :

./configure
--with-CC=/usr/local/bin/gcc
--with-includes=/usr/local/include
--with-libraries=/usr/local/lib

All seem to be ok, There just the geometry failure in regression test

But when I configure with --with-perl I have the following error :

Ok symbolic link between cc and gcc seem to be the better fix.

I have now tested the --with-CXX option to compile libpq++ and it
really don't work, here are the output :

ld: 0711-319 WARNING: Exported symbol not defined:
PgConnection::CloseConnection
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::Connect
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::ConnectionBad

ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::DBName
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::ErrorMessage
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::Exec
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::ExecCommandOk

ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::ExecTuplesOk
ld: 0711-319 WARNING: Exported symbol not defined: PgConnection::IntToString

....

ld: 0711-317 ERROR: Undefined symbol: basic_string<char,
string_char_traits<char>, __default_alloc_template<false, 0> >::nilRep
ld: 0711-317 ERROR: Undefined symbol:
__malloc_alloc_template<0>::__malloc_alloc_oom_handler
ld: 0711-317 ERROR: Undefined symbol: endl(ostream &)
ld: 0711-317 ERROR: Undefined symbol: cerr
ld: 0711-317 ERROR: Undefined symbol: .ostream::operator<<(char const *)
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_end_free
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_start_free
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_heap_size
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_free_list
ld: 0711-317 ERROR: Undefined symbol: .__out_of_range(char const *)
ld: 0711-317 ERROR: Undefined symbol: .__length_error(char const *)
collect2: ld returned 8 exit status
make[3]: *** [libpq++.so] Error 1
make[3]: Leaving directory
`/home/darold/postgresql-7.1beta6/src/interfaces/libpq++'
make[2]: *** [all] Error 2

I have change the Makefile.global and replace c++ by g++ but it the same
output.

Could you tell me what going wrong ? Is my GNU install not fully functionnal ?

I use :
gcc version 2.95.2.1 19991024 (release) libs for powerpc-ibm-aix4.3.2.0

Regards

Gilles DAROLD

Gilles DAROLD writes:

I have now tested the --with-CXX option to compile libpq++ and it
really don't work, here are the output :

ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_start_free
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_heap_size
ld: 0711-317 ERROR: Undefined symbol: __default_alloc_template<false,
0>::_S_free_list
ld: 0711-317 ERROR: Undefined symbol: .__out_of_range(char const *)
ld: 0711-317 ERROR: Undefined symbol: .__length_error(char const *)
collect2: ld returned 8 exit status
make[3]: *** [libpq++.so] Error 1
make[3]: Leaving directory
`/home/darold/postgresql-7.1beta6/src/interfaces/libpq++'
make[2]: *** [all] Error 2

This could be a name mangling problem. Maybe the linker needs to be
invoked specially when building C++ libraries. Maybe the C++ compiler
driver needs to be invoked directly. This could especially be a problem
if you're using the GNU compiler with system libraries, since those are
usually compiled by the system compiler.

I have change the Makefile.global and replace c++ by g++ but it the same
output.

I think the good C++ compiler on AIX is called xlC. In any case, make
sure that you don't mix different C++ compilers. You need to do 'gmake
clean' at least in the libpq++ directory if you're switching.

--
Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/

Peter Eisentraut wrote:

This could be a name mangling problem. Maybe the linker needs to be
invoked specially when building C++ libraries. Maybe the C++ compiler
driver needs to be invoked directly. This could especially be a problem
if you're using the GNU compiler with system libraries, since those are
usually compiled by the system compiler.

I have change the Makefile.global and replace c++ by g++ but it the same
output.

I think the good C++ compiler on AIX is called xlC. In any case, make
sure that you don't mix different C++ compilers. You need to do 'gmake
clean' at least in the libpq++ directory if you're switching.

AIX faq said that xlC compiler don't work with libpq++ but with g++ it
may works :

libpq++ does not work because xlC does not have the string and bool

classes.

compiling the few files, that fail, with g++ does work.

Humm, I have no xlC compiler installed.Here are the compilation lines :

make[3]: Entering directory
`/home/darold/postgresql-7.1beta6/src/interfaces/libpq++'
g++ -O2 -Wall -I../../../src/interfaces/libpq -I../../../src/include
-I/usr/local/include -c -o pgconnection.o p
gconnection.cc
g++ -O2 -Wall -I../../../src/interfaces/libpq -I../../../src/include
-I/usr/local/include -c -o pgdatabase.o pgd
atabase.cc
g++ -O2 -Wall -I../../../src/interfaces/libpq -I../../../src/include
-I/usr/local/include -c -o pgtransdb.o pgtr
ansdb.cc
g++ -O2 -Wall -I../../../src/interfaces/libpq -I../../../src/include
-I/usr/local/include -c -o pgcursordb.o pgc
ursordb.cc
g++ -O2 -Wall -I../../../src/interfaces/libpq -I../../../src/include
-I/usr/local/include -c -o pglobject.o pglo
bject.cc
ar crs libpq++.a pgconnection.o pgdatabase.o pgtransdb.o pgcursordb.o
pglobject.o
touch libpq++.a
../../../src/backend/port/aix/mkldexport.sh libpq++.a > libpq++.exp
/usr/local/bin/gcc -Wl,-H512 -Wl,-bM:SRE
-Wl,-bI:../../../src/backend/postgres.imp -Wl,-bE:libpq++.exp -o libpq++.
so libpq++.a -L/usr/local/lib -L../../../src/interfaces/libpq -lpq -lc
ld: 0711-224 WARNING: Duplicate symbol: __start

All work until it want to link with libraries, it seems to want the libc
(-lc) and I don't
have it installed. Is it normal or this realease is not portable with AIX
4.3.3 ?

I see nobody did a test of 7.1 on Linux 2.4.x ?

Would be nice to certify it is running on kernel 2.4.x as they claim this
is entreprise strength kernel...

Cheers.

Thomas Lockhart wrote:

Franck Martin <franck@sopac.org> writes:

Would be nice to certify it is running on kernel 2.4.x as they claim this
is entreprise strength kernel...

Lamar, if you send me your SRPM I can do that...

--
Trond Eivind Glomsr�d
Red Hat, Inc.

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

! FATAL 2: ZeroFill(logfile 0 seg 1) failed: No such file or directory
! pqReadData() -- backend closed the channel unexpectedly.

Is it possible you ran out of disk space?

Probably not.

The reason I was speculating that was that it seems pretty unlikely
that a write() call could return ENOENT, as the above appears to
suggest. I think that the errno = ENOENT value was not set by write(),
but is leftover from the expected failure of BasicOpenFile earlier in
XLogFileInit. Probably write() returned some value less than BLCKSZ
but more than zero, and so did not set errno.

Offhand the only reason I can think of for a write to a disk file
to terminate after a partial transfer is a full disk. What do you
think?

regards, tom lane

* Tom Lane <tgl@sss.pgh.pa.us> [010321 21:29]:

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

Tatsuo Ishii <t-ishii@sra.co.jp> writes:

! FATAL 2: ZeroFill(logfile 0 seg 1) failed: No such file or directory
! pqReadData() -- backend closed the channel unexpectedly.

Is it possible you ran out of disk space?

Probably not.

The reason I was speculating that was that it seems pretty unlikely
that a write() call could return ENOENT, as the above appears to
suggest. I think that the errno = ENOENT value was not set by write(),
but is leftover from the expected failure of BasicOpenFile earlier in
XLogFileInit. Probably write() returned some value less than BLCKSZ
but more than zero, and so did not set errno.

Offhand the only reason I can think of for a write to a disk file
to terminate after a partial transfer is a full disk. What do you
think?

What about hitting a quota?

LER

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org

--
Larry Rosenman http://www.lerctr.org/~ler
Phone: +1 972-414-9812 E-Mail: ler@lerctr.org
US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749

On Thu, Mar 22, 2001 at 12:31:03PM +1200, Franck Martin wrote:

I see nobody did a test of 7.1 on Linux 2.4.x ?

Would be nice to certify it is running on kernel 2.4.x as they claim this
is entreprise strength kernel...

I've been running the 7.1 betas on 2.4 for weeks without any problems.
I replied to the "call for platforms" e-mail, but it looks like it got
lost in the avalanche.
I'll run the regression tests with the latest CVS snapshot and submit
a report to the list.

	-Roberto
-- 
+----| http://fslc.usu.edu USU Free Software & GNU/Linux Club|------+
  Roberto Mello - Computer Science, USU - http://www.brasileiro.net 
      http://www.sdl.usu.edu - Space Dynamics Lab, Web Developer    

OK: Linux 2.4.2 i686 / gcc 2.95.2 / Debian testing/unstable

no problems.

OK?: NetBSD 1.5 i586 / egcs 2.91.66 / (netbsd-1-5 from Jan)

netbsd FAILED the geometry test, diff attached, dunno if its
critical or not.

--
marko