Обсуждение: PostgreSQL 7.3.4 to 4.2 Regresion Fails

Dear all,

I am migrating from PostgreSQL 7.3.4
to
PostgreSQL 7..4.2
Steps followed are :
1. Backup
2. Get the copy  >> extract >> own the extracted files by postgres
(SuperUser)
3. Remove the old data directoy as old and new data directory would be same
4. Configure with

 ./configure \--bindir=/usr/bin \--prefix=/postgres
\--libdir=/usr/lib/pgsql \--sysconfdir=/etc/sysconfig/pgsql
\--datadir=/postgres/data \--enable-multibyte \--with-perl
\--enable-syslog \--with-CXX \--with-tcl \--with-python \--enable-odbc
\--with-pgport=5432 \--with-unixodbc

5. Make with

gmake

6. Regress test with

gmake check

The test fails with following error

<code>

gcc -shared -o autoinc.so autoinc.o
rm refint.o autoinc.o
gmake[3]: Leaving directory `/usr/local/src/postgresql-7.4.2/contrib/spi'
/bin/sh ./pg_regress --temp-install --top-builddir=../../..
--schedule=./parallel_schedule --multibyte=SQL_ASCII
============== creating temporary installation        ==============
============== initializing database system           ==============
============== starting postmaster                    ==============
running on port 65432 with pid 3393
============== creating database "regression"         ==============
/usr/local/src/postgresql-7.4.2/src/test/regress/./tmp_check/install//usr/bin/createdb:
relocation error:
/usr/local/src/postgresql-7.4.2/src/test/regress/./tmp_check/install//usr/bin/createdb:
undefined symbol: get_progname
pg_regress: createdb failed
gmake[2]: *** [check] Error 2
rm regress.o
gmake[2]: Leaving directory
`/usr/local/src/postgresql-7.4.2/src/test/regress'
gmake[1]: *** [check] Error 2
gmake[1]: Leaving directory `/usr/local/src/postgresql-7.4.2/src/test'
gmake: *** [check] Error 2

</code>

Machine Specs are:
1. bison (GNU Bison) 1.875
2. gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)
3. Kernel 2.4.20-24.9
4. RH 9.0

What am I doing wrong ?

Thanks in advance.

--
Best Regards,
Vishal Kashyap
Director / Lead Developer,
Sai Hertz And Control Systems Pvt Ltd,
http://saihertz.rediffblogs.com
Jabber IM: vishalkashyap@jabber.org
ICQ :      264360076
Yahoo  IM: mailforvishal@yahoo.com
-----------------------------------------------
You yourself, as much as anybody in the entire
universe, deserve your love and affection.
- Buddha
---------------
pgsql=# select marital_status from vishals_life;

marital_status
------------------
Single not looking

1 Row(s) affected

                    ___
                   //\\\
                  ( 0_0 )
----------------o0o-----o0o---------------------

Just an update :

I ignored the Regression test results and went forward to install
PostgreSQL 7.4.2
After this I re ran there Regression test and it failed 1 test of 93 I
again ran the test and all 93 test passed

Now the only question I have is since initdb and all is working fiine

Must I  take the Regression test failures seriously ?



--
Best Regards,
Vishal Kashyap
Director / Lead Developer,
Sai Hertz And Control Systems Pvt Ltd,
http://saihertz.rediffblogs.com
Jabber IM: vishalkashyap@jabber.org
ICQ :      264360076
Yahoo  IM: mailforvishal@yahoo.com
-----------------------------------------------
You yourself, as much as anybody in the entire
universe, deserve your love and affection.
- Buddha
---------------
pgsql=# select marital_status from vishals_life;

marital_status
------------------
Single not looking

1 Row(s) affected

                    ___
                   //\\\
                  ( 0_0 )
----------------o0o-----o0o---------------------

"V i s h a l Kashyap @ [Sai Hertz And Control Systems]" <sank89@sancharnet.in> writes:
> /usr/local/src/postgresql-7.4.2/src/test/regress/./tmp_check/install//usr/bin/createdb:
> undefined symbol: get_progname

This happens because the old version of libpq.so is still in place and
is picked up in preference to the as-yet-uninstalled new version.  If
you are going to install 7.4 over 7.3, there's not a lot that "make
check" can do to prevent the dynamic linker from doing that :-(

            regards, tom lane

"V i s h a l Kashyap @ [Sai Hertz And Control Systems]" <sank89@sancharnet.in> writes:
> Must I  take the Regression test failures seriously ?

There are failures and there are failures.  You need to do at least some
investigation to understand why it's reporting the failure ...

            regards, tom lane