What does it mean?
One of my customer unable to install plpgsql
with next message:
www2(dms)~>psql -d www -c "select tst(); "
ERROR: Load of file /usr/local/pgsql/lib/plpgsql.so failed: dlopen
'/usr/local/pgsql/lib/plpgsql.so' failed. (/usr/local/pgsql/lib/plpgsql.so:
Undefined symbol "SPI_tuptable")
www2(dms)~>
what does it mean?
Env:
FreeBSD www2.sptimes.ru 3.1-RELEASE FreeBSD 3.1-RELEASE #4: Tue Mar 23 13:18:41
MSK 1999 wplus@www2.sptimes.ru:/usr/src/sys/compile/SPTIMES i386
Postgres 6.4.2 release
PS:
How about adding something like -a key to pg_version to report full
release and compile information?
---
Dmitry Samersoff, dms@wplus.net, ICQ:3161705
http://devnull.wplus.net
* there will come soft rains ...
One of my customer unable to install plpgsql
with next message:www2(dms)~>psql -d www -c "select tst(); "
ERROR: Load of file /usr/local/pgsql/lib/plpgsql.so failed: dlopen
'/usr/local/pgsql/lib/plpgsql.so' failed. (/usr/local/pgsql/lib/plpgsql.so:
Undefined symbol "SPI_tuptable")
www2(dms)~>what does it mean?
Env:
FreeBSD www2.sptimes.ru 3.1-RELEASE FreeBSD 3.1-RELEASE #4: Tue Mar 23 13:18:41
MSK 1999 wplus@www2.sptimes.ru:/usr/src/sys/compile/SPTIMES i386Postgres 6.4.2 release
Doesn't look to me like a v6.4.2. SPI_tuptable is a global
pointer which is defined in .../backend/executor/spi.c.
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#======================================== jwieck@debis.com (Jan Wieck) #
One of my customer unable to install plpgsql
with next message:
www2(dms)~>psql -d www -c "select tst(); "
ERROR: Load of file /usr/local/pgsql/lib/plpgsql.so failed: dlopen
'/usr/local/pgsql/lib/plpgsql.so' failed. (/usr/local/pgsql/lib/plpgsql.so:
Undefined symbol "SPI_tuptable")
www2(dms)~>
This problem exists onlty on FreeBSD 3.1
the same sources build on 2.2.8 works properly.
what does it mean?
Env:
FreeBSD www2.sptimes.ru 3.1-RELEASE FreeBSD 3.1-RELEASE #4: Tue Mar 23
13:18:41
MSK 1999 wplus@www2.sptimes.ru:/usr/src/sys/compile/SPTIMES i386
Postgres 6.4.2 release
PS:
How about adding something like -a key to pg_version to report full
release and compile information?
---
Dmitry Samersoff, dms@wplus.net, ICQ:3161705
http://devnull.wplus.net
* there will come soft rains ...
Dmitry Samersoff wrote:
One of my customer unable to install plpgsql
with next message:www2(dms)~>psql -d www -c "select tst(); "
ERROR: Load of file /usr/local/pgsql/lib/plpgsql.so failed: dlopen
'/usr/local/pgsql/lib/plpgsql.so' failed. (/usr/local/pgsql/lib/plpgsql.so:
Undefined symbol "SPI_tuptable")
www2(dms)~>This problem exists onlty on FreeBSD 3.1
the same sources build on 2.2.8 works properly.what does it mean?
It means that the dynamic loader isn't able to resolve a
reference to the global symbol "SPI_tuptable" from the
PL/pgSQL shared object into the backend. The symbol
"SPI_tuptable" is declared in .../src/backend/executor/spi.c
as
DLLIMPORT SPITupleTable *SPI_tuptable;
Since this symbol is referenced from another place in the
backend's static code (in ruleutils.c) I'm pretty sure the
symbol is there. It must be a problem with the FreeBSD 3.1
dynamic loader.
Env:
FreeBSD www2.sptimes.ru 3.1-RELEASE FreeBSD 3.1-RELEASE #4: Tue Mar 23
13:18:41
MSK 1999 wplus@www2.sptimes.ru:/usr/src/sys/compile/SPTIMES i386Postgres 6.4.2 release
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#======================================== jwieck@debis.com (Jan Wieck) #
On 10-May-99 Jan Wieck wrote:
Dmitry Samersoff wrote:
One of my customer unable to install plpgsql
with next message:www2(dms)~>psql -d www -c "select tst(); "
ERROR: Load of file /usr/local/pgsql/lib/plpgsql.so failed: dlopen
'/usr/local/pgsql/lib/plpgsql.so' failed. (/usr/local/pgsql/lib/plpgsql.so:
Undefined symbol "SPI_tuptable")
www2(dms)~>This problem exists onlty on FreeBSD 3.1
Since this symbol is referenced from another place in the
backend's static code (in ruleutils.c) I'm pretty sure the
symbol is there. It must be a problem with the FreeBSD 3.1
dynamic loader.
OK, I see.
Does any body have the same problem and is there known solution?
---
Dmitry Samersoff, dms@wplus.net, ICQ:3161705
http://devnull.wplus.net
* there will come soft rains ...