Integration of PVM into SGE 5.3

Co Thai Ngo <[email protected]> Thu, 16 Jun 2005 15:35:20 -0600
Newsgroups gmane.os.netbsd.devel.cluster,gmane.os.netbsd.current
Message-ID <[email protected]>
Hi,  =20
  =20
I'm trying to integrate PVM into SGE 5.3. But for some reasons, pvm is no=
t  =20
started. Here is what I did:  =20
- replaced SGE_ROOT/pvm with the supplied from SGE howto website  =20
(http://gridengine.sunsource.net/howto/pvm-integration/pvm-integration.ht=
ml)  =20
- Since NetBSD is not supported by PVM, I added the folloing line in to  =
=20
SGE_ROOT/pvm/src/aimk  =20
********  =20
case nbsd-i386:    =20
case glinux:    =20
case linux:    =20
set CC =3D gcc    =20
set CFLAGS =3D "-O -Wall -Werror -Wstrict-prototypes -DLINUX $DEBUG_FLAG =
 =20
$CFLAGS"    =20
set LFLAGS =3D "$DEBUG_FLAG $LFLAGS"  =20
...    =20
*******    =20
- Defined PE: =20
********   =20
arbutus# qconf -mp pvm  =20
pe_name           pvm =20
queue_list        all =20
slots             32 =20
user_lists        NONE =20
xuser_lists       NONE =20
start_proc_args   /usr/pkg/sge/pvm/startpvm.sh $pe_hostfile =20
$host /usr/pkg/pvm3 =20
stop_proc_args    /usr/pkg/sge/pvm/stoppvm.sh $pe_hostfile $host =20
allocation_rule   1 =20
control_slaves    FALSE =20
job_is_first_task TRUE =20
 =20
then I got the errors when I run the hello program: =20
********* =20
acacia: {36} more tester_loose.sh.pe481 =20
[pvmd pid25581] 06/16 14:21:08 readhostfile() iflist failed =20
startpvm: Couldn't get all of the 2 requested hosts =20
rm: /tmp/481.1.yucca.q/hostfile: No such file or directory =20
libpvm [pid27666] /tmp/pvmd.1024: No such file or directory =20
libpvm [pid27666] /tmp/pvmd.1024: No such file or directory =20
libpvm [pid27666]: pvm_halt(): Can't contact local daemon =20
******** =20
 =20
****** =20
acacia: {37} more tester_loose.sh.po481 =20
/usr/pkg/sge/default/spool/yucca/active_jobs/481.1/pe_hostfile =20
yucca.nmsu.edu /usr/pkg/pvm3 =20
/var/tmp/tmp.0.00025581aa =20
startpvm.sh: startup failed - invoking cleanup script =20
/usr/pkg/sge/default/spool/yucca/active_jobs/481.1/pe_hostfile yucca.nmsu=
.edu =20
/usr/pkg/sge/default/spool/yucca/active_jobs/481.1/pe_hostfile yucca.nmsu=
.edu =20
/usr/pkg/sge/default/spool/oenothera/active_jobs/481.1/pe_hostfile =20
oenothera.nmsu.edu /usr/pkg/pvm3 =20
/var/tmp/tmp.0.00026663aa =20
********/ =20
 =20
Does anyone know why pvm doesn't start. I've checked with SGE people and =
they =20
think the generated line "/var/tmp/tmp.0.00025581aa" shouldn't be there a=
nd =20
maybe PVM is compiled in a special way on NetBSD... But they seem not kno=
w how =20
to fix it. I highly appreciated if anyone could help me to fix that probl=
em. =20
Thank you very much,  =20
   =20
--    =20
Co Thai Ngo    =20
Dept. of Biology     =20
New Mexico State University     =20