@duy lee<br>i inserted the line #$ -V in my qsub script and the env variable $LD_LIBRARY_PATH is being set from script... thanks for that but still problem persists.<br><br>@rakshit<br><br>-----------------------------<br>
$find /usr/lib -name libgfortran* <br>-----------------------------<br>and output is <br>_______________________<br>/usr/lib/libgfortran.so.1.0.0<br>/usr/lib/libgfortran.so.1<br>/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/libgfortran.a<br>
/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/libgfortranbegin.a<br>/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/libgfortran.so<br>/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/32/libgfortran.a<br>/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/32/libgfortranbegin.a<br>
/usr/lib/gcc/x86_64-redhat-linux6E/4.3.2/32/libgfortran.so<br>_______________________<br>whereas on doing<br>-----------------------------<br>$find /usr/lib64 -name libgfortran*<br>-----------------------------<br>ouput is <br>
_________________<br>/usr/lib64/libgfortran.so.3.0.0<br>/usr/lib64/libgfortran.so.1.0.0<br>/usr/lib64/libgfortran.so.1<br>find: /usr/lib64/audit: Permission denied<br><b>/usr/lib64/libgfortran.so.3</b><br>_________________<br>
<br>it means my OS has the required library, i guess<br><br>i installed the QE on the head node<br>This cluster is a Rocks cluster with<br>
<br>
# of nodes: 10 (1 head node + 9 compute nodes)<br>
# of processors/node: 8<br>
# Total # of processors: 10X8 = 80<br><br>i tested the sample program for submitting batch jobs using SGE utility and it is working fine( <a href="http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html">http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html</a> )<br>
<br>sincerely,<br><div class="gmail_quote">sreekar guddeti<br>Dept. Physics<br>IIT Bombay<br>India<br><br><br>On Sun, Sep 6, 2009 at 12:28 PM, <span dir="ltr"><<a href="mailto:pw_forum-request@pwscf.org" target="_blank">pw_forum-request@pwscf.org</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Send Pw_forum mailing list submissions to<br>
<a href="mailto:pw_forum@pwscf.org" target="_blank">pw_forum@pwscf.org</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:pw_forum-request@pwscf.org" target="_blank">pw_forum-request@pwscf.org</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:pw_forum-owner@pwscf.org" target="_blank">pw_forum-owner@pwscf.org</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Pw_forum digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: error loading shared libraries on parallel execution (Duy Le)<br>
2. pw.x running but nothing happens (wangqj1)<br>
3. Re: error loading shared libraries on parallel execution<br>
(Bipul Rakshit)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Sat, 5 Sep 2009 19:39:17 -0400<br>
From: Duy Le <<a href="mailto:ttduyle@gmail.com" target="_blank">ttduyle@gmail.com</a>><br>
Subject: Re: [Pw_forum] error loading shared libraries on parallel<br>
execution<br>
To: PWSCF Forum <<a href="mailto:pw_forum@pwscf.org" target="_blank">pw_forum@pwscf.org</a>><br>
Message-ID:<br>
<<a href="mailto:8974d3b20909051639u575aed19xdf474c53f9a7d877@mail.gmail.com" target="_blank">8974d3b20909051639u575aed19xdf474c53f9a7d877@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<br>
Hi,I am not sure if this help. Could you please try to add<br>
#$ -V in your submitting script. Like this:<br>
<br>
#!/bin/bash<br>
#<br>
#$ -V<br>
#$ -cwd<br>
#$ -j y<br>
#$ -S /bin/bash<br>
#<br>
<br>
Good luck.<br>
D.<br>
<br>
On Sat, Sep 5, 2009 at 7:20 PM, sreekar guddeti<br>
<<a href="mailto:colonel.sreekar@gmail.com" target="_blank">colonel.sreekar@gmail.com</a>>wrote:<br>
<br>
> i know this issuehas been addressed and documented in troubleshooting<br>
> section of the users guide.<br>
> but i giveup in despair trying for a whole day to figure this problem<br>
> i run my jobs on rocks cluster by using SGE's facility of submitting batch<br>
> jobs<br>
><br>
> <a href="http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html" target="_blank">http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html</a><br>
><br>
> what i 'apparently' observe(or doubtfully infer) is that i can successfully<br>
> run a single parallel job, but on submitting a second job i get the error<br>
> ____________________________________________<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> ____________________________________________<br>
><br>
><br>
> i find out the path for the library and added to the LD_LIBRARY_PATH by<br>
> writing<br>
> _______________________________________<br>
> #set the library path to include gfortran libraries<br>
> export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64:/usr/lib<br>
><br>
><br>
> $ECHO<br>
> $ECHO "$LD_LIBRARY_PATH"<br>
> $ECHO<br>
> _______________________________________<br>
><br>
> in the file which acts as script for qsub ($qsub -pe orte 4<br>
> dosroutine.qsub) which is<br>
><br>
> dosroutine.qsub<br>
> -----------------------------------------------------------------<br>
><br>
> #!/bin/bash<br>
> #<br>
> #$ -cwd<br>
> #$ -j y<br>
> #$ -S /bin/bash<br>
> #<br>
><br>
> #extract the info about no of processors involved from command line<br>
> arguments of 'qsub'<br>
> PROCESSORS=$NSLOTS<br>
><br>
> #heuristically assign the no of processors per pool NPR<br>
> NPR=4<br>
> #as a result no of pools are give by<br>
> NPK=`expr $PROCESSORS / $NPR`<br>
><br>
> #!/bin/bash<br>
> #<br>
> #<br>
> #Script for performing a dos calculation on a parallel processor<br>
> WORKINGDIR=`pwd`<br>
> ECHO="echo"<br>
><br>
> #set the library path to include gfortran libraries<br>
> export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64:/usr/lib<br>
><br>
><br>
> $ECHO<br>
> $ECHO "$LD_LIBRARY_PATH"<br>
> $ECHO<br>
><br>
> # set the needed environment variables<br>
><br>
> PREFIX=`cd /home3/colonel/espresso-4.0.5 ; pwd`<br>
> $ECHO $PREFIX<br>
> BIN_DIR=$PREFIX/bin<br>
> PSEUDO_DIR=$PREFIX/pseudo<br>
> TMP_DIR=$HOME/tmp<br>
> PARA_PREFIX="/opt/openmpi/bin/mpirun -np $PROCESSORS"<br>
> PARA_POSTFIX="-npool $NPK"<br>
><br>
> # required executables and pseudopotentials<br>
> $ECHO<br>
> $ECHO " executables directory: $BIN_DIR"<br>
> $ECHO " pseudo directory: $PSEUDO_DIR"<br>
> $ECHO " temporary directory: $TMP_DIR"<br>
><br>
> #create results directory<br>
> for DIR in "$TMP_DIR" "$WORKINGDIR/results" ; do<br>
> if test ! -d $DIR ; then<br>
> mkdir $DIR<br>
> fi<br>
> done<br>
> cd $WORKINGDIR/results<br>
><br>
><br>
> # variables to represent programs<br>
> PW_COMMAND="$PARA_PREFIX $BIN_DIR/pw.x $PARA_POSTFIX"<br>
> DOS_COMMAND="$PARA_PREFIX $BIN_DIR/dos.x $PARA_POSTFIX"<br>
> PROJWFC_COMMAND="$PARA_PREFIX $BIN_DIR/projwfc.x $PARA_POSTFIX"<br>
><br>
><br>
> # DOS calculation for 0Ni0<br>
> cat > <a href="http://0ni0.dos.in" target="_blank">0ni0.dos.in</a> << EOF<br>
> &control<br>
> calculation='nscf'<br>
> restart_mode='from_scratch',<br>
> prefix='0ni0',<br>
> pseudo_dir = '$PSEUDO_DIR/',<br>
> outdir='$TMP_DIR/'<br>
> /<br>
> &system<br>
> ibrav=2, celldm(1) =6.48, nat=1, ntyp=1,<br>
> nspin = 2, starting_magnetization(1)=0.7,<br>
> ecutwfc = 24.0, ecutrho = 288.0, nbnd=8,<br>
> occupations='tetrahedra'<br>
> /<br>
> &electrons<br>
> conv_thr = 1.0e-10<br>
> mixing_beta = 0.7<br>
> /<br>
> ATOMIC_SPECIES<br>
> Ni 58.69 NiUS.RRKJ3.UPF<br>
> ATOMIC_POSITIONS<br>
> Ni 0.0 0.0 0.0<br>
> K_POINTS {automatic} !special points generated by tetrahedra method<br>
> 12 12 12 0 0 0<br>
> EOF<br>
><br>
> $ECHO " running DOS calculation for 0Ni0 ...\c"<br>
> $PW_COMMAND < <a href="http://0ni0.dos.in" target="_blank">0ni0.dos.in</a> > 0ni0.dos.out<br>
> $ECHO<br>
> $ECHO " done"<br>
><br>
> -------------------------------------------------------------<br>
><br>
> the output i get is<br>
><br>
> ************************************************<br>
> :/usr/lib64:/usr/lib<br>
><br>
> /home3/colonel/espresso-4.0.5<br>
><br>
> executables directory: /home3/colonel/espresso-4.0.5/bin<br>
> pseudo directory: /home3/colonel/espresso-4.0.5/pseudo<br>
> temporary directory: /home3/colonel/tmp<br>
> running DOS calculation for 0Ni0 ...\c<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
><br>
> done<br>
> *************************************************<br>
> plz suggest solutions which donot require root permissions , as i dont have<br>
> thanks in advance<br>
> --<br>
> Sreekar Guddeti<br>
> Department of Physics<br>
><br>
> _______________________________________________<br>
> Pw_forum mailing list<br>
> <a href="mailto:Pw_forum@pwscf.org" target="_blank">Pw_forum@pwscf.org</a><br>
> <a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
><br>
><br>
<br>
<br>
--<br>
--------------------------------------------------<br>
Duy Le<br>
PhD Student<br>
Department of Physics<br>
University of Central Florida.<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://www.democritos.it/pipermail/pw_forum/attachments/20090905/010fb1e3/attachment-0001.htm" target="_blank">http://www.democritos.it/pipermail/pw_forum/attachments/20090905/010fb1e3/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Sun, 6 Sep 2009 08:33:25 +0800 (CST)<br>
From: wangqj1 <<a href="mailto:wangqj1@126.com" target="_blank">wangqj1@126.com</a>><br>
Subject: [Pw_forum] pw.x running but nothing happens<br>
To: pw_forum <<a href="mailto:pw_forum@pwscf.org" target="_blank">pw_forum@pwscf.org</a>><br>
Message-ID:<br>
<<a href="mailto:25470012.218281252197205181.JavaMail.coremail@bj126app52.126.com" target="_blank">25470012.218281252197205181.JavaMail.coremail@bj126app52.126.com</a>><br>
Content-Type: text/plain; charset="gbk"<br>
<br>
<br>
Dear pwscf users<br>
When I run vc-relax on the computing cluster use one node which has 8 CPUs.<br>
The output file is as following:<br>
<br>
Program PWSCF v.4.0.1 starts ...<br>
Today is 6Sep2009 at 7:49:30<br>
Parallel version (MPI)<br>
Number of processors in use: 8<br>
R & G space division: proc/pool = 8<br>
For Norm-Conserving or Ultrasoft (Vanderbilt) Pseudopotentials or PAW<br>
.....................................................................<br>
Initial potential from superposition of free atoms<br>
starting charge 435.99565, renormalised to 436.00000<br>
Starting wfc are 254 atomic + 8 random wfc<br>
<br>
After one day ,it still like this and no iteration has completed ,there is also no error was turn up .There is no error in the input file because I have test it on anthoer computer which has 4 CPUs and it runs well .<br>
I can't find the reason about this ,any help will be appreciated .<br>
Best Regards<br>
Q.J.Wang<br>
XiangTan University<br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://www.democritos.it/pipermail/pw_forum/attachments/20090906/a4269bb4/attachment-0001.htm" target="_blank">http://www.democritos.it/pipermail/pw_forum/attachments/20090906/a4269bb4/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
Message: 3<br>
Date: Sun, 6 Sep 2009 07:58:49 +0100<br>
From: Bipul Rakshit <<a href="mailto:bipulrr@gmail.com" target="_blank">bipulrr@gmail.com</a>><br>
Subject: Re: [Pw_forum] error loading shared libraries on parallel<br>
execution<br>
To: PWSCF Forum <<a href="mailto:pw_forum@pwscf.org" target="_blank">pw_forum@pwscf.org</a>><br>
Message-ID: <<a href="mailto:3a749910909052358m6f23c3sf5bdbba1b595d8d5@mail.gmail.com" target="_blank">3a749910909052358m6f23c3sf5bdbba1b595d8d5@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<br>
hi,<br>
Just from root user type<br>
yum install libgfortran.so.3<br>
<br>
then it will install this files which is not present in your machine<br>
<br>
On Sun, Sep 6, 2009 at 12:20 AM, sreekar guddeti<br>
<<a href="mailto:colonel.sreekar@gmail.com" target="_blank">colonel.sreekar@gmail.com</a>>wrote:<br>
<br>
> i know this issuehas been addressed and documented in troubleshooting<br>
> section of the users guide.<br>
> but i giveup in despair trying for a whole day to figure this problem<br>
> i run my jobs on rocks cluster by using SGE's facility of submitting batch<br>
> jobs<br>
><br>
> <a href="http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html" target="_blank">http://www.rocksclusters.org/roll-documentation/sge/5.2/submitting-batch-jobs.html</a><br>
><br>
> what i 'apparently' observe(or doubtfully infer) is that i can successfully<br>
> run a single parallel job, but on submitting a second job i get the error<br>
> ____________________________________________<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> ____________________________________________<br>
><br>
><br>
> i find out the path for the library and added to the LD_LIBRARY_PATH by<br>
> writing<br>
> _______________________________________<br>
> #set the library path to include gfortran libraries<br>
> export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64:/usr/lib<br>
><br>
><br>
> $ECHO<br>
> $ECHO "$LD_LIBRARY_PATH"<br>
> $ECHO<br>
> _______________________________________<br>
><br>
> in the file which acts as script for qsub ($qsub -pe orte 4<br>
> dosroutine.qsub) which is<br>
><br>
> dosroutine.qsub<br>
> -----------------------------------------------------------------<br>
><br>
> #!/bin/bash<br>
> #<br>
> #$ -cwd<br>
> #$ -j y<br>
> #$ -S /bin/bash<br>
> #<br>
><br>
> #extract the info about no of processors involved from command line<br>
> arguments of 'qsub'<br>
> PROCESSORS=$NSLOTS<br>
><br>
> #heuristically assign the no of processors per pool NPR<br>
> NPR=4<br>
> #as a result no of pools are give by<br>
> NPK=`expr $PROCESSORS / $NPR`<br>
><br>
> #!/bin/bash<br>
> #<br>
> #<br>
> #Script for performing a dos calculation on a parallel processor<br>
> WORKINGDIR=`pwd`<br>
> ECHO="echo"<br>
><br>
> #set the library path to include gfortran libraries<br>
> export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64:/usr/lib<br>
><br>
><br>
> $ECHO<br>
> $ECHO "$LD_LIBRARY_PATH"<br>
> $ECHO<br>
><br>
> # set the needed environment variables<br>
><br>
> PREFIX=`cd /home3/colonel/espresso-4.0.5 ; pwd`<br>
> $ECHO $PREFIX<br>
> BIN_DIR=$PREFIX/bin<br>
> PSEUDO_DIR=$PREFIX/pseudo<br>
> TMP_DIR=$HOME/tmp<br>
> PARA_PREFIX="/opt/openmpi/bin/mpirun -np $PROCESSORS"<br>
> PARA_POSTFIX="-npool $NPK"<br>
><br>
> # required executables and pseudopotentials<br>
> $ECHO<br>
> $ECHO " executables directory: $BIN_DIR"<br>
> $ECHO " pseudo directory: $PSEUDO_DIR"<br>
> $ECHO " temporary directory: $TMP_DIR"<br>
><br>
> #create results directory<br>
> for DIR in "$TMP_DIR" "$WORKINGDIR/results" ; do<br>
> if test ! -d $DIR ; then<br>
> mkdir $DIR<br>
> fi<br>
> done<br>
> cd $WORKINGDIR/results<br>
><br>
><br>
> # variables to represent programs<br>
> PW_COMMAND="$PARA_PREFIX $BIN_DIR/pw.x $PARA_POSTFIX"<br>
> DOS_COMMAND="$PARA_PREFIX $BIN_DIR/dos.x $PARA_POSTFIX"<br>
> PROJWFC_COMMAND="$PARA_PREFIX $BIN_DIR/projwfc.x $PARA_POSTFIX"<br>
><br>
><br>
> # DOS calculation for 0Ni0<br>
> cat > <a href="http://0ni0.dos.in" target="_blank">0ni0.dos.in</a> << EOF<br>
> &control<br>
> calculation='nscf'<br>
> restart_mode='from_scratch',<br>
> prefix='0ni0',<br>
> pseudo_dir = '$PSEUDO_DIR/',<br>
> outdir='$TMP_DIR/'<br>
> /<br>
> &system<br>
> ibrav=2, celldm(1) =6.48, nat=1, ntyp=1,<br>
> nspin = 2, starting_magnetization(1)=0.7,<br>
> ecutwfc = 24.0, ecutrho = 288.0, nbnd=8,<br>
> occupations='tetrahedra'<br>
> /<br>
> &electrons<br>
> conv_thr = 1.0e-10<br>
> mixing_beta = 0.7<br>
> /<br>
> ATOMIC_SPECIES<br>
> Ni 58.69 NiUS.RRKJ3.UPF<br>
> ATOMIC_POSITIONS<br>
> Ni 0.0 0.0 0.0<br>
> K_POINTS {automatic} !special points generated by tetrahedra method<br>
> 12 12 12 0 0 0<br>
> EOF<br>
><br>
> $ECHO " running DOS calculation for 0Ni0 ...\c"<br>
> $PW_COMMAND < <a href="http://0ni0.dos.in" target="_blank">0ni0.dos.in</a> > 0ni0.dos.out<br>
> $ECHO<br>
> $ECHO " done"<br>
><br>
> -------------------------------------------------------------<br>
><br>
> the output i get is<br>
><br>
> ************************************************<br>
> :/usr/lib64:/usr/lib<br>
><br>
> /home3/colonel/espresso-4.0.5<br>
><br>
> executables directory: /home3/colonel/espresso-4.0.5/bin<br>
> pseudo directory: /home3/colonel/espresso-4.0.5/pseudo<br>
> temporary directory: /home3/colonel/tmp<br>
> running DOS calculation for 0Ni0 ...\c<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
> /home3/colonel/espresso-4.0.5/bin/pw.x: error while loading shared<br>
> libraries: libgfortran.so.3: cannot open shared object file: No such<br>
> file or directory<br>
><br>
> done<br>
> *************************************************<br>
> plz suggest solutions which donot require root permissions , as i dont have<br>
> thanks in advance<br>
> --<br>
> Sreekar Guddeti<br>
> Department of Physics<br>
><br>
> _______________________________________________<br>
> Pw_forum mailing list<br>
> <a href="mailto:Pw_forum@pwscf.org" target="_blank">Pw_forum@pwscf.org</a><br>
> <a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
><br>
><br>
<br>
<br>
--<br>
Dr. Bipul Rakshit<br>
Research Associate,<br>
S N Bose Centre for Basic Sciences,<br>
Salt Lake,<br>
Kolkata 700 098<br>
India<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://www.democritos.it/pipermail/pw_forum/attachments/20090906/d919f229/attachment.htm" target="_blank">http://www.democritos.it/pipermail/pw_forum/attachments/20090906/d919f229/attachment.htm</a><br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
Pw_forum mailing list<br>
<a href="mailto:Pw_forum@pwscf.org" target="_blank">Pw_forum@pwscf.org</a><br>
<a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
<br>
<br>
End of Pw_forum Digest, Vol 27, Issue 23<br>
****************************************<br>
</blockquote></div><br><br clear="all"><br>-- <br>Sreekar Guddeti<br><br>