From cwso@engin.umich.edu Wed Nov 20 11:11:59 1996 Received: from thuban.engin.umich.edu (cwso@thuban.engin.umich.edu [141.212.200.31]) by vishnu.hao.ucar.edu (8.7.3/8.7.3) with SMTP id LAA18719 for ; Wed, 20 Nov 1996 11:11:57 -0700 (MST) Received: (cwso@localhost) by thuban.engin.umich.edu (8.6.12/8.6.4) id NAA15012; Wed, 20 Nov 1996 13:11:54 -0500 Date: Wed, 20 Nov 1996 13:11:53 -0500 (EST) From: Chulwhan So To: foster@vishnu.hao.ucar.edu cc: cwso@engin.umich.edu Subject: Requesting help!!! Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: R Content-Length: 5162 Hello, let me introduce myself first briefly My name is Chulwhan So, grad. student at U. of Michigan working for Prof. Tim Killeen. Actually, he suggested me to study one of his former student, Wei Deng's thesis and his 1d Satellite Track Code which is based on Roble's 1d tgcm model to reproduce some of the results to prepare for near-future research. But I have very little computer experience with this kind of comprehensive model. My problem is that there is no one familiar with your code in the dept. now. Wei Deng suggested me to contact you to ask for help. Dr. Alan Burns helped me with some of the introductory materials but we still fail to run the code because of the problem in the submission file to CRAY (de_track1.sh);necessary files are missing including .tar files. I tried NCAR package but Wei Deng's account at NCAR now vanished, which contained some necessary .tar files He said he doesn't remember all the files in his NCAR account. Basically, those were some UNIX tar files which he got from youand he said those files are required to run Ray Roble's 1D TGCM model. We are kind of stuck. In fact, I don't know anything about the file contains. I think I really need your help to get everything on the right track. So, could you be kind enough to look over the file (de_track1.sh) that I send you here and to make some corrections or to instruct me with how to fix the mistakes in there and to send me back every needed information? Although the file has been changed slightly, I believe you could find what is wrong in very short time. So, please, do not ignore my request and help me, I have no one to turn to on this. Your help would be greatly appreciated. Thank you. ---Chulwhan So. ---de_track1.sh--- # # QSUB -s /bin/csh # QSUB -eo # QSUB -lT 800 # QSUB -q prem # QSUB -lM 3Mw # QSUB -nr # ja cd $TMPDIR # # glbmean model # (may be submitted via MASnet with netsh (aliased by shjob), # or via MIGS with nrnet) # set job = de_track # job name set rhome = /home/cedar/aburns # remote home set rsdir = "cedar.hao.ucar.edu:~aburns/$job" # remote dir with source and input set rtdir = "cedar.hao.ucar.edu:/d/aburns/$job" # remote temp dir to rcv metacode batchname $job.$$ # output file name (no path) # set rdstart = /ROBLE/GLBMEAN/GLBM03 # mss file to be read by restart set wrstart = /ABURNS/DE_TRACK/GLBM03 # mss file to be written by restart #set rdhist = /ROBLE/RGR91/MATRTS1 # hist to read from mss (optional) #set wrhist = /FOSTER/GLBMEAN/MATRTS2 # hist to write to mss (optional) # set host=gandalf.engin.umich.edu set login=tgcm set pw=tmodel. msread pltcva.o /EMERY/PLTCVA.O # plot lib from mss if ($?rdhist) then echo Doing msread of $rdhist ... msread -R ERADEX $rdhist # get read hist from mss assign -a ERADEX fort.7 # assign mss read hist to ft07 endif if ($?wrhist) assign -a ORADEX fort.12 # assign mss write hist to ft12 # echo Doing msread of $rdstart ... # msread $job.read $rdstart # get restart file from mss cp /u1/aburns/glbm03 $job.read assign -a $job.read fort.8 assign -a $job.write fort.9 assign -a vshdata.dat fort.11 assign -a geodat.dat fort.15 assign -a de1187.dat fort.14 assign -a lapi1187.dat fort.16 assign -a magb1187.dat fort.18 # rsh cedar.hao.ucar.edu "cd $rhome/$job; tarsrc" # tar source on sting #rcp $rtdir/$job.tar . # get src tar file from sting #tar xf $job.tar # untar source #msread $job.obj.tar /ABURNS/$job/obj.tar # get obj tar file from mss #tar xf $job.obj.tar # untar object code #checkincs # check includes ftp -n -v < $job.out || debug -s $job.ymp # execute rcp $job.out $rtdir/$job.out mswrite -t 367 $job.write $wrstart # write new restart file #rcp gmeta $rtdir/$job.cgm # return metacode to sting # ftp -n -v <