Dear Friend,
Thanks for all your help
I changed the parallel_option and now I am able to run parallel
I donot find words to thanks you and all others who helped me to solve the
problem
Regards and greetings
Raja
On Fri, Mar 22, 2013 at 11:18 AM, Lyudmila Dobysheva lyuk...@mail.ruwrote:
Dear developers and users,
Greetings!
Recently we installed recent Wien2k version with ubuntu OS in Dell server
It is running without any problem.
We have done the k-point parallelization.But it gives the following
error in lapw2.
we will be thankful to you if you can suggest how to overcome this
What is the error in lapw2?
The following command in a terminal while in the case directory might
give you more information on the cause of the error:
cat lapw2*.error
On 3/22/2013 12:03 AM, Mathrubutham Rajagopalan wrote:
Dear developers and users,
Greetings!
Recently we installed recent
Dear Dr Gavin
raja at ubuntu://home/raja/work/Cu$ cat lapw2* .error
2,'Cu.nsh','unknown','formatted',0
3,'Cu.in1', 'unknown','formatted',0
4,'Cu.inso', 'unknown','formatted',0
5,'Cu.in2', 'old','formatted',0
6,'Cu.output2','unknown','formatted',0
First, you entered the command incorrectly. There should be no space
between the '*' and '.error'. However, your lapw2 error files still
seem to be empty.
Second, did you compile Wien2k with -traceback in the Compiler options
line? If not, you need to add it, recompile, and run the scf
Dear Dr Gavin
Thanks for the response
The error file is empty
I included the -traceback option
current:FOPT:-FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML -traceback
current:FPOPT:-FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML -DFFTW3
-traceback
current:LDFLAGS:$(FOPT)
My guess:
Since this is such a short test calculation for a parallel run, the NFS
might not have written all necessary files ??
Does the error occur in a sequential run ?
or if you use only 2 parallel jobs ??
ls -als *errorwhich error files are not empty ??
cat *.error (without
Dear Dr Peter,
Thanks for the response
The error does not occur in sequential run
we have 8 processor But I used only four
I opened the following files All are empty
dstart.error
8 -rw-rw-r-- 1 raja raja 0 Mar 24 11:59 lapw0.error
8 -rw-rw-r-- 1 raja raja 0 Mar 24 11:59 lapw1_1.error
8
You do not even have lapw2*.error files (which should exist and be
empty).
What happens if you run:x lapw2 -pin a terminal ??
On 03/22/2013 08:35 AM, Mathrubutham Rajagopalan wrote:
Dear Dr Peter,
Thanks for the response
The error does not occur in sequential run
we have 8
Dear Dr Peter
This is the error message
raja at ubuntu://home/raja/work/Cu$ x lapw2 -p
running LAPW2 in parallel mode
FERMI - Error
cp: cannot stat `.in.tmp': No such file or directory
** LAPW2 crashed!
0.0u 0.0s 0:00.10 40.0% 0+0k 0+184io 0pf+0w
error: command /home/raja/wien2k/lapw2para
Now you have an error message !!
What is about ls-als *error and their contant ?
cat *.scf1 are there eigenvalues printed ???
cat *.in2 how does this file look like ?
On 03/22/2013 08:43 AM, Mathrubutham Rajagopalan wrote:
Dear Dr Peter
This is the error message
raja at
Your first post has:
lapw2 -p -c (11:23:12) running
I think you should have ran
x lapw2 -p -c
for a complex calculation
On 3/22/2013 1:50 AM, Peter Blaha wrote:
Now you have an error message !!
What is about ls-als *error and their contant ?
cat *.scf1 are there
Not really. The-c switch is not mandatory and will be added
automatically if inversion is not present in newer wien2k versions.
On 03/22/2013 08:54 AM, Gavin Abo wrote:
Your first post has:
lapw2 -p -c (11:23:12) running
I think you should have ran
x lapw2 -p -c
for a
22.03.2013 11:43, Mathrubutham Rajagopalan ?:
lapw1 -p -c (11:23:10) starting parallel lapw1 at Mon Mar 24 11:23:10 IST
2003
4 number_of_parallel_jobs ubuntu(11) ubuntu(11) ubuntu(11) ubuntu(11)
ubuntu(1)
ubuntu(1) ubuntu(1)Summary of lapw1para:
ubuntu k=11user=0
The order of your actions should be another:
1. Make a fresh directory.
2. Make in terminal the commands by hand
x lapw0
x lapw1 -c -p
x lapw2 -c -p
and send us the output in terminal and error files if nonzero
3. Check the output of lapw1: *.scf1 and *.output1
Best wishes
Lyudmila
Hello Dear
This is the output in terminal after lapw2
raja at ubuntu://home/raja/work/Cu$ x lapw2 -c -p
running LAPW2 in parallel mode
FERMI - Error
cp: cannot stat `.in.tmp': No such file or directory
** LAPW2 crashed!
0.0u 0.0s 0:00.11 45.4% 0+0k 96+184io 9pf+0w
error: command
I am waiting for your reply
Raja
Sent from my iPad
On Mar 22, 2013, at 5:21 AM, Lyudmila Dobysheva lyuka17 at mail.ru wrote:
The order of your actions should be another:
1. Make a fresh directory.
2. Make in terminal the commands by hand
x lapw0
x lapw1 -c -p
x lapw2 -c -p
and send us
I recommend checking for NaN's in your Cu.scf0 file.
This behaviour looks similar to what I've seen before in some cases when
using the broken FFT library in the early 12.1 release. It can be fixed
for the serial mode lapw0 by adding -DFFTW3 or -DFFTW2 to FOPT and
linking to an fftw in R_LIBS.
22.03.2013 13:38, Mathrubutham Rajagopalan wrote:
There is no scf1 and output1 files after running lapw1
This means that lapw1 already does not work.
So, again,
1. Check the file in directory /home/raja/wien2k/ parallel_options
there should be like:
setenv USE_REMOTE 0
setenv MPI_REMOTE 0
19 matches
Mail list logo
