Re: [Wien] System Error....Couldn't write to/.

[Gavin Abo]

That output file, case.outputd, as far as I know is not used an input to 
another calculation process. Thus, it seems safe to click the "initlapw 
without saving" button which would be another way to get around the 
"Couldn't write to /." error from "Save and initlapw" in this case.  
Although, if you are clicking the "initlapw without saving" or "Save and 
initlapw" button too quickly, you might have missed viewing the 
case.outputd to check if the condition gmax > gmin was meet.  Proceeding 
with a gmin value less then gmax could affect your results.


You didn't give the different file sizes of the case.outputd files for 
the BP, ZnSiP2 and other.  I suspect the file sizes are smaller than the 
case.outptud files produced for the C24H72N8In2Br14 and C16H20O7.


The TiC example in the usersguide [1] produces a case.outputd with a 
file size of about 21.8 kB.  That is a small unit cell and one of the 
calculation cases when the error is not observed.  The C16H20O7 has a 
relatively large unit cell [2] in comparison to that of TiC 
corresponding with the MB size case.outputd versus the kB one, respectively.


Whether you get the error "Couldn't write to /." or not with C16H20O7 
seems to also depend on the type of computer system(s) you are using 
(laptop, desktop, hpc? amount of RAM? SSD or HDD? how may CPU cores? 
using a virtual machine or not? Etc.).


For running calculation with WIEN2k, you have probably noticed how a 
single desktop computer is good for small unit cells but when moving to 
large structures such as slabs it can push researchers into using a hpc 
(e.g., the Roadrunner [3]).


At [4], you can probably see how there is quite a bit of difference in 
the computing ability and the computer resources between say a $1,700 
desktop and a $101,000,000 Roadrunner computer system.


[1] http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf
[2] https://doi.org/10.1007/s00894-022-05340-3
[3] https://en.wikipedia.org/wiki/Roadrunner_(supercomputer)
[4] 
https://insidehpc.com/2012/03/supercomputers-vs-your-computer-a-bang-for-the-buck-battle/ 




On 7/20/2023 10:23 PM, karina Khan wrote:
Thank you for your response. I want to ask you that if I complete the 
process by clicking continue initlapw without saving will it affect my 
results as doing so I didn't find any error in scf convergence.


One more thing this " System ErrorCouldn't write to/." didn't 
observe in other molecules like BP, ZnSiP2 and other.


On Fri, 21 Jul, 2023, 8:14 AM Gavin Abo,  wrote:

On an older, slower computer system, I was able to reproduce the
error displayed in w2web as:

*System Error*

Couldn't write to /.

_back to last page_

To produce the error, what I did in w2web was 1) I clicked the
"view case.outputd and cp case.in0_std case.in0" button, 2) paused
about 5 seconds, and 3) clicked the "Save and initlapw" button.

The file size of the case.outputd is 43.6 MB.  I suspect the error
results because my computer is slow in opening the case.outputd
file where I clicked on the save button before the file got opened up.

When I clicked "view case.klist" followed by clicking as fast as
possible "Save and initlapw", I do not encounter the same error. 
The file size of case.klist, however, is only 101 bytes.

I also 1) clicked the "view case.outputd and cp case.in0_std
case.in0" button, 2) paused about 1 minute, and 3) clicked the
"Save and initlapw" button.  When doing that, the error did not occur.

I put the 43.6 MB case.outputd on a new faster computer system
clicking the "Save and initlapw" button immediately after having
clicked the "view case.outputd and cp case.in0_std case.in0"
button and the error do not appear.

From that, it seems to be a computer system resource limitation
that is causing the error.

Kind Regards,

Gavin
WIEN2k user

On 7/14/2023 4:59 AM, karina Khan wrote:

Ok I will keep this thing in my mind for future
This command works well with different systems like Cs2AgInBr6,
BP but not with system having molecular formula C24H72N8In2Br14
and C16H20O7

On Thu, 13 Jul, 2023, 8:34 PM Peter Blaha,
 wrote:

I cannot reproduce this. It works perfectly fine with me.

Please try again in another case. Or use the recommended fast
mode. This will set automatically good parameters in
case.in0/1/2 and klist, and not the "stupid" defaults of the
step by step initialization.

PS: You cannot send large images to the mailing list. There
is a limit of 40 kB per message.


Am 13.07.2023 um 11:48 schrieb karina Khan:

Dear Wien2k users,
                       During initialisation, I stuck up in
the step *view case.outpd and cp case.in0_std case.in0,
*when I click on the step save and initlapw, I get a *System
Error couldn't write to/. * back to last page.

Re: [Wien] lapw0.def failed

[Laurence Marks]

They are all the same error, associated with fftw3.f03 not being found.

What you need to do is ls $WIENROOT/SRC_lapw0/fftw3.f03 (it is probably not
there, but just checking.)

Then look at the few lines above the first error (you don't need them all).
This will give you the compilation options. There should be something that
starts with a -I where your FFTW3 library include files are. This is
wrong on your system. To see what you currently have you can do
grep "-- fftw" $WIENROOT/SRC_lapw0/Makefile -A10

I cannot say exactly what is wrong as I dont know how you
configured fftw (not part of Wien2k). Something with them, which is also in
$WIENROOT/WIEN2k_OPTIONS for entries with FFTW in them


On Fri, Jul 21, 2023 at 9:53 AM Brik Hamida  wrote:

> hmd@hmd-host:~/wien2k23$ grep -i "error" */compile.msg | grep -v -e
> arguments -e xerror |grep -v know
>
> SRC_3ddens/compile.msg:fft_modules.F:174: Error: Can't open included file
> 'fftw3.f03'
>

-- 
Professor Laurence Marks (Laurie)
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
"Research is to see what everybody else has seen, and to think what nobody
else has thought", Albert Szent-Györgyi
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] lapw0.def failed

[Brik Hamida]

hmd@hmd-host:~/wien2k23$ grep -i "error" */compile.msg | grep -v -e
arguments -e xerror |grep -v know

SRC_3ddens/compile.msg:fft_modules.F:174: Error: Can't open included file
'fftw3.f03'
SRC_3ddens/compile.msg:make: *** [Makefile:84: fft_modules.o] Error 1
SRC_hf/compile.msg:modules_tmp_.F:108: Error: Can't open included file
'fftw3.f03'
SRC_hf/compile.msg:make[1]: *** [Makefile:281: modules.o] Error 1
SRC_hf/compile.msg:make: *** [Makefile:181: real] Error 2
SRC_hf/compile.msg:modules_tmp_.F:108: Error: Can't open included file
'fftw3.f03'
SRC_hf/compile.msg:make[1]: *** [Makefile:281: modules.o] Error 1
SRC_hf/compile.msg:make: *** [Makefile:185: complex] Error 2
SRC_lapw0/compile.msg:Fatal Error: Cannot open module file
‘xc_f03_lib_m.mod’ for reading at (1): No such file or directory
SRC_lapw0/compile.msg:make[1]: *** [Makefile:176: libxc_mod.o] Error 1
SRC_lapw0/compile.msg:make: *** [Makefile:125: seq] Error 2
SRC_lapw2/compile.msg:modules_tmp_.F:603: Error: Can't open included file
'fftw3.f03'
SRC_lapw2/compile.msg:make[1]: *** [Makefile:270: modules.o] Error 1
SRC_lapw2/compile.msg:make: *** [Makefile:169: real] Error 2
SRC_lapw2/compile.msg:modules_tmp_.F:603: Error: Can't open included file
'fftw3.f03'
SRC_lapw2/compile.msg:make[1]: *** [Makefile:270: modules.o] Error 1
SRC_lapw2/compile.msg:make: *** [Makefile:173: complex] Error 2
SRC_nlvdw/compile.msg:fft_modules.F:174: Error: Can't open included file
'fftw3.f03'
SRC_nlvdw/compile.msg:make[1]: *** [Makefile:159: fft_modules.o] Error 1
SRC_nlvdw/compile.msg:make: *** [Makefile:110: seq] Error 2
SRC_nmr/compile.msg:modules_tmp_.F:1345: Error: Can't open included file
'fftw3.f03'
SRC_nmr/compile.msg:make[1]: *** [Makefile:263: modules.o] Error 1
SRC_nmr/compile.msg:make: *** [Makefile:163: real] Error 2
SRC_nmr/compile.msg:modules_tmp_.F:1345: Error: Can't open included file
'fftw3.f03'
SRC_nmr/compile.msg:make[1]: *** [Makefile:263: modules.o] Error 1
SRC_nmr/compile.msg:make: *** [Makefile:167: complex] Error 2
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] lapw0.def failed

[Gavin Abo]

If you're building fftw3 from source with gfortran and are following my 
steps, the fftw3.f03 is installed in the ~/fftw-3.3.10/include directory 
as seen in step 4 of [1].


If you're using ifort instead of gfortran, fftw3 can also be built from 
source and the fftw3.f03 can be installed in the ~/fftw-3.3.10/include 
directory as seen in step 2 of [2].


If you are instead using fftw3 from the Ubuntu 22.04.2 LTS package, 
which likely only works when using gfortran, the library header file 
fftw3.f03 gets installed in the /usr/include/ directory as seen below:


username@computername:~/Desktop$ lsb_release -a
No LSB modules are available.
Distributor ID:    Ubuntu
Description:    Ubuntu 22.04.2 LTS
Release:    22.04
Codename:    jammy
username@computername:~/Desktop$ ls /usr/include/fftw3*
ls: cannot access '/usr/include/fftw3*': No such file or directory
username@computername:~/Desktop$ sudo apt install libfftw3-dev
...
username@computername:~/Desktop$ ls /usr/include/fftw3*
/usr/include/fftw3.f    /usr/include/fftw3.h /usr/include/fftw3q.f03
/usr/include/fftw3.f03  /usr/include/fftw3l.f03

Of note, the fftw3 library files (libfftw3.a and libfftw3.so), from the 
Ubuntu 22.04.2 LTS package get installed in the 
/usr/lib/x86_64-linux-gnu/ directory:


username@computername:~/Desktop$ ls /usr/lib/x86_64-linux-gnu/libfftw3.*
/usr/lib/x86_64-linux-gnu/libfftw3.a
/usr/lib/x86_64-linux-gnu/libfftw3.so
/usr/lib/x86_64-linux-gnu/libfftw3.so.3
/usr/lib/x86_64-linux-gnu/libfftw3.so.3.5.8

Depending on the install approach that you use for fftw3, you have to 
adjust the FFTW options and FFTW-LIBS in siteconfig appropriately.


[1] 
https://github.com/gsabo/WIEN2k-Docs/blob/main/WIEN2k23.2_Ubuntu22.04_Install_with_gfortran.pdf
[2] 
https://github.com/gsabo/WIEN2k-Docs/blob/main/WIEN2k23.2_Ubuntu22.04_Install_with_OneAPI(ifort).pdf


Kind Regards,

Gavin
WIEN2k user

On 7/21/2023 9:31 AM, Laurence Marks wrote:

They are all the same error, associated with fftw3.f03 not being found.

What you need to do is ls $WIENROOT/SRC_lapw0/fftw3.f03 (it is 
probably not there, but just checking.)


Then look at the few lines above the first error (you don't need them 
all). This will give you the compilation options. There should be 
something that starts with a -I where your FFTW3 library include 
files are. This is wrong on your system. To see what you currently 
have you can do

grep "-- fftw" $WIENROOT/SRC_lapw0/Makefile -A10

I cannot say exactly what is wrong as I dont know how you 
configured fftw (not part of Wien2k). Something with them, which is 
also in $WIENROOT/WIEN2k_OPTIONS for entries with FFTW in them



On Fri, Jul 21, 2023 at 9:53 AM Brik Hamida  wrote:

hmd@hmd-host:~/wien2k23$ grep -i "error" */compile.msg | grep -v
-e arguments -e xerror |grep -v know

SRC_3ddens/compile.msg:fft_modules.F:174: Error: Can't open
included file 'fftw3.f03'


--
Professor Laurence Marks (Laurie)
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 
"Research is to see what everybody else has seen, and to think what 
nobody else has thought", Albert Szent-Györgyi___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] calculation with lmbj potential

[Burhan Ahmed]

Dear experts, I am doing an scf calculation taking lmbj potential. My system is a slab of 6ql with a vacuum of 40 ang along c-axix. Whenever I try to run the scf calculation including SOC, after 999 iteration the scf is still not converged. I have analyzed the scf file, it shows the fluctuating nature. At first the selected Rmt was 2.5 then I reduces it to 2.34 and again convergence failed after 999 cycle. I have included -hdlo and -lvns 8 switch in init_lapw as it shows heavy atom. I am attaching the case.struct file. Hoping for any suggestion/solution.  RegardsBurhan AhmedResearch Scholar, AUS  

bi2te3lmbj.struct
Description: Binary data
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html