[Wien] error in mixer

Sat May 20 15:27:06 CEST 2006

This sounds like it may be a problem with your c++ library (used by 
fortran for I/O) and/or a problem with file names or permissions. If you 
look in mixer.f you will see that the error you are getting is because 
C.inc cannot be opened, so this has (probably) nothing to do with the 
iterations. Some suggestions.

a) Check permissions on C.inc (ls -l C.inc)
b) Do "cp C.inc Mytest". This will check for control characters embedded 
in the name which sometime don't show up otherwise. If this fails, it 
means that you have something after the name. If this is the case, did 
you use the web interface or init_lapw? If the web, try init_lapw.
c) Change your base name. Try (with a bash shell)
 	cp -r C Test
 	cd Test
 	ls C.* > tt; sed 's/C/mv C/' tt > t1 ; sed 's/C/Test/' tt > t2
 	paste t1 t2 > t3 ; sh t3
Then run again. Note that Test.inc has 8 bytes which is often better.
c) Check mixer.def to see if C.inc (or Test.inc) appears more than once. 
This would also give the problem.
e) In mixer.f, change the line
          OPEN (IUNIT,FILE=FNAME,STATUS=STATUS,FORM=FORM,ERR=920)
to
          OPEN (IUNIT,FILE=FNAME,STATUS=STATUS,FORM=FORM,ERR=920,IOSTAT=IOTEST)
then, after
   920 INFO = 2
!
!        file FNAME couldn't be opened
!
add
 	write(*,*)'I/O Error code ',IOTEST

This won't cure the problem, but might tell you more about it.

On Fri, 19 May 2006, Saeid Jalali wrote:

> Dear Sahu,
>  Did you comaper the *inc in both of your runsp and run calculations?
>  A suggestion:
>  You would run a more simple example like fcc Fe using both run_lapw and runsp_lapw to check whether the same error occurs only for the runsp or not. If yes, most likely something is wrong with your compilation. If not, something is wron with your initialization.
>  Your,
>  S. Jalali.
>
> brsahu at physics.utexas.edu wrote:
>  Hi,
>
> I compiled the latest source release 18.05.2006 on IBM P5 with xlf90
> xlc and mpxlf90 compilers:
>
> The libraries and linker options used were:
>
> Current settings:
> O Compiler options: -q64 -qarch=pwr5 -qtune=pwr5
> -qcache=auto -qstrict
> L Linker Flags: -q64 -L../SRC_lib -bmaxdata:0x80000000
> P Preprocessor flags '-WF,-DParallel'
> R R_LIB (LAPACK+BLAS): -lessl_r
> /usr/local/apps/lapack/lib64/liblapack.a
> /usr/local/apps/goto_blas/lib/libgoto_power5-r1.00.a
>
> The struct file used was:
>
> Graphite
> H LATTICE,NONEQUIV.ATOMS: 2194_P63/mmc
> MODE OF CALC= unit=bohr
> 4.648728 4.648728 14.550897 90.000000 90.000000120.000000
> ATOM -1: X=0.00000000 Y=0.00000000 Z=0.25000000
> MULT= 2 ISPLIT= 4
> -1: X=0.00000000 Y=0.00000000 Z=0.75000000
> C 1 NPT= 781 R0=0.00010000 RMT= 1.3400 Z: 6.0
> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
> 0.0000000 1.0000000 0.0000000
> 0.0000000 0.0000000 1.0000000
> ATOM -2: X=0.33333333 Y=0.66666667 Z=0.25000000
> MULT= 2 ISPLIT= 4
> -2: X=0.66666667 Y=0.33333333 Z=0.75000000
> C 2 NPT= 781 R0=0.00010000 RMT= 1.3400 Z: 6.0
> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
> 0.0000000 1.0000000 0.0000000
> 0.0000000 0.0000000 1.0000000
>
> (Saied's struct file has the z-cordinates of the two inequivalent
> atoms reversed
> but that should not matter)
>
> I get the error with and without reversing the z-coordinates of struct
> file in the first cycle in the MIXER step for a 4 processor
> coarse-grained parallel run:
>
>
> Calculating C in /gpfs/utexas/ee/eefz340/wien2k/C
> on champ01 with PID 119182C.broyd* files removed !
>
>
> start (Fri May 19 13:20:22 CDT 2006) with lapw0 (20/40 to go)
>
> cycle 1 (Fri May 19 13:20:22 CDT 2006) (20/40 to go)
>
>> lapw0 -p (13:20:22) starting parallel lapw0 at Fri May 19
>> 13:20:22 CDT 2006
> --------
> running lapw0 in single mode
> 4.6u 0.0s 0:04 98% 4+120k 0+0io 0pf+0w
>> lapw1 -up -p (13:20:26) starting parallel lapw1 at Fri May
>> 19 13:20:26 CDT 2006
> -> starting parallel LAPW1 jobs at Fri May 19 13:20:26 CDT 2006
> running LAPW1 in parallel mode (using .machines)
> 4 number_of_parallel_jobs
> champion1(5) 29.0u 0.1s 0:29 99% 4+939k 0+0io 0pf+0w
> champion1(5) 30.0u 0.1s 0:30 99% 4+947k 0+0io 0pf+0w
> champion1(5) 30.0u 0.1s 0:30 99% 4+949k 0+0io 0pf+0w
> champion1(5) 30.6u 0.1s 0:30 99% 4+956k 0+0io 0pf+0w
> Summary of lapw1para:
> champion1 k=20 user=119.6 wallclock=396
> 119.9u 0.8s 0:35 340% 4+943k 0+0io 1pf+0w
>> lapw1 -dn -p (13:21:02) starting parallel lapw1 at Fri May
>> 19 13:21:02 CDT 2006
> -> starting parallel LAPW1 jobs at Fri May 19 13:21:02 CDT 2006
> running LAPW1 in parallel mode (using .machines.help)
> 4 number_of_parallel_jobs
> champion1(5) 28.9u 0.1s 0:29 99% 4+937k 0+0io 0pf+0w
> champion1(5) 29.9u 0.1s 0:30 98% 4+956k 0+0io 0pf+0w
> champion1(5) 30.0u 0.1s 0:30 99% 4+948k 0+0io 0pf+0w
> champion1(5) 30.6u 0.1s 0:31 98% 4+962k 0+0io 0pf+0w
> Summary of lapw1para:
> champion1 k=20 user=119.4 wallclock=394
> 119.7u 0.8s 0:35 338% 4+946k 0+0io 0pf+0w
>> lapw2 -up -p (13:21:38) running LAPW2 in parallel mode
> champion1 2.5u 0.0s 0:02 99% 5+60k 0+0io 0pf+0w
> champion1 2.5u 0.0s 0:02 99% 5+59k 0+0io 0pf+0w
> champion1 2.5u 0.0s 0:02 97% 5+60k 0+0io 0pf+0w
> champion1 2.6u 0.0s 0:02 97% 5+58k 0+0io 0pf+0w
> Summary of lapw2para:
> champion1 user=10.1 wallclock=392
> 10.8u 0.3s 0:07 141% 5+55k 0+0io 0pf+0w
>> lapw2 -dn -p (13:21:46) running LAPW2 in parallel mode
> champion1 2.5u 0.0s 0:02 99% 5+59k 0+0io 0pf+0w
> champion1 2.4u 0.0s 0:02 98% 5+61k 0+0io 0pf+0w
> champion1 2.5u 0.0s 0:02 98% 5+59k 0+0io 0pf+0w
> champion1 2.3u 0.0s 0:02 98% 5+55k 0+0io 0pf+0w
> Summary of lapw2para:
> champion1 user=9.7 wallclock=393
> 10.4u 0.3s 0:07 144% 5+55k 0+0io 0pf+0w
>> lcore -up (13:21:53) 0.0u 0.0s 0:00 28% 0+5k 0+0io 0pf+0w
>> lcore -dn (13:21:53) 0.0u 0.0s 0:00 28% 0+0k 0+0io 0pf+0w
>> mixer (13:21:54) 0.0u 0.0s 0:00 45% 0+0k 0+0io 0pf+0w
>
>> stop error
>
>
>
> The mixer error file is
>
> 'MIXER' - can't open unit: 7
> 'MIXER' - filename: C.inc
>
> I checked the working directory. The file C.inc exists and is not empty.
>
>
> The spin-polarized graphite is not appropriate but it is only for test.
>
> I used 1 processor (serial run) but I get the same error in the first
> cycle and again in the MIXER step:
>
> Calculating C in /gpfs/utexas/ee/eefz340/wien2k/C
> on champ01 with PID 156606
>
> start (Fri May 19 13:45:32 CDT 2006) with lapw0 (20/40 to go)
>
> cycle 1 (Fri May 19 13:45:32 CDT 2006) (20/40 to go)
>
>> lapw0 (13:45:32) 4.5u 0.0s 0:04 93% 3+118k 0+0io 61pf+0w
>> lapw1 -up (13:45:37) 57.9u 0.3s 1:00 96% 4+725k 0+0io 54pf+0w
>> lapw1 -dn (13:46:37) 57.6u 0.3s 0:58 98% 4+722k 0+0io 0pf+0w
>> lapw2 -up (13:47:37) 7.6u 0.0s 0:09 83% 4+64k 0+0io 64pf+0w
>> lapw2 -dn (13:47:47) 7.4u 0.0s 0:08 89% 5+62k 0+0io 0pf+0w
>> lcore -up (13:47:56) 0.0u 0.0s 0:00 8% 0+0k 0+0io 16pf+0w
>> lcore -dn (13:47:57) 0.0u 0.0s 0:00 16% 1+17k 0+0io 0pf+0w
>> mixer (13:47:58) 0.0u 0.0s 0:00 8% 0+55k 0+0io 19pf+0w
>
>> stop error
>
> Pl. note I do not get the MIXER error and the cycles proceed for
> non-spinpolarized runs both parallel and serial.
>
>
> What could be wrong?
>
> Sahu
>
> Quoting Saeid Jalali :
>
>> The two C atoms are inequivalent that could be labeled as 1 and 2 as
>> follows. The following structure file does run with no error
>> within runsp_lapw as well as run_lapw.
>>
>> Title
>> H LATTICE,NONEQUIV.ATOMS: 2 194 P63/mmc
>> MODE OF CALC= unit=bohr
>> 4.648728 4.648728 14.550897 90.000000 90.000000120.000000
>> ATOM -1: X=0.00000000 Y=0.00000000 Z=0.75000000
>> MULT= 2 ISPLIT= 4
>> -1: X=0.00000000 Y=0.00000000 Z=0.25000000
>> C 1 NPT= 781 R0=0.00010000 RMT= 1.3300 Z: 6.0
>> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
>> 0.0000000 1.0000000 0.0000000
>> 0.0000000 0.0000000 1.0000000
>> ATOM -2: X=0.33333333 Y=0.66666667 Z=0.75000000
>> MULT= 2 ISPLIT= 4
>> -2: X=0.66666667 Y=0.33333333 Z=0.25000000
>> C 2 NPT= 781 R0=0.00010000 RMT= 1.3300 Z: 6.0
>> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
>> 0.0000000 1.0000000 0.0000000
>> 0.0000000 0.0000000 1.0000000
>>
>>
>>
>>
>>
>> Bhagawan Sahu
> wrote:
>>
>> Dear w2kusers,
>>
>> I compiled successfully the latest version of wien2k code on IBM power5.
>>
>> I tested the code 4 carbon atom/unit cell (in bulk graphite structure),
>> spin-unpolarized, spin-unpolarized+spin-orbit. It works.
>>
>> However, when I (just for testing) use "runsp_lapw' it gives error in
>> mixer.
>> (To do the spin-polarized calculation, I did the initilization again in a
>> new
>> directory to avoid the mix-up but this time with dstart -up and dstart-dn.
>> The outputs
>> of the initialization are generated properly)
>>
>>
>> 'MIXER' - can't open unit: 7
>> 'MIXER' - filename: C_HEX.inc
>>
>> I checked file C_HEX.inc exists in the working directory and it is not
>> empty.
>>
>> I tested it on 4 processors as well as serial mode but it stops with the
>> above error both ways.
>>
>> the :log file
>>
>> on May 15 20:42:40 CDT 2006> (x) nn
>> Mon May 15 20:42:50 CDT 2006> (x) sgroup
>> Mon May 15 20:42:53 CDT 2006> (x) symmetry
>> Mon May 15 20:43:12 CDT 2006> (x) lstart
>> Mon May 15 20:43:49 CDT 2006> (x) kgen
>> Mon May 15 20:44:09 CDT 2006> (x) dstart
>> Tue May 16 13:38:27 CDT 2006> (x) dstart -up
>> Tue May 16 13:38:30 CDT 2006> (x) dstart -dn
>>> (runsp_lapw) options: -p -i 20 -cc 0.0001
>> Tue May 16 13:39:28 CDT 2006> (x) lapw0 -p
>> Tue May 16 13:39:32 CDT 2006> (x) lapw1 -up -p
>> Tue May 16 13:39:50 CDT 2006> (x) lapw1 -dn -p
>> Tue May 16 13:40:08 CDT 2006> (x) lapw2 -up -p
>> Tue May 16 13:40:15 CDT 2006> (x) sumpara -up -d
>> Tue May 16 13:40:15 CDT 2006> (x) sumpara_vresp -up -d
>> Tue May 16 13:40:15 CDT 2006> (x) lapw2 -dn -p
>> Tue May 16 13:40:22 CDT 2006> (x) sumpara -dn -d
>> Tue May 16 13:40:22 CDT 2006> (x) sumpara_vresp -dn -d
>> Tue May 16 13:40:22 CDT 2006> (x) lcore -up
>> Tue May 16 13:40:22 CDT 2006> (x) lcore -dn
>> Tue May 16 13:40:22 CDT 2006> (x) mixer
>>
>> The struct file is
>>
>> raphite
>>
>> H LATTICE,NONEQUIV.ATOMS: 2194_P63/mmc
>>
>> MODE OF CALC= unit=bohr
>>
>> 4.648728 4.648728 14.550897 90.000000 90.000000120.000000
>>
>> ATOM -1: X=0.00000000 Y=0.00000000 Z=0.25000000
>> MULT= 2 ISPLIT= 4
>> -1: X=0.00000000 Y=0.00000000 Z=0.75000000
>> C NPT= 781 R0=0.00010000 RMT= 1.3400 Z: 6.0
>>
>> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
>> 0.0000000 1.0000000 0.0000000
>> 0.0000000 0.0000000 1.0000000
>> ATOM -2: X=0.33333333 Y=0.66666667 Z=0.25000000
>> MULT= 2 ISPLIT= 4
>> -2: X=0.66666667 Y=0.33333333 Z=0.75000000
>> C NPT= 781 R0=0.00010000 RMT= 1.3400 Z: 6.0
>>
>> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000
>> 0.0000000 1.0000000 0.0000000
>> 0.0000000 0.0000000 1.0000000
>> 24 NUMBER OF SYMMETRY OPERATIONS
>> -1 0 0 0.0000000
>> -1 1 0 0.0000000
>> 0 0-1 0.0000000
>> 1
>>
>> .....
>> and so on
>>
>> What am I doing wrong?
>>
>> Sahu
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>
>>
>> Sincerely yours,
>> S. Jalali.
>> /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
>> Saeid Jalali Asadabadi,
>> Department of Physics, Faculty of Science,
>> University of Isfahan (UI), Hezar Gerib Avenue,
>> 81744 Isfahan, Iran.
>> Phones:
>> Dep. of Phys. :+98-0311-793 2435
>> Office :+98-0311-793 2430
>> Fax No. :+98-0311-793 2409
>> E-mail :s_jalali_a at yahoo.com
>> www :http://www.ui.ac.ir
>> /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
>>
>> ---------------------------------
>> Blab-away for as little as 1¢/min. Make PC-to-Phone Calls using
>> Yahoo! Messenger with Voice.
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>
>
> Sincerely yours,
> S. Jalali.
> /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> Saeid Jalali Asadabadi,
> Department of Physics, Faculty of Science,
> University of Isfahan (UI), Hezar Gerib Avenue,
> 81744 Isfahan, Iran.
> Phones:
> Dep. of Phys.      :+98-0311-793 2435
> Office             :+98-0311-793 2430
> Fax No.            :+98-0311-793 2409
> E-mail             :s_jalali_a at yahoo.com
> www                :http://www.ui.ac.ir
> /_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
>
> ---------------------------------
> Be a chatter box. Enjoy free PC-to-PC calls  with Yahoo! Messenger with Voice.

Note: if you have an old email address for me, please note that "nwu" has
been changed to "northwestern".
-----------------------------------------------
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2225 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
http://www.numis.northwestern.edu
-----------------------------------------------