[Wien] LAPW2 crashed when running in parallel

[Wei Xie]

Dear Lyudmila,

On Nov 1, 2010, at 8:36 AM, Lyudmila V. Dobysheva wrote:

> 01 Nov 2010 02:56:47 Wei Xie wrote:
>> We encountered some problem when running in parallel (K-point, MPI or
>> both)--the calculations crashed at LAPW2. Note we had no problem running
>> it in serial.
>> This is a TiC example running
> 
> Dear Wei,
> 
> Isn't the error connected with spin-polarised - spin-UNpolarised cases? TiC is 
> to be calculated unpolarised, as far as I know.
TiC is non SP in the first example of UG, but we did it with SP here just for testing. We can run SP calculations in serial for TiC.
>> 1. stdout (abridged)
> ...
>> TiC.scf1up_1: No such file or directory.
> 
> Was lapw1 really successfull?
Thanks for your reminder. We checked and found that lapw1 was actually not successful either--there's no case.output1, case.output2, case.output…file in the case directory. My guess is that the computing nodes are not communicating well with the headnode, so that even the lapw1 finished ok, the output files are not written from computing nodes to the headnode. We are testing the communication now. 
> 
>> 3. uplapw2.error
>> Error in LAPW2
>> 'LAPW2' - can't open unit: 18
>> 'LAPW2' -        filename: TiC.vspup
>> 'LAPW2' -          status: old          form: formatted
> 
> It looks like your initialization was done without spin-polarization, but 
> runsp_lapw was run. But in this case lapw1 must also be unsuccessful (?).
see my answers above.

Your possible follow-up is appreciated in advance.

Thanks,
Wei
> 
> Best regards,
> Lyudmila Dobysheva
> ------------------------------------------------------------------
> Phys.-Techn. Institute of Ural Br. of Russian Ac. of Sci.
> 426001 Izhevsk, ul.Kirova 132
> RUSSIA
> ------------------------------------------------------------------
> Tel.:7(3412) 442118 (home), 218988(office), 250614(Fax)
> E-mail: lyu at otf.fti.udmurtia.su
>        lyuka17 at mail.ru
>        lyu at otf.pti.udm.ru
> http://fti.udm.ru/content/view/25/103/lang,english/
> ------------------------------------------------------------------
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien