[Wien] error in x spaghetti for parallel computing

[Peter Blaha]

When you do a parallel calculation you have to add    -p   to spaghetti

x spaghetti -p

-------------------------

I guess the problem for lapw2 occurred when you resubmitted another 
batch job (without x lapw1 -p -band).

This is clear, because only the lapw1 step creates suitable .machine* 
and .processes files, which are needed in lapw2 and they have to be 
compatible with the node-list of the current batch job.

Simply insert into your job script:

x lapw1 -p -d

before the    x lapw2 step.

On 02/17/2016 09:23 AM, Dr. K. C. Bhamu wrote:
> One more thing:
>
> When I checked the output file of x lapw2 -band then I got the following
> massage:
>
> running lapw2 in parallel mode
> calculating QTL's from parallel vectors
> ssh: connect to host nid01520 port: Connection refused^M
> **** lines are repeatitativelly written here for around 50times for this
> node
> ssh: connect to host nid01521 port: Connection refused^M
> **** lines are repeatitativelly written here for around 50 times for
> this node
> LAPW2 END
> 490.570u 24.737s 11.05.91 77.3% 0+0k 79259539+1284375io 8pf+0w
>
> Sincerely
> Bhamu
>
> *
> *
>
> On Wed, Feb 17, 2016 at 1:45 PM, Dr. K. C. Bhamu <kcbhamu85 at gmail.com
> <mailto:kcbhamu85 at gmail.com>> wrote:
>
>     Dear Wien2k Prof Blaha and Gavin
>
>     I ran my job on remote server using two nodes (64 cores).
>
>     Everything went fine till" x lapw2 -p -band -qtl (checked with cat *.error).
>
>     I inserted :FER into case.insp
>
>     Then I ran x spaghetti and the following error occured:
>
>
>     ERROR IN OPENING UNIT:           7
>              FILENAME:
>       orthorhombic_1.output1
>          STATUS: old          FORM:formatted
>     OPEN FAILED
>     0.0u 0.0s 0:00.08 0.0% 0+0k 2008+0io 10pf+0w
>     ----------------------
>
>     I checked the DIR but there was no case.output1 file while case.output2 was present.
>
>     What may be the error? I submitted the job trice: 1. for scf, 2. x lapw1 -band -p, 3. x lapw2 -band -p -qtl.
>
>     Should I have to bind the job in case.output1 file? If yes please suggest any command.
>     *******************
>
>
>
>     Yours
>
>     truly
>
>     Bhamu
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at 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
>

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/staff/tc_group_e.php
--------------------------------------------------------------------------