[Wien] job is crashing [lapw2] on cluster

[Dr. K. C. Bhamu]

Dear Prof. Peter and experts

Greetings!!!

This is error is severely reported previously but I could not find a
solution.
The error occurs after several scf cycles in 2Doptimization case for
hexagonal structure


**********
in cycle 17    ETEST: .0000062600000000   CTEST: -.000097
Fallback to compatibility mode with "old" save_lapw

************ for shortening the email****

starting parallel dstart at Wed Jan 18 11:41:59 IST 2017
-------- .machine0 : processors
running dstart in single mode
3.028u 0.067s 0:03.17 97.1% 0+0k 0+1056io 0pf+0w
0.028u 0.025s 0:00.06 66.6% 0+0k 0+920io 0pf+0w
clmextrapol_lapw has generated a new CuGaO2_993.clm
in cycle 12    ETEST: .0000207150000000   CTEST: .000134

>   stop error
>>> (min) status after runsp_c_lapw -I -p -orb -ec 0.00001 -cc 0.0001 -fc
1.0 -i 120\: 9 -> exit
ERROR status in 2D_V_-6.0_COA_-4.0

[kcbhamu at gu CuGaO2_993]$ cat *.dayfile


>   stop
Fallback to compatibility mode with "old" save_lapw


>   lapw2 -up -p   -orb (11:49:42) running LAPW2 in parallel mode
**  LAPW2 crashed!
0.063u 0.262s 0:00.66 48.4% 0+0k 0+216io 1pf+0w
error: command   /home/kcbhamu/Wien2k_16/lapw2para -up uplapw2.def   failed

>   stop error
[kcbhamu at gu CuGaO2_993]$


The job.err files tells about cluster connection problem,

ssh: Could not resolve hostname compute-11-20: Name or service not known
ssh: Could not resolve hostname compute-11-20: Name or service not known
ssh: Could not resolve hostname compute-11-20: Name or service not known
 LAPW1 END
 LAPW1 END
 LAPW1 END
 LAPW1 END
 LAPW1 END
FERMI - Error
cp: cannot stat `.in.tmp': No such file or directory
[kcbhamu at gu CuGaO2_993]$


I guess that it is cluster issue but when I saw the node in question, it
looks fine.


Could you please suggest me any possible solution?


Sincerely

Bhamu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20170118/a8419a11/attachment.html>