[Wien] Possible clash between different libiomp5.so library file

[Laurence Marks]

N.B., This may be due to you loading different modules, if that is how your
cluster is setup. Check that the same was used for compilation & execution.

On Wed, Jan 9, 2019, 20:16 Laurence Marks <L-marks at northwestern.edu wrote:

> Try "ldd lapw0" to see what shared libraries are being used, then edit
> your compilation options to do static compilation. Depending upon what
> compiler your have the options will change; do "ifort --help | more" or
> "ifort --help > Fort.txt" and read.... (or gfortran, etc).
>
> At least for the non-mpi you can build lapw0 & others without any shared
> libraries. I have never seen this fail, but it might.
>
> On Wed, Jan 9, 2019, 19:58 Lawal Mohammed <mohammedlawal08 at yahoo.com
> wrote:
>
>> Dear Developers and Users,
>>
>> I am facing a problem and need your help.
>> The issue is an error when I submitted a job on cluster machine. The
>> error message is:
>>
>> Calculating Bismuthtest1 in /home/username/WIEN2k/Bismuthtest1
>> on node5 with PID 21645
>> using WIEN2k_18.2 (Release 17/7/2018) in /software/wien2k-18.2
>>
>>     start       (Thu Jan 10 09:48:12 CST 2019) with lapw0 (100/99 to go)
>>
>>     cycle 1     (Thu Jan 10 09:48:12 CST 2019)  (100/99 to go)
>>
>> >   lapw0  -p   (09:48:12) starting parallel lapw0 at Thu Jan 10 09:48:12
>> CST 2019
>> -------- .machine0 : processors
>> running lapw0 in single mode
>> 0.022u 0.036s 0:00.08 62.5%     0+0k 0+56io 0pf+0w
>> error: command   /software/wien2k-18.2/lapw0para lapw0.def   failed
>> >   stop error
>>
>> In the output file the error is:
>> lapw0: relocation error: lapw0: symbol kmp_aligned_malloc, version
>> VERSION not defined in file libiomp5.so with link time reference
>> grep: *scf1*: No such file or directory
>> grep: lapw2*.error: No such file or directory
>>
>> Many thanks to Gavin for pointing out that the error was due to clash
>> between other libraries with the  Intel Fortran library. However, after
>> I reported the issue, system admin confirmed that the environmental
>> variables was okay. Any help to rectify this issue is highly appreciated.
>>
>> I will provide addition information if needed.
>>
>> Thanks alot for your time.
>> With kind regards.
>>
>>
>> *Lawal *
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20190109/bada8176/attachment.html>