×
If you have problems during the execution of MRCC, please attach the output with an adequate description of your case as well as the followings:
This information really helps us during troubleshooting :)
- the way mrcc was invoked
- the way build.mrcc was invoked
- the output of build.mrcc
- compiler version (for example: ifort -V, gfortran -v)
- blas/lapack versions
- as well as gcc and glibc versions
This information really helps us during troubleshooting :)
Seg fault in CCSDTQP NMR calculation
5 years 4 months ago #473
by d2wilson
Seg fault in CCSDTQP NMR calculation was created by d2wilson
Hi,
I've been using MRCC together with CFOUR (fantastic combination), but have come across an error when trying to run CCSDTQP NMR calculations. It works fine with CCSDTQ and other methods, but all my attempts with CCSDTQP fail.
I have tried using different versions of MRCC (including the latest Sept 2017 version) both compiled and as the pre-compiled binary version. I've tried running in serial and OMP parallel but get the same error. Error message is given below. I've tried increasing stack size, etc, but the problem remains. Any help/advice would be most welcome!
MRCC completes the CCSDTQP energy, then fails in the first iteration of the gradient (LR-CC).
Starting left-hand LR-CC iteration for root 1 ...
======================================================================
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Fatal error in mrcc.
Program will stop.
Compile info:
Compilers: Intel 2017.4
OpenMPI: openmpi_ib/1.8.4
System info:
uname -a
Linux flashlite1.local 2.6.32-696.el6.x86_64 #1 SMP Tue Mar 21 19:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Attached:
Input, Output and example PBS script.
Best wishes
Dave
I've been using MRCC together with CFOUR (fantastic combination), but have come across an error when trying to run CCSDTQP NMR calculations. It works fine with CCSDTQ and other methods, but all my attempts with CCSDTQP fail.
I have tried using different versions of MRCC (including the latest Sept 2017 version) both compiled and as the pre-compiled binary version. I've tried running in serial and OMP parallel but get the same error. Error message is given below. I've tried increasing stack size, etc, but the problem remains. Any help/advice would be most welcome!
MRCC completes the CCSDTQP energy, then fails in the first iteration of the gradient (LR-CC).
Starting left-hand LR-CC iteration for root 1 ...
======================================================================
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Fatal error in mrcc.
Program will stop.
Compile info:
Compilers: Intel 2017.4
OpenMPI: openmpi_ib/1.8.4
System info:
uname -a
Linux flashlite1.local 2.6.32-696.el6.x86_64 #1 SMP Tue Mar 21 19:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Attached:
Input, Output and example PBS script.
Best wishes
Dave
Please Log in or Create an account to join the conversation.
5 years 4 months ago #474
by Nike
Replied by Nike on topic Seg fault in CCSDTQP NMR calculation
Dear Dave,
I tried your input file and got the same thing:
Luckily it only takes a few minutes so hopefully someone who knows the code better can try.
With best wishes!
Nike
I tried your input file and got the same thing:
Norm of residual vector: 0.00000004
CPU time [min]: 95.663 Wall time [min]: 4.916
Iteration 15 CC energy: -100.26225538 Energy decrease: 2.3833E-10
======================================================================
Iteration has converged in 15 steps.
Final results:
Total CCSDTQP energy [au]: -100.262255380299
Total MP2 energy [au]: -100.254036682967
Starting left-hand LR-CC iteration for root 1 ...
======================================================================
Fatal error in mrcc.
Program will stop.
************************ 2017-11-23 03:09:44 *************************
Error at the termination of mrcc.
**********************************************************************
--executable dmrcc finished with status 256 in 334.41 seconds (walltime).
This computation required 336.07 seconds (walltime).Luckily it only takes a few minutes so hopefully someone who knows the code better can try.
With best wishes!
Nike
Please Log in or Create an account to join the conversation.
5 years 4 months ago #475
by kallay
Best regards,
Mihaly Kallay
Replied by kallay on topic Seg fault in CCSDTQP NMR calculation
Dear Dave,
Thank you very much for reporting this problem.
You can find three Fortran files, xpert.f, xmrcc.f, and xlambda.f, in the download area. Please compile the program with these files. This will resolve the problem.
Thank you very much for reporting this problem.
You can find three Fortran files, xpert.f, xmrcc.f, and xlambda.f, in the download area. Please compile the program with these files. This will resolve the problem.
Best regards,
Mihaly Kallay
The following user(s) said Thank You: Nike
Please Log in or Create an account to join the conversation.
Time to create page: 0.023 seconds