- Posts: 3
- Thank you received: 0
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
OMP MRCC hanging process
- lixpas
- Topic Author
- Offline
- New Member
Less
More
9 years 3 days ago - 9 years 3 days ago #245
by lixpas
OMP MRCC hanging process was created by lixpas
Dear MRCC developers,
I am currently running CCSDT(Q) and CCSDTQ calculations on orbitals from DIRAC with OMP version of MRCC. Several times I ran into a problem of the calculation hanging (randomly) in the middle of the CC iterations. I receive no error message, the mrcc and dmrcc processes are still alive, but not accessing any files or consuming any CPU resources for a long period of time (days).
If I attach strace to the mrcc process, it gives back:
futex(0x2aaaaf76be84, FUTEX_WAIT_PRIVATE, 3625, NULL
(of course, the numbers can differ from one instance to another)
gdb backtrace for the same process is attached as a file. So is the output file. Input files are generated from DIRAC via dirac_mointegral_export utility.
It appears to be a problem with parallelization. Possibly multiple threads finish their execution at the exact same time and are therefore waiting for one another indefinitely. Have you encountered similar problems before? Do you have any suggestions what to look for?
Thank you in advance
Lukas
I am currently running CCSDT(Q) and CCSDTQ calculations on orbitals from DIRAC with OMP version of MRCC. Several times I ran into a problem of the calculation hanging (randomly) in the middle of the CC iterations. I receive no error message, the mrcc and dmrcc processes are still alive, but not accessing any files or consuming any CPU resources for a long period of time (days).
If I attach strace to the mrcc process, it gives back:
futex(0x2aaaaf76be84, FUTEX_WAIT_PRIVATE, 3625, NULL
(of course, the numbers can differ from one instance to another)
gdb backtrace for the same process is attached as a file. So is the output file. Input files are generated from DIRAC via dirac_mointegral_export utility.
It appears to be a problem with parallelization. Possibly multiple threads finish their execution at the exact same time and are therefore waiting for one another indefinitely. Have you encountered similar problems before? Do you have any suggestions what to look for?
Thank you in advance
Lukas
Attachments:
Last edit: 9 years 3 days ago by lixpas. Reason: attaching files with correct txt extension
Please Log in or Create an account to join the conversation.
- kallay
- Offline
- Administrator
- Mihaly Kallay
9 years 3 days ago #246
by kallay
Best regards,
Mihaly Kallay
Replied by kallay on topic OMP MRCC hanging process
Dear Lukas,
It is strange. I have never encountered this problem. Unfortunately I have no idea what the solution is.
It is strange. I have never encountered this problem. Unfortunately I have no idea what the solution is.
Best regards,
Mihaly Kallay
Please Log in or Create an account to join the conversation.
Time to create page: 0.038 seconds