Hi Mark, I cannot answer your question, but this error message is raised for exactly the reason it says, and for some reason, synchronous submission mode is the default. It is probably better to switch this over to asynchronous, which is more generally supported, and I have just made this change. You can either wait for the next build, or edit /usr/cac/rhel6/lsa/phenix/phenix-dev-1624/cctbx_project/libtbx/easy_mp.py, line 417, and change: asynchronous=False to asynchronous=True and hopefully, it will get further. BW, Gabor On 2014-02-18 20:37, Mark A Saper wrote:
Hi,
This is a technical question for cluster experts.
I tried to run phenix.composite-omit-map which can divide the calculations into separate jobs, on a very large cluster controlled by the PBS system. I used this command of which the first 2 lines specify to divide the work up into 16 processes, and each process will use 1 of the CPU's in the cluster.
phenix.composite_omit_map nproc=16 background=False technology=pbs qsub_command="qsub -V -l procs=1,walltime=00:30:00,mem=2000mb,qos=flux -j oe -M [email protected] -q flux -m abe " omit_type=anneal map_type=2mFo-Dfc output.file_name=composite_omit_map9.mtz reflections3_5percent.mtz Refine_122.pdb
This is the end of the log file before it aborted:
----------Calculating individual omit maps----------
|-----------------------------------------------------------------------------|
| target_work(ml) = 4.65539 r_work = 0.1750 r_free = 0.2066 |
|-----------------------------------------------------------------------------|
And this is the error log:
Traceback (most recent call last): File "/usr/cac/rhel6/lsa/phenix/phenix-dev-1624/build/intel-linux-2.6-x86_64/../../phenix/phenix/command_line/composite_omit_map.py", line 238, in <module> run(sys.argv[1:]) File "/usr/cac/rhel6/lsa/phenix/phenix-dev-1624/build/intel-linux-2.6-x86_64/../../phenix/phenix/command_line/composite_omit_map.py", line 166, in run out=out) File "/usr/cac/rhel6/lsa/phenix/phenix-dev-1624/phenix/phenix/automation/composite_omit_map.py", line 395, in __init__ preserve_exception_message=True) File "/usr/cac/rhel6/lsa/phenix/phenix-dev-1624/cctbx_project/libtbx/easy_mp.py", line 502, in parallel_map display_stderr = not preserve_exception_message, File "/usr/cac/rhel6/lsa/phenix/phenix-dev-1624/cctbx_project/libtbx/queuing_system_utils/processing/__init__.py", line 405, in PBS raise RuntimeError, "PBS does not support synchronous submission" RuntimeError: PBS does not support synchronous submission
I know that PBS doesn't support synchronous submission. Perhaps each job (or map region) requires more than one processor?
Thanks, Mark
_________________________________
Mark A. Saper, Ph.D. Associate Professor of Biological Chemistry, University of Michigan c/o Biophysics, Chem 3040 930 N University Ave Ann Arbor MI 48109-1055 U.S.A.
[email protected] phone (734) 764-3353 fax (734) 764-3323
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb