Hi,
Your job crashed generating plots for B-factors or scale factors.
Please look at bfactors.star. Does it contain 'nan' or 'inf'?
If so, please make sure that you are using the latest code in our
repository and you are NOT continuing failed runs.
We made several updates to fix this issue recently. If the problem persists,
please let us know.
Best regards,
Takanori Nakane
> Dear list (or better dear Relioners),
>
> I am still having problems with bayesian polishing. I have the same issue
> on two independent datasets now (first one worked like a charm). The first
> problem was already documented in my previous email to this newsletter
> (link: https://goo.gl/DjC8a4). Anyway, I have tried to bypass the training
> step and using the default parameters to perform particle polishing.
> Unfortunately I always get a segmentation fault (I have tried every
> possible combination of MPI/threads, even on single CPU).
>
> Does that mean there's something wrong with input movies? Anything I
> should check or control?
>
> command:
> mpirun -n 20 `which relion_motion_refine_mpi` --i
> Refine3D/job138/run_data.star --f PostProcess/job141/postprocess.star
> --corr_mic MotionCorr/job135/corrected_micrographs.star --m1
> ./Refine3D/job138/run_half1_class001_unfil.mrc --m2
> ./Refine3D/job138/run_half2_class001_unfil.mrc --mask
> ./MaskCreate/job140/mask.mrc --first_frame 1 --last_frame -1 --o
> Polish/job145/ --s_vel 0.2 --s_div 5000 --s_acc 2 --combine_frames
> --bfac_minfreq 20 --bfac_maxfreq -1 --only_do_unfinished --j 1
>
> run.out:
> [...]
> + Reading Refine3D/job138/run_data.star...
> - Using pixel size calculated from magnification and detector pixel
> size in the input STAR file: 0.889
> + Using movie pixel size from
> MotionCorr/job135/raw_images/EXPERIMENT-FOLDER/FoilHole_16123192_Data_16124420_16124421_20180716_1625_Fractions.star:
> 0.889 A
> + Using coord. pixel size from
> MotionCorr/job135/raw_images/EXPERIMENT-FOLDER/FoilHole_16123192_Data_16124420_16124421_20180716_1625_Fractions.star:
> 0.889 A
> + Using dose per frame from
> MotionCorr/job135/raw_images/EXPERIMENT-FOLDER/FoilHole_16123192_Data_16124420_16124421_20180716_1625_Fractions.star:
> 1 e/A^2
> + Reading references ...
> + Masking references ...
> + Transforming references ...
> + Initializing motion estimator ...
> + Performing loop over all micrographs ...
> 1.86/1.86 hrs
> ............................................................~~(,_,">
> + Fitting B/k-factors between 13 and 81 pixels, or 20 and 3.24869
> Angstrom ...
> --------------------------------------------------------------------------
> mpirun noticed that process rank 7 with PID 18716 on node cryo1 exited on
> signal 11 (Segmentation fault).
>
> run.err:
> [cryo1:18721] *** Process received signal ***
> [cryo1:18721] Signal: Segmentation fault (11)
> [cryo1:18721] Signal code: (128)
> [cryo1:18721] Failing at address: (nil)
> [cryo1:18721] [ 0] /lib64/libpthread.so.0(+0xf130)[0x7f50779f4130]
> [cryo1:18721] [ 1]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(_ZN7CPlot2D25DrawXAxisLabelsPostScriptEv+0x22c)[0x46633c]
> [cryo1:18721] [ 2]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(_ZN7CPlot2D20OutputPostScriptPlotESs+0x1c8)[0x46a3a8]
> [cryo1:18721] [ 3]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(_ZN15FrameRecombiner14weightsFromFCCERKSt6vectorI13MetaDataTableSaIS1_EE+0x1e4f)[0x4d6e9f]
> [cryo1:18721] [ 4]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(_ZN15FrameRecombiner4initERKSt6vectorI13MetaDataTableSaIS1_EEiiidiSsbP16ObservationModelP17MicrographHandler+0x368)[0x4d7a78]
> [cryo1:18721] [ 5]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(_ZN16MotionRefinerMpi3runEv+0x10a)[0x4617ca]
> [cryo1:18721] [ 6]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi(main+0x45)[0x42ae85]
> [cryo1:18721] [ 7]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f5077645af5]
> [cryo1:18721] [ 8]
> /XXX/relion-3.0_beta/build/bin/relion_motion_refine_mpi[0x42ba61]
> [cryo1:18721] *** End of error message ***
>
>
> Thanks a lot,
>
> Paolo
> ------------------------------------------------
> Paolo Swuec, PhD
> Cryo-Electron Microscopy Lab ? Facility Manager
> Dep. BioSciences
> University of Milano
> Via Celoria 26 - 20133 (Italy)
> Tel. +39 02 503 147 78
>
> Follow on twitter @PS_CryoEM
>
>
> ########################################################################
>
> To unsubscribe from the CCPEM list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
>
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
|