Error running CosmoMC with Planck Likelihoods

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Carter Bartram
Posts: 3
Joined: May 05 2017
Affiliation: Dartmouth College

Error running CosmoMC with Planck Likelihoods

Post by Carter Bartram » May 05 2017

Hello, I've installed the Planck data and linked everything correctly (at least I think so, as I used to get the "cannot stat file" error but no longer do). However, now when I try to run CosmoMC on test_planck.ini I get the following error:
~/CosmoMC-Nov2016$ ./cosmomc test_planck.ini
Number of MPI processes: 1
file_root:test5
Random seeds: 15355, 7524 rand_inst: 1
compile with CLIK to use clik - see Makefile
MpiStop: 0
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 128.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
I've checked the Makefile, and the only reference to clik I could find was under camspec. I tried running make camspec, but the problem persists. What am I doing wrong?

Jason Dossett
Posts: 97
Joined: March 19 2010
Affiliation: The University of Texas at Dallas
Contact:

Error running CosmoMC with Planck Likelihoods

Post by Jason Dossett » May 05 2017

Can you show us the output when you make, that is indicating that when you built, -DCLIK was not in the FFLAGS, so you may not have your environment set up correctly.

Did you do this:

Code: Select all

source path_to_plc-2.0/bin/clik_profile.sh
echo -e "\nsource path_to_plc-2.0/bin/clik_profile.sh" >> ~/.bashrc
as described in the Planck Readme

Carter Bartram
Posts: 3
Joined: May 05 2017
Affiliation: Dartmouth College

Error running CosmoMC with Planck Likelihoods

Post by Carter Bartram » May 06 2017

This is the output from make camspec:

Code: Select all

~/CosmoMC-Nov2016$ make camspec
cd ./source && make highL=../highL PLANCKLIKE=cliklike_CamSpec
make[1]: Entering directory '/home/ubuntu/CosmoMC-Nov2016/source'
mkdir -p ReleaseMPI
cd ../camb && \
make --file=Makefile_main libcamb OUTPUT_DIR=ReleaseMPI \
RECOMBINATION=recfast EQUATIONS=equations_ppf NONLINEAR=halofit_ppf
make[2]: Entering directory '/home/ubuntu/CosmoMC-Nov2016/camb'
mkdir -p ReleaseMPI
ar -r ReleaseMPI/libcamb_recfast.a ReleaseMPI/constants.o  ReleaseMPI/utils.o   ReleaseMPI/subroutines.o ReleaseMPI/inifile.o  ReleaseMPI/power_tilt.o  ReleaseMPI/recfast.o ReleaseMPI/reionization.o ReleaseMPI/modules.o ReleaseMPI/bessels.o ReleaseMPI/equations_ppf.o ReleaseMPI/halofit_ppf.o ReleaseMPI/lensing.o ReleaseMPI/SeparableBispectrum.o ReleaseMPI/cmbmain.o ReleaseMPI/camb.o
make[2]: Leaving directory '/home/ubuntu/CosmoMC-Nov2016/camb'
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c cliklike.f90 -o ReleaseMPI/cliklike.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c CMB.f90 -o ReleaseMPI/CMB.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c DataLikelihoods.f90 -o ReleaseMPI/DataLikelihoods.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c calclike.f90 -o ReleaseMPI/calclike.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c ImportanceSampling.f90 -o ReleaseMPI/ImportanceSampling.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c MCMC.f90 -o ReleaseMPI/MCMC.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c minimize.f90 -o ReleaseMPI/minimize.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c SampleCollector.f90 -o ReleaseMPI/SampleCollector.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c GeneralSetup.f90 -o ReleaseMPI/GeneralSetup.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c CalcLike_Cosmology.f90 -o ReleaseMPI/CalcLike_Cosmology.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c Calculator_CAMB.f90 -o ReleaseMPI/Calculator_CAMB.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c CosmologyConfig.f90 -o ReleaseMPI/CosmologyConfig.o
mpif90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/ -c driver.F90 -o ReleaseMPI/driver.o
mpif90 -o ../cosmomc ReleaseMPI/MiscUtils.o ReleaseMPI/ArrayUtils.o ReleaseMPI/StringUtils.o ReleaseMPI/MpiUtils.o ReleaseMPI/FileUtils.o ReleaseMPI/RandUtils.o ReleaseMPI/ObjectLists.o ReleaseMPI/Interpolation.o ReleaseMPI/IniObjects.o ReleaseMPI/ObjectParamNames.o ReleaseMPI/Matrix_utils_new.o  ReleaseMPI/settings.o ReleaseMPI/samples.o ReleaseMPI/IO.o ReleaseMPI/likelihood.o ReleaseMPI/GeneralTypes.o ReleaseMPI/BaseParameters.o ReleaseMPI/propose.o ReleaseMPI/ParamSet.o ReleaseMPI/PowellConstrainedMinimize.o ReleaseMPI/ImportanceSampling.o ReleaseMPI/MCMC.o ReleaseMPI/minimize.o ReleaseMPI/SampleCollector.o ReleaseMPI/GeneralSetup.o ReleaseMPI/CosmologyTypes.o ReleaseMPI/CosmoTheory.o ReleaseMPI/bbn.o ReleaseMPI/Calculator_Cosmology.o ReleaseMPI/CalcLike_Cosmology.o ReleaseMPI/Likelihood_Cosmology.o ReleaseMPI/Calculator_CAMB.o ReleaseMPI/CMB_BK_Planck.o ReleaseMPI/cliklike.o ReleaseMPI/mpk.o ReleaseMPI/wigglez.o ReleaseMPI/bao.o ReleaseMPI/supernovae_Union2.o ReleaseMPI/supernovae_SNLS.o ReleaseMPI/supernovae_JLA.o ReleaseMPI/szcounts.o ReleaseMPI/supernovae.o ReleaseMPI/HST.o  ReleaseMPI/CMB.o ReleaseMPI/CMBlikes.o ReleaseMPI/ElementAbundances.o ReleaseMPI/wl.o ReleaseMPI/CosmologyParameterizations.o ReleaseMPI/DataLikelihoods.o ReleaseMPI/calclike.o ReleaseMPI/CosmologyConfig.o ReleaseMPI/driver.o -L../camb/ReleaseMPI -lcamb_recfast -lblas -llapack   -L/home/ubuntu/Downloads/plc-2.0/lib -lclik_f90 -cpp -O3 -ffast-math -ffree-line-length-none -fopenmp -fmax-errors=4 -march=native -DMPI -DCLIK -I../camb/ReleaseMPI -I/home/ubuntu/Downloads/plc-2.0/include -JReleaseMPI -IReleaseMPI/
make[1]: Leaving directory '/home/ubuntu/CosmoMC-Nov2016/source'
I did the two lines from the readme again and double checked that my bashrc contains the path to clik_profile.sh.

Antony Lewis
Posts: 1936
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: Error running CosmoMC with Planck Likelihoods

Post by Antony Lewis » May 08 2017

You don't need to do "make camspec" (the option may be present in the Makefile as a relic of our Planck-internal version). Just use default make.

Carter Bartram
Posts: 3
Joined: May 05 2017
Affiliation: Dartmouth College

Error running CosmoMC with Planck Likelihoods

Post by Carter Bartram » May 09 2017

Well, I just reinstalled CosmoMC and everything now works fine. Thanks for your help.

Post Reply