Hi Bruno, It seems that I met the similar problem. My question is how to turn METIS on? Need I write something in the pbs job file or contact with the guy who is in charge of the cluster? Thank you very much! Please refer to my question https://groups.google.com/forum/#!topic/dealii/116cFgS6EiE
Best, Chenchen 在 2015年7月15日星期三 UTC-4上午10:25:50，Bruno Turcksin写道： > > Weixiong, > > the error that you get is normal. By default, candi turns off METIS but > step-17 and step-18 require METIS to run in parallel. PETSc is installed > correctly that's why when you use one processor everything works fine. If > you want to use METIS just turn it ON in candi.cgf > > Best, > > Bruno > > On Wednesday, July 15, 2015 at 12:58:13 AM UTC-5, Weixiong Zheng wrote: >> >> Sorry for the ambiguity, let me clarify here: >> Those error messages came from using candi-PETSc with mpirun in step-18 >> (step-17 has the same problem). >> >> mpirun -np 1 ./step-18 was fine. but mpirun -np x ./step-18 (x>=2) gave >> me the errors. >> >> Thanks, >> Weixiong >> >> 在 2015年7月15日星期三 UTC-5上午12:51:07，Weixiong Zheng写道： >>> >>> Dear all, >>> >>> After struggling on installing parallel stuffs with deal.ii for 4 days >>> as an Ubuntu rookie (the previous discussion), the final compromise was >>> made as: I used candi to make deal.ii + trilinos and built deal.ii + petsc >>> by myself from source. The reason for building PETSc without candi is that >>> when using -np>1, the following error manifested as shown in the end of >>> this post. >>> >>> This time, I didn't install libumfpack, but directly installed >>> libsuitesparse-dev. And recall that last time both candi and self-compiled >>> Trilinos did not work, yet, candi-trilinos works well this time, I would >>> assume the undefined reference blabla was caused by "libumfpack" >>> >>> Yet, I don't have luck with candi+petsc but the self-compiled petsc >>> works well with deal.ii with MPI. I didn't change anything in candi.cfg >>> expect the PROC. >>> >>> Though i now have working PETSc (self-compiled) and Trilinos (candi), it >>> is still interesting to know what's going on with candi-PETSc (well, I >>> would assume this could be my problem,) >>> >>> Thanks in advance, >>> Weixiong >>> >>> >>> >>> >>> >>> ERROR: Uncaught exception in MPI_InitFinalize on proc 1. Skipping >>> MPI_Finalize() to avoid a deadlock. >>> >>> >>> ---------------------------------------------------- >>> Exception on processing: >>> >>> -------------------------------------------------------- >>> An error occurred in line <70> of file >>> </home/weixiong/apps/candi/unpack/deal.II-v8.2.1/source/lac/sparsity_tools.cc> >>> >>> in function >>> void dealii::SparsityTools::partition(const >>> dealii::SparsityPattern&, unsigned int, std::vector<unsigned int>&) >>> The violated condition was: >>> false >>> The name and call sequence of the exception was: >>> ExcMETISNotInstalled() >>> Additional Information: >>> (none) >>> -------------------------------------------------------- >>> >>> Aborting! >>> ---------------------------------------------------- >>> ERROR: Uncaught exception in MPI_InitFinalize on proc 0. Skipping >>> MPI_Finalize() to avoid a deadlock. >>> >>> >>> ---------------------------------------------------- >>> Exception on processing: >>> >>> -------------------------------------------------------- >>> An error occurred in line <70> of file >>> </home/weixiong/apps/candi/unpack/deal.II-v8.2.1/source/lac/sparsity_tools.cc> >>> >>> in function >>> void dealii::SparsityTools::partition(const >>> dealii::SparsityPattern&, unsigned int, std::vector<unsigned int>&) >>> The violated condition was: >>> false >>> The name and call sequence of the exception was: >>> ExcMETISNotInstalled() >>> Additional Information: >>> (none) >>> -------------------------------------------------------- >>> >>> Aborting! >>> ---------------------------------------------------- >>> >>> -------------------------------------------------------------------------- >>> mpirun has exited due to process rank 1 with PID 13739 on >>> node Berserker exiting improperly. There are two reasons this could >>> occur: >>> >>> 1. this process did not call "init" before exiting, but others in >>> the job did. This can cause a job to hang indefinitely while it waits >>> for all processes to call "init". By rule, if one process calls "init", >>> then ALL processes must call "init" prior to termination. >>> >>> 2. this process called "init", but exited without calling "finalize". >>> By rule, all processes that call "init" MUST call "finalize" prior to >>> exiting or it will be considered an "abnormal termination" >>> >>> This may have caused other processes in the application to be >>> terminated by signals sent by mpirun (as reported here). >>> >> -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
