Getting error while running example code

[subhk]

hi!

I've installed Slepc (version- 0.1.3) and trying to do some example. I'm trying demo1.jl in the example that is given. Although I'm getting an error (relevant part is posted here)

ERROR: LoadError: AssertionError: iszero(error) Stacktrace: [1] EPSSetUp(eps::SlepcEPS) @ SlepcWrap ~/.julia/packages/SlepcWrap/VVZOZ/src/eps.jl:186 [2] top-level scope

I think it is throwing the error at these line of the code which is line 83 of the example: EPSSetUp(eps);

I'm trying to understand origin of the above error. Any help would be highly aprreciated.

Thank you, Subhajit

[bmxam]

Hi Subhajit,

Sorry for the late answer. I've just tested the master branch of SlepcWrap with Slepc-3-13-1 and it works fine on my side for demo1.jl. Which Slepc version are you using (I don't mean the SlepcWrap version, but the SLEPc library version? Additional question: what command did you run to execute the example?

[bmxam]

Hi Matthew,Somehow I can't see your comment on Github so I'm answering by email. From your error, it seems that you're trying to apply to solve a linear system (with PETSc) with a direct method, in parallel. PETSc doesn't have a native solver for that. As suggested by the error message, you will have to compile PETSc with an additionnal package such as MUMPS. -------- Message d'origine --------De : Matthew Thomas @.> Date : 08/05/2024 02:46 (GMT+01:00) À : "bmxam/SlepcWrap.jl" @.> Cc : bmxam @.>, Comment @.> Objet : Re: [bmxam/SlepcWrap.jl] Getting error while running example code (Issue #4) Hello bmxam, I am getting the same error when I run the Helmholtz example. I do not get the same issue running demo1.jl and I have identified the issued coming from EPSSetOperators(eps, A, B), changing to EPSSetOperators(eps, A) removes the issue. I get the following error, [0]PETSC ERROR: --------------------- Error Message -------------------------------------------------------------- [0]PETSC ERROR: See https://petsc.org/release/overview/linear_solve_table/ for possible LU and Cholesky solvers [0]PETSC ERROR: Could not locate a solver type for factorization type LU and matrix type mpiaij. I am running Slepc-3-21.1 and was successfully able to run this code in serial. Thanks.

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: @.***>

[Syndrius]

Hello,

Using Mumps did solve my issues, thank you for the help.

I should also let you know that I am running this on Mac, and SlepcWrap/PetscWrap will not compile out of the box. In my case this was because the petsc file libpetsc.so is libpetsc.dylib on mac. I was able to get this to work by modifying the build.jl file to look for .dylib instead. Obviously, quite the hack, but it worked for me.

Thanks again for your help.

On Tue, May 14, 2024 at 1:01 AM bmxam @.***> wrote:

Hi Matthew,Somehow I can't see your comment on Github so I'm answering by email. From your error, it seems that you're trying to apply to solve a linear system (with PETSc) with a direct method, in parallel. PETSc doesn't have a native solver for that. As suggested by the error message, you will have to compile PETSc with an additionnal package such as MUMPS. -------- Message d'origine --------De : Matthew Thomas @.> Date : 08/05/2024 02:46 (GMT+01:00) À : "bmxam/SlepcWrap.jl" @.> Cc : bmxam @.>, Comment @.> Objet : Re: [bmxam/SlepcWrap.jl] Getting error while running example code (Issue #4) Hello bmxam, I am getting the same error when I run the Helmholtz example. I do not get the same issue running demo1.jl and I have identified the issued coming from EPSSetOperators(eps, A, B), changing to EPSSetOperators(eps, A) removes the issue. I get the following error, [0]PETSC ERROR: --------------------- Error Message

[0]PETSC ERROR: See https://petsc.org/release/overview/linear_solve_table/ for possible LU and Cholesky solvers [0]PETSC ERROR: Could not locate a solver type for factorization type LU and matrix type mpiaij. I am running Slepc-3-21.1 and was successfully able to run this code in serial. Thanks.

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/bmxam/SlepcWrap.jl/issues/4#issuecomment-2107890172, or unsubscribe https://github.com/notifications/unsubscribe-auth/AONRAFYOV2TCDLUTTVVBOX3ZCDIWFAVCNFSM6AAAAAAUQVMRO2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMBXHA4TAMJXGI . You are receiving this because you commented.Message ID: <bmxam/SlepcWrap. @.***>