Regarding Moose installation on HPC Cluster

[rakeshm22]

Hello everyone,
I am trying to install MOOSE on our HPC cluster. I followed all the instructions mentioned on the MOOSE HPC Installation Guide.

Here are the steps I followed:

(a) Currently Loaded Modules (as required for installing MOOSE on the HPC cluster):

1. python/conda-python/3.9
2. compiler/gcc/13.1.0
3. cmake/3.28.1
4. compiler/openmpi/gcc/4.0.2

(b) Activate Environment: Activated the desired MPI environment.
(c) Cloning MOOSE: Successfully cloned the MOOSE repository.
(d) PETSc, libMesh, and WASP:

Successfully ran the PETSc command.
However, while installing libMesh, I encountered the following error:

 *********************************************************************
                                        ERROR
********************************************************************
make[4]: Leaving directory `/home/rakeshm22/projects/moose/libmesh/build/contrib/netcdf/v4/libhdf5'
Making all in liblib
make[4]: Entering directory `/home/rakeshm22/projects/moose/libmesh/build/contrib/netcdf/v4/liblib'
  CC       libnetcdf_la-nc_initialize.lo
  CCLD     libnetcdf.la
/usr/bin/ld: cannot find -ludev
collect2: error: ld returned 1 exit status
make[4]: *** [libnetcdf.la] Error 1
make[4]: Leaving directory `/home/rakeshm22/projects/moose/libmesh/build/contrib/netcdf/v4/liblib'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/home/rakeshm22/projects/moose/libmesh/build/contrib/netcdf/v4'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/rakeshm22/projects/moose/libmesh/build/contrib/netcdf/v4'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/rakeshm22/projects/moose/libmesh/build/contrib'
make: *** [all-recursive] Error 1
**********************************************************************************************************

[GiudGiud]

Hello

/usr/bin/ld: cannot find -ludev

This is a missing dependency in your environment.
you need to install it locally or get your cluster admins to make it available to you in another module
if you have admin rights and apt-get is installed (unlikely) you can just run

sudo apt-get install --reinstall libudev-devel

[rakeshm22]

Hello GuidGiud,

Thank you for your response.

I am not a sudo user, so I kindly request your guidance on how to install it locally and resolve this issue. Your assistance would be greatly appreciated.

Thank you again for your help.

[GiudGiud]

I have never installed libudev from source. I m not sure how easy it is.
can you see if you can find it in the modules?

module avail

should show everything you can load.
Your HPC team might also know if you they already have it with modules for different compiler stack

[rakeshm22]

Hello everyone,
I am trying to reproduce grain growth for a bicrystal system in the presence of a temperature gradient as an external field. I wrote a .i file based on the equations mentioned in the paper.
I am not able to find any output. Kindly help in this regards.

(https://github.com/user-attachments/assets/796e1709-6867-4548-955e-a7407dd1719a)
(https://github.com/user-attachments/files/18279534/Michael.R_2013.pdf)

[Mesh]
type = GeneratedMesh
dim = 2
nx = 20
ny = 10
xmin = 0
xmax = 20
ymin = 0
ymax = 10
[]

[GlobalParams]
block = 0
[]

[Variables]
[./eta1]
[../]

[./eta2]
[../]

[./T]
order = FIRST
family = LAGRANGE
[../]
[]

[ICs]

[./IC_eta1]
x1 = 0
y1 = 0
x2 = 20
y2 = 10
inside = 1.0
outside = 0.0
type = BoundingBoxIC
variable = eta1
int_width = 0.5
[../]

[./IC_eta2]
x1 = 10
y1 = 0
x2 = 20
y2 = 10
inside = 1.0
outside = 0.0
type = BoundingBoxIC
variable = eta2
int_width = 0.5
[../]

[./T]
x1 = 0
y1 = 0
x2 = 18
y2 = 10
inside = 0.01
outside = 0.01
type = BoundingBoxIC
variable = T
int_width = 0.5
[../]

[]

[AuxVariables]
[./T_x]
order = CONSTANT
family = MONOMIAL
[../]
[./T_y]
order = CONSTANT
family = MONOMIAL
[../]

[./eta1_x]
order = CONSTANT
family = MONOMIAL
[../]
[./eta1_y]
order = CONSTANT
family = MONOMIAL
[../]

[./eta2_x]
order = CONSTANT
family = MONOMIAL
[../]
[./eta2_y]
order = CONSTANT
family = MONOMIAL
[../]

[]

[AuxKernels]
[./T_x]
type = VariableGradientComponent
variable = T_x
component = x
gradient_variable = T
[../]
[./T_y]
type = VariableGradientComponent
variable = T_y
component = y
gradient_variable = T
[../]
[./eta1_x]
type =VariableGradientComponent
variable = eta1_x
component = x
gradient_variable = eta1
[../]

[./eta1_y]
type =VariableGradientComponent
variable = eta1_y
component = y
gradient_variable = eta1
[../]

[./eta2_x]
type =VariableGradientComponent
variable = eta2_x
component = x
gradient_variable = eta2
[../]

[./eta2_y]
type =VariableGradientComponent
variable = eta2_y
component = y
gradient_variable = eta2
[../]

[]

[Kernels]

[./AC1_bulk]
type = AllenCahn
variable = eta1
f_name = F
args = 'eta2 T '
[../]
[./AC1_int]
type = ACInterface
variable = eta1
kappa_name = kappa_s
[../]
[./e1_dot]
type = TimeDerivative
variable = eta1
[../]

[./AC2_bulk]
type = AllenCahn
variable = eta2
f_name = F
args = 'eta1 T '
[../]
[./AC2_int]
type = ACInterface
variable = eta2
kappa_name = kappa_s
[../]
[./e2_dot]
type = TimeDerivative
variable = eta2
[../]

[./diff2]
type = Diffusion
variable = T
[../]

[]

[Materials]

[./FreeEng]
type = DerivativeParsedMaterial

args = 'eta1 eta2 T_x T_y eta1_x eta1_y eta2_x eta2_y  T'

f_name = F

constant_names = ' ww'  
constant_expressions = ' 1.14 '

function = 'f2:=(1/4eta1^4-1/2eta1^2+1/4eta2^4-1/2eta2^2);
f3:=s*(eta1^2eta2^2);
f4:=eta14.0*(T_xeta1_x+T_yeta1_y);
f5:=eta24.0(T_xeta2_x+T_yeta2_y);

        F:= ww*(f2 + f3 + 0.25) + f4 + f5;F'

derivative_order = 2
outputs = exodus

[../]

[./const]
type = GenericConstantMaterial
prop_names = 'kappa_s L '
prop_values = '1.0 1.0 '
outputs = exodus
[../]

[]

[BCs]

[./Periodic]
[./all]
auto_direction = 'y'
[../]
[../]

[./neumaan1]
type = NeumannBC
boundary = 'left right'
variable = eta1
value = 0
[../]
[./neumaan2]
type = NeumannBC
boundary = 'left right'
variable = eta2
value = 0
[../]

[./Dirichlet1]
type = DirichletBC
boundary = 'left'
variable = T
value = 0.1
[../]

[./Dirichlet2]
type = DirichletBC
boundary = 'right'
variable = T
value = 1.5
[../]

[]

[Postprocessors]
[./dt]
type = TimestepSize
[../]

[]

[Preconditioning]
active = ' '
[./SMP]
type = SMP
full = true
petsc_options_iname = '-ksp_gmres_restart -snes_atol -snes_rtol -ksp_rtol -pc_type -build_twosided'
petsc_options_value = ' 160 1e-10 1e-8 1e-6 bjacobi allreduce'
[../]
[]

[Executioner]
type = Transient
scheme = BDF2
solve_type = newton

petsc_options_iname = '-pc_type -ksp_grmres_restart -sub_ksp_type -sub_pc_type -pc_asm_overlap'
petsc_options_value = 'asm 31 preonly lu 1'
l_max_its = 30
nl_abs_tol = 1e-7
l_tol = 1e-7
nl_max_its = 30
nl_rel_tol = 1e-7
end_time = 30000000
dt=0.50

[]

[Outputs]
execute_on = 'timestep_end'
exodus = true
interval=2
[]

(https://github.com/user-attachments/assets/796e1709-6867-4548-955e-a7407dd1719a)
(https://github.com/user-attachments/files/18279534/Michael.R_2013.pdf)

[GiudGiud]

hello

please open a different post, the topic of this discussion is HPC installation not phase field