Rocky Linux 8

by

Configuring ABAQUS Linux Client for New License Server | Step-by-Step

If you have migrated to a new ABAQUS License Server and is planning to point your ABAQUS Linux Client to the new License Server. Here are the simple steps you will need to do

Step 1: Find the custom_v6.env files to edit.

Depending on how you install, for me I like to put everything in /usr/local so my abaqus will be placed there.

# cd /usr/local/abaqus-2023/SIMULIA/EstProducts/2023/linux_a64/SMA/site/
# vim custom_v6.env

Step 2: Edit the custom_v6.env file

Add the abaquslm_license_file = “27398@XXX.XXX.XXX.XXX”

# Installation of Established Products 2023
# Wed Oct 11 13:30:54 2023
plugin_central_dir="/usr/local/abaqus-2023/DassaultSystemes/SIMULIA/CAE/plugins/2023"
# retrieve licensing configuration from EstablishedProductsConfig.ini
importEnv('licensing.env')
abaquslm_license_file = "27398@XXX.XXX.XXX.XXX"

References:

  1. Manually Configuring Licensing for ABAQUS
by

SSH Error – Receive Packet Type 51 on Rocky Linux 8

If you are having SSH issues and if you turned on high verbosity and the following output is generated

# ssh -vvv XXX.XXX.XXX.XXX
.....
.....
debug1: Offering public key: debug3: send packet: type 50
debug2: we sent a publickey packet, wait for reply
debug3: receive packet: type 51
.....
.....
debug2: we did not send a packet, disable method
debug1: No more authentication methods to try.
user1@192.168.0.1: Permission denied (publickey,gssapi-with-mic,password)

According to SSH protocol (RFC 4252), these are the general authentication message codes

SSH_MSG_USERAUTH_REQUEST            50
SSH_MSG_USERAUTH_FAILURE            51
SSH_MSG_USERAUTH_SUCCESS            52
SSH_MSG_USERAUTH_BANNER             53

It means that the authentiation methods mentioned in the “Permission Denied” was not accepted. What are some of the common issues.

Type 1 (Most Common Mistake): Permission Errors on the .ssh Folder and files inside .ssh

Type 2: Incorrect Configuration Settings on the /etc/ssh/sshd_config
(Assuming you are using Password Authentication) Inside /etc/ssh/sshd_config, you should have something like

PermitRootLogin no
.....
PasswordAuthentication yes
.....
ChallengeResponseAuthentication no
.....
GSSAPIAuthentication yes
GSSAPICleanupCredentials no
.....
UsePAM yes

Type 2: Incorrect Configuration Settings on the /etc/ssh/ssh_config
In Rocky Linux 8, everything should be commented except the last line “Include /etc/ssh/ssh_config.d/*.conf”

References:

  1. How to Fix SSH Failed Permission Denied (publickey,gssapi-keyex,gssapi-with-mic)
  2. ssh receive packet: type 51
by

Building LAMMPS using CMAKE with OpenMPI on Rocky Linux 8

What is LAMMPS (briefly)?

LAMMPS is a classical molecular dynamics code with a focus on materials modeling. It’s an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. More Information on the software, do take a look at https://www.lammps.org/

Where to Download?

You can download the latest stable from Download LAMMPS

Step 1: Ensure Prerequisites are present

Step 2: Download and build LAMMPS

For more information, do take a look at https://docs.lammps.org/Install_tarball.html

$ tar -zxvf lammps-stable.tar.gz
$ cd lammps-2Aug203
$ mkdir build
$ touch setup.sh
$ vim setup.sh

Inside the setup.sh

cmake   -C ../cmake/presets/most.cmake ../cmake             \
        -D CMAKE_INSTALL_PREFIX=/usr/local/lammps-2Aug2023  \
        -D BUILD_MPI=on                                     \
        -D BUILD_SHARED_LIBS=yes                            \
        -D FFT=FFTW3                                        \
        -D FFTW3_INCLUDE_DIRS=${FFTW_INC}                   \
        -D FFTW3_LIBRARIES=${FFTW_LIB}/libfftw3_mpi.a

Notes:
The -C ../cmake/presets/most.cmake command adds the packages that don’t need extra libraries.

Make and Compile……

$ make -j 16 
$ make install

References:

  1. Compiling LAMMPS and using it with Python
  2. Building LAMMPS using CMake
by

Compiling FFTW-3.3.10 with OpenMPI on Rocky Linux 8

For detailed explanation and information, do take a look at FFTW Installation on UNIX. For my installation.

We will be focusing on using Nvidia hpcx only for this blog. To compile Nvidia hpcx, do take a look at Installing and using Mellanox HPC-X Software Toolkit

You may want to module use which come in the hpcx installation

export HPCX_HOME=/usr/local/hpcx-v2.15-gcc-MLNX_OFED_LINUX-5-redhat8-cuda12-gdrcopy2-nccl2.17-x86_64
module use $HPCX_HOME/modulefiles

Next, I used the following parameters that suit my HPC Environment. The default installation is already double-precision. I needed MPI, OPenMPI and needs AVX512….. 

# ./configure --prefix=/usr/local/fftw-3.3.10 --enable-threads --enable-openmp --enable-mpi --enable-avx512
# make && make install

References:

  1. FFTW Installation on UNIX

by

Installing ORCA-5.0.4 on Rocky Linux 8 with OpenMPI

ORCA is a general-purpose quantum chemistry package that is free of charge for academic users. The Project and Download Website can be found at ORCA Forum. The current version is 5.0.4.

The current prerequisites that I have used were OpenMPI-4.1.1 and System GNU which is 8.5.

Unless I have missed something, the packages of ORCA-5.0.4 has been split into 3 different packages which you have to untar and combine together

  • orca_5_0_4_linux_x86-64_openmpi411_part1
  • orca_5_0_4_linux_x86-64_openmpi411_part2
  • orca_5_0_4_linux_x86-64_openmpi411_part3

How do I untar the packages?

The first thing is to untar all the packages separately first. Assuming you are untarring at the /usr/local/

$ tar -xf orca_5_0_4_linux_x86-64_openmpi411_part1.tar.xz
$ tar -xf orca_5_0_4_linux_x86-64_openmpi411_part2.tar.xz
$ tar -xf orca_5_0_4_linux_x86-64_openmpi411_part3.tar.xz

How do I do with all the untarred packages?

Copy all the untar files into /usr/local/orca-5.0.4. 

cp -rv ../orca_5_0_4_linux_x86-64_openmpi411_part1/* .
cp -rv ../orca_5_0_4_linux_x86-64_openmpi411_part2/* .
cp -rv ../orca_5_0_4_linux_x86-64_openmpi411_part3/* .

How to Compile OpenMPI-4.1.1?

Although the Compiling OpenMPI-4.1.5 for ROCEv2 with GNU-8.5 is of a higher version of OpenMPI, the principle and parameters can still be used.

How do I Put them Together?

If you are not using the Module Environment, you can consider installing. For more information do take a look at Installing Environment Modules on Rocky Linux 8.5. All you need to do is then is to load the additional module such as OpenMPI as a prerequisites. Alternatively, you can set the PATH, LD_LIBRARY_PATH of OpenMPI something like this.

export OPENMPI_HOME=/usr/local/openmpi-4.1.1
export PATH=$PATH:$OPENMPI_HOME/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$OPENMPI_HOME/lib:$OPENMPI_HOME/lib64
export MANPATH=$MANPATH:$OPENMPI_HOME/share
export PATH=$PATH:/usr/local/orca-5.0.4

If you are using without Module Environment, you may want to 

orca $INPUT > $OUTPUT

References:

  1. Installing ORCA
by

Errors with sirius-7.5.2 packages during installation

If you are installing based on Installing CP2K with Nvidia HPCX on Rocky Linux 8.5 and if you encounter the issue

.....
Installing from scratch into /usr/local/software/cp2k/tools/toolchain/install/sirius-7.5.2

         for (auto it : unit_cell_.spl_num_paw_atoms()) {

             ^

/usr/local/software/cp2k/tools/toolchain/build/SIRIUS-7.5.2/src/potential/potential.hpp:710:9: error: expected iteration declaration or initialization

         for (auto it : unit_cell_.spl_num_paw_atoms()) {

         ^~~

/usr/local/software/cp2k/tools/toolchain/build/SIRIUS-7.5.2/src/potential/potential.hpp:717:5: warning: no return statement in function returning non-void [-Wreturn-type]

     }
.....

The reason is that the sirius package is installed by default. The issue can be issued if you put the parameters “–with-sirius=no” 

% cd cp2k
% cd /usr/local/software/cp2k/tools/toolchain
% ./install_cp2k_toolchain.sh --no-check-certificate --with-openmpi --with-sirius=no
by

/usr/bin/ld: cannot find -liberty on Rocky Linux 8

I was trying to install CP2K on Rocky Linux 8 using the Installing CP2K with Nvidia HPCX on Rocky Linux 8.5 and I encountered an issue 

libtool: warning: library '/usr/local/hpcx-v2.17.1-gcc-mlnx_ofed-redhat8-cuda12-x86_64/ucc/lib/libucc.la' was moved.
/usr/bin/ld: cannot find -liberty
collect2: error: ld returned 1 exit status

The Resolution can be done by installing binutils-devel

# dnf install binutils-devel
by

Guide to Creating Symbolic Links with Ansible

You can use the ansible.builtin.file module. In my example below, I wanted to link the Module Environment profile.csh and profile.sh to be placed on the /etc/profile.d so that it will load on startup. Do take a look at the Ansible Document ansible.builtin.file module – Manage files and file properties

- name: Check for CUDA Link
  stat: path=/usr/local/cuda
  register: link_available

- name: Create a symbolic link for CUDA
  ansible.builtin.file:
    src: /usr/local/cuda-12.2
    dest: /usr/local/cuda
    owner: root
    group: root
    state: link
  when:
    - ansible_os_family == "RedHat"
    - ansible_distribution_major_version == "8"
    - link_available.stat.isdir is not defined and link_available.stat.isdir == False
by

Unable to Install hdf5, hdf5-devel and hdf5-static on Rocky Linux 8.7

If you are doing a dnf install on hdf5 packages, you will notice errors like the one below

nothing provides libsz.so.2()(64bit) needed by hdf5-1.10.5-4.el8.x86_64
(try to add '--skip-broken' to skip uninstallable packages or '--nobest' to use not only best candidate packages)

To resolve the issue, you will need to install and enable PowerTools

Step 1: Install DNF plugins package

dnf install dnf-plugins-core

Step 2: Install EPEL

The reason is that some software from its source code requires some dependencies that are available in EPEL

dnf install epel-release

Step 3: Enable PowerTools repository on Rocky Linux 8

dnf config-manager --set-enabled powertools

Step 4: Now try installing HDF5

dnf install hdf5 hdf5-devel hdf5-static
by

Setting up 2 Gateways with a Default Gateway for most Traffic and the 2nd Gateway for selected Subnet Traffic on Rocky Linux 8

Issues:

Suppose you have 2 network cards and their own gateway. The challenge is that you can only have 1 default gateway. How do we work this out?

Solution:

Type the following command

$ ip route show
default via 192.168.1.254 dev eno0 proto static metric 104
192.168.2.0/24 via 192.168.2.254 dev eno1 proto static metric 103
10.10.1.0/24 via 192.168.2.254 dev eno1 proto static metric 103

That means the default route for traffic is via eno1. All traffic except 192.168.2.0 and 10.10.1.0 will pass through the second gateway. How do we do it?

Set Default Route for all traffic

To set all traffic through the default gateway, do the following

$ ip route add default via 192.168.1.254 dev eno0 proto static metric 104

Set Selected IP Subnet for 2nd Gateway

$ ip route add 192.168.2.0/24 via 192.168.2.254 dev eno1 proto static metric 103
$ ip route add 10.10.1.0/24 via 192.168.2.254 dev eno1 proto static metric 103

Setting the DNS Correctly for each Network Card

If each of the Network Cards requires a different DNS, do make sure you put in the /etc/sysconfig/network-scripts

$ vim /etc/sysconfig/network-scripts/ifcfg-eno0
....
....
DEVICE=eno0
ONBOOT=yes
IPADDR=192.168.1.1
GATEWAY=192.168.1.254
DNS1=192.168.1.252
DNS2=192.168.1.253
NETMASK=255.255.255.0

$ vim /etc/sysconfig/network-scripts/ifcfg-eno1
....
....
DEVICE=eno1
ONBOOT=yes
IPADDR=192.168.2.1
GATEWAY=192.168.2.254
DNS1=192.168.2.252
DNS2=192.168.2.253
NETMASK=255.255.255.0

Deleting Route from Table

ip route delete 192.168.2.0/24 via 192.168.2.254 dev eno1 proto static metric 103

Different DNS Servers and Different Domains (For RHEL 8)

You can configure dnsmasq service and NetworkManager to send DNS queries for a specific domain to a selected DNS server. The Information can be found in Chapter 38. Using different DNS servers for different domains

By default, Red Hat Enterprise Linux (RHEL) sends all DNS requests to the first DNS server specified in the /etc/resolv.conf file. If this server does not reply, RHEL uses the next server in this file.

In environments where one DNS server cannot resolve all domains, administrators can configure RHEL to send DNS requests for a specific domain to a selected DNS server. For example, you can configure one DNS server to resolve queries for example.com and another DNS server to resolve queries for example.net. For all other DNS requests, RHEL uses the DNS server configured in the connection with the default gateway.

Procedure 1: Install dnsmasq package

# dnf install dnsmasq

Procedure 2: Edit the /etc/NetworkManager/NetworkManager.conf file, and set the following entry in the [main] section:

dns=dnsmasq

Procedure 3: Reload the NetworkManager service:

# systemctl reload NetworkManager

Procedure 4: Verify that the nameserver entry in the /etc/resolv.conf file refers to 127.0.0.53:

# cat /etc/resolv.conf
nameserver 127.0.0.1

Procedure 5a: Verify using TCPDump Packet Sniffer

# dnf install tcpdump

Procedure 5b: On one terminal, start tcpdump to capture DNS traffic on all interfaces:

# tcpdump -i any port 53

Procedure 5c: On a different terminal, resolve host names for a domain for which an exception exists and another domain, for example:

# host -t A www.redhat.com
# host -t A www.MyInternalDomain.com

Verify in the tcpdump output that Rocky Linux sends only DNS queries for the http://www.redhat.com domain to the designated DNS server and through the corresponding interface and vice versa for the Internal Domain

References:

  1. Chapter 38. Using different DNS servers for different domains
  2. Two Default Gateways on One System
  3. Linux Set up Routing with IP Command