Base software kit

These software packages should be installed on every machine we provision, regardless of their purpose. A script on GitLab has a dirt-simple script to do this, though it should be improved and/or developed (including making an Ansible playbook for easy install on a bunch of cluster nodes).

• nano - text editor
• vim - text editor
• emacs - text editor
• nfs-common for Debian or nfs-utils for CentOS - allow NFS mounts
• wget - wget a file from the Internet
• iproute2 - more modern instantiations of ifconfig and a lot of other network tools
  • net-tools for those among us who like ifconfig
• gcc - most recent C compiler (modules can support as many versions as we'd like but we need at least one)
• ntp - get a time server
• modules
• rsync
• rclone
• openldap
• openldap-clients
• openssh-ldap
• pam_ldap
• nss-pam-ldapd
• unattended-upgrades or yum-conf or dnf-automatic (also configure to automatically apply security patches)

On a high-performance computing server or cluster, add these:

• C3 tools (see the C3 tools page) - possibly deprecated, Ansible is a good replacement
• Slurm
• OpenMPI (latest version)

If the server will be used for bioinformatics, add these:

• ea-utils
• AdapterRemoval
• sga
• BBMapTool
• bowtie2
• samtools
• NCBI mitochondrial and plastid databases (full URLs, they’re stable)
• Python scripts to subset databases and SAM files by species
• PMD tools
• mapdamage2

If the server will host surveying software, add these:

• Docker
• ODM
• WebODM

Python libraries (Python 3):

• Pandas
• Numpy
• Scipy