Skip to content

Latest commit

 

History

History
141 lines (101 loc) · 3.75 KB

first_install.md

File metadata and controls

141 lines (101 loc) · 3.75 KB

First install

Ubuntu 20.04

Basics

  • sudo apt-get install build-essential for build tools
  • sudo apt-get install net-tools for ip tools
  • Open file manager and
    • add permissions to columns in 'list columns' (under preferences)
    • check 'show hidden files'
    • change to list view
  • Edit .bashrc
    • alias ll='ls -alFG'
    • add shortcut aliases for ssh etc (stored offline)
  • install tweaks for gnome apt install gnome-tweaks

Basic utilities

sudo apt-get install libsecret-1-0 libsecret-1-dev
cd /usr/share/doc/git/contrib/credential/libsecret
sudo make
git config --global credential.helper /usr/share/doc/git/contrib/credential/libsecret/git-credential-libsecret
  • svn sudo apt install subversion
    • check and update ssh settings in .subversion/
    • basic version in dotfiles repo
  • Redshift sudo apt-get install redshift redshift-gtk
    • save config file (see GH)
    • run redshift-gtk to enable toolbar widget and set to autostart
  • tmux

move /home and /srv to new disks

  • use blkid to identify disks (sdc1 - will be /srv and sdb1 - will be home)
  • rm /srv and edit /etc/fstab to add mount point for dev/sdc1
  • add temp mount point for /dev/sdb1 on /mnt
  • copy /home/* to /mnt
  • mv /home -> /home.bak
  • mkdir /home
  • unmount /dev/sdb1
  • edit /etc/fstab/ to point /dev/sdb1 to /home
  • sudo mount -a
  • reboot and check it all holds
  • rm /home.bak

tensorflow - in cuda/python etc

I'm going to use miniconda - conda is designed for different software (so can manage cuda/python/R in the same environment. reticulate plays nicely with it, and miniconda avoids downlowding the bloatware of anaconda.

python/conda setup

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
chmod +x Miniconda3-latest-Linux-x86_64.sh
./Miniconda3-latest-Linux-x86_64.sh

# I don't want to activate the base environment everytime I open a shell
conda config --set auto_activate_base false

Install tensorflow (could it really be this easy)

conda create --name tf_gpu 
activate tf_gpu
conda install tensorflow-gpu

R and Rstudio

sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/'
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9
sudo apt update

sudo apt install r-base r-base-core r-recommended r-base-dev

# to help with spatial packages
sudo apt install libgdal-dev libproj-dev libgeos-dev libudunits2-dev libnode-dev libcairo2-dev libnetcdf-dev
sudo apt install libglu1-mesa-dev freeglut3-dev mesa-common-dev

# aws cli

Following  https://docs.aws.amazon.com/cli/latest/userguide/install-cliv2-linux.html

curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" unzip awscliv2.zip sudo ./aws/install


## aws-cdk

Needs nodejs first

VERSION=v14.15.5 DISTRO=linux-x64 sudo mkdir -p /usr/local/lib/nodejs sudo tar -xJvf node-$VERSION-$DISTRO.tar.xz -C /usr/local/lib/nodejs

The add to .basrc VERSION=v14.15.5 DISTRO=linux-x64 export PATH=/usr/local/lib/nodejs/node-$VERSION-$DISTRO/bin:$PATH

sudo npm -g install aws-cdk


# rsudio

sudo apt install gdebi-core
wget https://download1.rstudio.org/desktop/bionic/amd64/rstudio-1.4.1103-amd64.deb
sudo gdebi rstudio-1.4.1103-amd64.deb

# start up rstudio and run script for package install

Rstudio configs