cheatsheet.md

cheatsheet

• you can keep this super handy by adding this alias:
  • alias cheatsheet="less ~path/to/cheatsheet.md
• get external ip:
  • curl http://ipecho.net/plain
  • Or, this oneliner (put in an alias): wget -q -O - http://zacanger.com/ip && echo " "
• Getting Help:
  • View the manual for a command: man command
  • Get help without using man: command --help (or, frequently, command -h)
  • Can't remember the name of a command, but you know a relevant word? apropos word
  • Index of help pages: info
• Basic file and directory operations:
  • pwd shows your current directory
  • ls shows files in your current directory
    • ls -a shows all files (including hidden)
    • ls -R shows files recursively
    • ls -lt shows files by modification time (most recent first)
    • mv moves a file or directory (also for renaming), eg mv foo bar or mv ~/bin ~/Dropbox/bin
    • rm removes a file (this is permanent)
      • rmdir or rm -r will remove a directory, recursively
      • npm i -g trash-cli empty-trash-cli gives you a really useful trash in the command line
        • 'alias rm='trash' and alias erm='empty-trash' are two aliases I use for safety
    • cp file copies file
      • cp -r directory copies directory, recursively
    • mount /dev/device/name /media/device/name mounts a filesystem
      • unmount /media/device/name unmounts it
    • dd clones filesystems
      • It's probably best not to use dd until you really know it well
    • parted, fdisk, cfdisk, and mkfs are tools for working with filesystems and partitions
      • Again, probably best not to touch these until you're sure you know what's up
• Administration:
  • sudo foo executes foo as a temporary administrator account
    • su or sudo su or sudo -s lets you become root (administrator)
      • exit exits that business
    • sudo !! executes previous command with sudo
  • Installing things from source:
    • Unpack the archive (I highly recommend the utilities atool or unp, because tar commands are ridiculous)
    • cd into the unpacked directory
    • less README (or cat README) -- always check for a README! (and an INSTALL file, if one exists!)
    • There may be a script to run that will generate some files you'll need, ./autogen.sh
    • ./configure checks for appropriate configs and generates a Makefile (if there's a configure file
    • make compiles
      • make install will install
      • alias makelist="make -rpn | sed -n -e '/^$/ { n ; /^[^ .#][^ ]*:/p ; }' | egrep --color '^[^ ]*:'" is a useful alias for listing all make targets
      • make clean cleans up all the miscellaneous cruft from the configure and make processes
  • Installing things using package managers:
    • Debian (and its descendants):
      • sudo apt-get update updates information on available packages
      • sudo apt-get upgrade upgrades out-of-date things
      • apt-cache search word searches for 'word' in available packages
      • apt-cache show packagename shows details on 'packagename'
      • sudo apt-get install packagename installs packagename
      • sudo apt-get remove packagename uninstalls packagename
      • dpkg --get-selections shows currently installed packages
      • sudo add-apt-repository adds a PPA (Ubuntu)
      • sudo dpkg -i file.deb installs file.deb
        • sudo dpkg -r programname uninstalls programname
    • Python:
      • https://bootstrap.pypa.io/ Download these three, and run them with python3 thing.py
      • pip install -U pip installs pip (or updates)
      • pip search pip searches for pip
      • virtualenv dirname creates a virtual environment
        • source dirname/bin/activate connects to that venv
        • deactivate disconnects
        • pip install pip==versionnumber -E dirname installs pip at version number into venv
        • pip freeze -E dirname > requirements.txt exports venv info into shareable format
        • pip install -r requirements.txt installs from requirements.txt
        • pip install -E dirname -r requirements.txt imports venv from requirements.txt
    • Node (NPM)
• Basic commands:
  • thing | less views the output of thing in a paged format
    • less filename views that filename in a paged format
    • G jumps to the bottom of the view
    • g jumps to the top
    • / starts a search
    • more and pg are two earlier pagers
    • pager will start whichever one happens to be your default
    • most is a very different alternative, without the vi-like keybinds, which can also view binary files easily
  • cat file will print the contents of file to the terminal
  • locate filename will search (everywhere) for that filename
    • sudo updatedb will update locate's database (this is usually set as a daily cronjob by default)
  • which programname will show the location of an executable in your PATH
  • grep query will search everything below cwd for that query
    • grep query filename will search only in the file filename
    • command | grep query will search the output of command for query
    • ack is a popular alternative to grep
    • grep itself has a few aliases usually built in: egrep, fgrep, maybe vgrep
    • ag (AG, The Silver Searcher) is the grep to be using:
      • Less characters to type
      • Better highlighting, colour options, and defaults
      • MUCH faster than grep (or ack)
      • Plus the guy who wrote it is just a really swell person
      • https://github.com/ggreer/the_silver_searcher
  • ps -e lists all running processes
    • ps aux | sort -nk +4 | tail lists the top ten processes by memory usage
    • echo " "; ps -eo pcpu,pid,user,args | sort -rk1 | head -6 | column -t; echo " "; lists the top five processes by CPU usage
  • top is an interactive system monitor
    • htop is a much nicer alternative:
      • colour
      • settings
      • much cleaner
      • more friendly keys
      • http://hisham.hm/htop/
    • npm i -g vtop will install a (javascript) version
      • very much simplified
      • nifty little graphs
  • pkill processname kills processname
    • pkill -NUM name kills name using the signal NUM
      • pkill -9 foo kills foo with SIGKILL (kill it dead, right now, do it!)
      • pkill -15 foo kills foo a little bit more gently, allowing time for cleanup, etc.
  • renice process stops the processes from hogging all resources and making computer lag
  • command & starts command in the background (so you don't need to waste a terminal on it)
    • nohup command & starts command in the background and keeps it running after you've logged off
  • tar is a mess. you could memorize all of its commands and flags, or...
    • you could use unp, which is in Python, and is as simple as typing unp archivename.extension
      • https://github.com/mitsuhiko/unp
    • you could use atool (which I recommend):
      • aunpack foo.ext unpacks the archive foo.ext
      • als foo.ext lists all its files
      • apack to make an archive
      • http://www.nongnu.org/atool/
  • gpg -o outputname.gpg -c targetfile to encrypt a file
    • gpg -o outputname -d target.gpg to decrypt a file
• Shell things (mostly Bash, but Zsh is usually pretty compatible):
  • Directories:
    • ~/ means user's home
    • ./ means cwd (current working directory)
    • ../ is parent directory
    • ../../../../../ means the great-great-great-grandparent dierctory (Etc)
    • * means all files in current directory (use caution, always)
  • Output redirects:
    • | is the heart of UNIX.
      • foo | bar sends the output of foo to the input of bar
      • foo | bar | quux > baz sends the output of foo into bar, out to quux, and then writes that output to baz
    • things > stuff sends the output of things to a file, stuff (warning, this WILL overwrite the content of stuff)
      • blah >> stuff will append the output of blah to stuff (so, it won't overwrite)
    • tee works like |, but it also writes its output to the terminal (it's a T shaped output!)
      • so cat foo.txt | sort | uniq tee bar.txt will cat all of foo.txt, sort it (alphabetically), remove duplicate lines, and send the output to both bar.txt and the terminal
  • ;:
    • one ; two will execute one fully, then execute 2
  • & will execute simultaneously
    • && will only execute if the previous command was completed successfully
  • || only executes if the previous command completed unsuccessfully
  • * matches zero or more characters
    • ? matches any one character
    • [chars] matches any characters inside brackets
    • [a-z] matches any characters inside that range
• Some more basic utilities:
  • ifconfig and iwconfig to configure network interfaces and wireless network interfaces (specifically)
  • ssh username@ipaddress to connect to a server
    • ssh -X username@ipaddress to forward X (the display server) from the target to your current machine
  • scp -r sourcefilename:user@ip targetfilename:targetuser@targetip copies files/directories from one machine to another (recursively)
  • rsync source target copies changes between files/directories (very useful for syncing, can work remotely like ssh)
  • ping address checks if address (which can be an IP address or a domain) is available, and reports response times
  • traceroute ipaddress views the full network route to ipaddress
  • iptables -L shows firewall rules
  • nmap localhost shows open ports on localhost
  • wget http://zacanger.com/ downloads http://zacanger.com/
    • wget -c completes a partial download
    • -b runs in the background
    • -ftp-user=username --ftp-password=password ftp://example.com/directory/file downloads over FTP
    • --mirror does a full-on mirroring download
    • -i file reads urls in file as a list to download
  • netcat listens for input from the network, and dumps to a file
  • chown username file changes the owner of a file (may need to be run as root, depending on the file)
    • chown -R username directory for directories, recursively
  • chmod changes file permissions
    • chmod +x makes a file executable
    • chmod -x makes it not executable
  • users shows current users
    • adduser adds a user
    • usermod changes user priveleges
    • deluser removes a user
  • groups shows all user groups
    • groupadd creates a new group
    • groupmod changes priveleges
    • delgroup removes group
  • lsof shows what processes are using what files
  • diff shows the differences between two files
  • head -n NUM file shows the top NUM lines of file
  • tail -n NUM file shows the last NUM lines of file
  • md5sum file and md5deep directory checksum file/all in directory
  • sha1sum and sha1deep are the same, but with sha1
  • watch -d -n NUM command calls command every NUM of seconds, shows difference in output
  • time command executes command and shows how long it took
  • du -a directory | sort -n -r | less shows files in directory from largest to smallest
  • date shows date and time
  • dmidecode shows motherboard info
    • inxi shows system info in a much more friendly format
      • https://github.com/smxi/inxi
  • echo $HOSTNAME shows current hosname
  • lsb release -a shows current (Linux) distro info
    • cat /etc/issue does similar
    • uname -a shows (Linux) system/kernel info
    • lsmod shows info about kernel modules
    • modprobe to configure kernel modules (caution!)
    • printenv shows environment variables
      • (so does env on most systems)
    • lspci shows PCI-connected hardware
      • lsusb shows USB-connected hardware
    • dumpcap dumps wireless card data
    • dumpkeys dumps keyboard driver data
• Git:
  • git init creates new repo (project)
  • git config user.name "username"
  • git config user.email "email"
  • git clone target clones to local
  • git commit -m "message" commits with message
  • git status shows curernt status
  • git log shows full log
  • git pull
  • git push
  • git branch creates
  • git checkout branchname switches to branchname
  • git merge branchname merges branchname
  • pip install -U legit will give you some fantastic shortcuts
    • legit help and legit install to get started
  • npm i -g gh gives you some nifty Github specific commands (try gh to get started)
  • hub is Github's official client
    • Adds some Github-specific things to Git, like:
      • Pull Requests
      • Issues
      • Cloning from Github with the shorthand hub clone user/repo
      • https://github.com/github/hub/releases
      • or, brew install hub (on a Mac)
  • See my .gitignore_global
  • See my .gitconfig
  • Much, much more
• MySQL:
  • help
  • show database shows databases
    • use databasename
    • show tables shows schemas
    • DROP DATABASE databasename
    • CREATE DATABASE databasename
    • CREATE USER username@hostname IDENTIFIED BY 'password';
    • select * from mysql.user; shows users
    • delete from mysql.user WHERE User='username';
    • mysqldump databasename > dumpfilename.txt exports text with commands to rebuild all tables
    • mysql -u username -p < dumpfilename.txt restores from a dump
    • mysqldump -u username -p --opt databasename > dumpfilename.sql dumps the entire database
    • mysql -u username -p --database=databasename < dumpfilename.sql restores from the full dump