#!/usr/bin/env bash
#
# JbTecWiz Support Centre -- generated fix script
#
# Fault : Docker: "no space left on device" with plenty of disk free
# Fix   : Reclaim Docker's storage safely
# Source: https://jbtecwiz.com/support/lnx-docker-space
#
# Run as     : Shell as root
# Expect     : 25 minutes
# Risk       : medium
# Reversible : NO -- read the undo note below
#
# WHEN THIS IS THE RIGHT FIX
#   Docker is out of space. Work from the safest reclaim to the most
#   aggressive and stop when you have enough.
#
# HOW TO UNDO IT
#   None -- pruned images must be pulled again, pruned volumes are gone.
#
# Walks the fix one step at a time and asks before each. Steps with no
# command are yours to do -- it prints those and waits. DRYRUN=1 prints
# without executing; UNATTENDED=1 does not ask.
#
# --------------------------------------------------------------------
# NO WARRANTY - USE AT YOUR OWN RISK
#
# This script is provided by JbTecWiz as-is and with no warranty of any
# kind, express or implied. You run it entirely at your own risk.
#
# JbTecWiz accepts no liability for any loss or damage arising from its
# use, including but not limited to data loss, downtime, or configuration
# changes that turn out to be wrong for your system.
#
# You are responsible for reading this script before running it, for
# satisfying yourself that it suits the machine in front of you, and for
# having a working backup first. Some steps cannot be undone.
# --------------------------------------------------------------------

set -uo pipefail
DRYRUN="${DRYRUN:-0}"
UNATTENDED="${UNATTENDED:-0}"
failed=0

if [ "$(id -u)" -ne 0 ]; then
  echo "  This fix is documented as needing root. Re-run with sudo." >&2
  exit 3
fi

rule() { printf "\n%s\n" "$(printf '-%.0s' $(seq 1 70))"; if [ $# -gt 0 ]; then echo "$1"; fi; }
prose() { echo "$1" | fold -s -w 74 | sed "s/^/  /"; }

# Returns 0 when the caller should run the command, 1 when it should not.
# A manual step always returns 1 -- there is nothing for the caller to run.
step() {  # step <n> <do> <why> <manual|cmd> [command lines...]
  local n="$1" dotext="$2" why="$3" mode="$4"; shift 4
  rule "  Step $n of 7"
  prose "$dotext"
  if [ -n "$why" ]; then echo; prose "$why"; fi
  if [ "$mode" = "manual" ]; then
    echo; echo "  -> Do this yourself, then press Enter to carry on."
    if [ "$UNATTENDED" = "0" ] && [ "$DRYRUN" = "0" ]; then read -r _; fi
    return 1
  fi
  echo; printf "    %s\n" "$@"; echo
  if [ "$DRYRUN" = "1" ]; then echo "  (dry run -- not executed)"; return 1; fi
  if [ "$UNATTENDED" = "0" ]; then
    read -r -p "  Run this step? [Y]es / [S]kip / [Q]uit " a
    case "$a" in
      [Qq]*) echo "  Stopped at your request."; exit 0 ;;
      [Ss]*) echo "  Skipped."; return 1 ;;
    esac
  fi
  return 0
}

rule
echo "  Docker: "no space left on device" with plenty of disk free"
echo "  Reclaim Docker'\''s storage safely"
echo
echo "  Risk: medium   NOT REVERSIBLE   25 minutes"
echo
prose 'No warranty. Use at your own risk - JbTecWiz accepts no liability. Read it before you run it, and have a backup.'
rule
echo
if [ "$UNATTENDED" = "0" ] && [ "$DRYRUN" = "0" ]; then
  read -r -p "  Ready? [y/N] " go
  case "$go" in [Yy]*) ;; *) echo "  Nothing was changed."; exit 0;; esac
fi

if step 1 'See where the space has gone before deleting anything.' '' cmd 'docker system df -v | head -40' 'sudo du -sh /var/lib/docker/*'; then
  docker system df -v | head -40
  sudo du -sh /var/lib/docker/*
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 1 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 2 'Check whether /var/lib/docker is even on the filesystem you think it is.' 'Overlay2 creates enormous numbers of small files, so a Docker host can exhaust inodes while showing free blocks -- check -i as well as -h.' cmd 'df -h /var/lib/docker' 'df -i /var/lib/docker'; then
  df -h /var/lib/docker
  df -i /var/lib/docker
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 2 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 3 'Remove stopped containers, unused networks and dangling images. This is the safe one.' '' cmd 'docker system prune'; then
  docker system prune
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 3 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 4 'Clear the build cache, which is usually the biggest single consumer on a CI host.' '' cmd 'docker builder prune -a'; then
  docker builder prune -a
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 4 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 5 'Remove images not used by any container. Read the list before confirming.' '-a removes every image without a running container, including ones you intend to run later. On a machine with slow internet that is an expensive mistake.' cmd 'docker image prune -a'; then
  docker image prune -a
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 5 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 6 'Volumes last, and only after checking what is in them -- this is where data lives.' 'A dangling volume is one no container currently references. That includes the database volume of a container you stopped ten minutes ago.' cmd 'docker volume ls -f dangling=true' 'docker volume prune'; then
  docker volume ls -f dangling=true
  docker volume prune
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 6 failed. The rest of the fix may depend on it." >&2
  fi
fi

if step 7 'Stop it recurring by capping the log driver, which grows without limit by default.' '' cmd 'printf '\''{\n  "log-driver": "json-file",\n  "log-opts": { "max-size": "50m", "max-file": "3" }\n}\n'\'' | sudo tee /etc/docker/daemon.json' 'sudo systemctl restart docker'; then
  printf '{\n  "log-driver": "json-file",\n  "log-opts": { "max-size": "50m", "max-file": "3" }\n}\n' | sudo tee /etc/docker/daemon.json
  sudo systemctl restart docker
  if [ $? -ne 0 ]; then
    failed=$((failed+1))
    echo "  Step 7 failed. The rest of the fix may depend on it." >&2
  fi
fi

rule "  Confirm it worked"
prose 'Docker reports reclaimed space and pulls succeed.'
if [ "$DRYRUN" = "0" ]; then
  docker system df
  df -h /var/lib/docker
fi

rule
if [ "$failed" -gt 0 ]; then
  echo "  Finished with $failed failed step(s)."
  echo "  Read the full write-up at https://jbtecwiz.com/support/lnx-docker-space"
else
  echo "  Finished."
fi
echo
prose 'To undo: None -- pruned images must be pulled again, pruned volumes are gone.'
rule