Docker garbage collection based on when an image was cached/downloaded

docker-gc

#!/bin/bash

# Root used by Docker runtime
ROOT="/var/lib/docker"

# Device for Docker root
#ROOT_DEV="/dev/vg_root/lv_var"
ROOT_DEV=$(df -T ${ROOT} | awk '/dev/ { print $1 }')

# Docker thinpool device
# POOL_DEV="/dev/vg_root/docker-pool"
POOL_LV=$(lvs --read-only | awk '/docker/ { print $1 }')
POOL_VG=$(lvs --read-only | awk '/docker/ { print $2 }')
POOL_DEV="/dev/${POOL_VG}/${POOL_LV}"

# Cleanup images older then x sec.
OLDER_THEN=$((7 * 24 * 60 * 60)) # 1 week

# High-water mark
HWMARK="90"

# Cleanup images older then x sec. once high-water mark is reached
HWMARK_OLDER_THEN=$((24 * 60 * 60)) # 1 day

NOW=$(date -u +"%s")

# Parse date to sec.
parse_date() {
    echo $(date -u --date "${1}" "+%s")
}

# Time since in sec.
time_since() {
    before=$(parse_date "${1}")
    echo $((${NOW} - ${before}))
}

# Human readable time
time_hr() {
    local T=${1}
    local D=$((T/60/60/24))
    local H=$((T/60/60%24))
    local M=$((T/60%60))
    local S=$((T%60))
    (( $D > 0 )) && printf '%d days ' $D
    (( $H > 0 )) && printf '%d hours ' $H
    (( $M > 0 )) && printf '%d minutes ' $M
    (( $D > 0 || $H > 0 || $M > 0 )) && printf 'and '
    printf '%d seconds\n' $S
}

# Cleanup images older then x sec.
gc_images_older_then() {
    local older_then=${1}

    images=""
    short_images=""
    for id in $(docker images -q --no-trunc); do
        inode=$(stat --format '%i' ${ROOT}/devicemapper/metadata/${id})
        ctime=$(debugfs -R "stat <${inode}>" ${ROOT_DEV} 2>/dev/null | awk -F '-- ' '/ctime:/ { print $2 }')
        elapsed=$(time_since "${ctime}")
	short_id=${id:0:12}
        [[ ${elapsed} > ${older_then} ]] && images="${images} ${id}" && short_images="${short_images} ${short_id}"
        echo -n "."
    done
    echo
    if [ -n "${images}" ]; then
        log "Cleanup images older then $(time_hr ${older_then}), image id(s): ${short_images}"
        docker rmi $images
    fi
}

# Log to syslog
log() {
    logger -s -t $(basename $0) $1
}

log "Started"

# Cleanup stopped containers
containers=$(docker ps -a -q -f "status=exited")
if [ -n "${containers}" ]; then
    log "Cleanup stopped container id(s): ${containers}"
    docker rm $containers
fi

# Cleanup dangling images
images=$(docker images -q --filter "dangling=true")
if [ -n "${images}" ]; then
    log "Cleanup dangling image id(s): ${images}"
    docker rmi $images
fi

# Cleanup images older then x sec.
gc_images_older_then ${OLDER_THEN}

usage=$(lvs ${POOL_DEV} | tail -1 | awk '{ print $5 }')
if [[ ${usage%.*} -lt 90 ]]; then
    log "Done"
    exit 0
fi

log "High-water mark ${HWMARK}% for docker-pool reached"

# Cleanup images older then x sec.
gc_images_older_then ${HWMARK_OLDER_THEN}

usage=$(lvs ${POOL_DEV} | tail -1 | awk '{ print $5 }')
if [[ ${usage%.*} -lt 90 ]]; then
    log "Done"
    exit 0
fi

log "Still reaching high-water mark ${HWMARK}%"

# Cleanup all cached images
images=$(docker images -q)
if [ -n "${images}" ]; then
    log "Cleanup all cached image id(s): ${images}"
    docker rmi $images
fi

log "Done"