VDEV Rebalancing

[NickF]

Hi All,

My datasets have grown to the point where doing send/receive is no longer a viable option for me to easily rebalance the data on my pool.

Several others have recommended this script

GitHub - markusressel/zfs-inplace-rebalancing: Simple bash script to rebalance pool data between all mirrors when adding vdevs to a pool.

Simple bash script to rebalance pool data between all mirrors when adding vdevs to a pool. - markusressel/zfs-inplace-rebalancing

github.com

But it seems a missing prerequisite breaks it in Scale (but it reportedly present in CORE)

Is there another way to accomplish what my goal of rebalancing my VDEVs? They are pretty whacked out right now, as I've added two new VDEVs in the past year or so.

Commenting out lines 60 and 61 seems to get the script to run, but I lose the telemetry of how far it's gotten:

PLEASE SEE MY LATER REPLY.

 

[Arwen]

The missing command is:

bc - An arbitrary precision calculator language

Debian, (used by TrueNAS SCALE), uses a different command to install a new package. But, TrueNAS SCALE is considered an appliance, thus you are discouraged from installing random software.

However, this might be a useful addition to SCALE, as Core already has bc. Please submit a feature request. (aka "Report a Bug" at the top of the forums). SCALE is under heavy development and adding such a simple package to the base OS might be worth while.

 

[crkinard]

I ran this script from inside my Plex docker. It already had RW permissions to a bulk of the data on my NAS. I just logged into the shell, apt-get installed bc and a few outer dependencies, then ran the script.

Once done I just restarted the plex pod to clear out anything I added.

 

[NickF]

This is the fixed script with working progress.

Code:

#!/usr/bin/env bash

# exit script on error
set -e
# exit on undeclared variable
set -u

# file used to track processed files
rebalance_db_file_name="rebalance_db.txt"

# index used for progress
current_index=0

## Color Constants

# Reset
Color_Off='\033[0m'       # Text Reset

# Regular Colors
Red='\033[0;31m'          # Red
Green='\033[0;32m'        # Green
Yellow='\033[0;33m'       # Yellow
Cyan='\033[0;36m'         # Cyan

## Functions

# print a help message
function print_usage() {
  echo "Usage: zfs-inplace-rebalancing --checksum true --skip-hardlinks false --passes 1 /my/pool"
}

# print a given text entirely in a given color
function color_echo () {
    color=$1
    text=$2
    echo -e "${color}${text}${Color_Off}"
}


function get_rebalance_count () {
    file_path=$1

    line_nr=$(grep -xF -n "${file_path}" "./${rebalance_db_file_name}" | head -n 1 | cut -d: -f1)
    if [ -z "${line_nr}" ]; then
        echo "0"
        return
    else
        rebalance_count_line_nr="$((line_nr + 1))"
        rebalance_count=$(awk "NR == ${rebalance_count_line_nr}" "./${rebalance_db_file_name}")
        echo "${rebalance_count}"
        return
    fi
}

# rebalance a specific file
function rebalance () {
    file_path=$1

    # check if file has >=2 links in the case of --skip-hardlinks
    # this shouldn't be needed in the typical case of `find` only finding files with links == 1
    # but this can run for a long time, so it's good to double check if something changed
    if [[ "${skip_hardlinks_flag,,}" == "true"* ]]; then
        hardlink_count=$(stat -c "%h" "${file_path}")

        if [ "${hardlink_count}" -ge 2 ]; then
            echo "Skipping hard-linked file: ${file_path}"
            return
        fi
    fi

    current_index="$((current_index + 1))"
    progress_percent=$((current_index * 100 / file_count))
    color_echo "${Cyan}" "Progress -- Files: ${current_index}/${file_count} (${progress_percent}%)"

    if [[ ! -f "${file_path}" ]]; then
        color_echo "${Yellow}" "File is missing, skipping: ${file_path}"
    fi

    if [ "${passes_flag}" -ge 1 ]; then
        # check if target rebalance count is reached
        rebalance_count=$(get_rebalance_count "${file_path}")
        if [ "${rebalance_count}" -ge "${passes_flag}" ]; then
        color_echo "${Yellow}" "Rebalance count (${passes_flag}) reached, skipping: ${file_path}"
        return
        fi
    fi
 
    tmp_extension=".balance"
    tmp_file_path="${file_path}${tmp_extension}"

    echo "Copying '${file_path}' to '${tmp_file_path}'..."
    if [[ "${OSTYPE,,}" == "linux-gnu"* ]]; then
        # Linux

        # -a -- keep attributes
        # -d -- keep symlinks (dont copy target)
        # -x -- stay on one system
        # -p -- preserve ACLs too
        cp -adxp "${file_path}" "${tmp_file_path}"
    elif [[ "${OSTYPE,,}" == "darwin"* ]] || [[ "${OSTYPE,,}" == "freebsd"* ]]; then
        # Mac OS
        # FreeBSD

        # -a -- Archive mode.  Same as -RpP.
        # -x -- File system mount points are not traversed.
        # -p -- Cause cp to preserve the following attributes of each source file
        #       in the copy: modification time, access time, file flags, file mode,
        #       ACL, user ID, and group ID, as allowed by permissions.
        cp -axp "${file_path}" "${tmp_file_path}"
    else
        echo "Unsupported OS type: $OSTYPE"
        exit 1
    fi

    # compare copy against original to make sure nothing went wrong
    if [[ "${checksum_flag,,}" == "true"* ]]; then
        echo "Comparing copy against original..."
        if [[ "${OSTYPE,,}" == "linux-gnu"* ]]; then
            # Linux

            # file attributes
            original_md5=$(lsattr "${file_path}" | awk '{print $1}')
            # file permissions, owner, group
            # shellcheck disable=SC2012
            original_md5="${original_md5} $(ls -lha "${file_path}" | awk '{print $1 " " $3 " " $4}')"
            # file content
            original_md5="${original_md5} $(md5sum -b "${file_path}" | awk '{print $1}')"

            # file attributes
            copy_md5=$(lsattr "${tmp_file_path}" | awk '{print $1}')
            # file permissions, owner, group
            # shellcheck disable=SC2012
            copy_md5="${copy_md5} $(ls -lha "${tmp_file_path}" | awk '{print $1 " " $3 " " $4}')"
            # file content
            copy_md5="${copy_md5} $(md5sum -b "${tmp_file_path}" | awk '{print $1}')"
        elif [[ "${OSTYPE,,}" == "darwin"* ]] || [[ "${OSTYPE,,}" == "freebsd"* ]]; then
            # Mac OS
            # FreeBSD

            # file attributes
            original_md5=$(lsattr "${file_path}" | awk '{print $1}')
            # file permissions, owner, group
            # shellcheck disable=SC2012
            original_md5="${original_md5} $(ls -lha "${file_path}" | awk '{print $1 " " $3 " " $4}')"
            # file content
            original_md5="${original_md5} $(md5 -q "${file_path}")"

            # file attributes
            copy_md5=$(lsattr "${tmp_file_path}" | awk '{print $1}')
            # file permissions, owner, group
            # shellcheck disable=SC2012
            copy_md5="${copy_md5} $(ls -lha "${tmp_file_path}" | awk '{print $1 " " $3 " " $4}')"
            # file content
            copy_md5="${copy_md5} $(md5 -q "${tmp_file_path}")"
        else
            echo "Unsupported OS type: $OSTYPE"
            exit 1
        fi

        if [[ "${original_md5}" == "${copy_md5}"* ]]; then
            color_echo "${Green}" "MD5 OK"
        else
            color_echo "${Red}" "MD5 FAILED: ${original_md5} != ${copy_md5}"
            exit 1
        fi
    fi

    echo "Removing original '${file_path}'..."
    rm "${file_path}"

    echo "Renaming temporary copy to original '${file_path}'..."
    mv "${tmp_file_path}" "${file_path}"

    if [ "${passes_flag}" -ge 1 ]; then
        # update rebalance "database"
        line_nr=$(grep -xF -n "${file_path}" "./${rebalance_db_file_name}" | head -n 1 | cut -d: -f1)
        if [ -z "${line_nr}" ]; then
        rebalance_count=1
        echo "${file_path}" >> "./${rebalance_db_file_name}"
        echo "${rebalance_count}" >> "./${rebalance_db_file_name}"
        else
        rebalance_count_line_nr="$((line_nr + 1))"
        rebalance_count="$((rebalance_count + 1))"
        sed -i "${rebalance_count_line_nr}s/.*/${rebalance_count}/" "./${rebalance_db_file_name}"
        fi
    fi
}

checksum_flag='true'
skip_hardlinks_flag='false'
passes_flag='1'

if [[ "$#" -eq 0 ]]; then
    print_usage
    exit 0
fi

while true ; do
    case "$1" in
        -h | --help )
            print_usage
            exit 0
        ;;
        -c | --checksum )
            if [[ "$2" == 1 || "$2" =~ (on|true|yes) ]]; then
                checksum_flag="true"
            else
                checksum_flag="false"
            fi
            shift 2
        ;;
        --skip-hardlinks )
            if [[ "$2" == 1 || "$2" =~ (on|true|yes) ]]; then
                skip_hardlinks_flag="true"
            else
                skip_hardlinks_flag="false"
            fi
            shift 2
        ;;
        -p | --passes )
            passes_flag=$2
            shift 2
        ;;
        *)
            break
        ;;
    esac
done;

root_path=$1

color_echo "$Cyan" "Start rebalancing:"
color_echo "$Cyan" "  Path: ${root_path}"
color_echo "$Cyan" "  Rebalancing Passes: ${passes_flag}"
color_echo "$Cyan" "  Use Checksum: ${checksum_flag}"
color_echo "$Cyan" "  Skip Hardlinks: ${skip_hardlinks_flag}"

# count files
if [[ "${skip_hardlinks_flag,,}" == "true"* ]]; then
    file_count=$(find "${root_path}" -type f -links 1 | wc -l)
else
    file_count=$(find "${root_path}" -type f | wc -l)
fi

color_echo "$Cyan" "  File count: ${file_count}"

# create db file
if [ "${passes_flag}" -ge 1 ]; then
    touch "./${rebalance_db_file_name}"
fi

# recursively scan through files and execute "rebalance" procedure
# in the case of --skip-hardlinks, only find files with links == 1
if [[ "${skip_hardlinks_flag,,}" == "true"* ]]; then
    find "$root_path" -type f -links 1 -print0 | while IFS= read -r -d '' file; do rebalance "$file"; done
else
    find "$root_path" -type f -print0 | while IFS= read -r -d '' file; do rebalance "$file"; done
fi

echo ""
echo ""
color_echo "$Green" "Done!"

Replaced line 72 with:

Code:

progress_percent=$((current_index * 100 / file_count))

Which was:

Code:

progress_percent=$(echo "scale=2; ${current_index}*100/${file_count}" | bc)

Commented on existing PR:

remove bc dependency since TrueNAS Scale doesn't include it. by StraightfaceStudios · Pull Request #20 · markusressel/zfs-inplace-rebalancing

Simple bash script to rebalance pool data between all mirrors when adding vdevs to a pool. - remove bc dependency since TrueNAS Scale doesn't include it. by StraightfaceStudios · Pull Request #20 · markusressel/zfs-inplace-rebalancing

github.com

Hope this helps someone.

 

[Brandito]

Helped me, even though I never bothered with the original script, this modification appears to work as expected on my scale install.

 

[vaschthestampede]

Since the indicated script doesn't work with dedup enabled, can you give me an alternative?

EDIT:
In my specific case I need rebalancing to rewrite the dedump tables in the dedicated SSD disks instead of the spinning disks since I ran out of space.