GitHub/ansitest
1
0
Fork 0
mirror of https://github.com/kneutron/ansitest.git synced 2025-01-16 04:42:55 +08:00
Code Issues Packages Projects Releases Wiki Activity
ansitest/ZFS/zfs-space-calculator-simulator.sh
kneutron b13c16e61f
Add files via upload
2023-05-17 00:03:48 -05:00

360 lines
16 KiB
Bash
Raw Blame History

#!/usr/bin/env bash5
# 2023 Kingneutron
# Simulate a large zpool in GB vs TB to do free-space-available calc rough approximations
# REQUIRES: seq , truncate , sort , grep , sed , column , xargs
# NOTE cd to where you want the output files to be first BEFORE running this, and
# NOTE some filesystems like OSX JHFS+ do NOT support sparse files!
# APFS filesystem should be fine
# PROTIP - Quick variable mods between runs - Example:
# cd /Volumes/sgtera2/zfs-tmpdisks && \
# export numdisks=8 disksize=12 zptype=raidz3 vdevs=1; source zfs-space-calculator-simulator.sh
# zpool destroy zspacecalc # when done with it
# NOTE subsequent consecutive runs will remove the backing-store files, so dont store anything important
# on the temporary zpool!
# failexit.mrg
# REF: https://sharats.me/posts/shell-script-best-practices/
function failexit () {
echo '! Something failed! Code: '"$1 $2" >&2 # code (and optional description)
exit $1
}
# Check for root priviliges
if [ "$(id -u)" -ne 0 ]; then
echo "Please run $0 as root."
exit 1
fi
# xxx TODO EDITME
#numdisks=6
#disksize=18 # GB
# Check for override values on commandline, if run with ' source ' the user can provide vars directly w/o editing
# 24-bay disk shelf should be achievable for homelab, even if all slots not used
# https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2380057.m570.l1313&_nkw=24+disk+sas&_sacat=0
[ "$numdisks" = "" ] && numdisks=24
[ "$disksize" = "" ] && disksize=16 # GB, multiply by ~10 for TB
[ "$zptype" = "" ] && zptype=raidz2
#zptype=raidz2
#zptype=raidz3
#zptype=raidz
#zptype=mirror
[ "$vdevs" = "" ] && vdevs=3
zp=zspacecalc
zpool export -f $zp
[ $(df |grep -c $zp) -gt 0 ] && failexit 101 "$zp still not exported"
[ $(zpool list |grep -c $zp) -gt 0 ] && failexit 101 "$zp still not exported"
# Sparse files
rm -vf zsizetest* # had to remove existing backing-store files in case numdisks went down from previous run
echo "Creating $numdisks zpool disks"
# pad nums
for i in $(seq -w 1 $numdisks); do
echo -n "$i "
truncate -s ${disksize}G zsizetest${i}
done
echo ''
ls -lh z*
# disks per vdev
let dperv=$numdisks/$vdevs
# array
declare -a arr
IFS=$'\n' arr=( $(ls -1 ./z* |sort --human-numeric-sort |xargs -n $dperv) )
# Separate at newline; we need the preceding './' to replace it later with PWD
# Do some string replacing, precede with topology and provide path to backing-store virt disk
let numelements=${#arr[@]}-1 # start at 0
for line in $(seq 0 $numelements); do
arr[$line]="$zptype ${arr[$line]}"
arr[$line]=$(echo ${arr[$line]} |sed "s%./%$PWD/%g")
done
#echo ${arr[*]} # debugg
#echo "Press any key to continue"
#read -n 1
echo "Creating $zp"
set -x
cmd="zpool create -f -o ashift=12 -o autoexpand=on -O atime=off -O compression=lz4 $zp \
$(echo ${arr[*]})"
eval "$cmd"
# ${arr[*]})" # NOTWORK
# Dont ask me why we have to do this insane shit, refused to create it otherwise due to quoting!
# $PWD/zsizetest*
# DONE use driveslicer for vdevs
# no empty lines
zpool status -v $zp |awk 'NF>0'
zpool list -o name,size,alloc,free $zp
zfs list $zp
df -h |egrep "ilesystem|zspace" |column -t
echo "Created zpool $zp , $zptype with $numdisks disks / $vdevs vdevs // $dperv disks per vdev"
echo "Press any key to continue or ^C if you Sourced $0"
read -n 1
exit;
ls |xargs -n 6
zsizetest1 zsizetest10 zsizetest11 zsizetest12 zsizetest13 zsizetest14
zsizetest15 zsizetest16 zsizetest17 zsizetest18 zsizetest19 zsizetest2
.
zsizetest42 zsizetest43 zsizetest44 zsizetest45 zsizetest46 zsizetest47
zsizetest48 zsizetest5 zsizetest6 zsizetest7 zsizetest8 zsizetest9
After pad:
ls |sort --human-numeric-sort |xargs -n 6
zsizetest01 zsizetest02 zsizetest03 zsizetest04 zsizetest05 zsizetest06
.
zsizetest43 zsizetest44 zsizetest45 zsizetest46 zsizetest47 zsizetest48
echo ${ar[0]}
zsizetest01 zsizetest02 zsizetest03 zsizetest04 zsizetest05 zsizetest06
ar[0]="raidz2 ${ar[0]}"
echo ${ar[0]}
raidz2 zsizetest01 zsizetest02 zsizetest03 zsizetest04 zsizetest05 zsizetest06
ar[1]="raidz2 ${ar[1]}"
echo ${ar[*]}
raidz2 zsizetest01 zsizetest02 zsizetest03 zsizetest04 zsizetest05 zsizetest06 raidz2 zsizetest07 zsizetest08 zsizetest09 zsizetest10 zsizetest11 zsizetest12 zsizetest13 zsizetest14 zsizetest15 zsizetest16 zsizetest17 zsizetest18 zsizetest19 zsizetest20 zsizetest21 zsizetest22 zsizetest23 zsizetest24 zsizetest25 zsizetest26 zsizetest27 zsizetest28 zsizetest29 zsizetest30 zsizetest31 zsizetest32 zsizetest33 zsizetest34 zsizetest35 zsizetest36 zsizetest37 zsizetest38 zsizetest39 zsizetest40 zsizetest41 zsizetest42 zsizetest43 zsizetest44 zsizetest45 zsizetest46 zsizetest47 zsizetest48
echo ${arr[0]} |sed 's%./%$PWD/%g'
raidz2 $PWD/zsizetest01 $PWD/zsizetest02 $PWD/zsizetest03 $PWD/zsizetest04 $PWD/zsizetest05 $PWD/zsizetest06
===== Various configs / output results
# NOTE zpool list is all addressable disks with parity,
# zfs list is "usable" space - always take the LOWER number for free space
# 48x16G raidz2, 6 vdevs
+ zpool list zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 762G 1.20M 762G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 717K 525G 205K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 525Gi 205Ki 525Gi 1% 6 1100594534 0% /Volumes/zspacecalc
Created zpool raidz2 zspacecalc with 48 disks / 6 vdevs
# 8 disks per vdev, 2 can fail per vdev or 12 total
# 48x16G raidz1, 6 vdevs = more free space but more risk
+ zpool list zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 764G 816K 764G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 537K 591G 153K /Volumes/zspacecalc
+ df -h |egrep 'ilesystem|zspace' |column -t
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
/dev/disk12s1 591Gi 154Ki 591Gi 1% 6 1239903502 0% /Volumes/zspacecalc
Created zpool raidz zspacecalc with 48 disks / 6 vdevs
# 48x16G raidz3, 4 vdevs, more available space than ^ raidz2 with 6 vdevs (538G vs 525G)
# I/O will likely be slower due to parity
+ zpool list zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 764G 1.64M 764G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 976K 538G 279K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 538Gi 279Ki 538Gi 1% 6 1127734972 0% /Volumes/zspacecalc
Created zpool raidz3 zspacecalc with 48 disks / 4 vdevs
# 3/12 disks per vdev can fail (18 total for pool) with no dataloss
# 48x16G mirror, 24 vdevs - low amount of free space per # of disks used
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 372G 456K 372G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 372K 360G 96K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
/dev/disk12s1 360Gi 96Ki 360Gi 1% 6 755760264 0% /Volumes/zspacecalc
Created zpool zspacecalc , mirror with 48 disks / 24 vdevs
# 48x16G triple mirror, 3x16 vdevs with 16G/ea, waste of space but good survivability - 2/3 disks per vdev can fail
# Might be good for hospital / 99% uptime requirement
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 248G 420K 248G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 348K 240G 96K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 240Gi 96Ki 240Gi 1% 6 503839952 0% /Volumes/zspacecalc
Created zpool zspacecalc , mirror with 48 disks / 16 vdevs
mirror-15 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest46 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest47 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest48 ONLINE 0 0 0
# Same config as triple mirror, 3x/vdev but raidz = LOTS more free space
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 760G 840K 760G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 464K 490G 128K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 490Gi 128Ki 490Gi 1% 6 1028345793 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz with 48 disks / 16 vdevs
raidz1-15 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest46 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest47 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest48 ONLINE 0 0 0
# raidz2, 24x16G disks, 8 disks per vdev, achievable for homelab with proper budget
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 381G 1.18M 381G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 691K 262G 205K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 262Gi 205Ki 262Gi 1% 6 550296474 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz2 with 24 disks / 3 vdevs
raidz2-2 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest17 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest18 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest19 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest20 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest21 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest22 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest23 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest24 ONLINE 0 0 0
# For ~100TB of "usable space" pool and hopefully decent I/O with 2 vdevs:
# 14x12GB raidz2, 7/vdev, up to (4) disks in the pool can fail (2 per vdev) with no dataloss
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 167G 1.12M 167G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 623K 108G 192K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
/dev/disk12s1 108Gi 192Ki 108Gi 1% 6 225964209 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz2 with 14 disks / 2 vdevs
raidz2-1 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest08 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest09 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest10 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest11 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest12 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest13 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest14 ONLINE 0 0 0
# For ~100TB of "usable space" with larger disks - but slower i/o due to only 1 vdev
# 9x16GB raidz2 , 2/9 disks can fail
# Probably suitable for a "backup" pool
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 143G 1.12M 143G
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 713K 106G 219K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
/dev/disk12s1 106Gi 220Ki 106Gi 1% 6 221293659 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz2 with 9 disks / 1 vdevs
raidz2-0 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest1 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest2 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest3 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest4 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest5 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest6 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest7 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest8 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest9 ONLINE 0 0 0
# If you really want to run a "Petabyte" pool
# numdisks=72; disksize=20; zptype=raidz2; vdevs=8
# 2/9 disks per vdev can fail, or up to (16) total
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 1.40T 1.27M 1.40T
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 823K 1.03T 219K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
/dev/disk12s1 1.0Ti 216Ki 1.0Ti 1% 6 277005840 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz2 with 72 disks / 8 vdevs
raidz2-7 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest64 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest65 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest66 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest67 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest68 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest69 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest70 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest71 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest72 ONLINE 0 0 0
# same "petabyte" config but with RAIDZ3:
# numdisks=72; disksize=20; zptype=raidz3; vdevs=6
# Up to 3/12 disks per vdev can fail, or (18) total for the pool with no dataloss
+ zpool list -o name,size,alloc,free zspacecalc
NAME SIZE ALLOC FREE
zspacecalc 1.39T 1.69M 1.39T
+ zfs list zspacecalc
NAME USED AVAIL REFER MOUNTPOINT
zspacecalc 1.02M 1005G 279K /Volumes/zspacecalc
+ df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted
zspacecalc 1.0Ti 279Ki 1.0Ti 1% 6 2107862646 0% /Volumes/zspacecalc
Created zpool zspacecalc , raidz3 with 72 disks / 6 vdevs
raidz3-5 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest61 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest62 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest63 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest64 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest65 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest66 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest67 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest68 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest69 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest70 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest71 ONLINE 0 0 0
/Volumes/sgtera2/zfs-tmpdisks/zsizetest72 ONLINE 0 0 0
Reference in New Issue View Git Blame Copy Permalink