fs

Compare

https://en.wikipedia.org/wiki/Comparison_of_file_systems

File system
Host OS
Online grow
Offline grow
Online shrink
Offline shrink

FAT32(X)

misc.

No

3rd-party

No

3rd-party

NTFS

Windows

Yes

Yes

Yes

Yes

ReFS

Windows

Yes

?

No

?

Btrfs[51]

Linux

Yes

No

Yes

No

ext4[52]

Linux

Yes

Yes

No

Yes

HFS+

Linux

No

No

No

No

HFS+

macOS

Yes

No

Yes

No

APFS

macOS

?

?

?

?

ZFS

misc.

Yes

No

No

No

ReiserFS[59]

Linux

Yes

Yes

No

Yes

XFS[60]

Linux

Yes

No

No

No

Check info

lshw -short -C disk
lshw -class disk -class storage
lsblk -f # check file system type
tune2fs -l /dev/sda1 | grep -i count

gdisk -l /dev/sda  #fdisk many give wrong GPT partiton

# file -s /dev/vda
    /dev/vda: DOS/MBR boot sector
# file -s /dev/vda1
    /dev/vda1: Linux rev 1.0 ext4 filesystem data, UUID=... (needs journal recovery) (extents) (large files) (huge files)

Convert between MBR and GPT

sudo sgdisk -g /dev/sda
sudo sgdisk -m /dev/sda
sudo partprobe -s

mkpart, format

https://wiki.archlinux.org/index.php/Parted

parted -s /dev/sdb mklabel gpt
parted -s /dev/sdb unit mib mkpart primary 0% 100% / 1MiB 512MiB

mkpart primary fat32 1MiB 512MiB
set 1 esp on

mkfs.ext4 /dev/sdb1

mount/umount

mount -o rw,remount /   # recovery
umount -l /PATH/OF/BUSY-DEVICE

mount -o loop,ro $path $mount_point # ISO

mount.nfs $1:$2 $3
mount -tnfs4 -ominorversion=1 server_nfs_4.1:/dir\

echo $path $mount_point cifs username=$user,password=$passwd 0 0 >>  /etc/fstab # SMB

fstab

Loss NAS mount may cause hang!

cat /proc/mounts    # list mounted

/dev/vdb1               /root/data       ext4    defaults,noatime 0 0
/dev/cdrom              /media/CentOS           auto    user,noauto,exec,utf8        0    0
//192.168.88.10/_ISO /mnt/ISO/ cifs username=user,password=pwd 0 0
//servername/sharename /media/windowsshare cifs guest,uid=1000,iocharset=utf8 0 0
/dev/mapper/x--vg--root /home           btrfs   defaults,subvol=@home 0       2
/dev/sda2       /mymnt/win   ntfs-3g  rw,umask=0000,defaults 0 0

is checked by the dump(ext2/3 filesystem backup) utility. This field is usually set to 0, which disables the check. / sets the order for filesystem checks at boot time; see fsck(8). 1 for the root device, 2 for other partitions, 0 to disable checking. [Arch] If the root file system is btrfs, set to 0 instead of 1.

LVM

pvdisplay -v -m

            Wiping internal VG cache
            Wiping cache of LVM-capable devices
        --- Physical volume ---
        PV Name               /dev/sde1
        VG Name               wd500-vg
        PV Size               <422.53 GiB / not usable 0   
        Allocatable           yes 
        PE Size               4.00 MiB
        Total PE              108167
        Free PE               5523
        Allocated PE          102644
        PV UUID               nPoenO-vin9-wkZF-4Dxh-5WT2-yU3f-EtAMs5
        
        --- Physical Segments ---
        Physical extent 0 to 5522:
            FREE
        Physical extent 5523 to 95122:
            Logical volume	/dev/wd500-vg/data
            Logical extents	0 to 89599
        Physical extent 95123 to 95366:
            Logical volume	/dev/wd500-vg/swap_1
            Logical extents	0 to 243
        Physical extent 95367 to 108166:
            Logical volume	/dev/wd500-vg/data
            Logical extents	89600 to 102399

create vg+lv & format & mount:

vgcreate vg-name /dev/sdc3
lvcreate -L 80G wd500-vg -n data
mkfs.btrfs /dev/mapper/wd500--vg-data
mount /dev/mapper/wd500--vg-data /data2
echo $(cat /proc/mounts | tail -n 1) >> /etc/fstab ; ls /etc/fstab ;

vg operations

vgrename $vg_uuid new-vg-name

vgchange -ay # Activate vg

Add disk to vg

pvcreate /dev/sdb   # delete all partitions first
vgextend ubuntu-vg /dev/sdb

Remove disk from vg: https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/5/html/logical_volume_manager_administration/disk_remove_ex

pvmove /dev/sdb1
vgreduce myvg /dev/sdb1

lv operations

lvremove sandisk-u2/mint

Resize fs

PV thrink: gparted

lvresize -L +20G /dev/debian9-vg/root # -r, --resizefs
resize2fs /dev/dlvebian9-vg/root

yum install e4fsprogs
resize4fs /dev/debian9-vg/root # resize ext4 if resize2fs error: Filesystem has unsupported feature(s)

lvextend --resize-fs --extents +100%FREE vg/lv # not support btrfs

btrfs

https://btrfs.wiki.kernel.org/index.php/Getting_started#Creating_a_filesystem

btrfs filesystem show /mnt/intel160
btrfs filesystem usage /


btrfs device usage  /mnt/intel160G  # get {device ID}
btrfs filesystem resize 7:max /mnt/point # {device ID}
btrfs filesystem resize +60G /data # use gparted for shrink

dmesg | grep crc32c # verify if Btrfs checksum is hardware accelerated, e.g.: crc32c-intel

snapshot

btrfs subvolume delete snapshot1/ 
# mv snapshot folder will be very slow

btrfs subvolume snapshot /var/lib/docker /var/lib/docker/btrfs/snapshot_20201230 

btrfs subvolume list /var/lib/docker

mount -o subvolid=298 /dev/sda5 /var/lib/docker
subvolid=298,subvol=/btrfs/snapshot_20201230 

f2fs

https://www.kernel.org/doc/Documentation/filesystems/f2fs.txt

NTFS

https://wiki.archlinux.org/index.php/NTFS-3G

Swap

swapoff -v /dev/mapper/ubuntu--vg-swap_1
lvreduce --size -22G /dev/debian9-vg/swap
mkswap /dev/mapper/ubuntu--vg-swap_1
swapon -va

echo /dev/VG/LV swap swap defaults 0 0 >> /etc/fstab

Benchmark

dd

    dd if=/dev/zero of=./test_iops bs=4k count=10000 oflag=direct
    dd if=/dev/zero of=./test_bw   bs=1G count=1     oflag=direct

fio

git clone https://github.com/axboe/fio.git && cd fio/examples/

fio --name=randwrite --ioengine=libaio --group_reporting \
 --iodepth=1 --rw=randwrite \
 --direct=1 --bs=4k --numjobs=8 \
 --size=512M --runtime=5

fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=random_read.fio --bs=4k --iodepth=64 --size=4G --readwrite=randread
fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=random_write.fio --bs=4k --iodepth=64 --size=4G --readwrite=randwrite

SMART

apt install smartmontools

https://wiki.archlinux.org/index.php/S.M.A.R.T.

smartctl --info /dev/sdf

smartctl --test=short /dev/sdf     

smartctl --all /dev/sdf

smartctl -l selftest /dev/sdf
smartctl -l xerror /dev/sdf
-l TYPE, --log=TYPE
        Show device log. TYPE: error, selftest, selective, directory[,g|s],
                            xerror[,N][,error], xselftest[,N][,selftest],
                            background, sasphy[,reset], sataphy[,reset],
                            scttemp[sts,hist], scttempint,N[,p],
                            scterc[,N,M], devstat[,N], ssd,
                            gplog,N[,RANGE], smartlog,N[,RANGE]

SAMBA

docker

docker run  --restart=unless-stopped --net host --name samba  \ 
    -v /data:/data -d \
    dperson/samba -p -s "public;/data;yes;no"

apt

apt install samba
vi /etc/samba/smb.conf

[public]
    comment = public anonymous access
    path = /data
    browsable =yes
    create mask = 0660
    directory mask = 0771
    writable = yes
    guest ok = yes
  
service smbd restarttart

Last updated