Skip to content

Latest commit

 

History

History
284 lines (210 loc) · 6.64 KB

zfs.md

File metadata and controls

284 lines (210 loc) · 6.64 KB

zfs

create filesystem

zpool create NAME DEVICE

create zfs slot

zfs create NAME/path1/path2

set acl

zfs set acltype=posix NAME

set mountpoint

zfs set mountpoint=MOUNTPOINT NAME

take snapshot

zfs snapshot NAME@TAG

list snapshots

zfs list -t snapshot

mount snapshot

mount -t zfs NAME@TAG MOUNTPOINT

destroy snapshot

zfs destroy NAME@TAG

rollback to a snapshot

zfs rollback NAME@TAG

get compression

zfs get compression NAME

set compression

zfs set compression=on NAME

get compression ratio

zfs get compressratio

send snapshots to other device

given follow example

NAME                                       USED  AVAIL  REFER  MOUNTPOINT
bk/test/current@one                         64K      -    96K  -
bk/test/current@two                         56K      -    96K  -
bk/test/current@three                        0B      -   104K  -

given a bk2 pool exists

  • send initial snapshot
zfs send bk/test/current@one | zfs recv bk2/test/current
  • this gives
NAME                                       USED  AVAIL  REFER  MOUNTPOINT
bk/test/current@one                         64K      -    96K  -
bk/test/current@two                         56K      -    96K  -
bk/test/current@three                        0B      -   104K  -
bk2/test/current@one                        64K      -    96K  -
  • update bk2 with other bk snapshots ( two, three ) by specifying from latest existing also in bk2 @one toward the latest to send @three
zfs send -I bk/test/current@one bk/test/current@three | zfs recv bk2/test/current
  • this gives
NAME                                       USED  AVAIL  REFER  MOUNTPOINT
bk/test/current@one                         64K      -    96K  -
bk/test/current@two                         56K      -    96K  -
bk/test/current@three                        0B      -   104K  -
bk2/test/current@one                        64K      -    96K  -
bk2/test/current@two                        56K      -    96K  -
bk2/test/current@three                       0B      -   104K  -

build zfs raid with ssd cache

  • suppose follow layout
    • /dev/sdc3 is a 200GB ssd cache disk
    • /dev/sda, /dev/sdb, /dev/sdd, /dev/sde, /dev/sdf are 4tb disks
zpool create tank raidz1 sda sdb sdd sde sdf -f
  • add /dev/sdc3 as cache
zpool add tank cache /dev/sdc3  -f
  • get status
$ zpool status -v tank
  pool: tank
 state: ONLINE
config:

        NAME        STATE     READ WRITE CKSUM
        tank        ONLINE       0     0     0
          raidz1-0  ONLINE       0     0     0
            sda     ONLINE       0     0     0
            sdb     ONLINE       0     0     0
            sdd     ONLINE       0     0     0
            sde     ONLINE       0     0     0
            sdf     ONLINE       0     0     0
        cache
          sdc3      ONLINE       0     0     0

errors: No known data errors

save disk identifiers mapping

know the serial belonging a device allow you to know which of the disk is to remove and replace when degraded

lsblk -o PATH,SERIAL

zfs test redundancy

  • physical disk removing 1 disk ( or without physical detach can be offlined with zpool offline POOL DEVICE )
$ zpool status
  pool: tank
 state: DEGRADED
status: One or more devices has been removed by the administrator.
        Sufficient replicas exist for the pool to continue functioning in a
        degraded state.
action: Online the device using zpool online' or replace the device with
        'zpool replace'.
config:

        NAME        STATE     READ WRITE CKSUM
        tank        DEGRADED     0     0     0
          raidz1-0  DEGRADED     0     0     0
            sde     ONLINE       0     0     0
            sdf     ONLINE       0     0     0
            sdb     ONLINE       0     0     0
            sdc     ONLINE       0     0     0
            sdd     REMOVED      0     0     0
        cache
          sda3      ONLINE       0     0     0

errors: No known data errors

  • wipe 1GB of the disk wipefs -af /dev/DEVICE ( carefully check which DEVICE is the one to destroy )

  • reattach 1 disk ( or without physical action can be onlined with zpool online POOL DEVICE )

zpool replace POOL DEVICEPOOL

or if device changed /dev identifier

zpool replace POOL DEVICEPOOL NEWDEVICEPOOL
  • now zpool status
  pool: tank
 state: ONLINE
status: One or more devices is currently being resilvered.  The pool will
        continue to function, possibly in a degraded state.
action: Wait for the resilver to complete.
  scan: resilver in progress since Sat Feb  1 13:37:15 2025
        25.4G / 926G scanned, 453M / 926G issued at 64.7M/s
        96.3M resilvered, 0.05% done, 04:04:10 to go
config:

        NAME        STATE     READ WRITE CKSUM
        tank        ONLINE       0     0     0
          raidz1-0  ONLINE       0     0     0
            sde     ONLINE       0     0     0
            sdf     ONLINE       0     0     0
            sdb     ONLINE       0     0     0
            sdc     ONLINE       0     0     0
            sdd     ONLINE       0     0     0  (resilvering)
        cache
          sda3      ONLINE       0     0     0

errors: No known data errors

send snapshot through ssh

  • note: i instead of I parameter ( need to configure ~/.ssh/config to allow access DESTHOST )
zfs send -i bk/test/current@one bk/test/current@three | pv | ssh DESTHOST zfs recv bk2/test/current

create zfs encrypted

  • given tank an existing pool
zfs create -o encryption=on -o keylocation=prompt -o keyformat=passphrase tank/encrypted
  • to open the encrypted pool after reboot
zfs load-key -r tank/encrypted
  • to mount all datasets
zfs mount -a