FreeNAS

New way using FreeBSD only

This is my new bible: https://www.freebsd.org/doc/handbook/zfs.html

Start with downloading freebsd memstick image: https://download.freebsd.org/ftp/releases/amd64/amd64/ISO-IMAGES/12.1/FreeBSD-12.1-RELEASE-amd64-memstick.img

Write it to a USB stick using Win32DiskImager.

Insert this USB stick and a secondary USB stick (installation destination) in the server. Boot and install FreeBSD on the secondary USB stick.

Then I did (to create a two-disk zfs mirror):

zpool create tank mirror /dev/ada0 /dev/ada1
zfs create tank/media
zfs create tank/backup
zfs set compression=gzip tank/backup

This creates two datasets, one media without compression, and one backup with compression.

Now install want Samba https://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-samba.html (yes it will pull in some X libs for cups).

pkg install samba410

Old stuff

Properly init and add disks to your pool via command line

This is a great start for ZFS newbies: https://forums.freenas.org/index.php?threads/slideshow-explaining-vdev-zpool-zil-and-l2arc-for-noobs.7775/

In this quest of mine I got great help from this article: https://forums.freenas.org/index.php?threads/building-pools-from-the-cli.17540/

I have a FreeNAS server with 1x2TB disk. I want to replace it with 2x4TB disks in a mirror configuration. This is how I did it!

# zpool status
  pool: BigData
 state: ONLINE
  scan: none requested
config:

	NAME                                          STATE     READ WRITE CKSUM
	BigData                                       ONLINE       0     0     0
	  gptid/a37b301e-eb46-11e7-b559-6cf049956cad  ONLINE       0     0     0 

errors: No known data errors

1) Find out the dev nodes for your new disks

# ls -l /dev/ada*
# zpool status
# glabel list

The ls command will show you all your disks dev-id:s. The zpool command will show you the gpt-id of all disks currently included in your pools. "glabel list" will give you the mapping between gpt-id and dev nodes. With this information you know which /dev/adaX devices that are your new/unused disks.

2) If any partitions currently exists on the disks, clear them out (ada0 and ada1 in my case, but may be different for you)

# gpart destroy -F /dev/ada0
# gpart destroy -F /dev/ada1

3) Init the disk as a GPT and create one swap partition and one zfs partition on each disk

# gpart create -s gpt /dev/ada0
# gpart add -a 4096 -i 1 -s 2g -t freebsd-swap /dev/ada0
# gpart add -a 4096 -i 2 -t freebsd-zfs /dev/ada0

# gpart create -s gpt /dev/ada1
# gpart add -a 4096 -i 1 -s 2g -t freebsd-swap /dev/ada1
# gpart add -a 4096 -i 2 -t freebsd-zfs /dev/ada1

4) The disks are now ready to be used in a pool! Now run glabel and note down the new gpt-id:s for your disks

# glabel status
                                      Name  Status  Components
gptid/fcb8a69a-eb13-11e7-8eaf-6cf049956cad     N/A  ada2p1
gptid/a37b301e-eb46-11e7-b559-6cf049956cad     N/A  ada3p2
gptid/a35ca845-eb46-11e7-b559-6cf049956cad     N/A  ada3p1
gptid/8d0dbf30-f6ce-11e7-a931-6cf049956cad     N/A  ada0p1
gptid/96286393-f6ce-11e7-a931-6cf049956cad     N/A  ada0p2
gptid/b791e4cb-f6ce-11e7-a931-6cf049956cad     N/A  ada1p1
gptid/bbb96c09-f6ce-11e7-a931-6cf049956cad     N/A  ada1p2

5) Now you can either create a new pool using zpool create, or you can add them to an existing pool with zpool attach. I will start by attaching one of them as a mirror to my existing BigData pool:

# zpool attach BigData gptid/a37b301e-eb46-11e7-b559-6cf049956cad gptid/bbb96c09-f6ce-11e7-a931-6cf049956cad

6) Now wait for the resilvering to complete

# zpool status
  pool: BigData
 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 Thu Jan 11 05:03:53 2018
	220G scanned at 423M/s, 118M issued at 228K/s, 1.67T total
	109M resilvered, 0.01% done, no estimated completion time
config: 

	NAME                                            STATE     READ WRITE CKSUM
	BigData                                         ONLINE       0     0     0
	  mirror-0                                      ONLINE       0     0     0
	    gptid/a37b301e-eb46-11e7-b559-6cf049956cad  ONLINE       0     0     0
	    gptid/bbb96c09-f6ce-11e7-a931-6cf049956cad  ONLINE       0     0     0  (resilvering)

errors: No known data errors

7) Now we are ready to replace the old drive

# zpool replace BigData gptid/a37b301e-eb46-11e7-b559-6cf049956cad gptid/96286393-f6ce-11e7-a931-6cf049956cad

Wait again for resilvering to complete. Now our pool is completely replaced and up and running on the new disks:

# zpool status
  pool: BigData
 state: ONLINE
  scan: resilvered 1.67T in 0 days 09:50:37 with 0 errors on Fri Jan 12 09:23:46 2018
config:

	NAME                                            STATE     READ WRITE CKSUM
	BigData                                         ONLINE       0     0     0
	  mirror-0                                      ONLINE       0     0     0
	    gptid/96286393-f6ce-11e7-a931-6cf049956cad  ONLINE       0     0     0
	    gptid/bbb96c09-f6ce-11e7-a931-6cf049956cad  ONLINE       0     0     0

errors: No known data errors

8) The size of the pool is still only 2TB, it should be 4T since we replaced our original 2TB with 2x4TB drives. So now tell ZFS to expand the drives, and then reboot the system.

# zpool online -e BigData gptid/96286393-f6ce-11e7-a931-6cf049956cad
# zpool online -e BigData gptid/bbb96c09-f6ce-11e7-a931-6cf049956cad
# reboot

9) Now the pool should be fully operational and have correct size!

# zpool list
NAME           SIZE  ALLOC   FREE  EXPANDSZ   FRAG    CAP  DEDUP  HEALTH  ALTROOT
BigData       3.62T  1.67T  1.96T         -     2%    46%  1.00x  ONLINE  /mnt