ZFS
Jump to navigation
Jump to search
ZFS on Ubuntu 18.04
sudo apt-get update && sudo apt-get install zfsutils-linux lsblk --ascii -o NAME,PARTUUID,LABEL,PATH,FSTYPE sudo zpool create -f tank /dev/disk/by-partuuid/9c179fff-7b03-794e-8650-f1d507a885b6 # # zpool create pool02 /dev/disk/by-partuuid/c8e0c300-5ec9-714c-aef9-fa0dc3f0cab6 sudo zfs create tank/local-path sudo zpool create -f tank /dev/sdc zpool status sudo zfs set mountpoint=/zfs/local-path tank/local-path sudo zfs snapshot tank/local-path@snap1 zfs list -t snapshot sudo zfs rollback tank/local-path@snap1
sudo zpool import sudo zpool import tank -f
https://docs.oracle.com/cd/E19253-01/819-5461/gbcya/index.html
vim /etc/default/zfs
ZFS_INITRD_POST_MODPROBE_SLEEP='15' # increase longer/shorter if needed 5-15
sudo update-initramfs -u sudo update-grub
Setup zpool with cache/logs from vdevs
sudo zpool create -f tank mirror nvme3n1 nvme4n1 mirror nvme5n1 nvme6n1 zpool create -f tank mirror sdd sde mirror sdf sdg mirror sdh sdi mirror sdj sdk sudo zpool add tank mirror nvme2n1 nvme3n1 sudo zpool add tank log mirror nvme0n1 nvme1n1 sudo zpool add tank cache nvme2n1 nvme3n1
sys prep to have default docker and lxd on zfs datasets
sudo systemctl stop lxd lxd.socket sudo rm -Rf /var/lib/lxd sudo zfs create tank/lxd sudo zfs create tank/lxd sudo zfs set mountpoint=/var/lib/lxd tank/lxd sudo zfs create tank/libvirt sudo zfs set mountpoint=/var/lib/libvirt tank/libvirt sudo zfs create tank/docker sudo zfs set mountpoint=/var/lib/docker tank/docker sudo zfs mount -a sudo apt-get install docker-ce # after you setup from get docker site
install qemu-kvm libvirt
sudo apt-get install libguestfs-tools qemu-kvm libvirt-clients libvirt-daemon-system bridge-utils virt-manager lintian curl wget git
Other
sudo zpool import tank sudo zfs mount -a sudo apt-get -y install docker-ce
ZFS Trouble Shooting
nvme smart-log /dev/nvme0n1 nvme smart-log /dev/nvme3n1
both reported no errors
https://docs.huihoo.com/opensolaris/solaris-zfs-administration-guide/html/ch09s06.html
zpool clear ... didn't do anything
zpool status
errors: No known data errors
pool: tank
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: http://zfsonlinux.org/msg/ZFS-8000-4J
scan: none requested
config:
NAME STATE READ WRITE CKSUM
tank DEGRADED 0 0 0
mirror-0 ONLINE 0 0 0
sdc ONLINE 0 0 0
sdd ONLINE 0 0 0
mirror-1 ONLINE 0 0 0
sde ONLINE 0 0 0
sdf ONLINE 0 0 0
mirror-2 ONLINE 0 0 0
sdg ONLINE 0 0 0
sdh ONLINE 0 0 0
mirror-3 ONLINE 0 0 0
sdi ONLINE 0 0 0
sdj ONLINE 0 0 0
logs
mirror-4 DEGRADED 0 0 0
4221213393078321817 FAULTED 0 0 0 was /dev/nvme3n1
nvme1n1 ONLINE 0 0 0
cache
nvme2n1 ONLINE 0 0 0
nvme0n1 FAULTED 0 0 0 corrupted data
Use with caution. This example would be mirror
You could remove disk from cache and use that for mirror
zpool remove tank cache nvme0n1
replace drive using singular(not two dev names) command
zpool replace tank nvme3n1
auto used free nvme0n1 from pool if none available it may just replace itself
Failed Drives
ZFS Intent Log & ARC/L2Arch
- https://pthree.org/2013/04/19/zfs-administration-appendix-a-visualizing-the-zfs-intent-log/
- https://www.ixsystems.com/blog/o-slog-not-slog-best-configure-zfs-intent-log/
- https://www.ixsystems.com/blog/zfs-zil-and-slog-demystified/
- https://www.45drives.com/wiki/index.php?title=FreeNAS_-What_is_ZIL%26_L2ARC
- https://en.wikipedia.org/wiki/Adaptive_replacement_cache
linux disk hot-swapping take a disk offline: echo offline > /sys/block/sda/device/state put a disk online: echo running > /sys/block/sda/device/state delete a disk before unplugging: echo 1 > sys/block/sda/device/delete scan for new hotplugged disks: echo "0 0 0" > /sys/class/scsi_host/host/scan display the controller IDs: omreport storage controller display the IDs for the physical disks attached to the controller: omreport storage pdisk controller=0 displays property information for all virtual disks on all controllers omreport storage vdisk make sure that I’m dealing with the proper disks: omconfig storage pdisk action=blink controller=0 pdisk=0:0:2 omconfig storage pdisk action=unblink controller=0 pdisk=0:0:2 omconfig storage pdisk action=blink controller=0 pdisk=0:0:3 omconfig storage pdisk action=unblink controller=0 pdisk=0:0:3 clear the configuration: omconfig storage controller action=clearforeignconfig controller=0 check for the new virtual disk: omreport storage vdisk initialize a disk: omconfig storage vdisk action=fastinit controller=0 vdisk=id how to replace a disk on a dell server with several disks in a jbod configuration umount /dev/sdg1 echo offline > /sys/block/sdg/device/state echo 1 > sys/block/sdg/device/delete install replacement disk, then: echo "0 0 0" > /sys/class/scsi_host/host/scan # # WARNING!! WARNING!! Danger Will Robinson!! # verify that the disk is the same device # fdisk -l /dev/sdg # echo running > /sys/block/sdg/device/state parted /dev/sdg mkfs.ext3 -m0 -L /disk6 /dev/sdg1 tune2fs -i0 -c0 /dev/sdg1 mount /dev/sdg1 fdisk /dev/sdg >>> Command (m for help): n (create new partition) >>> Partition number (1-4): 1 >>> Command (m for help): p (create primary partition) >>> Command (m for help): w (Write and save partition table) mkfs.ext3 -L disk2 -m 0 /dev/sdg1 tune2fs -c 0 -i 0 /dev/sdg1 cat >>/etc/fstab <<EOM LABEL=disk2 /disk2 ext3 defaults 1 2 EOM mkdir /disk2 mount -L disk2