0

I am using ZFS on Ubuntu 16.04 which uses systemd. My boot drive is an SSD and sometimes I have USB drives attached.

The ZFS volume is 10 raidz2 drives with 2 hot spares. On my most recent reboot ZFS did not come up and I did a zpool import command to bring it back. But the result is this:

# zpool status
  pool: sbn
 state: ONLINE
  scan: scrub repaired 0 in 69h21m with 0 errors on Tue Dec 25 23:21:25 2018
config:

    NAME        STATE     READ WRITE CKSUM
    sbn         ONLINE       0     0     0
      raidz2-0  ONLINE       0     0     0
        sdb     ONLINE       0     0     0
        sdc     ONLINE       0     0     0
        sdd     ONLINE       0     0     0
        sde     ONLINE       0     0     0
        sdf     ONLINE       0     0     0
        sdg     ONLINE       0     0     0
        sdh     ONLINE       0     0     0
        sdi     ONLINE       0     0     0
        sdj     ONLINE       0     0     0
        sdk     ONLINE       0     0     0
    spares
      sdl       FAULTED   corrupted data
      sdk       FAULTED   corrupted data

errors: No known data errors

In other words ZFS did not recognize my two spare drives. I can get them back with zpool remove then zpool add but last time I did that I got a drive letter wrong and trashed the partition table on my boot drive! (to say the least: argh)

Can someone advise me as how to avoid this problem?

AlanObject
  • 101
  • 2
  • did you add sdk once as a pool disk and once as a spare disk? This might be partially the source of the problem – mmusante Dec 27 '18 at 02:24
  • @mmusante I didn't notice before sdk was listed as both a spare and a raidz2 member until you just pointed this out. However this is the result of the zpool import process. The drive letters are not necessarily the same on each reboot sometimes these drives start at sda and not sdb. very confusing I will remove the spares now. – AlanObject Dec 27 '18 at 05:02
  • 3
    You also should use the `/dev/disk/by-id` device reference, especially when your device names are changing. Check the *Workaround* from [this question](https://unix.stackexchange.com/questions/288599/forcing-zpool-to-use-dev-disk-by-id-in-ubuntu-xenial). – Thomas Dec 27 '18 at 10:13

0 Answers0