Did I just screw up my RAIDZ1 ?

[thed00d]

I am currently running :

FreeBSD freenas.MATRIX 9.1-STABLE FreeBSD 9.1-STABLE #0 r+16f6355: Tue Aug 27 00:38:40 PDT 2013 root@build.ixsystems.com:/tank/home/jkh/src/freenas/os-base/amd64/tank/home/jkh/src/freenas/FreeBSD/src/sys/FREENAS.amd64 amd64

One of my drives were faulty so I shutdown, removed the drive, inserted the new drive and ran the zpool replace command:

2013-10-11.19:53:03 zpool replace FreeNAS /dev/gptid/cb7be29d-a945-11e2-8102-00005a11c112 /dev/ada0

The zpool resilvered (apologies if I am using the incorrect terminology here...) and all appears fine. However when running zpool status I see the following:

[root@freenas] ~# zpool status
pool: FreeNAS
state: ONLINE
scan: scrub in progress since Sat Oct 12 17:45:36 2013
2.42T scanned out of 5.52T at 165M/s, 5h28m to go
0 repaired, 43.79% done
config:

NAME STATE READ WRITE CKSUM
FreeNAS ONLINE 0 0 0
raidz1-0 ONLINE 0 0 0
ada0 ONLINE 0 0 0
gptid/32793470-b5a5-11e2-b4cf-00005a11c112 ONLINE 0 0 0
gptid/cd5cb1ab-a945-11e2-8102-00005a11c112 ONLINE 0 0 0
gptid/ce33a33c-a945-11e2-8102-00005a11c112 ONLINE 0 0 0

errors: No known data errors

Is it a problem that drive is showing as ada0 and not gptid ? I rebooted the server afterwards and it did pick everything up correctly. (ignore the scrub, I started that afterwards).

Thanks !

 

[cyberjock]

It might be a problem in the future if you need someone to help you with data recovery because things went horribly wrong and you don't know the actual partition size(you have no partition right now on ada0).

If you had followed the manual for disk replacement this wouldn't be an issue either ;)

Lastly, RAIDZ1 is not reliable. There's a link in my sig if you want to read why. But be warned that about 90% of all users that lose their data had a RAIDZ1.

 

[thed00d]

Thanks for the reply. I doubt I will need data recovery - this is just my personal stuff at home and I have it backed up elsewhere as well.

I don't know what I was thinking when I was replacing this drive... I had previously replaced another drive correctly :(

Is there anything I can do now to remedy this ? I'm thinking remove drive from chassis, power on. "Remove" drive using FreeNAS gui, power off, then re-insert disk ?

I will make a plan in the future to implement RAIDZ2 !

 

[cyberjock]

You could go into the GUI and do a disk replacement per the manual and just don't use a "new" disk as you can reuse the old. But keep in mind that since you went with RAIDZ1 if one of your disks fails or starts going bad during the resilver you will lose data. :(

If it won't let you reuse the disk then you'll have to wipe the disk(quick wipe from the GUI) and potentially reboot to let it use the new disk. FreeNAS protects disks that are assigned to pools so you don't unintentionally corrupt them, hence the need to do the quick wipe and reboot.

 

[thed00d]

Sorted. Busy resilvering onto disk correctly now.

Although now I am really paranoid about having multiple drive failures at the same time .

 

[cyberjock]

Sorted. Busy resilvering onto disk correctly now.

Although now I am really paranoid about having multiple drive failures at the same time .

You should be. Raid 5 was declared unreliable years ago.....