My Pool is setup improperly as 512b and not 4k? What to do?

[datahoarderz]

I've been running this pool for 4 years now. I created it long time ago when I ran OpenIndiana and later migrated to FreeNAS. Admittedly I was really bad about this and ever since I set it up I haven't really looked into the back end. I've just been using this as a NAS to dump old photos. Shame on me I know.

Here's what happens when I do a "zpool status"

(also ignore the fact that I haven't scrubbed this pool in forever. I've been moving around and traveling for work so I moved the NAS to my parents' place while I was moving)

Code:

  pool: freenas-boot
state: ONLINE
  scan: scrub repaired 0 in 0h0m with 0 errors on Sat Sep 26 03:45:15 2015
config:

  NAME  STATE  READ WRITE CKSUM
  freenas-boot  ONLINE  0  0  0
  da0p2  ONLINE  0  0  0

errors: No known data errors

  pool: tank
state: ONLINE
status: One or more devices are configured to use a non-native block size.
  Expect reduced performance.
action: Replace affected devices with devices that support the
  configured block size, or migrate data to a properly configured
  pool.
  scan: scrub repaired 0 in 5h13m with 0 errors on Sat Jan  9 21:28:29 2016
config:

  NAME  STATE  READ WRITE CKSUM
  tank  ONLINE  0  0  0
  raidz2-0  ONLINE  0  0  0
  gptid/e93757c1-ccae-2765-bb43-ad992c38e23e  ONLINE  0  0  0  block size: 512B configured, 4096B native
  gptid/f4a58805-8e11-e243-e177-9c57577afc8f  ONLINE  0  0  0  block size: 512B configured, 4096B native
  gptid/b0d51fe8-d45c-1fc9-becc-fbf898bd225e  ONLINE  0  0  0  block size: 512B configured, 4096B native
  gptid/31ccc716-d877-3bcb-c9ca-dc36f697dec7  ONLINE  0  0  0  block size: 512B configured, 4096B native
  gptid/cfd2de3c-f47b-06e5-d538-ee37260bc0e7  ONLINE  0  0  0  block size: 512B configured, 4096B native
  gpt/zfs  ONLINE  0  0  0  block size: 512B configured, 4096B native

errors: No known data errors

In further reading it looks like my ashift is set incorrectly as 9. These are Samsung F4 2TB HDs.

So is my only choice to wipe the pool and start over?

 

[danb35]

So is my only choice to wipe the pool and start over?

That's correct. Or just don't worry about it.

 

[toadman]

Yep. As stated by the zpool command:

status: One or more devices are configured to use a non-native block size.
Expect reduced performance
action: Replace affected devices with devices that support the
configured block size, or migrate data to a properly configured
pool.

danb35 is right. Either don't worry about it or move the data off to another pool and recreate a pool with ashift 12 with your disks.

 

[Yatti420]

@danb35 I didn't wanna hijack this but what about pools that report ashift=9 but have no error.. Mine does this...

zdb -C | grep ashift
ashift: 9

I know back in the day I had gotten this error.. I swear I recreated and addressed the issue..

 

[TooMuchData]

I use two 8GB SataDOMs as mirrored boot devices. After installing the latest 9.3.1 updates (20160517) zpool status reports "one or more devices...". But, the devices didn't change, and both were formatted at the same time by the 9.3.1 installer. So, why would only one device be identified as having the 512 block size? Any way to fix this?

 

[Mirfster]

zdb -C | grep ashift
ashift: 9

I don't believe it hurts anything as long as you are not getting any status messages about

One or more devices are configured to use a non-native block size.

 

[TooMuchData]

My boot pool is healthy and there are no known data errors. However zpool status reports the blocksize problem:

########## ZPool status report for freenas-boot ##########

pool: freenas-boot
state: ONLINE
status: One or more devices are configured to use a non-native block size.
Expect reduced performance.
action: Replace affected devices with devices that support the
configured block size, or migrate data to a properly configured
pool.
scan: scrub repaired 0 in 0h0m with 0 errors on Tue May 31 07:15:02 2016
config:

NAME STATE READ WRITE CKSUM
freenas-boot ONLINE 0 0 0
mirror-0 ONLINE 0 0 0
gptid/61e0e833-352e-11e5-8a66-d050995044d1 ONLINE 0 0 0 block size: 512B configured, 4096B native
ada0p2 ONLINE 0 0 0

errors: No known data errors​

Given the above, I conclude the problem is with ada1 (shown by gptid), not ada0. I also doubt there is any real problem - the sataDOM devices will clearly perform better than USB2 sticks. But, I would like to understand this better. So, I ran the command "geom disk list" and see the following:

Geom name: ada0
Providers:
1. Name: ada0
Mediasize: 8012390400 (7.5G)
Sectorsize: 512
Mode: r1w1e2
descr: SATADOM-MV 3ME
ident: 20150209AA5061316005
fwsectors: 63
fwheads: 16

Geom name: ada1
Providers:
1. Name: ada1
Mediasize: 8012390400 (7.5G)
Sectorsize: 512
Stripesize: 4096
Stripeoffset: 0
Mode: r1w1e3
descr: SATADOM-MV 3ME
ident: 20150209AA506131600B
fwsectors: 63
fwheads: 16​

Both devices actually use 512B sectors, but ada1 seems to have a stripesize (that is being interpreted as a native block size). Do I understand correctly?

Can I remove ada1 from the mirror, reformat it and add it back in to the mirror? What would be the procedure, please?