ZFS drive replacement guidance

[ncsa]

Hello

I am in need of some guidance in replacing a HDD which ZFS is reporting as having CRC errors.

System is:
- FreeNAS-8.0-RELEASE-amd64
- ZFS2
- 8x 1.5TB HDD
- 2x 3ware 4-port controllers
- each HDD is configured as a single device on the controllers to allow ZFS to see each individual drive.
- ZFS pool was upgraded from version 13 to 15 via console upgrade command

Code:

nom# zpool get all pool
NAME  PROPERTY       VALUE       SOURCE
pool  size           10.9T       -
pool  used           6.89T       -
pool  available      3.99T       -
pool  capacity       63%         -
pool  altroot        -           default
pool  health         ONLINE      -
pool  guid           1890640775006263064  default
pool  version        15          default
pool  bootfs         -           default
pool  delegation     on          default
pool  autoreplace    off         default
pool  cachefile      none        local
pool  failmode       wait        default
pool  listsnapshots  off         default

At the console running a ZPOOL Status I see that da3 has the CRC errors.

Code:

nom# zpool status pool
  pool: pool
 state: ONLINE
status: One or more devices has experienced an unrecoverable error.  An
	attempt was made to correct the error.  Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
	using 'zpool clear' or replace the device with 'zpool replace'.
   see: http://www.sun.com/msg/ZFS-8000-9P
 scrub: scrub completed after 14h54m with 0 errors on Sun Oct 30 08:01:10 2011
config:

	NAME        STATE     READ WRITE CKSUM
	pool        ONLINE       0     0     0
	  raidz2    ONLINE       0     0     0
	    da0     ONLINE       0     0     0
	    da1     ONLINE       0     0     0
	    da2     ONLINE       0     0     0
	    da3     ONLINE       0     0  269K  5.53G repaired
	    da4     ONLINE       0     0     0
	    da5     ONLINE       0     0     0
	    da6     ONLINE       0     0     0
	    da7     ONLINE       0     0     0

errors: No known data errors

I had assumed that this would be the 4th drive on the first controller but when taking that drive out ZPOOL stated it was device da2.

Is there a way to ensure I am taking the correct drive out?

Also I was looking at upgrading to the latest nightly build, 8.1 as I read this has improved on the ZFS tools - should I wait before applying this update?

 

[ProtoSD]

The FAQ is your friend. The Unofficial one has more info, but here's a link that should answer your question.

I_have_a_bad_drive,_How_do_I_check_the_serial_#_so_I_can_make_sure_I_remove_the_correct_disk.

 

[ncsa]

Thanks protosd - I had just discovered your FAQ and will first upgrade to 8.1 and then look at replacing the HDD. Unfortunately the Serial numbers are not displayed in the GUI - the HDD are behind a 3Ware controller and smartctl is currently not able to extract this detail.

 

[survive]

Hi ncsa,

I wouldn't upgrade to 8.1, go to 8.0.2.

8.0.2 was released a couple of weeks ago, to get 8.1 you would have to run a nightly build which may or may not work as intended. The big problem you might run into is if you find 8.1 isn't working you you may have to do a full restore of all your setting if the config database doesn't import on 8.0.2.

Either way be sure you save a copy of your config before you update!

-Will

 

[ProtoSD]

Hi ncsa,

Will is correct, you should upgrade to 8.0.2. The "Official FAQ" hadn't been updated and said 8.0.1, version 8.1 is not even beta yet.

 

[ncsa]

Thanks guys - yes I found that out as went to 8.1 but the ZFS library would not load so could not see the storage. Went back to a fresh install of 0.0.2 and reloaded the config - running scrubs again as this version seems to have the pool reporting clean & healthy.

The field for the HDD serial numbers now appear but it looks like I need to load the 3ware smart tool to get the GUI to read these, I can run the CLI query but still not sure which is correct for da3 (not sure how to load the 3ware tool). I also have a rocket raid 4320 controller in the same NAS and the drives on this show their serial numbers in the GUI so must be more native support.

 

[ncsa]

Having upgraded to FreeNAS-8.0.2-RELEASE-amd64 (8288), the suspected drive has no longer reported any errors and FN is functioning as well as it did before, so for now looks like it may have been a data corruption which has now been cleaned up - thanks for the help.

 

[Forssux]

Hi There,

It seems the Faq is still empty concerning replacing bad hard disk...http://doc.freenas.org/index.php/FAQs#How_do_I_replace_a_bad_drive_-_Can_it_be_larger.2Fsmaller.3F

I also didn't find this info in the very good powerpoint presentation of http://forums.freenas.org/showthrea...explaining-VDev-zpool-ZIL-and-L2ARC-for-noobs!

So for me The question still stands..

paleoN informed me with a

Code:

smartctl -a -d 3ware,YY /dev/twe0

but it generates a error