How to resolve a sector warning?

[Gruselgurke]

I currently get this error on my FreeNAS:
WARNING: The volume Data (ZFS) status is UNKNOWN: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected.Determine if the device needs to be replaced, and clear the errors using 'zpool clear' or replace the device with 'zpool replace'.

The console says /dev/ada1 has "1 Offline uncorrectable" and "1 Currently unreadable". I thought bad sectors would be resolved by the HDD or either the OS. How can I resolve this manually?

 

[tingo]

The message about bad sectors on ada1 is not necessarily related to your zpool error.
Now, do a 'zpool status' and paste the output here.

 

[Gruselgurke]

Will do on monday! I sadly have no access to the NAS from home.

 

[Yatti420]

Will do on monday! I sadly have no access to the NAS from home.

I had a whole series of problems.. Finally stable now but haven't really pushed alot of data yet.. Gonna have backup/cycle stuff eventually..

 

[Joshua Parker Ruehlig]

I got the same error, tried spinrite on the drive and it errored and wouldn't complete. gonna try another SMART test on my linux system (probably gonna not complete like it did on freenas).
Then I'm gonna RMA the drive. That's why i love raidz2, still got redundancy even though I'm missing a drive =]

 

[Yatti420]

I got the same error, tried spinrite on the drive and it errored and wouldn't complete. gonna try another SMART test on my linux system (probably gonna not complete like it did on freenas).
Then I'm gonna RMA the drive. That's why i love raidz2, still got redundancy even though I'm missing a drive =]

I had to use Hiren's boot CD to find a bad drive which i RMAd.. This wasn't the end of me getting these errors though (bad nics) etc.. I have a post awhile back about it..

http://forums.freenas.org/search.php?searchid=535722


See ya,

 

[Gruselgurke]

I got ill sorr for the late response.
So here is what zpool status is giving me:

using 'zpool clear' or replace the device with 'zpool replace'.
see: http://www.sun.com/msg/ZFS-8000-9P
scan: scrub repaired 208G in 1h43m with 0 errors on Thu Oct 18 21:39:04 2012
config:

NAME STATE READ WRITE CKS
UM
Data ONLINE 0 0
0
mirror-0 ONLINE 0 0
0
gptid/b859f474-07e2-11e2-b487-5404a6ccafee ONLINE 3 26.1M 1.9
1M
gptid/b8ddf3be-07e2-11e2-b487-5404a6ccafee ONLINE 0 0
0
mirror-1 ONLINE 0 0
0
gptid/dc69f84d-07e2-11e2-b487-5404a6ccafee ONLINE 0 0
0
gptid/dd055c87-07e2-11e2-b487-5404a6ccafee ONLINE 0 0
0

Or for a better view a picture:

This seems ok to me actually.
Oh and btw. how do I scroll in the web shell? This is really bugging me that I get always cut off when a command prints something and I can't see what.

 

[Yell]

your first disk in the first mirror doesn't look that good,
check the drive with "smartctl -a device" and look for bad block, pending sectors and stuff that is labeled as failed.

If you haven't configured smart tests, start a short one (smartctl -t short device).

Based on smart i would RMA the drive

 

[Yatti420]

I got ill sorr for the late response.
So here is what zpool status is giving me:


Or for a better view a picture:

This seems ok to me actually.
Oh and btw. how do I scroll in the web shell? This is really bugging me that I get always cut off when a command prints something and I can't see what.

I'd be worried.. it looks like you have read/write/checksum errors which could be a bad drive.. My errors were only checksum when I was having issues.. It was a new drive and turned 2 be a lemon so I rma'd..f What kind of hardware/drives are you using etc?

Edit: Depending on what version oh Hiren's people use.. There is certain tools (for TB drives or greater) which work really good for sector repair.. I'd backup just in case..

 

[Gruselgurke]

Well the drives are mirrored and don't hold sensivite data, thats ok I guess.
My FreeNAS system broke after upgrading to 8.3-RELEASE but I'll try the smart test when I get it back up.
But couldn't it be the sector error I mentioned above in the first post?

 

[Gruselgurke]

So I tested it with a HDD Regenerator boot stick. No bad sectors were found. Can this be that FreeNAS is giving me a false alarm?

Oh and @Yatti420 I'm using an AMD E-450 board with 8GB RAM. The hard drives are quite old I doubt I can still RMA those.

 

[paleoN]

Gruselgurke stop being lazy and setup SSH. :p

Then from a SSH session as root the output of the following for the two drives in the first mirror set:

Code:

smartctl -q noserial -a /dev/adaX

Throw the output in some [code][/code] tags.

So I tested it with a HDD Regenerator boot stick. No bad sectors were found. Can this be that FreeNAS is giving me a false alarm?

Or the drive spared the bad sector(s) out. The above will tell us.

Also, an updated:

Code:

zpool status -v

Would be useful.

 

[Gruselgurke]

Will do tomorrow!

 

[Saeed A Siddiki]

hey man -- wanted to read your link for those BAD nic's -- the result is not in the archives apparently

I had to use Hiren's boot CD to find a bad drive which i RMAd.. This wasn't the end of me getting these errors though (bad nics) etc.. I have a post awhile back about it..

http://forums.freenas.org/search.php?searchid=535722


See ya,

 

[Yatti420]

hey man -- wanted to read your link for those BAD nic's -- the result is not in the archives apparently

Hello Saeed,

The NICs I was using/testing was the D-Link gigabit ethernet cards.. The one that was bad for me had a scratch on the pins that I didn't notice at first.. I'm using onboard currently.. Seems to be ok.. The harddrives I had purchased were brand new dlink green drives.. Worked good initially.. Upon upgrade to freenas 8 everything went to hell and 1 specific drive with majority of use started acting up... RMA worked good and now everything seems ok after I upgraded some of my hardware.... Im running 8.3.0 right now non upgraded (zfs v28) pools.. If you click my name and view all the threads I created you should be able to get a little more info..

Thanks,