Degraded pool after resilvering

[Mallabu]

~# zpool status

pool: NASBOX01

state: DEGRADED

scrub: scrub completed after 5h21m with 0 errors on Mon Jan 12 06:47:42 2015

config:



NAME STATE READ WRITE CKSUM

NASBOX01 DEGRADED 0 0 0

raidz1 DEGRADED 0 0 0

ada1p2 ONLINE 0 0 0

ada2p2 ONLINE 0 0 0

replacing DEGRADED 0 0 0

9878694156672902476 OFFLINE 0 0 0 was /dev/ada3p2/old

ada3p2 ONLINE 0 0 0

ada4p2 ONLINE 0 0 0



errors: No known data errors



What do I ned to do to remove the degraded status so NASBOX01 is ONLINE?

History:

I removed ada3p2 from the pool, since I did not have a spare SATA port I added a new drive of same capacity into the same SATA port as the failing drive and resilvered. It completed but now I am still left with a degraded status. when I use the -x it tells me the pool is healthy. How do I fix this?

 

[dlavigne]

Does a reboot fix it?

 

[Mallabu]

I have rebooted 3 times and it just stays the same. I have to do something with zpool command to reinstate the new drive as the old drive. How can I achieve this or am I barking up the wrong tree? I don't understand why FreeNas does not have documentation covering redundancy (part of disaster recovery) , its terrible that the outcomes for a straight no error but failing hard disk is not documented anywhere. You would have thought that by now version 9.3 documentation for redundancy would have been written up for (1) a system with no problems under raidZ1 replacing a fully functioning hard disk and (2) a system with catastrophic failure on one hard disk and step by step recovery. I pulled out a Hard disk because it was having Off-line uncorrectable sectors. It hadn't failed, I needed the hard disk out and replaced so I could send it back for warranty. But now the server works as it should without complaint but in the degraded state which I cannot understand why the OS doesn't understand whereabouts the data is and make the necessary adjustments??? Its so frustrating that such a problem is not documented comprehensively somewhere. I am sure its a one line command that I have to run, but what could it be? :(

 

[cyberjock]

First, repost the output of zpool status from pastebin. The text formatting is crucial.

Second, did you do the last step of the procedure for replacing a disk per the FreeNAS manual?

Third, considering everything is being listed by adaXp2, I'm going to go on a limb and say you didn't replace the disk per the manual to start with, nor did you create the pool per the manual?

 

[Mallabu]

The zpool was created using the GUI interface via web browser. All has been well for three weeks until I noticed that the log was showing that 140 Off-line uncorrectable sectors where showing up on ada3p2 and after finding out which physical drive that was I followed the documentation and off-lined the drive and removed the physical hard disk, then replaced it with the same capacity and make of drive into the same SATA port (No free available ports so I couldn't do a hard disk to hard disk replace) . Upon boot I used the replace command to re-sliver the new drive and it did that for 13 hours. Then I made a little mistake because the manual said nothing other than after re-silvering the system will returned to a healthy state. It was at this point I used the detach command to remove the unavailable drive but this took away the ada3p2 and I panicked. Long story short I had to re-silver the drive again which was another 13 hours. Now I'm at this point. The zpool is in the degraded state. Something is still outstanding. Would you kindly send me a link for the documentation in the manual covering the re-silvering process as I may have referenced the wrong documentation. Phew! I got that off my chest!!

 

[Mallabu]

The server is working as it should with all drives online and healthy, but the zpool is still showing it is in the degraded state which is not good. Can anyone tell me what I must do to change the status of the zpool to online? The spool status shows the following:

[root@freenas] ~# zpool status
pool: NASBOX01
state: DEGRADED
scrub: resilver completed after 0h0m with 0 errors on Wed Jan 14 02:09:04 2015
config:

NAME STATE READ WRITE CKSUM
NASBOX01 DEGRADED 0 0 0
raidz1 DEGRADED 0 0 0
ada1p2 ONLINE 0 0 0
ada2p2 ONLINE 0 0 0
replacing DEGRADED 0 0 0
9878694156672902476 OFFLINE 0 0 0 was /dev/ada3p2/old
ada3p2 ONLINE 0 0 0 512 resilvered
ada4p2 ONLINE 0 0 0

errors: No known data errors
[root@freenas] ~#

 

[SweetAndLow]

use pastbin or code tags for your output. when you remove the formatting it means nothing to people trying to help you.

 

[Mallabu]

I have some great news, I found the answer that was deluding me, it was their in front of me all along. You can, build with one drive redundant! Lets recap, I did not have a spare SATA port so all I could do is replace the failing one with a brand new one of same capacity. After re-silvering had completed I was left with a degraded zpool which is shown in my previous entry thread. What I did next was to type zpool detach NASBOX01 9878694156672902476. And hey presto the ghost drive was gone and the zpool changed from degraded to Online. Ho-ray!!!

Now, this made me think about the use of ZFS1 which is okay if one drive where to show SMART consistency errors, but what if corruption or damage across the strip, this would mean a degraded scenario. At some point I will need to install a fifth hard disk and restore from backup onto a NEW ZFS2 pool. But for now I will work on my next project to upgrade from freenas 2.35 to 9.3 BETA. I will need a 64bit processor, just trying to find out if Intel Celeron D 3.06GHz @ 3.08 GHz processor is able to run freenas 9.3 BETA 64bit OS?

 

[Mallabu]

I have some great news, I found the answer that was deluding me, it was their in front of me all along. You can, build with one drive redundant! Lets recap, I did not have a spare SATA port so all I could do is replace the failing one with a brand new one of same capacity. After re-silvering had completed I was left with a degraded zpool which is shown in my previous entry thread. What I did next was to type zpool detach NASBOX01 9878694156672902476. And hey presto the ghost drive was gone and the zpool changed from degraded to Online. Ho-ray!!!

Now, this made me think about the use of ZFS1 which is okay if one drive where to show SMART consistency errors, but what if corruption or damage across the strip, this would mean a degraded scenario. At some point I will need to install a fifth hard disk and restore from backup onto a NEW ZFS2 pool. But for now I will work on my next project to upgrade from freenas 2.35 to 9.3 BETA. I will need a 64bit processor, just trying to find out if Intel Celeron D 3.06GHz @ 3.08 GHz processor is able to run freenas 9.3 BETA 64bit OS?

 

[cyberjock]

Well:

1. There is no FreeNAS 2.35.. so no clue where you are getting that number.
2. If you are truly going to ask about running a beta when the release version came out almost 2 months ago, then you deserve to hit every bug that is there. ;)
3. Feel free to read our documentation that lists minimum hardware... you'll realize your hardware is inadequate. :P

 

[Mallabu]

Well:

1. There is no FreeNAS 2.35.. so no clue where you are getting that number.
2. If you are truly going to ask about running a beta when the release version came out almost 2 months ago, then you deserve to hit every bug that is there. ;)
3. Feel free to read our documentation that lists minimum hardware... you'll realize your hardware is inadequate. :p

Your right! its a typo, should have been 8.35. And your right again with the upgrade to BETA, better upgrade to the nearest stable version.
As for my hardware, the CPU needs to be upgraded before I even go into 9x territory as the current one is 32 bit and is not 64-bit capable. I did upgrade from 8.01 p3 to 8.35 and upgraded the ZFS from version 13 to 28. ;)

 

[cyberjock]

There is no 8.35 either. :P

 

[Mallabu]

Your attention to detail must be commended. Freenas 8.3.