[sOLVED] - It was because of the pending sector raw value on disk was at 1 which worked in 4.7 but not in 5.0 so after a pre-clear of the drive it changed pending sector to raw value zero and then everything went smoothly

 

UPDATE: This is definitely a 5.0rc8 Issue because after someone told me my drive was dead I restored my Flash Drive with my 4.7 backup I had made before attempting the upgrade and I am back to normal as I was before... my suspicion still is on that 7-8 jumper since on bootup in console I see the logical errors and other things but in 4.7 drive shows up, passes all tests and so on, yet in 5.0 its "Dead" or missing

 

I am not sure if it is something to do with the WDC_WD20EARS disk that has the 7-8 jumper in it, but that disk is showing missing after following the upgrade instructions.  Here is a screenshot, not sure if I should power down, reboot and see if it comes up or if there is something I need to do, I am scared to mess anything up and go back to what I had before on 4.7 when all the disks were just fine.

Screenshot:

http://o7.no/UAbrKB

 

Any help is appreciated please

 

Here is log:

Dec 11 15:53:12 Tower kernel: md: import disk0: [8,16] (sdb) ST32000542AS_5XW0XS0M size: 1953514552

Dec 11 15:53:12 Tower kernel: mdcmd (2): import 1 0,0

Dec 11 15:53:12 Tower kernel: md: disk1 missing

Dec 11 15:53:12 Tower kernel: mdcmd (3): import 2 8,0 1953514552 ST2000DM001-9YN164_Z1E08F0J

Dec 11 15:53:12 Tower kernel: md: import disk2: [8,0] (sda) ST2000DM001-9YN164_Z1E08F0J size: 1953514552

Dec 11 15:53:12 Tower kernel: mdcmd (4): import 3 0,0

Dec 11 15:53:12 Tower kernel: mdcmd (5): import 4 0,0

Dec 11 15:53:12 Tower kernel: mdcmd (6): import 5 0,0

Dec 11 15:53:12 Tower emhttp: shcmd (90): /usr/local/sbin/emhttp_event driver_loaded

Dec 11 15:53:12 Tower emhttp_event: driver_loaded

Does it appear in the drop down list?

No it does not (at least not under the same name tag), all that appears is (sdc) 0, here is a screenshot of the entire dropdown:

 

http://o7.no/XQ4l9T

 

After my first boot I did assign the cache drive to cache as per the instructions, so that one was in the dropdown menu originally but now it is just that (sdc) 0, which if you try to select it for Disk1 it just goes right back to "No Device"

 

Really not sure what to do here. Let me know what logs or anything I can provide because it was working great on 4.7 well over a year.

I have done a test just to make sure it is not the SATA port or something relating to it, so I swapped the cable from the cache drive to the Disk1 and vice versa, and again the only one that is detected is my Cache drive, so it must be something with that drive, perhaps with the jumpers and the 4k-aligned or unaligned?  I it just reads it at (sdd) 0 or (sdc) 0.

 

Not sure what else I can do to get my array back up...

Would there be some possible harm or would it be even worth the effort to try the following with my situation of the missing disk before I go back to 4.7 and restore old files on flash:

 

1.  Replace existing go script with the default one to remove any other start-up processes that I currently have on my old go script

2. Change the hard drive settings to unaligned from 4k-aligned  (since the one missing is a jumpered EARS drive perhaps this has something to do with why it is not getting read?)

 

Any help is appreciated, I have been without my NAS for a day and I need to find a solution soon because I need it live.

Definitely try #1. See my sig. Number 2 is not the cause of your issue. Can you get a SMART report?

Should I let it boot up as it is (before I restore original go file) and go inside unmenu and get smart report there?  Or what is the best way to get the smart report that would help you see what you need to see?

 

From the upgrade instructions:

http://lime-technology.com/wiki/index.php/UnRAID_Server_Version_5.0-beta_Release_Notes

 

"Disable or un-install all non-stock unRAID add-ons."

Ok, I have put int he stock go file so it should be just booting stock now yet disk is still missing with exact same thing showing (sdc) 0, below is my old go file.  What is the best way to pull up the SMART report? 

 

I finally plugged in a monitor to my unRaid and on the screen during boot-up i kept seeing sdc and I/O error on logical sector or something along those lines that shows up on the system log I am about to attach. 

 

System Log: http://pastebin.com/Bxk0UY6W

 

Old Go File: http://pastebin.com/VT1Tni1C

 

Please let me know how I can get a SMART report to supply

I have ran the Smart Report by running the following command I found in wiki:

smartctl  -a  -d  ata  /dev/sda >/boot/smart.txt    I also then ran it with sdc I am attaching both.

 

smart.txt = code above

smart1.txt = I replaced sda with sdc

smart.txt

smart1.txt

The disk is dead. It need to be replaced.

But how could this be if only minutes before it was perfectly fine working on 4.7?  I have a screenshot of it with a green light and all and everything was rosy, then I power down to change a couple files on the flash drive and the disk is dead?  Should I try to go back to 4.7 now and see if it powers up?  Am I at risk of anything for at least trying?

 

This is minutes (perhaps seconds) before I upgraded to 5 as I was powering down 4.7 and stopping array: http://dl.dropbox.com/u/38304/Screenshots/4.png

 

What is the replacement procedure since I am also going through the upgrade to 5.0 and I have yet to fix permissions, or set any users etc?  I am still at #3 of the Upgrade Instructions in release notes.

 

Thanks

Ok, so I think the drive is in fact not dead, I went to 4.7 and it seems to have come back to life:

http://o7.no/Xfw7aN

 

What reports/logs can I run for you (in unmenu or other forms) in order to see what is happening that is preventing me to go to 5.0rc8?

Smart Report (Works this time) for drive from unmenu: http://pastebin.com/XBUV7Ruf

 

Right now I am just relieved and satisfied that 1. Drive is not dead, 2. My array and nas is back online and working as it was

But I do want to update to 5 soon to take advantage of plugins and need to transition my installed sab/sickbeard/couchpotato over, so I would like to figure out what could possibly be causing this to upgrade asap.  I did also want to swap out the parity drive with a 3tb drive so I can expand with 3tb Drives in the future, but maybe I should wait until I am stable at 5 before I look at this? 

 

More Screenshots showing drive is read:

http://o7.no/XfwAtv

 

 

Any ideas on how I could successfully upgrade to 5.0rc8 due to my issues above?  Obviously the disk is fine, it just is missing on 5.0 and I really want to upgrade to be able to take advantage of new plugins.

The drive has a pending sector. Pending sectors represent unreadable sectors. The drive needs to be rebuilt to clear the pending sectors. If you have a pre-cleared spare use it now and preclear the current disk. If the Current Pending Sector RAW_VALUE goes to zero and the self-assessment is PASSED then the drive becomes the spare. Otherwise the drive can be rebuilt in place with the following procedure:

 

Stop the array and unassign the drive.

Start the array.

Stop the array and assign the drive.

Start the array and click Rebuild.

 

After the drive is rebuilt the Current Pending Sector RAW_VALUE should be zero.

 

Thanks for the help, I actually do not have a spare drive so I will follow your directions, I was waiting to upgrade so I could then change to 3TB drives in the future so I did not want to buy another 2TB unless I absolutely needed to.  Is this procedure pretty safe?  Should I leave or remove the 7-8 Jumper when executing your instructions?  Do I do this at current 4.7 I presume?

I was waiting to upgrade so I could then change to 3TB drives in the future so I did not want to buy another 2TB unless I absolutely needed to.

You can buy a 3TB, short stroke it using HPA to exactly the same size as a 2TB. You would need a second 3TB drive to get any capacity increase anyway, so if you can afford it I'd go ahead and get a new drive now.

My idea was once I was on 5.0 to get a 3tb and set it as my parity drive and use my current parity drive for extra storage.  If I were to buy a 3tb now and not use it as parity then I am not sure if it is the smoothest transition.

My idea was once I was on 5.0 to get a 3tb and set it as my parity drive and use my current parity drive for extra storage.  If I were to buy a 3tb now and not use it as parity then I am not sure if it is the smoothest transition.

It would be exactly the same. Until you have a second 3TB drive, it makes no difference if you have a 2 or 3 TB parity drive. Once you do transition to a version that supports 3TB drives, you can buy that 2nd 3TB drive, assign it as parity, and use your current parity drive as data. After everything is stable, you can remove the HPA from the 3TB you short stroked, and gain another 1TB, as well as the 2TB from your old parity drive.

Thanks for the help, I actually do not have a spare drive so I will follow your directions, I was waiting to upgrade so I could then change to 3TB drives in the future so I did not want to buy another 2TB unless I absolutely needed to.  Is this procedure pretty safe?  Should I leave or remove the 7-8 Jumper when executing your instructions?  Do I do this at current 4.7 I presume?

 

If the default partition format is 4K-aligned then you can remove the jumper; but it is safer to leave the jumper and change the setting to UNaligned for this drive only. We want to change as little as possible. The procedure is identical to replacing a failed drive. As long as all of the other drives are healthy the rebuild will succeed.

Ok, I have done the steps and it is now rebuilding, my only concern is it never prompted/asked me to set it as unaligned for this drive and my default is the 4k-Aligned, so I hope that is not going to be a problem.  I never saw the option under devices for this particular drive

The default setting should have been changed to un-aligned for this drive only.

Crud, It is now rebuilding with my previous attempt so I am not sure if I should let it finish, or if I should cancel (Is there risk?) and change to unaligned and start the process over again or would that compromise my data?  Am I ok to cancel the process and then I presume:

stop array & unassign drive

start array

stop array and assign drive (with unaligned set)

Start Array and let it rebuild?

 

Yes.

Ok great, thank you!! It is now rebuilding with the unaligned setting, now after this is done what test do I run to make sure that it will be good to upgrade to 5.0 with a Raw Value to Zero?