Hi,

 

I built a 10 disk unRaid server about 3 years ago with 2Tb Hitachi drives, just got my first drive fail so pretty good going all told.

 

I bought a WD Caviar Green 2Tb as a replacement and started the latest version of Preclear a few hours ago.

 

For the first 10minutes or so (Disk Pre- read 4%) the read speed was ~140 MB/s but for the last 2.5 hrs it has been running at around 1MB/s.

 

I know the expected completion time is around 30 hrs but wanted to check this speed was normal as it seems very slow from what I have been reading?

 

Also is it possible to start the array whilst one disk is pre-clearing as I keep my work on the unRaid server as well as my media and may need access to it?

 

Thanks very much.

 

Adam

post syslog and system specs.

Thanks for reply.

 

I just noticed it has jumped back up to ~137 MB/s and is at 17% so maybe just normal.

 

Syslog is attached, lots of red! 

 

System Specs I don't have written down so this is what I can see from old Scan orders and poking around inside

 

Not 100% sure of the Processor unfortunately.

 

Antec Nine Hundred Two Case

ASUS P5E-VM HDMI

Ram 2 x 2GB PC2-6400

Intel Quadcore Q6600 2400Mhz ??

2 x IB-555SK 3.5 SATA/SAS 5 BAY

10 x 2TB HIT SATA OF10311 32m 7200

HDD Controllers 6 onboard SataII

2 x PCIEx1 2 channel controllers

 

The HDD I am preclearing is running on an onboard Sata port.

 

Can you help with the question about using the array?

 

Cheers.

 

Adam

syslog-2013-08-29.zip

is the drive you are clearing sdi perhaps???

Yes

ok if the speeds are back to normal let if finish and post the smart report it outputs (it saves a before/after).

 

my guess is that the drive had numerous bad sectors and the system was doing a bunch of error handling and it finally got past them.

 

speeds normal or in the bucket again??

I had to reboot my workstation so lost the telnet connection and am not sure how to find the status now?

 

If I reconnect via telnet is there a command to see the progress of the preclear?

 

Also, can I start up the array whilst the 'missing' disk preclears?

 

Thanks.

 

Adam

Oh dear.

 

Checked the syslog and it showed the server had rebooted around 4am so i am guessing something went wrong with the Preclear.

 

I tried again this morning but the disk doesn't seem to be responding:

 

 

root@Server:~# cd /boot
root@Server:/boot# preclear_disk.sh -l
====================================1.13
Disks not assigned to the unRAID array
  (potential candidates for clearing)
========================================
     /dev/sdi = ata-WDC_WD20EZRX-00D8PB0_WD-WMC4M0240182
root@Server:/boot# preclear_disk.sh -A /dev/sdi
BLKRRPART: Input/output error
Sorry: Device /dev/sdi is not responding to an fdisk -l /dev/sdi command.
You might try power-cycling it to see if it will start responding.
root@Server:/boot#

 

I tried rebooting the server but I get the same thing. Any ideas?

 

Thanks,

 

Adam

are you able to get a smart report for the drive?

 

if yes, post it, if not, check cable connections and try again.

Checked and reseated all connectors, powered up and was able to get a smart report attached.

 

Thanks for your help!

 

Adam

 

P.S. Have started a Long Smart Test

smart.txt

this seems troubling, none of my drives experience this.

 

 

  7 Seek_Error_Rate        0x002e  001  001  000    Old_age  Always      -      38693

 

 

i'd try a new sata cable, then rma the drive personally if that doesnt solve it.

*sigh* I don't seem to be having much luck.

 

I changed the SATA cable and tried again but couldn't get it to write a smart report. Not sure why...

 

I have purchased another drive and set that going about 24hrs ago. Expected completion is around 29hrs but when I check putty it shows this:

 

================================================================== 1.13
=                unRAID server Pre-Clear disk /dev/sdi
=               cycle 1 of 1, partition start on sector 63
= Disk Pre-Read in progress: 96% complete
= ( 1,923,070,464,000  bytes of  2,000,398,934,016  read ) 70.3 MB/s
=
=
=
=
=
=
=
=
=
=
Disk Temperature: 43C, Elapsed Time:  5:04:09

 

Which seems to imply that after 5 hrs the process stopped or probably more likely Putty disconnected. (Spindown delay is set to never).

 

How do I reconnect Putty so I can view the current progress?

 

Thanks.

 

Adam 

Are you pre-clearing with screen, If not i advise you to do so, details are on the wiki.

 

When you pre-clear with screen you able to disconnect and reconnect to the session as and when ever you like to check on progress, if done with putty when you close the putty window the pre-clear stops.

Hi,

 

I just followed the instructions here, I am not sure of the 'screen' option?

 

http://lime-technology.com/forum/index.php?topic=2817.0

 

I'll see if I can find it.

 

So assuming putty dropped the connection for some reason , it says 'inactive' in the window title , I need to start the preclear again?

 

Cheers.

 

Adam

http://lime-technology.com/wiki/index.php/Configuration_Tutorial#Preclearing_With_Screen

 

Found it, thanks very much

yes use screen because when putty disconnects (without using screen) it terminates the pre-clear.

 

 

on the plus side, 5hrs and 90+% done pre-read is a good sign!!

Ok quick question before I start the preclear again.

 

The guide suggests -A but all the rest of my drives and default partition format is MBR unaligned.

 

Should I stick to: preclear_disk.sh /dev/sdi

 

Also my array has been unprotected now for a week and I would rather get my new drive in ASAP, should I really run it 3 times?

 

Thanks

 

Adam

On all my drives i used the -A option i believe its used if need and not it not, but your best bet is wait for someone more experienced to chime in, still new to this.

 

Regarding the 3 passes, I always do 3 no matter the size of the drive 500GB to 3TB as once it passes these 3 passes i trust the drive.

 

It gives me piece of mind, but its up to you at the end of the day.

Thanks guys!

 

Not sure what an offset jumper is so will avoid and use -A

Ok, kind of hard to add a jumper by accident so all good, cheers!

 

Adam

Hi Chaps,

 

I came down this morning to this:

 

================================================================== 1.13
=                unRAID server Pre-Clear disk /dev/sdi
=               cycle 1 of 1, partition start on sector 64
= Disk Pre-Clear-Read completed                                 DONE
= Step 1 of 10 - Copying zeros to first 2048k bytes             DONE
= Step 2 of 10 - Copying zeros to remainder of disk to clear it DONE
= Step 3 of 10 - Disk is now cleared from MBR onward.           DONE
= Step 4 of 10 - Clearing MBR bytes for partition 2,3 & 4       DONE
= Step 5 of 10 - Clearing MBR code area                         DONE
= Step 6 of 10 - Setting MBR signature bytes                    DONE
= Step 7 of 10 - Setting partition 1 to precleared state        DONE
= Step 8 of 10 - Notifying kernel we changed the partitioning   DONE
= Step 9 of 10 - Creating the /dev/disk/by* entries             DONE
= Step 10 of 10 - Verifying if the MBR is cleared.              DONE
= Post-Read in progress: 55% complete.
(  1,110,412,800,000  of  2,000,398,934,016  bytes read ) 109 MB/s
Disk Temperature: 40C, Elapsed Time:  15:48:33

 

I assume again that putty lost connection at some point so I created a new putty connection and tried reconnecting to screen (screen -r) and got this:

 

root@Server:~# screen -r
There are several suitable screens on:
        21603.pts-0.Server      (Detached)
        27289.pts-0.Server      (Detached)
        27271.pts-0.Server      (Detached)
Type "screen [-d] -r [pid.]tty.host" to resume one of them.
root@Server:~#

 

I am not really sure what to do now?

 

Thanks.

 

Adam

screen -r pid

 

screen -r 21603

screen -r 27289

screen -r 27271

 

One of them should be the one your after, if there not just type exit to close out of them and detach.

 

Some screen commands

 

Ctrl + A + c = To open a New Windows

Ctrl + A + n = next window (if you have multiple windows running)

Ctrl + A + 0…9 = go to window number x where it should come with 0 to 9.

Ctrl + A + d = detach from a window. Say, if you want to close your session and return later.

Ctrl + d = detach and clear the screen session,

 

To re-attach to a screen, you may need to check the number of screen is running.

 

screen -ls

 

Mainly use Ctrl + A + c  to create another instance of screen so you can run multiple pre clears and Ctrl + A + d to detach your current session and then screen -r to reconnect when you return later.

 

Ok I tried a few things below but no joy so far:

 

root@Server:~# screen -ls
There are screens on:
        21603.pts-0.Server      (Attached)
        27289.pts-0.Server      (Detached)
        27271.pts-0.Server      (Detached)
3 Sockets in /root/.screen.

root@Server:~# screen -r 21603
There is a screen on:
        21603.pts-0.Server      (Attached)
There is no screen to be resumed matching 21603.
root@Server:~# screen -r 27289
Cannot find terminfo entry for 'vt100'.
root@Server:~# screen -r 27271
Cannot find terminfo entry for 'vt100'.

 

Sorry to be a PITA!

 

Adam

your attached to screen on 21603.pts-0.Server (Attached) now use Ctrl + A  n = to get to the next session