Hello,

 

My server have been run for months, year+ on all brand new motherboard and hardrive Western Digital WD20EARS,  6 slots licences. 5.0 beta 6a.

 

It were in operational service without any problem.

Today I saw a red ball.

 

I consult the forum and get the smart log file on the parity (sdb), disk1(sdc), disk2 (sde).

 

Disk1(sdb)  is red balled.

log file name as:

parity martb.txt

disk1 smartc.txt

disk2 smarte.txt

 

I read the forum and look at the log file at Reallocated_Sector_Ct.

 

I'm not sure what to decipher. is 155 0  7  mean bad ? really bad or ?.

 

the parity and disk2 is show 0 o 7

 

I need help on reading the log file  how to evaluate a bad situation.

 

Should I replace the drive now should I bring the array down and power off ?.

 

Thank you for helping.

Darn,

 

Forgot the log file

smartc.txt

smartb.txt

smarte.txt

Disk 1 does not look healthily. Monitor the Reallocated_Sector_Ct and the Current_Pending_Sector values. I'd rebuild disk1 on a spare drive and run pre-clear on the current disk1.

 

Thank dgaschk

 

Hi I'm thinkn the same too.

 

1) Get a new drive WD20EARX or WD20EARS. run 1 preclear cycle.

2) Replace disk1 with the new precleared drive.

3) Once the array is rebuild and operational, run preclear on the red ball drive 1 cycle to see what happen. then run a smart test to comfirm that it have heath issue.

4) if it is a health issue then try to RMA from WD.

 

 

Darn when the 2T was $79 ea I was short on cash. Now that I have some cash they are $129 ea.

 

 

Does any know how to start an RMA process with WD ?

 

I brought the drive from Micro Center.

 

Thanks

Hi Dgaschk,

 

Here is what I'm thinking:

 

1) Get a new drive WD20EARX or WD20EARS. run 1 preclear cycle.

2) Replace disk1 with the new precleared drive.

3) Once the array is rebuild and operational, run preclear on the red ball drive 1 cycle to see what happen. then run a smart test to comfirm that it have heath issue.

4) if it is a health issue then try to RMA from WD.

5) I can't RMA it then I would like to use it as a cold backup unit as follow:

That could work.

http://websupport.wdc.com/warranty/rmainfo.asp?custtype=end〈=en

 

Thanks for helping.

 

one more thing. should I RMA the redball ? or can I RMA the redball. have any one done that base on unraid red ball and Reallocated_Sector_Ct counter ?

 

 

 

They will take it back. You can probably get it to fail completely with enough pre-clear cycles.

Thanks,

 

Brand new preclear 2T completed. I'm follow the below link for drive replacement.

 

http://lime-technology.com/wiki/index.php?title=Replacing_a_Data_Drive

 

Every thing go by the book. ETA to complete rebuild: 9 hours.

 

 

a few questions:

 

  1) is there a way or in a new release to tell exactly when the redball occur ?. This is my first red ball experience so my new procedure is check the console BEFORE and AFTER a files transaction.

 

  2) I can't believe how much my family and I take the server operational usefulness for granted so much past year. When I saw the redball, I sorta "Oh no!" and slap the alarm buttom to activate the klaxon and "That it's we shutting down". Anyway is there some one out there have come up with a scheme to keep the server running without risking a total meltdown ? .i.e a second redball.

 

3) I'm wondering about having 2 full data drive (2T ea) against 4 (1Tea) incase of a full melt down (ie 2 redball). a 2ea 1T redball still sorta 50% data loss compare to 100% data loss of 2ea 2T 2 redball.

 

4) what ind of files backup system unraid user have these day?

 

Once the replacement disk is rebuild, parity sync. , data file spot check then I'll run a preclear on the redball and post here. most likely Friday of this week before the preclear on the redball is completed.

 

Thanks dgaschk, Joe, BRit for hand holding.