• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Identify clicking drive in array?

C

ChineseStunna

Limp Gawd
2FA
Joined
Jan 11, 2002
Messages
442
Hi all,

I've recently updated my array to 10x3TB drives in a MDADM RAID6 array but it seems like one of the drives might be on its way out. I hear clicking for a couple seconds during which time array grinds to a halt, this happens every 5-10 min. MDADM shows array is clean and healthy and smartctl reports no errors for any of the drives

I've checked /dmesg and /var/log/messages but neither has any direct indicators of a failed/failing drive, any experts have ideas on how to ID the faulty drive out of the array?

I obviously can't tell acoustically since they are all mounted in close proximity in the drive cages and the sound only comes intermittently.

Thanks!
 
Have you tried something like a dd dump of the first 8GB or so of each drive and check performance?

Code: 
dd if=/dev/sda of=/dev/null bs=8M count=1000

smartctl reports no errors for any of the drives
Click to expand...

Current_Pending_Sector and Offline_Uncorrectable both have raw values of 0.

Is Load_Cycle_Count low on all drives? Perhaps some drives are parking the heads after some delay.
 
Pretty simple.

iostat -x 1

look for the disk that constantly has high %util when you hear the clicking.
 
Current_Pending_Sector and Offline_Uncorrectable both are at 0 for all drives.

Load_Cycle_Count is low <10 for all drives except /dev/sdd and it already shows a reallocate count of 86 after only being up and running for a week vs. 0 for all other drives so I'm pretty confident it's this one.

Thanks for the guidance!
 
Last edited:
drescherjm said:
Have you tried something like a dd dump of the first 8GB or so of each drive and check performance?

Code: 
dd if=/dev/sda of=/dev/null bs=8M count=1000



Current_Pending_Sector and Offline_Uncorrectable both have raw values of 0.

Is Load_Cycle_Count low on all drives? Perhaps some drives are parking the heads after some delay.
Click to expand...

I think I've found the disk per my last post, but I doubt it's due to the delay as the drives are running full tilt right now sync data back to my workstation. I haven't heard it for the past hour or so, weird.
 
You can also look into the dmesg output whether there are any error messages. In fact, this is the first thing I do on all Linux or Unix-related systems in case of problems.
 
In fact, this is the first thing I do on all Linux or Unix-related systems in case of problems.
Click to expand...

Same for me. The OP mentioned that there was nothing suspicious in the dmesg in the first post.
 
You must log in or register to reply here.
Back
Top