What would cause a drive to never spin down? I installed a new parity drive and am building parity on the new drive. All other drives spun down hours ago, but for some reason something is preventing disk1 from spinning down. Is there a way to tell which files are open on individual drives? I remember being able to do this on 4.7, but I can't find it in 5.0. The "open files" list on 5.0 is a giant list showing every process compared to the 4.7 list which only showed actual open files on the array. Array drives are all <2TB and the new parity is 6TB, so the array drives should all spin down after the rebuild passes 2TB. Ideas?

 

 

lsof /mnt/*

Thanks! I'll use that command the next time I'm doing a parity check to see what the culprit is. Is there a plugin or another way to get this functionality form the web interface? Right now all that shows up is sleep and cache_dir.

Thanks! I'll use that command the next time I'm doing a parity check to see what the culprit is. Is there a plugin or another way to get this functionality form the web interface? Right now all that shows up is sleep and cache_dir.

http://lime-technology.com/forum/index.php?topic=27051.msg237095#msg237095

http://lime-technology.com/forum/index.php?topic=27051.msg237095#msg237095

 

Just what I'm looking for, thanks again!

So I'm running another check and sure enough disk1 is not spinning down and the check is at 4.9TB location when all data drives are 2TB or less. Looks like there are no open files on any disks. What do you guys think? What would keep the drive from spinning down?

 

COMMAND    PID USER  FD  TYPE DEVICE SIZE/OFF NODE NAME

cache_dir 3025 root  cwd    DIR    9,1      104    2 /mnt/disk1

cache_dir 3026 root  cwd    DIR    9,2      176    2 /mnt/disk2

cache_dir 3027 root  cwd    DIR    9,3      176    2 /mnt/disk3

cache_dir 3028 root  cwd    DIR    9,4      128    2 /mnt/disk4

cache_dir 3029 root  cwd    DIR    9,5      152    2 /mnt/disk5

cache_dir 3030 root  cwd    DIR    9,6      152    2 /mnt/disk6

sleep    3127 root  cwd    DIR    9,1      104    2 /mnt/disk1

sleep    3128 root  cwd    DIR    9,6      152    2 /mnt/disk6

sleep    3129 root  cwd    DIR    9,2      176    2 /mnt/disk2

sleep    3130 root  cwd    DIR    9,5      152    2 /mnt/disk5

sleep    3131 root  cwd    DIR    9,3      176    2 /mnt/disk3

sleep    3132 root  cwd    DIR    9,4      128    2 /mnt/disk4

You sure you don't have an override on the spin-down time  for that disk?

 

There actually could be an issue that disk1 doesn't spin down during a parity check for whatever reason. Most people would never notice it since it isn't that common historically to have such a size disparity between parity and data.

You sure you don't have an override on the spin-down time for that disk?

 

There actually could be an issue that disk1 doesn't spin down during a parity check for whatever reason. Most people would never notice it since it isn't that common historically to have such a size disparity between parity and data.

 

It spins down a second after the parity drive spins down after the check is complete.

So disk1 and the parity drive are both set to the same spinup group. I'll disable them after the check is complete.