No disks will spin down

[aitf311]

I am having trouble getting any of my disks to spin down. On their own, they will never spin down, but I can click the disk on the main page to spin them down. When they do manage to show the spun down icon, the drive is back up and spinning within a matter of minutes...even with no use.

 

My Setup:

• Unraid 5.0.5 Pro

• Motherboard: BIOSTAR Hi-Fi A85W

• CPU:  AMD A10-5800K

• Ram: 8GB

 

Addons:

Dynamic with the following plugins

• WebGui

• Active Streams

• Cache Dirs

• Disk Health

• System Info

• Plugin Control

 

PhAzE Plugins

• CouchPotato

• Plex Media Server

• SABnzbd

• Sick-Beard

 

Disk Setup:

• Unraid Flash Drive: 16GB / 15.6GB free

• Parity: 4TB

• Disk 1: 4TB / 1.58TB free

• Disk 2: 2 TB / 118GB free

• Disk 3: 3TB / 625GB free

• Disk 4: 2TB / 121GB free

• Disk 5: 2TB / 70.1GB free

• Disk 6: 4TB / 1.03TB free

• Disk 7: 2TB / 147GB free

• Disk 8: 1TB / 72.4GB free

• Cache Drive: 120GB SSD

 

Shares:

SMB Security Settings: Export: Yes, Security: Public

• Applications

o Allocation: High-water

o Minimum free space: 0kb

o Split level: not set

o Included disks: All

o Excluded disks: none

o Use cache disk: Yes

• Documents

o Allocation: High-water

o Minimum free space: 0kb

o Split level: not set

o Included disks: All

o Excluded disks: none

o Use cache disk: Yes

 

• Media

o Allocation: High-water

o Minimum free space: 0kb

o Split level: not set

o Included disks: All

o Excluded disks: none

o Use cache disk: Yes

 

• Apps (on cache drive – used with PhAzE plugins)

o Allocation: High-water

o Minimum free space: 0kb

o Split level: not set

o Included disks: All

o Excluded disks: none

o Use cache disk: no

 

Users:

root

 

Disk Settings:

• Enable auto start: Yes

• Default spin down delay: 30 minutes

• Force NCQ disabled: Yes

• Enable spinup groups: No

• Default partition format: MBR: 4K-aligned

• Md_num_stripes: 1280

• Md_write_limit: 768

• Md_sync_window: 384

 

Network Settings:

• Enable bonding: no

• Bonding mode: active-backup (1)

• Obtain IP address automatically: yes

• Obtain DNS server address automatically: yes

 

Share Settings:

• Enable user shares: Yes

• Included disks: All

• Excluded disks: none

• Use cache disk: Yes

• Min free space: 2000000

• Mover schedule: Daily

• Time of day: 03:40

• Mover logging: Disabled

 

SMB Settings:

• Enable SMB: Yes (Workgroup)

• Workgroup: WORKGROUP

• Local Master: no

 

Folder Caching:

• Folder caching function: enabled

• Wait until array is online: Yes

• Force disks busy: no

• Suspend during ‘Mover’ process: yes

• Scan user shares: Yes

 

syslog.txt

[sureguy]

Try booting into safe mode, see what happens.

[aitf311]

my go file has not been modified and is identical to the go-safe file. I assume this means I'm essentially always running the equivalent of safe mode, is this correct?

 

Anyone have any ideas what this can be? These disks just keep spinning all day long.

[JonathanM]

my go file has not been modified and is identical to the go-safe file. I assume this means I'm essentially always running the equivalent of safe mode, is this correct?

 

Anyone have any ideas what this can be? These disks just keep spinning all day long.

Safe mode is accessed when booting the server by selecting that option in the boot menu with the local keyboard. The go file is only one small part of it. Boot in safe mode and see if the issue is still there.

[Frank1940]

In Dynamix, on the "Settings" tab, what is the "Disk Health" setting for 'Enable background polling for spun-down disks:'?

[aitf311]

OK so I boot into safe mode and then spun down all the disks using the spindown button. They all spun down for about 10 seconds and then gradually started to spinup again, even without any interaction with the tower.

 

I have three pcs in my house, all Windows 8.1, and all three map my user shares as network drives (movies, music, documents). I shut down two of the pcs and removed the shares on this PC and have been going for about 10-15 minutes without the drives spinning up. That's the longest I've seen the drives stay spun down. Is this expected behavior from mapped network drives? Any way around this?

[Frank1940]

One thing that can happen with Windows is that it goes through the entire directory tree to attached the 'open with' program (may have the wrong phase here) that is associated with the file extension.  (Example:  .DOC files with MS Word.)  If you have a lot of files in the share this can take hours...

[aitf311]

Frank, background polling is set to 'no'.

[Frank1940]

Frank, background polling is set to 'no'.

 

There is where mine is set and I don't have the issue.

 

Since the problem seems to be occurring when you connect from Win8 machines, I suspect that your problem has its roots there.  Do you connect or reconnect those computers regularly or are they connected for long period of time? 

 

Have you tried waiting for a few hours to see if the drives spin down if Win8 has decided that it must 'catalog' some data from those shares and it takes a long period of time to do so? 

[aitf311]

Well it turns out I got a little excited too soon. I left for an hour with just no pcs connected to the tower, and yet all my drives are still spinning. My spin down time is set to 30 minutes.

[trurl]

Try

lsof /mnt/disk*

 

[aitf311]

trurl, I ran the command last night, and had 1 thing listed, an images folder in an installation folder for some mac software I downloaded at least a year ago and haven't touched since. I deleted the entire parent folder.

 

I checked the tower first thing this morning and all of my drives were still spinning but the command didn't show anything being accessed.

[Frank1940]

On the 'Main' tab, click on each disk name  ('Parity', 'Disk 1', etc.) and see what the 'Spin Down Delay' is set to.  (Mine is set to 'Use default'.)

[aitf311]

All 8 of mine are set to default as well.

[dgaschk]

Look for the "inotifywait" package.