Hi all.  My unraid server has been running flawlessly for 8 months.  I just noticed that my cache drive hasn't been moving files to the correct shares.  Looking at the system log it looks like the mover is attempting to place them on a disk that doesn't have enough space.  I've marked this disk as "excluded" on the share, I've changed the allocation method to "Most-Free" but the mover still tries to put the files onto that disk.

 

Has anyone else experienced this?  Am I doing something wrong?

FYI - you haven't told us anything about your hardware, version running or attached the syslog....

 

I assume you have manually tried to run the mover and same thing happens?

 

Shawn

Sorry for not including that information to begin with.  I did try running the mover manually and the same thing happens.

 

Unraid version: 4.7

Cache drive: Maxtor 300GB SATA drive (old one that I had lying around)

All other hard drives (including parity) are WD Green 2TB drives (EARS)

Motherboard: Gigabyte AMD 760G + SB710

Processor: AMD Sempron AM3 Sargas

 

Here's an excerpt from the syslog.  I'm trying to move 20 or so files and each have the same error.  I did successfully move other files off of it this morning, but now the mover keeps going to disk 1, even after excluding it from the share.

 

 

Nov  5 09:40:55 Tower logger: ./TV/Enterprise/Season 4/ENTERPRISE s4e9 s4e10 s4e11 s4e12.iso

Nov  5 09:40:55 Tower logger: .d..t...... TV/Enterprise/Season 4/

Nov  5 09:40:55 Tower logger: >f+++++++++ TV/Enterprise/Season 4/ENTERPRISE s4e9 s4e10 s4e11 s4e12.iso

Nov  5 09:40:56 Tower shfs0: shfs_write: write: (28) No space left on device

Nov  5 09:40:56 Tower logger: rsync: writefd_unbuffered failed to write 4 bytes [sender]: Broken pipe (32) (Minor Issues)

Nov  5 09:40:56 Tower logger: rsync: write failed on "/mnt/user0/TV/Enterprise/Season 4/ENTERPRISE s4e9 s4e10 s4e11 s4e12.iso": No space left on device (28) (Minor Issues)

Nov  5 09:40:56 Tower logger: rsync error: error in file IO (code 11) at receiver.c(298) [receiver=3.0.2] (Errors)

Nov  5 09:40:56 Tower logger: rsync: connection unexpectedly closed (31 bytes received so far) [sender]

Nov  5 09:40:56 Tower logger: rsync error: error in rsync protocol data stream (code 12) at io.c(635) [sender=3.0.2] (Errors)

I'm having a similar issue. Posted mine here: http://lime-technology.com/forum/index.php?topic=16301.0

 

I think I may have stumbled upon the solution - running the mover now to see.

 

I would welcome any suggestions.

Unfortunately I tried using midnight commander to move files from the cache to a specific disk but they didn't move and MC zeroed out the bytes on the cache drive.  I lost the files that were cached but I'm going to try to restore them later today.  I think it's time to replace the cache drive.

I wonder if updating the min-max to most free, after the files are copied, might not affect them as they have already been "allocated" when you copied them on - just a thought. Best bet, maybe just map from another system to \\tower\disk1 (or whatever disk), and then to the cache drive and manually move them off? Try that... werid error with MC though - have never had any issues using it.

 

Have you run a SMART report on the cache drive?

 

Shawn

Not yet.  It's one of the oldest drives I have so I'm not really surprised something like this is happening.  I'm just curious why Disk1 (full) was the ONLY disk the mover was attempting to move files to.  I need to take a look at my settings for the entire system to make sure this doesn't happen again.

File structure?

Split level setting?

Any empty directories still remaining on disk1?