I have been setting up a server for a while now. After installing a couple of packages, I just got around to preclearing 3x2TB hitachis yesterday. I didn't quite grasp the output from the preclears after 24 hours runtime, but I left the server up and running to check into it today. I found out the reports for the disks probably showed everything fine. After a while I tried to rerun preclear and get back into Unmenu, but it didn't work. Looking into it I could not find the mount for the disk. Later after looking into it I see the USB has unmounted itself, several times even. It looks like it has jumped from sdd at boottime to sde after the first unmount, and that's of course why it wont reattach. Right now I am keeping the server up and running for answering anything you'd want to know.

 

I am running 5.0b2 (hoping to see the AFP soon) on a Asus P7H55D-M Evo with Ci3 and 2 Gb RAM, currently only the three Hitachis connected.

 

The running Syslog is attached below (had to zip full file, .txt has a bunch of preclear truncated), but a few highlights might be appropriate:

 

First attaching the USB:

Nov  7 19:46:00 Tower kernel: scsi 8:0:0:0: Direct-Access     USB Mass Storage Device        PQ: 0 ANSI: 0 CCS
Nov  7 19:46:00 Tower kernel: usb-storage: device scan complete
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] 7698432 512-byte logical blocks: (3.94 GB/3.67 GiB)
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Write Protect is off
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Mode Sense: 03 00 00 00
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Assuming drive cache: write through
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Assuming drive cache: write through
Nov  7 19:46:00 Tower kernel:  sdd: sdd1
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Assuming drive cache: write through
Nov  7 19:46:00 Tower kernel: sd 8:0:0:0: [sdd] Attached SCSI removable disk

 

Then the first bomb:

Nov  9 10:32:44 Tower kernel: usb 2-1.1: USB disconnect, address 3
Nov  9 10:32:45 Tower kernel: usb 2-1.1: new high speed USB device using ehci_hcd and address 4
Nov  9 10:32:45 Tower kernel: usb 2-1.1: configuration #1 chosen from 1 choice
Nov  9 10:32:45 Tower kernel: scsi9 : SCSI emulation for USB Mass Storage devices
Nov  9 10:32:45 Tower kernel: usb-storage: device found at 4
Nov  9 10:32:45 Tower kernel: usb-storage: waiting for device to settle before scanning
Nov  9 10:32:50 Tower kernel: scsi 9:0:0:0: Direct-Access     USB Mass Storage Device        PQ: 0 ANSI: 0 CCS
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] 7698432 512-byte logical blocks: (3.94 GB/3.67 GiB)
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Write Protect is off
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Mode Sense: 03 00 00 00
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Assuming drive cache: write through
Nov  9 10:32:50 Tower kernel: usb-storage: device scan complete
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Assuming drive cache: write through
Nov  9 10:32:50 Tower kernel:  sde: sde1
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Assuming drive cache: write through
Nov  9 10:32:50 Tower kernel: sd 9:0:0:0: [sde] Attached SCSI removable disk

 

The subsequent messages is from some apps trying to read from the orphan mount, I guess.

Nov  9 10:41:20 Tower kernel: FAT: Directory bread(block 4318160) failed

 

This is how the system still think the mounts look like:

root@192:~# mount
fusectl on /sys/fs/fuse/connections type fusectl (rw)
usbfs on /proc/bus/usb type usbfs (rw)
/dev/sdd1 on /boot type vfat (rw,noatime,nodiratime,umask=0,shortname=mixed)
root@192:~# 

 

Well, would someone agree with me that this USB is probably not reliable? I didn't touch anything on the server between 22.00 last night and 12.00 today, when this problem started appearing. And to think I ordered (and got delivered) a pro key last weekend. I hope I have a chance to reregister the license to another USB if this one is screwed.

 

My first guess: Someone asks me to detach the USB and run checkdisk from windows. I guess that will not matter since it's already detached from the mountpoint, but I'll wait for a while if someone want me to check something while it's still there.

 

/Lars Olof

unraidlog.txt.zip

unraidlog2.txt

Could it be an intermittent connection to the USB drive?

Could it be an intermittent connection to the USB drive?

 

Wouldn't really think so. It's in the back panel of my motherboard. No cables, just straight into the port. Weight or leverage is no issue, since this USB is only protruding a few mm outside the jack (mini sized).

Weird... my 4.5.6 unRAID has been running 24/7 for many months now and today from the CLI I tried to "cd /boot" but it was empty!

Dmesg returned a bunch of:

 FAT: Directory bread(block 7679) failed 

 

I stopped the array and restarted. Everything looks ok so far but I'm worried about the flash drive now. For information, the USB flash drive is plugged directly on the motherboard inside the case.

Put it in your window's PC, run scandisk/checkdisk on it.

Full scandisk under WinXP didn't report any problem.

 

My hardware configuration can be found here: http://lime-technology.com/forum/index.php?topic=2031.msg59139#msg59139

I had the pretty much the same problem you are reporting when I first setup unRAID. I checked it with XP and didn't see any issues, but was a little paranoid. I ended up just formatting the USB, creating a new build and copying over my config files and its been running for months just fine.