TomatoUSB

I have an ftp server setup usb hard drive but I've found that occasionally logging into the ftp server produces no directories. I've also found this to be the case when I try to access the hard drive from Windows Explorer via a network drive.

Unmounting and mounting the drive in the tomato interface solves the problem for both the FTP and windows explorer.

How do I prevent the directories from disappearing like that?

RT-N16 Tomato Firmware v1.28.9054 MIPSR2-beta K26 USB vpn3.6

Any tips?

Similar problem here. I have external HD in NFTS mounted on Shiby's Tomato Firmware 1.28.0000 MIPSR2-108 K26 USB AIO on Netgear WNR 3500 L 2. I started Samba server and mapped the network drive in Windows 7. I do have problems accessing the drive from Windows 7. Did some research from the Windows 7 point of view with partial success (most of the time I can access the drive), but from time to time the HD just disapeears. It is not only unmounted but it is not seen by the router. Even if I unplug it and plug it again it is still not seen until I restart the router.
When I check the logs, I see that the drive is continously dismounted and mounted like that:
Mar 17 18:44:29 ksiega user.info kernel: usb 1-1: USB disconnect, address 2
Mar 17 18:44:29 ksiega user.info kernel: scsi 0:0:0:0: [sda] Unhandled error code
Mar 17 18:44:29 ksiega user.info kernel: scsi 0:0:0:0: [sda] Result: hostbyte=0x01 driverbyte=0x00
Mar 17 18:44:29 ksiega user.info kernel: scsi 0:0:0:0: [sda] CDB: cdb[0]=0x28: 28 00 00 00 00 3f 00 00 20 00
Mar 17 18:44:29 ksiega user.warn kernel: end_request: I/O error, dev sda, sector 63
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 0
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 1
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 2
Mar 17 18:44:29 ksiega daemon.notice ntfs-3g[1332]: ntfs_attr_pread_i: ntfs_pread failed: Input/output error
Mar 17 18:44:29 ksiega daemon.notice ntfs-3g[1332]: Failed to read index block: Input/output error
Mar 17 18:44:29 ksiega daemon.notice ntfs-3g[1332]: Unmounting /dev/sda1 (halas)
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 3
Mar 17 18:44:29 ksiega user.err kernel: scsi 0:0:0:0: rejecting I/O to dead device
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 0
Mar 17 18:44:29 ksiega user.err kernel: scsi 0:0:0:0: rejecting I/O to dead device
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 44
Mar 17 18:44:29 ksiega user.err kernel: scsi 0:0:0:0: rejecting I/O to dead device
Mar 17 18:44:29 ksiega user.err kernel: Buffer I/O error on device sda1, logical block 44
Mar 17 18:44:29 ksiega user.info hotplug[2728]: USB partition unmounted from /tmp/mnt/halas
Mar 17 18:44:36 ksiega user.info kernel: usb 1-1: new high speed USB device using ehci_hcd and address 3
Mar 17 18:44:36 ksiega user.debug hotplug[2749]: Attached USB device 1-1:1.0 [INTERFACE=8/6/80 PRODUCT=1058/704/105]
Mar 17 18:44:36 ksiega user.info kernel: usb 1-1: configuration #1 chosen from 1 choice
Mar 17 18:44:36 ksiega user.info kernel: usb-storage: Quirks match for vid 1058 pid 0704: 8000
Mar 17 18:44:36 ksiega user.info kernel: scsi1 : SCSI emulation for USB Mass Storage devices
Mar 17 18:44:38 ksiega user.notice kernel: scsi 1:0:0:0: Direct-Access WD 2500BEV External 1.05 PQ: 0 ANSI: 4
Mar 17 18:44:38 ksiega user.notice kernel: sd 1:0:0:0: [sda] 488397168 512-byte hardware sectors (250059 MB)
Mar 17 18:44:38 ksiega user.notice kernel: sd 1:0:0:0: [sda] Write Protect is off
Mar 17 18:44:38 ksiega user.debug kernel: sd 1:0:0:0: [sda] Mode Sense: 21 00 00 00
Mar 17 18:44:38 ksiega user.err kernel: sd 1:0:0:0: [sda] Assuming drive cache: write through
Mar 17 18:44:38 ksiega user.err kernel: sd 1:0:0:0: [sda] Assuming drive cache: write through
Mar 17 18:44:38 ksiega user.info kernel: sda: sda1
Mar 17 18:44:38 ksiega user.notice kernel: sd 1:0:0:0: [sda] Attached SCSI disk
Mar 17 18:44:39 ksiega user.info hotplug[2759]: USB ntfs fs at /dev/sda1 mounted on /tmp/mnt/halas

Would appreciate any help with that.

This sounds like my issue. I'm not entirely through diagnosing it, but the symptoms sound the same.

I have 2 EXT3 drives in an enclosure. They work most of the time, however (at random) during my nightly backups when the drives are hit hard they will dismount and re-mount.

I usually have 2 1TB ext3 mounts sda1 and sdb1. The failure manifests in one of a few ways:

1) sda1 becomes sdc1 and is still accessable.
2) sda1 simply goes away and won't re-mount ever again. I have to wipe the drive
3) all drives disappear and won't re-appear until I reboot tomato.

I'll try to get some logs. I usually panic and reboot when this happens and lose any information that might be there. This is biting me hard because I rely on this for backups and it's turning into a very unreliable system.

I know this is Tomato combined with the ASUS RT-N16 since it has happened with 2 different HDD enclosures from 2 different vendors.

There is hope: I have a USB protocol analyzer. As a last resort I can let that run overnight on the backup and probably be able to point the finger one way or the other.

This issue may be related (or the same) as this thread:

USB drive disappears (Sorry, can't post a link due to being a low karma user)

I've inserted "echo 128 >/sys/block/sda/device/max_sectors" in my Init script like so:

—- snip —

sleep 30

echo 128 > /sys/block/sda/device/max_sectors
mkdir /tmp/mnt/sda1
mount /dev/sda1 /tmp/mnt/sda1

- snip -

So far so good!

I was able to run a full copy of my backups last night with this fix and no spontaneous dismounts! This was a "cp" of one 3TB drive to another on the same USB port through a hub. I've never been able to do this before.

After further research, it appears that this is a necessary change in order to successfully use a whole bunch of USB storage devices. I'd say that this should be front and center on the list of required modifications to make this kernel reliable with external storage.

Could an admin or someone please save others from this suffering and stick a note in the install to do this? I know it looks like there are no more releases, but think of your poor suffering fellow geeks. How many countless man hours are spent rediscovering this many years old issue?

Thanks! Hope this helps someone out there.

I experience the same problem as above. But I don't know where the init script is located or how to change it? Thanks in advance!