padavanfirmware / rt-n56u

Padavan Firmware
http://padavanfw.net/
6 stars 4 forks source link

usb hard disk access via samba / dlna fails after short time #220

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago
What steps will reproduce the problem?
1.samba file copy of a larger file fails after short time with disk I/O error, 
no files visible then anymore
2.same when playing a movie via dlna from the router's usb disk on an external 
rendering device, no files on device then anymore

What is the expected output? What do you see instead?
file copying or playing a movie should simply proceed ;-)

What version of the product are you using? On what operating system?
rt-n56u with 1.1.1.8e-b2 + Samsung Bluray BD-D5100 + laptop with kubuntu 
11.10/64

Please describe the problem as detailed as it's possible.
If you have connection problem, then syslog file is required. (please do attach 
it as a file)
Note that if there will be a poor problem description the issue status will be 
changed to 'Invalid'!
after initial connection of a usb hard disk (tried with 2 different ones, a 
500GB WD with ntfs, and a 1TB Samsung with ext3) everything looks fine. All 
files are visible via smb or dlna. But a few seconds after starting a larger 
file copy from that usb disk to my laptop via samba (wireless or wired), the 
copying stops with some disk I/O error, and the disk isn't accessible anymore. 
Same happens when playing a movie from that usb disk via dlna, just takes 
longer, but also the router isn't accessible on the bluray anymore afterwards. 
The general log and the dhcp leases log are completely emptied. It's not 
possible to remount the usb disk, only a reboot of the router brings it to life 
again. Normal router operations still seem to work. This effect definitely is 
new with the latest version, and it's 100% reproducible!

Original issue reported on code.google.com by wolfgang.griech@gmail.com on 19 Apr 2012 at 5:06

GoogleCodeExporter commented 9 years ago
This regression already fixed in 1.1.1.8e-b3.

Original comment by andy.pad...@gmail.com on 19 Apr 2012 at 7:17

GoogleCodeExporter commented 9 years ago
wow thx, that was quick, and the issue seems to be fixed too!
and btw, my Epson Stylus DX8400 USB printer works fine too with this update..

Original comment by wolfgang.griech@gmail.com on 19 Apr 2012 at 8:04

GoogleCodeExporter commented 9 years ago
have to correct myself, the issue isn't really fixed, just takes longer now! 
Played a movie via DNLA from the USB disk (Samsung 1TB Ext3) connected to the 
router, and after about 15 minutes I got a disconnect. The router folder is 
still there  on my Samsung Bluray player, but when I go to the Videos folder, I 
get the message that the folder is empty. And the ASUS web interface tells the 
USB disk is unmounted. The syslog / general log is completely wiped out, as 
before. Unplugging the USB disk and plugging it back in doesn't solve the 
issue, it's not possible to renew the system status, e.g. mount the disk. Also 
"transmission" tells on a formerly running download: "No such file or 
directory". Everything else seemed to be working as normal. After a "soft 
reboot" from the web interface the USB disk could be mounted and was working 
again..

Original comment by wolfgang.griech@gmail.com on 19 Apr 2012 at 8:25

GoogleCodeExporter commented 9 years ago
some more hopefully helpful information when this happens, again after 10-15 
minutes playing a movie via DLNA from that USB disk:

- ASUS web interface shows the USB disk is unmounted

- mount shows:
/dev/sda2 on /media/AiDisk_a2 type ext3 (ro,noatime,data=ordered)
/dev/sda2 on /opt type ext3 (ro,noatime,data=ordered)

- ls /media/AiDisk_A2 shows there isn't any file on that disk

- ls -alR /var shows owner and group as 0:0, instead of admin:admin, OK after 
reboot

anything else I could supply?

Original comment by wolfgang.griech@gmail.com on 21 Apr 2012 at 5:09

GoogleCodeExporter commented 9 years ago
OK guys, and went back to 1.0.1.8e-p2 just to see if a similar problem is there 
too. And it is, I just never ran with that optware USB disk extensively before. 
And I guess I figured out what goes wrong, but still no clue why:

as I wrote above, my USB disk normally shows up as /media/AiDisk_a2, e.g. 
/dev/sda2

well, whenever such a timeout or disconnect happens, my USB disk suddenly has 
disappeared as /dev/sda*, but shows up as /dev/sdb* now, and is properly 
mounted as /media/AiDisk_b2! And optware is still working fine, even the 
actually running ssh session continues without problems, but is also mounted as 
/dev/sdb2 now:

/opt/home # mount
...
/dev/sdb2 on /media/AiDisk_b2 type ext3 (rw,noatime,data=ordered)
/dev/sdb2 on /opt type ext3 (rw,noatime,data=ordered)

/opt/home # ls /media
AiDisk_b2

so someone or somewhat causes an unmount of /dev/sda* and mounts it as /dev/sdb*
and I almost bet it has to do with optware and maybe the shell scripts 
opt-mount.sh and opt-umount.sh
on the firmware versions before I never used optware, and could play movies 
without any problems. And btw, also a normal ftp transfer of a bigger file 
fails after a certain time, with the same effect..

Original comment by wolfgang.griech@gmail.com on 21 Apr 2012 at 5:10

GoogleCodeExporter commented 9 years ago
Btw, why have you decided to divide the disk if there is one file system?

Original comment by ser...@soulblader.com on 21 Apr 2012 at 10:06

GoogleCodeExporter commented 9 years ago
yeah thx, a good idea too! And is just one big 1TB ext3 and a swap, so not so 
easy, as it is about 80% full, but will give it a try if I can't find anything 
else. Maybe a 100GB ext3 for optware would be the idea..

tried last night with 1.0.1.8e-p2 with optware disabled from the web interface, 
after a factory default reset of the router and reloading saved configuration. 
Exactly same results, playing a movie, and it stopped after about 15 minutes, 
and had /dev/sdb* instead of /dev/sda* when starting..

next step this morning will be to go back to original ASUS 1.0.1.8e

Original comment by wolfgang.griech@gmail.com on 22 Apr 2012 at 3:27

GoogleCodeExporter commented 9 years ago
I've got the same 1TB WD hard drive in Tsunami rack with the Optware running. 
But there is 1 partition and it is ext2 formatted. I do not have any problems 
with it!
I also use file streaming, read and write large files from it.
I do not use any downloaders in it! And I don't use samba at all. It is 
switched off. I use NFS.
I think there is a key to its stable work described above.

Original comment by d...@soulblader.com on 22 Apr 2012 at 4:57

GoogleCodeExporter commented 9 years ago
hmm interesting, so some more options to play with..
btw, I also tried ftp and sftp, and same things happening there..
and in the meantime I'm back to a test with original ASUS 1.0.1.8.e, and also 
had a disconnect there when playing a movie, but no change of the device name 
from sda* to sdb*
also thought that my USB disk might be defective, so ran a fsck, and also 
badblocks check from my laptop, and everything is OK, and the badblocks check 
really runs a long time, about 10h, and not a single problem, also no 
disconnect or alike..

Original comment by wolfgang.griech@gmail.com on 22 Apr 2012 at 5:11

GoogleCodeExporter commented 9 years ago
In my opinion, there is a problem with the kernel ext driver or with the hard 
disk controller. FTP, sFTP and, samba are applications, running by different 
daemons. Is there any way to update your disk firmware in order to check it?..

Original comment by d...@soulblader.com on 22 Apr 2012 at 6:31

GoogleCodeExporter commented 9 years ago
yes thx again for your valuable input, and I fully agree, just trying to nail 
it down step by step, because I clearly love this project and the RT-N56U, and 
would like to participate and contribute to the extend I can!

so my USB disk is pretty new, fall last year, Samsung M2 Portable 3 it says, 1 
TB, and I have no clue if there are any firmware updates for it, but will take 
a look..

and I'll also connect this USB disk again to my i7 laptop running kubuntu 
11.10/64, also having the minidlna there, so kind of similar conditions, then 
it should be clear whether it's the USB disk, or some problem with the ASUS 
firmware..

the disconnect with the original ASUS 1.0.1.8e firmware I wouldn't overrate, 
the minidlna was still in the middle of building its database (~600 movies), 
and had this before, my Samsung Bluray player BD-D5100 is a bit sensitive on 
data coming not in time and then simply disconnects, this is a known issue too, 
Samsung improved, but couldn't really fix it..

Original comment by wolfgang.griech@gmail.com on 22 Apr 2012 at 7:13

GoogleCodeExporter commented 9 years ago
so it's not the Samsung 1TB USB disk and not the Samsung Bluray player, all 
perfectly ran several movies via minidlna from the USB disk connected to my 
laptop, wireless to the RT-N56U and wired to the player, so about the same as 
if the USB disk would be connected directly to the RT-N56U, except that the 
internal USB and minidlna isn't being used..

Original comment by wolfgang.griech@gmail.com on 22 Apr 2012 at 5:07

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
OK guys, here are my latest findings:

my problem with the 1TB USB (3.0) disk at the ASUS RT-N56U is independent of 
the firmware version, so also happens with old original ASUS firmware, as well 
as with your latest and greatest stuff..

so I cloned this USB disk to a 500GB WD USB 2.0 disk with ext3 file system, 
which I know was running fine before on the router, but formatted with ntfs. 
And all is running fine now, no disconnects and strange change of the device 
name.

looks like it is true what d... said in comment #10, there seems to be a severe 
compatibility issue with the 1TB USB (3.0)disk and the ASUS. Of course this is 
connected as a normal USB 2.0 disk, and perfectly runs as this with my i7 
laptop, which also just has USB 2.0 ports.. hardware? firmware? driver? And 
there is no firmware upgrade for this "Samsung M2 Portable 3.0" USB disk, I 
checked it as good as I could..

Original comment by wolfgang.griech@gmail.com on 24 Apr 2012 at 5:38

GoogleCodeExporter commented 9 years ago
I have a similar issue also when I connect a 2TB samsung HD with NTFS it is not 
discovered by upnp devices. 

Does UPNP support HD with more than 1TB? 

Will issue be resolved if using ext3 or if I spilt the drive in 2 partitions? 

Original comment by borgrodr...@gmail.com on 23 May 2012 at 6:29

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
Hi,

I think this unresolved thread is related to a "spin-down" issue: I bought an 
ASUS RT-n56u and connected my HDs (WD Elements Desktop 1TB, 2TB). These disks 
are equipped with the „WD GreenPower Technology™”, which switch the HD 
into a sleep(?) mode in case of inactivity (?). I use these external disks for 
about 2 and 1 yeas without any problem (with variety of laptops: Dell, HPs,..., 
desktops..)+ Samsung TV and blueray HTC. 

However, this green feature with RT-n56u produces a fatal error: the HD becomes 
disconnected instantly/very frequently, specifically, both in the WAN and LAN 
case, and both with computers and Samsung HTC blueray, the HD become 
disconnected for example in video streaming(!), and re-connect only very slowly 
and totally randomly (sometimes reconnect within a sec, but struggles: 
connect/disconnect, but sometimes lost for x10 minutes.

This error  never happens if the HD is directly connected to the USB of the 
computers or of USB blueray/TV.

The forums are full with complaints about this „obligatory-green” 
technology of WD, which simply makes this drives useless in such settings. 
Althoguh WD provided a tool to disable this feature altogether: WdIdleDisable, 
but it does not work for WD Elements Desktop. Another solutions is to modify 
the Advanced Power Management (APM), spin-down value with CrystalDiskInfo or 
HDDScan, but this cannot be changed for WD Elements Desktop....

In summary: WD hard-wired this feature into WD Elements Desktop, and it really 
works nice if it is connected directly to the computer/TV/blueray, but it does 
not work, if it is connected to TO ASUS RT-N56U!

My guess is that probably there is some „green” control for the USB in 
RT-n56u, and/or probably there is a so efficient caching when reading large 
files, that the  HD USB connected gets no requests and this inactivity triggers 
the spin-down, which in turn disconnect the drive.. 

Please, is there any option to correct this problem?

Thanks in advance!

Original comment by aph...@gmail.com on 15 Sep 2012 at 8:15

GoogleCodeExporter commented 9 years ago
hmm guess I can subscribe to the theory with the spin-down, although I'm not 
really sure, but from the symptoms I can see at my Samsung 1TB it could be a 
possible explanation. Just stops after a while even when streaming a movie from 
it. Never happens if the very same USB-HD is directly connected to my laptop..

have you already played with the spin-down setting? It's there since the latest 
FW update, pls see the attached file. But haven't spent too much time on 
investigations there, kind of gave up on this and am just using a 500GB WD 
blue, no issues there..

am kind of disappointed by the USB interface of the RT-N56U in general, pls 
also see issue 222 about the USB printers, also not really working, so I'm kind 
of wondering if there is something wrong with the USB stack in general..

Original comment by wolfgang.griech@gmail.com on 15 Sep 2012 at 8:30

Attachments:

GoogleCodeExporter commented 9 years ago
wow, I am shocked (some hope returns :), but you attached a FW with version 
1.1.2.3010?! where could you access it?? sorry for being so dumb, but by 
official ASUS checking I see that the latest FW is 1.0.1.8j..

Original comment by aph...@gmail.com on 15 Sep 2012 at 9:21

GoogleCodeExporter commented 9 years ago
now you lost me, you are in this forum and don't know about the downloads???

just go to the "Download" tab of this website, and download the latest from 
there:

http://code.google.com/p/rt-n56u/downloads/detail?name=RT-N56U_1.1.2.3-010_newgu
i.zip&can=2&q=

this is what at least what I installed.. Make sure you saved your settings, 
because all will be lost when installing this update, so you can restore them 
afterwards ;-)

Original comment by wolfgang.griech@gmail.com on 15 Sep 2012 at 9:33

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
..and your comments motivated me to give my Samsung 1TB USB-HD with the new 
firmware another try, but nothing has changed, except that due to the changed 
GUI I have an error log now, opposed to previous versions, it's attached, but 
the important lines are just shown here:

Sep 15 14:49:16 opt-umount.sh: started [/dev/sda2 /media/AiDisk_a2]
Sep 15 14:49:45 kernel: usb 1-1: USB disconnect, address 2
Sep 15 14:50:07 kernel: usb 1-1: new high speed USB device using rt3xxx-ehci 
and address 3
Sep 15 14:50:07 kernel: usb 1-1: configuration #1 chosen from 1 choice
Sep 15 14:50:07 kernel: scsi1 : SCSI emulation for USB Mass Storage devices
Sep 15 14:50:07 kernel: usb-storage: device found at 3
Sep 15 14:50:07 kernel: usb-storage: waiting for device to settle before 
scanning
Sep 15 14:50:09 kernel: usb-storage: device scan complete
Sep 15 14:50:11 kernel: scsi 1:0:0:0: Direct-Access     Samsung  M2 Portable 3  
  3    PQ: 0 ANSI: 6
Sep 15 14:50:13 kernel: SCSI device sdb: 1953504688 512-byte hdwr sectors 
(1000194 MB)
Sep 15 14:50:13 kernel: sdb: Write Protect is off
Sep 15 14:50:13 kernel: sdb: Mode Sense: 27 00 00 00
Sep 15 14:50:13 kernel: sdb: assuming drive cache: write through
Sep 15 14:50:13 kernel: SCSI device sdb: 1953504688 512-byte hdwr sectors 
(1000194 MB)
Sep 15 14:50:13 kernel: sdb: Write Protect is off
Sep 15 14:50:13 kernel: sdb: Mode Sense: 27 00 00 00
Sep 15 14:50:13 kernel: sdb: assuming drive cache: write through
Sep 15 14:50:13 kernel:  sdb: sdb1 sdb2
Sep 15 14:50:13 kernel: sd 1:0:0:0: Attached scsi disk sdb
Sep 15 14:50:13 kernel: sd 1:0:0:0: Attached scsi generic sg0 type 0
Sep 15 14:50:14 kernel: Adding 4335608k swap on /dev/sdb1.  Priority:-2 
extents:1 across:4335608k
Sep 15 14:50:14 automount: Activate swap partition /dev/sdb1 SUCCESS!
Sep 15 14:50:16 kernel: kjournald starting.  Commit interval 5 seconds
Sep 15 14:50:16 kernel: EXT3 FS on sdb2, internal journal
Sep 15 14:50:16 kernel: EXT3-fs: mounted filesystem with ordered data mode.
Sep 15 14:50:16 opt-mount.sh: started [/dev/sdb2 /media/AiDisk_b2]
Sep 15 14:50:21 FTP server: daemon is started
Sep 15 14:50:22 Samba Server: daemon is started
Sep 15 14:50:22 rpc.mountd[3489]: Kernel does not have pseudo root support.
Sep 15 14:50:22 rpc.mountd[3489]: NFS v4 mounts will be disabled unless fsid=0
Sep 15 14:50:22 rpc.mountd[3489]: is specfied in /etc/exports file.
Sep 15 14:50:22 NFS server: daemon is started
Sep 15 14:50:22 Media Server: daemon is started
Sep 15 14:50:22 Transmission: daemon is started
Sep 15 15:14:34 kernel: usb 1-1: USB disconnect, address 3
Sep 15 15:14:34 kernel: Buffer I/O error on device sdb2, logical block 38012269
Sep 15 15:14:34 kernel: lost page write due to I/O error on sdb2
Sep 15 15:14:34 kernel: Aborting journal on device sdb2.
Sep 15 15:14:34 kernel: journal commit I/O error

my disk has 2 EXT3 partitions, just for testing purposes, and both get mounted 
properly, and work properly. But after running a movie for about 25 minutes I 
get completely disconnected, the important line is as follows:

Sep 15 15:14:34 kernel: usb 1-1: USB disconnect, address 3

so clearly something very bad happens here!!!

and btw, I set the "HDD spindown timeout (h:m):" to "never" and the "HDD 
disable APM (head parking)?" to "on" (was wrong in my previous post..), but 
apparently not related to my issue at all..

Original comment by wolfgang.griech@gmail.com on 15 Sep 2012 at 10:12

Attachments:

GoogleCodeExporter commented 9 years ago
..and to complete the picture a bit, a warm reboot of the ASUS brought the 
Samsung USB-HD back to life again, and running the same HD and playing the same 
movie on my laptop perfectly works, already over an hour by now..

so I guess your idea might be right, in a way at least, still am not sure about 
the spin-down, but something causes a hard disconnect of the USB, either the 
device itself, or the ASUS, for the latter speaks the successful reboot, but 
maybe also causes the USB HD to reinitialize, don't know enough about this 
(yet)..

Original comment by wolfgang.griech@gmail.com on 15 Sep 2012 at 10:52

GoogleCodeExporter commented 9 years ago
you are right ;-), I was so surprised by your screenshot, particularly with 
"HDD spindown timeout/APM", that the similarity of this GUI tricked me, I 
thought it was the ASUS official sw. Of course, you answered my question: I 
should try this FW definitely, I start ASAP.

(checking your log, but my #1 candidate remains this cache/spin-down mixture..) 

Original comment by aph...@gmail.com on 15 Sep 2012 at 12:59

GoogleCodeExporter commented 9 years ago
still cannot believe it, but it seems that this FW solved my problem with the 
"WD Elements Desktop" HDs: now they never disconnect.

Thank you for your comments, which brought me here, and congrat for the 
developers for the  project, seems to be a delicate  sw... ;-), it needs time 
to explore all these new options.

I still have a minor problem, perhaps as another Samsung blueray user :), you 
can help: the blueray shows only small part of the directories of the HD. 
Earlier I thought that the disconnection disrupts its exploration process, but 
now the connection is stable, and I still cannot see all the dirs. I set access 
without password, upnp, enforce full update, etc. But it is still partial. If I 
check the "web" link on the GUI, the miniDLNA(?) summary also contains much 
less numbers as expected. Any idea??

Original comment by aph...@gmail.com on 15 Sep 2012 at 3:29

GoogleCodeExporter commented 9 years ago
hey good to hear, at least your issues are solved!

ok, and I have an idea, just be patient, this minidlna media server is a slow 
pig, pls take a look at the CPU load on top of the web GUI, as long as it's 
close to 100% it's still digging for new media. Also set the "Database content 
update mode:" to "Update for new files only". Guess minidlna is just using 
sqlite3 as database, and this really isn't made for handling 10000 entries.. ha 
ha ha

Original comment by wolfgang.griech@gmail.com on 15 Sep 2012 at 3:43

GoogleCodeExporter commented 9 years ago
patience was the solution :D, thnx! (now I have some strange WAN 
disconnections, i guess in heavy CPU load, digging the archive now for this, 
perhaps a new theory... :)

Original comment by aph...@gmail.com on 16 Sep 2012 at 5:46

GoogleCodeExporter commented 9 years ago
maybe we can move this discussion to another place, am willing to share my 
experience with this excellent router and the great firmware of this project 
(yes it is, despite a few annoying things..), you may contact me directly, it's 
wolfgang.griech@gmail.com

Original comment by wolfgang.griech@gmail.com on 16 Sep 2012 at 5:54

GoogleCodeExporter commented 9 years ago
my USB hard disk problem still persists with the latest firmware upgrade 
3.0.2.4-011

Original comment by wolfgang.griech@gmail.com on 21 Sep 2012 at 5:23

GoogleCodeExporter commented 9 years ago
The same problem happen to my WD Elements 1TB usb3 too.
I'm currently on 1.1.2.3-010.

It will happen while i'm copying large file or too many small files.
My WD Elements will disconnect from being sda1 then reconnect as sdb1.

Here is the log:

Sep 29 02:27:45 kernel: usb 1-1: USB disconnect, address 11
Sep 29 02:27:45 kernel: printk: 1 messages suppressed.
Sep 29 02:27:45 kernel: Buffer I/O error on device sda1, logical block 3116496
Sep 29 02:27:45 kernel: lost page write due to I/O error on sda1
Sep 29 02:27:45 kernel: Buffer I/O error on device sda1, logical block 3116497
Sep 29 02:27:45 kernel: lost page write due to I/O error on sda1
Sep 29 02:27:45 kernel: scsi 5:0:0:0: rejecting I/O to dead device

Original comment by chiram...@gmail.com on 28 Sep 2012 at 7:57

GoogleCodeExporter commented 9 years ago
ah thanks, and good to know I'm not alone, and really seems to be related to 
USB3.0 then...

and ran another test by supplying my USB HD from 2 USB ports to exclude 
possible power supply limitations of the RT-N56U's USB port(s), but didn't 
change a bit...

and also opened the HD box and looked inside because I thought it might be a 
SATA2 disk, so that I could connect to it using an USB2.0 interface, but wasn't 
the case, already the HD's PCB shows up with a genuine USB3.0 interface...

Original comment by wolfgang.griech@gmail.com on 30 Sep 2012 at 9:33

GoogleCodeExporter commented 9 years ago
Hi Guys,

I was having similar issues with both factory firmware and the very latest 
custom one (RT-N56U_1.1.2.6-018_full.trx). I have a Samsung portable USB 2.0 
500GB hard drive formatted ntfs. 
The kind of solution was to use a double USB cable, it seems the router cannot 
give enough juice to the hard drive. 

Original comment by Gabor.Ba...@gmail.com on 27 Oct 2012 at 9:58

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
Dec 17, 2012:
"I also have "unmounts" after a certain period of use. No green drives or 
anything but 2.5" 60GB Seagate in an enclosure, both USB's slot's used with a 
y-adapter cable for dual power. Using 3.0.3.0-026 Padavan."

Update:
I solved my problem! Of course this is a personal case but it happens that 
after a week of head scratching and having the same symptoms as all of you 
mentioned, the USB Y-Cable supplied along with the NexStar TX 2.5" SATA 
enclosure was faulty. The 60GB never ran fine on laptop either, despite having 
both ports plugged for supposed "extra power", turns out that despite the cable 
looked fine, by switching out the cable with another one I had laying around, 
the drive is mounted flawlessly for 2 weeks now, no unmounts, hiccups or 
anything abnormal. The spindown works perfect too. The router is perfect [Full 
Build 3.0.3.1-027], USB ports are proper 2.0 spec and certainly runs fine.

Original comment by ampd...@gmail.com on 7 Jan 2013 at 7:37

GoogleCodeExporter commented 9 years ago
Hello guys! Today after half a year of suffering the same problem with WD 
external 2 TB HDD i found the solution. BTW its not a simple one.
Notice that it works only for WD drives and you are doing everything on your 
own risk, so backup somehow before operating.
The way to stop the APM (auto heads parking or WD GreenPower Technology) is to 
use MS-DOS utility form WD called 'wdidle3' which is the older version of 
'WdIdleDisable_v6' utility, mentioned in this topic, but it works for every WD 
HDD.
To disable the heads parking timer you need to follow these steps:

1) Download the utility, MS-DOS image and the utility to take it all to the 
flash drive here: 
https://docs.google.com/file/d/0B8NAfZJXviA1aGRQUVloeDJRSTg/edit

2) Insert empty flash drive and format it with loading the MS-DOS on it by the 
HPUSBFW utility. (MS-DOS files may appear to be invisible in Windows 7 or 8).

3) Copy wdidle3 files to the root of a flash drive.

4) Take your WD HDD out of the black plastic case. It is easy, skrewdriver is 
the only tool you need here.

5) Connect it to your PC via SATA and (!!!) disconnect every other WD HDD from 
your PC in order not to affect them.

6) Boot from the flash drive to MS-DOS and in terminal use the following 
commands:
wdidle3 /r - to check the existing value of the timer
wdidle3 /d - to disable the timer
wdidle /? - for help

6.1) reboot several times and check that timer is disabled using wdidle /r 
command.

7) turn off the pc and disconnect your cured WD HDD from it. Insert it in the 
plastic case and use it with rt-n56u.

This is how I cured my 2TB WD HDD (model WD 20EARX-00PASB0).

Original comment by Denis.Za...@gmail.com on 24 Jan 2013 at 9:50

GoogleCodeExporter commented 9 years ago
also sorry for my English and many thanks to developers of this awesome 
firmware!
Hope my answer helped someone. Good luck!

Original comment by Denis.Za...@gmail.com on 24 Jan 2013 at 9:53

GoogleCodeExporter commented 9 years ago
[deleted comment]
GoogleCodeExporter commented 9 years ago
I was having same issue with an external enclosure with 160 GB 2.5" Toshiba 
HDD. after flashing latest firmware (3.0.3.3-046) and changing the spin down 
time, the hard disk is stable and accessible for last day or two. Thanks a lot 
for your work guys. 

Original comment by Adarsha...@gmail.com on 18 Mar 2013 at 1:29

GoogleCodeExporter commented 9 years ago
Very useful reading - Im having similar issues, but with a SEAGATE 2TB - anyone 
come across a solution for the SEAGATE drive?

Original comment by darinkra...@gmail.com on 28 Jul 2013 at 2:11

GoogleCodeExporter commented 9 years ago
One last comment - are you running on 2.4 or 5ghz?  I cant get the samsung blu 
ray (BBD5700) to see the 5ghz, (laptop can see it fine) - any suggestions?

Original comment by darinkra...@gmail.com on 28 Jul 2013 at 2:22

GoogleCodeExporter commented 9 years ago
same issue 3.4.3.6-064 with WD MyBook 2tb (usb 3.0) and a 16gb usb stick (usb 
2.0) both are ntfs and works flawless on pc
disconnect on copy (large/many) files with smb and instand reconnect but with 
another name (stick change name tonight patriot to AiDisk_b1 without any use)
after a router reboot the name chanced back 

a fix was amazing 

Jul 27 22:03:42 kernel: usb 1-1: new high-speed USB device number 2 using 
rt3xxx-ehci
Jul 27 22:03:42 kernel: usb 1-1: New USB device found, idVendor=13fe, 
idProduct=3100
Jul 27 22:03:43 kernel: usb 1-1: Product: Patriot Memory
Jul 27 22:03:43 kernel: usb 1-1: Manufacturer:         
Jul 27 22:03:43 kernel: usb 1-1: SerialNumber: *****************
Jul 27 22:03:43 kernel: scsi0 : usb-storage 1-1:1.0
Jul 27 22:03:46 kernel: scsi 0:0:0:0: Direct-Access              Patriot Memory 
  PMAP PQ: 0 ANSI: 0 CCS
Jul 27 22:03:46 kernel: sd 0:0:0:0: Attached scsi generic sg0 type 0
Jul 27 22:03:46 usb_modeswitch: no rule for device 13fe:3100
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] 31293440 512-byte logical blocks: 
(16.0 GB/14.9 GiB)
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Write Protect is off
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Mode Sense: 23 00 00 00
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] No Caching mode page present
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] No Caching mode page present
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 27 22:03:47 kernel:  sda: sda1
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] No Caching mode page present
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through
Jul 27 22:03:47 kernel: sd 0:0:0:0: [sda] Attached SCSI removable disk
Jul 27 22:03:48 kernel: ufsd: driver 8.5 (Jun 21 2013 17:08:01) LBD=ON with 
delayalloc with ioctl loaded at c22e7000
Jul 27 22:03:48 kernel: NTFS read/write support included
Jul 27 22:03:48 kernel: Hfs+/HfsX read/write support included
Jul 27 22:03:48 kernel: Built for: realtek_new_ic_2011-04-29
Jul 27 22:03:48 kernel: ufsd: use builtin utf8
Jul 27 22:03:52 kernel: ufsd: sda1 native journal need replay
Jul 27 22:03:57 FTP server: daemon is started

Jul 28 01:26:20 kernel: usb 1-1: USB disconnect, device number 2
Jul 28 01:26:20 kernel: usb 1-1: new high-speed USB device number 3 using 
rt3xxx-ehci
Jul 28 01:26:20 kernel: usb 1-1: New USB device found, idVendor=13fe, 
idProduct=3100
Jul 28 01:26:20 kernel: usb 1-1: Product: Patriot Memory
Jul 28 01:26:20 kernel: usb 1-1: Manufacturer:         
Jul 28 01:26:20 kernel: usb 1-1: SerialNumber: *****************
Jul 28 01:26:20 kernel: scsi1 : usb-storage 1-1:1.0
Jul 28 01:26:23 kernel: scsi 1:0:0:0: Direct-Access              Patriot Memory 
  PMAP PQ: 0 ANSI: 0 CCS
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] 31293440 512-byte logical blocks: 
(16.0 GB/14.9 GiB)
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] Write Protect is off
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] Mode Sense: 23 00 00 00
Jul 28 01:26:23 kernel: sd 1:0:0:0: Attached scsi generic sg0 type 0
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] No Caching mode page present
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] Assuming drive cache: write through
Jul 28 01:26:23 kernel: sd 1:0:0:0: [sdb] No Caching mode page present
Jul 28 01:26:24 usb_modeswitch: no rule for device 13fe:3100
Jul 28 01:26:24 kernel: sd 1:0:0:0: [sdb] Assuming drive cache: write through
Jul 28 01:26:24 kernel:  sdb: sdb1
Jul 28 01:26:24 kernel: sd 1:0:0:0: [sdb] No Caching mode page present
Jul 28 01:26:24 kernel: sd 1:0:0:0: [sdb] Assuming drive cache: write through
Jul 28 01:26:24 kernel: sd 1:0:0:0: [sdb] Attached SCSI removable disk
Jul 28 01:26:24 kernel: ufsd: use builtin utf8
Jul 28 01:26:24 kernel: ufsd: sdb1 native journal need replay
Jul 28 01:26:54 kernel: Buffer I/O error on device sda1, logical block 29973306
Jul 28 01:26:54 kernel: lost page write due to I/O error on sda1
Jul 28 01:26:54 kernel: Buffer I/O error on device sda1, logical block 29973309
Jul 28 01:26:54 kernel: lost page write due to I/O error on sda1
Jul 28 01:26:54 kernel: Buffer I/O error on device sda1, logical block 29973311
Jul 28 01:26:54 kernel: lost page write due to I/O error on sda1

Original comment by p4a...@hotmail.com on 28 Jul 2013 at 4:00

GoogleCodeExporter commented 9 years ago
I solved this problem a year ago by simply do the following:

1. Scan and fix file system error on my WD (at that time chkdsk for ntfs, and 
later fsck for ext4)

2. upgrade my usb wifi client from dlink DWA-120 wireless G to ASUS USB-N53 
wireless N

I believe that the I/O error in my case causing by file system error.
Upgrading wifi client  may be irrelevant.

Original comment by chiram...@gmail.com on 30 Jul 2013 at 6:18

GoogleCodeExporter commented 9 years ago
I have the exact same problem. Surprised that nothing was fixed after more than 
a year? My USB drive disconnects when I try to copy a 4Gb file. And then 
reconnects under a new drive name.

Original comment by edmond.p...@gmail.com on 17 Aug 2013 at 2:52

GoogleCodeExporter commented 9 years ago
Same problem. USB mount point changes on its own from sda1 to sdb1 for no 
reason at all. 

Original comment by kolsl...@gmail.com on 9 Sep 2013 at 3:46

GoogleCodeExporter commented 9 years ago
Same issues here. I have two Hard drives connected to ASUS USB-N53.

When you first set it up it uses the correct name - WD_Drive_01 (RT-N56U 
(RT-N56U). But it randomly drops out and reverts to AiDisk_c1 (RT-N56U). Drives 
formatted to NTFS.

A router reboot resets the name, or it sometimes fixes itself. This is a real 
pain when using a WD Live Hub, as it loses the shares a lot.

Using latest Firmware version 3.4.3.6-069

Original comment by briancre...@googlemail.com on 20 Oct 2013 at 9:08

GoogleCodeExporter commented 9 years ago
I'm currently experiencing this issue with the latest firmware version 
3.4.3.7-072
Has any progress been done to help resolve this issue?

My log:
==============
Dec 18 22:47:07 kernel: usb 1-1: USB disconnect, device number 12
Dec 18 22:47:07 kernel: usb 1-1: new high-speed USB device number 13 using 
rt3xxx-ehci
Dec 18 22:47:07 kernel: usb 1-1: New USB device found, idVendor=0951, 
idProduct=1642
Dec 18 22:47:07 kernel: usb 1-1: Product: DT 101 G2
Dec 18 22:47:07 kernel: usb 1-1: Manufacturer: Kingston
Dec 18 22:47:07 kernel: usb 1-1: SerialNumber: **********************
Dec 18 22:47:07 kernel: scsi11 : usb-storage 1-1:1.0
Dec 18 22:47:08 kernel: EXT4-fs error (device sda2): ext4_find_entry:932: inode 
#26625: comm sh: reading directory lblock 0
Dec 18 22:47:08 kernel: EXT4-fs error (device sda2): ext4_find_entry:932: inode 
#26628: comm sh: reading directory lblock 0
Dec 18 22:47:10 kernel: scsi 11:0:0:0: Direct-Access     Kingston DT 101 G2     
   1.00 PQ: 0 ANSI: 4
Dec 18 22:47:10 kernel: sd 11:0:0:0: [sdc] 60437492 512-byte logical blocks: 
(30.9 GB/28.8 GiB)
Dec 18 22:47:10 kernel: sd 11:0:0:0: [sdc] Write Protect is off
Dec 18 22:47:10 kernel: sd 11:0:0:0: [sdc] Mode Sense: 45 00 00 00
Dec 18 22:47:10 kernel: sd 11:0:0:0: Attached scsi generic sg0 type 0
Dec 18 22:47:10 kernel: sd 11:0:0:0: [sdc] Write cache: disabled, read cache: 
enabled, doesn't support DPO or FUA
Dec 18 22:47:10 kernel:  sdc: sdc1 sdc2
Dec 18 22:47:10 kernel: sd 11:0:0:0: [sdc] Attached SCSI removable disk
Dec 18 22:47:10 usb_modeswitch: no rule for device 0951:1642
Dec 18 22:47:11 kernel: EXT4-fs (sdc2): mounted filesystem with ordered data 
mode. Opts: (null)
Dec 18 22:47:11 opt-mount.sh: started [/dev/sdc2 /media/AiDisk_c2]
Dec 18 22:47:16 kernel: EXT4-fs error (device sda2): ext4_find_entry:932: inode 
#26625: comm sh: reading directory lblock 0

Original comment by jgio...@gmail.com on 19 Dec 2013 at 3:52

GoogleCodeExporter commented 9 years ago
I'm using the latest firmware 3.4.3.7-075, and having the same issue.
My hard drive is WD 1 TB, in a USB enclosure.

Original comment by barnitar...@gmail.com on 29 Jan 2014 at 3:56

GoogleCodeExporter commented 9 years ago
I had lot of trouble with these disconnections.
Tried different file system (ext3). Still the same.
Reading the posts I saw different things could help:
- use different HDD
- disable spin-down and head parking (for WD drives)
- USB port power is not string enough (but I use an externally powered 
enclosure)
- not to use USB3 but USB2 interface (change the HDD enclosure)

After digging more, I saw a comment #3 for Issue 884
   "Please use farther USB port (from Front panel). This is known hardware issue."
So I just plugged in to the second USB port, and it works great. I hope I'm not 
speaking too early. It is running fine for a day. Usually I got disconnections 
after 1-2 hour.

I hope it helps.

Original comment by barnitar...@gmail.com on 30 Jan 2014 at 5:10

GoogleCodeExporter commented 9 years ago
I have the same issue with my new Samsung M3 1TB usb 3.0. I try to use the 
second usb port of my N56U and seems like work some times. Any news about that? 

Original comment by horst.ma...@gmail.com on 26 Mar 2014 at 11:35