Disk Read/Write Error

[QNAPJason]

Hi all,
Update:
We are now working with WD for this issue.
For those who are still experiencing sudden removal of Western Digital HDDs & Disk Read/Write Error, please email me (jasonhsu@qnap.com) with the following information.
1. the serial numbers of those WD HDDs (affected HDDs)
2. model of HDDs
3. HDD firmware version (The S.M.A.R.T. tool from NAS admin. page)
4. NAS model
5. Nas Firmware version(s)
6. Detailed description of error
7. Current Status
8. Forum ID
9. System Log file
...and anything you want to add.

Sincerely,

Jason

[QNAPJason]

Hi guys,
If you experience the drive loss problem again, please PM or email me (jasonhsu@qnap.com) the output of the following command:

SSH or telnet into your NAS with admin account (Windows users can download & install "PuTTY"), then run the following command:

Code: Select all

#dmesg

Copy the entire output and send it to me.

This only works before the reboot of the NAS. Therefore, as soon as you discover the drive issue, please run the command.

Thank you.

Jason

[QNAPJason]

Hi guys,
Update on 09/22: QNAP will release a beta firmware next week for users who are experiencing this with their WD HDDs. Moreover, WD will officially release the new firmware for their HDDs targeting this issue in Q4.


If you're still having the Disk Read/Write error, please try the latest beta firmware and let us know if it works for you. We need your help.
-----------------------------------------
TS-239:
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

TS-439:
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

TS-509
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

TS-639
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

TS-809
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

TS-809U
ftp://csdread:csdread@ftp.qnap.com/NAS/ ... -3.1.1.img

Modified:
1. wakeup dalay
2. turn on debug message in SATA driver

If you still have the same problem, please login to your NAS by ssh/telnet and use the following procedures to get the detail message.

# cd /share/Public
# touch kmsg.txt
# cat /proc/kmsg > kmsg.txt

Then press the Ctrl -C to stop and go to the NAS Public folder to find the
kmsg.txt. and Send us the file to check (jasonhsu@qnap.com).

Best regards,

Jason

[clx]

After having my NAS die a few weeks ago, and subsequently getting a new one from QNAP then having their engineer fix the persisting problems caused to the disks from the first crash (and losing/corrupting a large chunk of my data), i will be following this topic eagerly.

My confidence is currently shaken in my 509 and the disks currently in it, I am currently re-formatting with EXT4 but i wont hold my breath until a firm update from QNAP is out stating what the problem is with these dropped drives, at which point i might get new drives, or i might change product.

[ftan]

Sept. 2, installed Beta Firmware 3.1.1 Build 0828T
Rescanned the drives, rebuilt raid EXT3.
Previous symptom usually happened during "quiet time" when the unit dropped drives, but so far so good.
I will leave it alone for another week and put the unit through its paces again.
Will keep everyone posted.

[NightFlash]

I still have massive errors with latest firmware one disk is still giving me read/write errors and refuses to get online, if I change the disc to a new one it gives the same error really don't know what to do except to send em back and buy something else soon.

This ** =(

[karlwest]

I appear to also be a lucky winner of this issue, with a QNAP TS-639 PRO & 4 x WD2002FYPS-01U1B04.0 drives in a RAID 5 array. Current QNAP firmware is 3.1.0 Build 0627T. I bought and setup the device on July 20th, and it has been running fine since then, up until the Disk Read/Write Error occured on September 8th. I have a seperate email off to Jason with all the rest of the info he asks for.

[clx]

This does seem to be an increasingly common problem with a wide range of drive makes and NAS models.
Could Qnap comment on what the problem is? whether its a hardware issue or a problem with configs becoming corrupt ?

[QNAPJason]

This does seem to be an increasingly common problem with a wide range of drive makes and NAS models.
Could Qnap comment on what the problem is? whether its a hardware issue or a problem with configs becoming corrupt ?

Hi clx,
This is caused by the characteristics of this type of HDDs. We're still trying different ways to provide a fix for it.

[wabasaja]

Hi guys,

my problem was as I wrote here at page 2 of this thread (post wabasaja on Mon Aug 03, 2009 11:06 pm)...

After 7 days of scanning bad blocks via SATA scan within the QNAP admin Tools...I stopped the scan....which
found already more than 1 Mio. bad blocks...

I replaced the problem causing HDD whit a new one ( Seagate ST31500341AS CC1H (1,5TB) and made a rebuild...
LED's went back to green....after some hours i shut the 509 down because I had to left for some days.

The first RAID5 config was 5 x Seagate ST31500341AS CC1G (1,5TB)

I checked the problem causing HHD where QNAP SATA scan recorded more than 1 Mio. bad blocks with all available seatools scans...

seatools found nothing at all....all tests passed without problems...

So what to do....send the HDD to seagate...which I guess is senseless because they do the same scan with their tool...
What would you do??

Tonight I turn the NAS on and let it run thru for the next days to check if I lose another or the same HHD again.

Thx to all!!
Have a nice time everyone out there!!
walt

[ftan]

Spoke too soon, I unplugged the trunked NICs to quiet the Qnap and it dropped 2 drives (~6 Hours after the nics were unplugged and 2 Hours withing each other).
Will be sending QNAP the error logs.

[QNAPJason]

Hi all,
If you're still having Disk Removal, please also change the HDD jumper setting to OPT1 for further test.

[wabasaja]

Hi QNAPJason,

Thanks for your Informations!!

Are you talking in context to the jumpers only about some HDD's lets say WD's >= 1,5 TB or about all HDD's with
capacity ??
Also about Seagate Barr. 1,5TB??

Can you have a look at my posts within that thread...and let me know your opinion..would be nice!

Thx in advance!!!
nice day 4all of you!!!
walt

[karlwest]

Bummer - I tried upgrading my firmware to the debug version referenced earlier ( U_TS-639_20090902_wakeup_delay_debug-3.1.1.img ) but it failed after the progress bar made it to 6%. I redownloaded it several times and tried again just to confirm I hadn't rec'd a corrupted copy but no dice...failed at 6% each time. I also tried upgrading to the latest non-Beta build ( TS-639_3.1.1_Build0815 ) and then trying to get the Beta version to install, but that didn't help either.

It did upgrade fine to TS-639_3.1.1_Build0815, but I still have the Disk Read/Write Error. I also tried removing the drive, readding it, but it was redetected as failed and thus the RAID5 entry is still degraded and won't allow me to make any changes. (the options are greyed out, anyway - I think it goes into read-only mode when degraded)

I'll try the jumper thing on the 'bad' drive next. After that, will probably see about redoing the whole RAID5 config from scratch.


EDIT: Set jumper setting to OPT1 enabled, removed/readded drive - no joy, still see Disk Read/Write Error. I've avoided breaking and recreating the RAID5 array mainly becuase I'm lazy, but I suppose I'll do that next. Its like there is some bit of QNAP OS that just refuses to 'reset' the disk read/write error after it has first occurred. It still seems to me extremely unlikely the drive itself is really bad...these are brand-new enterprise class drives, after all. Oh well, guess I'll reset everything and try rebuilding it.

[NightFlash]

I tried to change the disc but same slots still gives me read/write error so it seems like qnap os is killing the whole sataport and not just that disc cause the WD tool says disc is ok.