"Disk Failed", then "Disk Unplugged" errors

[boogie1313]

Reporting my issue. I owned a TVS473-8G upgraded to 32GB with 4 Ironwolf 4TB in Raid 5 configuration. Latest firmware. I owned the NAS since Jul 2017

2 week ago, my slot 4 had the Disk Unplugged error. Performed a diagnostic using another computer which reported no issue.
Plug in again to rebulid, but failed. So I RMA the Disk. A week later, replacement disk back and still failed when I tried to rebuild it.
Just 2 days ago, i sent in the NAS for RMA. Forgot to check the forum for similar issue.

This was the 2nd time that i sent in my NAS for RMA for a short time of 4 months.. the 1st was due to eth4 issue.
I owned another synology NAS DS413 which i owned it for the last 5 years.. never sent in for RMA before.. sad..

[gdcorson]

I have had this issue on my TVS-473 in bay 3. [#RNU-673-80775]

I lodged a support ticket 1 week ago but have not heard anything back apart from the automated response.

Initially though it was a genuine disk fault and bought 2 Westen Digital drives because I though I had made a bad choice going with Seagate Ironwolfs.

The new drive started rebuilding but failed at 1.8% and I got the same "Disk Failed", then "Disk Unplugged" errors again.

Contacted the vendor and was told to raise a ticket with Qnap and get an RMA if it cannot be resolved by online supoort; which I doubt it can be.

Now I wait.....

[gdcorson]

I have to apologise and correct my previous post.

The responses to my ticket from Qnap were in a spam folder. Beginner mistake by me.

Unfortunately the ticket was closed automatically (which is fair enough) so I have asked to have it re-opened.

[Jumpy07]

Just experienced this on our 1282T, latest firmware, running raid 5, Disk 1 Failed then Disk Unplugged, (7 month old system with 10gb Seagate Drives)

Jacked the drive out and re-insterted and its started a rebuild, SMART says all good... will order a new drive as wont trust it, but niggle in the back of my mind is firmware upgrade we did on Fri to 4.3.3.0396

[boogie1313]

RMA back 2 days ago, first rebuild took 5 hours and was successful. however this issue occurred again today .. qnap ticket raised.. qnap quality, i have nothing to say...

Host: Disk 4 Read I/O error, ILLEGAL REQUEST sense_key=0x5, asc=0x21, ascq=0x4, CDB=88 00 00 00 00 00 00 10 2d b0 00 00 00 08 00 00 .

Reporting my issue. I owned a TVS473-8G upgraded to 32GB with 4 Ironwolf 4TB in Raid 5 configuration. Latest firmware. I owned the NAS since Jul 2017

2 week ago, my slot 4 had the Disk Unplugged error. Performed a diagnostic using another computer which reported no issue.
Plug in again to rebulid, but failed. So I RMA the Disk. A week later, replacement disk back and still failed when I tried to rebuild it.
Just 2 days ago, i sent in the NAS for RMA. Forgot to check the forum for similar issue.

This was the 2nd time that i sent in my NAS for RMA for a short time of 4 months.. the 1st was due to eth4 issue.
I owned another synology NAS DS413 which i owned it for the last 5 years.. never sent in for RMA before.. sad..

[syrit]

Also getting this kind of issue, QNAP is a TS-869 Pro in RAID 6 so i accept it is slightly aged, however the NAS has been working fabulously for over a year, always kept up to date, then a month or so ago Disk 5 and 6 'unplugged' along with the 'detected but is inaccessible' (Running 8 x WD RED WD30EFRX) errors. Took is as coincidence that the drives had just failed simultaneously and purchased 2 new WD30EFRX drives. RAID rebuilt, all is functioning fine and then TODAY disk 5 and 6 'unplugged' and 'detected but is inaccessible'.

I would have put it down to the QNAP failing (and it still could be) but seeing as others here have been seeing similar issues with the latest or recent firmware seems a bit coincidental and the drives that have become 'unplugged' are literally weeks old. Rebooted the NAS and the raid has begun rebuilding itself without any input in terms of ejecting or replacing drives. Just our two cents.

[buffalowolff]

Another here. I have a TS-253A with seagate ironwolf 4TB drives in raid 1. Since December I've had 4 host disk unplugged incidents, where simply pulling and inserting the same drive rebuilt without issue. It has happened on both of the 2 disks, luckily at different times. I had the NAS in a cold basement, I've moved it upstairs today, we'll see if it fares better for me.

[michael_canny]

My ability to search the forum seems to be lacking so I started a new topic on this issue, however I was kindly directed here and thought it might be prudent to add my woes....

Out of approximately 50 QNAP NAS boxes of various ages and models we currently have five, approximately one year-old, TVS-463s that, in the last week, have all reported the same issue, disk 3 has disappeared. All 5 are installed in different geographical locations around the world but all perform identical jobs, albeit with data from their own individual "offices". None of them are connected to each other from a network point of view.

Each unit has four identical disks installed, 2TB WD Red Pro WD2001FFSX's, in RAID 10 and up until last week had no issues.

On the first instance of this happening we thought the disk had failed and swapped it out for a new one, this had no effect.

On 4 of the 5 units a reboot will prompt the NAS to see disk 3 and a RAID rebuild begins, a few minutes later the system reports "HDD 3 ejected" with no intervention from anyone. We also see similar if the disk is removed and inserted again. All instances have been checked and disks are installed correctly.

Firmware on all was 4.3.3.0262 build 20170727 and updated around August last year when we standardised the whole environment, no changes made to configs or anything since then, I updated one to 4.3.4.0435 build 20171230 today to see if it would have any impact on this issue but it hasn't.

Any thoughts on this would be really appreciated, when the first one happened it was strange, with five all suffering the same things (one of which has now had disk 4 fail so we've lost the lot) it's a really concern, we have another 20 of these

I have a ticket open with QNAP and they haven't responded in 48 hours, I'll call them tomorrow but am really happy to have been directed here, I'm not alone!

[michael_canny]

I have a response from QNAP - apparently me adding additional logs & information as and when I had new instances of the issue effectively reset their 24 hour SLA on the ticket so nobody looked at it until today.

Their summary is:

We have checked the logs., seems to be there is an issue with the hardware.

Kernel logs snippet:
------------------
[48486.608925]
ata3: hard resetting link <3>[48488.814515] ata3:
COMRESET failed (errno=-32) &lt;3&gt;<br />[48488.818944]
ata3: reset failed, giving up &lt;3&gt;<br
/>[48488.822961] ata3: EH pending after 5 tries, giving up
&lt;6&gt;<br />[48488.828011] ata3: EH complete
&lt;3&gt;[48489.015272] ata3: exception Emask 0x10 SAct 0x0 SErr
0x190002 action 0xe frozen &lt;3&gt;<br />[48489.022786]
ata3: irq_stat 0x80400000, PHY RDY changed
&lt;3&gt;[48489.027916] ata3: SError: { RecovComm PHYRdyChg
10B8B Dispar } &lt;6&gt;<br />[48489.033897] ata3: hard
resetting link &lt;3&gt;[48491.317774] ata3: COMRESET failed
(errno=-32) &lt;4&gt;[48491.322263] ata3: reset failed
(errno=-32), retrying in 8 secs &lt;4&gt;<br
/>[48499.028344] ata3: limiting SATA link speed to 3.0 Gbps
&lt;6&gt;<br />[48499.033528] ata3: hard resetting link
&lt;3&gt;[48501.420728] ata3: COMRESET failed (errno=-32)
&lt;4&gt;<br />[48501.425172] ata3: reset failed
(errno=-32), retrying in 8 secs

Even after you have replace the new compatible drive, still the same issue. Please send the NAS for repair. 

In a conventional environment an RMA might not be so bad, in mine it's a nightmare due to the location of each one, going forward we're going to have to seriously reconsider using QNAP hardware.

[dulcifium]

Can someone summarise which firmware/device combos are affected by this issue?

[Neilo]

Has Qnap acknowledged this is an issue? What have they done to resolve it? Is the issue a manufacturing defect in the Backplane or is it software related? I'm a bit worried that replacing the backboard isn't resolving the issue and is just a band-aid until it happens again.

[michael_canny]

Has Qnap acknowledged this is an issue? What have they done to resolve it? Is the issue a manufacturing defect in the Backplane or is it software related? I'm a bit worried that replacing the backboard isn't resolving the issue and is just a band-aid until it happens again.

The "official" response I got from my QNAP ticket was to log an RMA - I now have 6 of these with the same issue that all need to be sent back to QNAP.

They gave no clue as to the underlying issue but there doesn't seem to be a pattern to the NAS models or firmware versions based on the content here, for me it's all TVS-463's though.

[IIGTech]

BEWARE of update: 4.3.4.0435 on Qnap TS-563. Exact same thing happened to me. Upgraded Firmware to 4.3.4.0435. Soon after disk #3 showed offline and disconnected. Although running a RAID 5 with 4 disks, my Vol immediately became unavailable and could not access data for the 12 hours it took to rebuild, after we reseated the hard drive (twice) to get it back online. Thank god we had just made a backup to external drive. Had to share external backup drive in order to allow business to continue. Emailed QNAP and tried to chat support, without any success, no one ever responded. I read in some of the comments in this posts that this will happen again, I am shocked that QNAP would put out an update, and keep out, an update with this symptoms. I am also disappointed that QNAP support is not available and responsive for this kind of situation. Based on this situation and lack of business grade support, I don't feel confident to continue using QNAP as my NAS solution.

My conclusion from experience and this thread, is that it is NOT a hardware issue, but a QNAP firmware issue. What I am deducting is that it started happening after version 4.3.3.0188. I know my QNAP was running solid on version 4.2.2(20160901). Can anyone confirm the latest firmware that can be used, that does not produce: "Host: Disk 3 failed" followed by "Host: Disk 3 unplugged" and then "RAID Group 1 is in degraded mode"

[Trexx]

QNAP's technical support hours are listed here: https://www.qnap.com/en-us/about/con_sh ... wone&cid=1 Their online support generally has a 24-hour SLA (M-F).

For users who need a higher level of support, the general recommendation would be to purchase your QNAP product through a QNAP VAR/Reseller who offers a higher level of support to meet your business needs.

All that being said, I have reached out to a contact at QNAP (in Taiwan) so hopefully that will help get some extra visibility to this issue. It is night time there, so may take a day or so before he is able to see the message.

[Neilo]

Hopefully, they respond to you TREXX. One person is already confirmed having to send their unit in twice for the same issue. I'm too worried about the reliability. Of course, i have backups in my strategy so I'm safe. Very frustrating when my older qnap I sold never had a single issue all the years and now only having the TVS-473 for six months and this happens. Yikes!

Runinng 4.3.4.0435 build 20171230