TR-004 keeps ejecting by itself 451+

[lgottschald]

Hi,

I bought a TR-004 and keeps ejecting by itself. NAS keeps running, when I checked the notice board, received the message: Enclosure Ejected (4) TR-004-N344 has been ejected.

451+ (Firmware 5.0.0.1858 16GB RAM)
[Disk1 - WD RED Plus 8TB]
[Disk2 - WD RED Plus 8TB]
[Disk2 - WD RED Plus 8TB]
[Disk4 - WD RED Plus 8TB]

TR-004 (Set individual, USB 3.0 Port, Firmware 1.1.1)
[Disk1 - Empty]
[Disk2 - WD RED Plus 10TB]
[Disk3 - WD RED Plus 8TB ]
[Disk4 - WD RED Plus 8TB ]

Both using an UPS. Is there a way to see why he is ejecting? Logs? Configuration?

I research a little and found a post from an user with the same setup, his explanation was the USB 3.0, he changed to 2.0 with a small cable and said it was fixed, that TR-004 doesn't work well with high speed ports. But USB 2.0 is too slow to transfer large files. The empty bay is because the other 10TB disk arrived DOA from WD and I sent back. Both 10TB disks came together, the other two 8TB I bought in person last Friday. Is possible a disk is causing this? I tried using support chat but no one answered, then I tried to open a ticket. Still waiting. Any ideas?

Thanks in advance

[dolbyman]

Poking in the dark here .. but the USB ports are on the Super I/O chip... the LPC bug of your device makes the USB (and other devices) fail, leading to a sudden/sudden death of the NAS

[lgottschald]

Poking in the dark here .. but the USB ports are on the Super I/O chip... the LPC bug of your device makes the USB (and other devices) fail, leading to a sudden/sudden death of the NAS

I am accepting any suggestions. I was using a Seagate Expansion 6TB USB 3.0 before the TR-004 (3-4 months) and never ejected. Worked flawless all the time, do you think maybe the TR-004 is more sensible to changes and eject to avoid dmg or thinks that the USB connection was lost/disconnected?

Thanks.

[dolbyman]

The issue would be with your NAS not the TR004

Only way to find out would be to scope out the CLK Signal

[Crusher]

The issue would be with your NAS not the TR004

Only way to find out would be to scope out the CLK Signal

Wouldn't that depend on the build age of his TS-451+ as that LPC bug was fixed a couple of years ago in the J series processors.

[dolbyman]

dont know starting what revision (and if) it was fixed

[Crusher]

dont know starting what revision (and if) it was fixed

It was fixed from June 2018 according to the Intel Document Number: 335864-002
Intel® Celeron® Processor J1800, J1900,
N2807, and N2930 for Internet of Things
Specification Update Addendum
June 2018 - which you can find here:
https://www.intel.com/content/dam/www/p ... dendum.pdf
In Section 5.1 and 5.2,they acknowledge the issue and then in section 3.0, it is listed as fixed.

I researched this about a month or two ago because I was using a TS-451+ with the TR-004 enclosure but it was just too limited in it's throughput and not capable of 4K video, which my two TS-653D's handle with ease now.

[dolbyman]

QNAP board revision I meant ..not intel

[Crusher]

QNAP board revision I meant ..not intel

So is it a QNAP board problem and not an Intel problem? The way around it was to short out two headers on the board but I would have thought that that related more to the issue with the Intel processors, rather than the motherboard? Happy to be corrected as I still have a TS-451+ which is now not being used but I'd like to know more details on this if you can offer any.

Thanks.

[dolbyman]

QNAP must know internally if they ever build versions with the fixed processor and marked it by revision or at least starting with a certain date or serial code .. but only QNAP would know what.

Maybe they bought tons of CPU in stock (or the OEM did that they use) and they ALL have the degrading processors. I don't know.

Only QNAP would know the date code/ revision when and if they have implemented the fixed processors.

Similar to the revision of the new x72X boards (revision 1.7 or 2.0 is the fixed one .. all other are susceptible to sudden death)

So, yes it's an intel issue, but just because intel has introduced a fixed revision of the processor doesn't mean that they immediately made it into production of all manufacturers

[jasonwoerner]

Chiming in. Also experiencing this issue. I have a storage pool on the extended storage, so that goes into read-only mode right after the eject occurs. It seems the storage is reconnected automatically, or at least detected as available to reconnect. But I don't know what the root cause is, either, and the problem occurs every 10 days or so, perhaps depending on the throughput between hardware. I've opened a ticket today.

[gate49]

Chiming in. Also experiencing this issue. I have a storage pool on the extended storage, so that goes into read-only mode right after the eject occurs. It seems the storage is reconnected automatically, or at least detected as available to reconnect. But I don't know what the root cause is, either, and the problem occurs every 10 days or so, perhaps depending on the throughput between hardware. I've opened a ticket today.

I have the identical issue...

[dolbyman]

and you opened a ticket as well?

[qrusher]

try to downscale memory size