Odyssey X86J4125: USB timeouts + port freezes rendering it unusable

Products & Technology ODYSSEY Serials
#1

Hi,

I am experiencing severe USB stability issues on my Odyssey X86J4125; ports keep disconnecting and the system needs to be reset. Both USB3 and USB2 are affected, although the latter appears to be more stable (i.e, can run for weeks w/o issue, presumably due to the lower 480Mbit/s bandwidth on USB2). Equipment connected to USB3 however will only work for a few minutes. The affected equipment has been tested on an identical OS running the same versions, and cable.

This started being noticable with an external storage system, connected via USB3 directly to the front USB3 (A) port. I have not previously used the Odyssey for this (my 2nd, older system referenced above, was the primary storage system until now)

I have have tried different cables – 2x USB-A to USB-B (the storage unit has a USB-B/host connector) and 1x USB-C to USB-B – using the USB-C input on the front (connecting to the same USB-B port on the storage system – an IcyBox RAID.
The front USB-C port behaves the same–presumably as it uses the same bus as the USB-A.

When this happens the port is effectively dead. USB2 works better, but this also occurred when it was connected to a USB2 port in the back.

Example kernel errors when this occurs (note: many different kernels have also been tried, including the exact same version being used on a different system where the same storage system works perfectly over USB3 (running for months and years). The issue appears to be specific to the Odyssey X86 – either bandwidth/voltage, USB controller chip or BIOS issue.

The behaviour is repeatable, and consistently happens within 20 minutes of disk i/o. Rebooting + reverting to a USB2 port fixes the issue, and it can then run for days – even weeks – until a similar USB port hang happens.

And to clarify: no USB hubs or any other equipment is connected. Issue has also been seen with USB mice connected to the USB3 port (laggy, or unresponsive. Fixed by reconnecting, or moving to a USB2 port)

[2692175.334899] scsi 2:0:0:0: Direct-Access     HzW RAID 5                     PQ: 0 ANSI: 5
[2692175.336208] sd 2:0:0:0: [sdb] Very big device. Trying to use READ CAPACITY(16). 
[2692175.336441] sd 2:0:0:0: [sdb] 117190754304 512-byte logical blocks: (60.0 TB/54.6 TiB)
[2692175.336784] sd 2:0:0:0: [sdb] Write Protect is off
[2692175.336792] sd 2:0:0:0: [sdb] Mode Sense: 28 00 00 00
[2692175.337273] sd 2:0:0:0: [sdb] No Caching mode page found 
[2692175.337282] sd 2:0:0:0: [sdb] Assuming drive cache: write through
[2692175.619205] sd 2:0:0:0: [sdb] Attached SCSI disk
]: CODE_FILE=log/log.c:CODE_LINE=686:_vprint_log:PV /dev/sdb online, VG backup is complete.

[2692425.244112] sd 2:0:0:0: [sdb] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK cmd_age=0s
[2692425.244134] sd 2:0:0:0: [sdb] tag#0 CDB: Read(16) 88 00 00 00 00 15 9e 04 9a c0 00 00 08 00 00 00 
[2692425.244141] I/O error, dev sdb, sector 92845415104 op 0x0:(READ) flags 0x80700 phys_seg 16 prio class 0 
[2692425.244155] usb 2-1: USB disconnect, device number 3
[2692425.244285] sd 2:0:0:0: [sdb] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK cmd_age=0s
[2692425.244296] sd 2:0:0:0: [sdb] tag#0 CDB: Read(16) 88 00 00 00 00 15 e3 e1 d8 00 00 00 00 08 00 00 
[2692425.244301] I/O error, dev sdb, sector 94017542144 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
[2692425.244459] device offline error, dev sdb, sector 94017542144 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[2692425.244482] Buffer I/O error on dev dm-3, logical block 477898752, async page read 
[2692425.244575] device offline error, dev sdb, sector 94017542144 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[2692425.244593] Buffer I/O error on dev dm-3, logical block 477898752, async page read 
[2692425.245593] Buffer I/O error on dev dm-3, logical block 201326593, async page read 
[2692425.245609] Buffer I/O error on dev dm-3, logical block 201326593, async page read 
[2692425.245721] Buffer I/O error on dev dm-3, logical block 603979777, async page read 
[2692425.245733] Buffer I/O error on dev dm-3, logical block 603979777, async page read 
[2692425.245803] Buffer I/O error on dev dm-3, logical block 201326594, async page read 
[2692425.245817] Buffer I/O error on dev dm-3, logical block 201326594, async page read 
[2692425.245879] Buffer I/O error on dev dm-3, logical block 201326595, async page read 
[2692425.245893] Buffer I/O error on dev dm-3, logical block 201326595, async page read 
[2692425.276582] scsi 2:0:0:0: rejecting I/O to dead device
[2692430.116542] usb 2-1: new SuperSpeed USB device number 4 using xhci_hcd
[2692430.134216] usb 2-1: New USB device found, idVendor=152d, idProduct=0539, bcdDevice= 1.00

Any assistance and pointers would be immensely helpful.

#2

Seems the user support has gone a bit downhill; let me know whether this is not the correct place to post it.

I was one of the first users to buy the J4125, and a previous issue with the first-gen Odysssey (defective fan) was quickly dealt with, so I am hoping this issue can be resolved as well.

Posting BIOS information in the event this is helpful, and the issue is not hardware related.

        Version: SD-BS-CJ41G-300-101-F
        Release Date: 04/09/2021
#3

Hi there,
I would LQQK to the power supply, They are very medium units. Requires good power to work properly and run cooler.
Low DC ripple.
HTH
GL :slight_smile: PJ
:v:

#4

Perhaps, but I should probably be seeing other issues if such were the case, as there are two M.2 disks connected, together with a SATA drive and various USB2 peripherals – which work fine. The issue appears to be related to the USB3 ports (e.g, the same device which has issues disconnecting works fine when connected on the SATA port, but not on the USB3 on the Odyssey)

#5

Hi there,
Even more so the power supply , USB 3.0 is more power capable. I would unplug everything else and shut it off for a period then , repower and take some measurements,use a USB port meter also put a fan on it. see how far it is able to get.
My .02
GL :slight_smile: PJ
:v:

#6

Ok, thanks. If the power supply is assumed defective I would expect Seeed to respond and send a replacement, like they did previously with the fan. I do not have any way to measure this, unfortunately, only from what I am experiencing wrt. how the system behaves. I have disconnected peripherals.

Again, the issue is reproducible and the external disk system will disconnect after 10-20 minutes regardless of using the USB3-A and USB3.1-C port. With USB2/480Mbps it runs much more stable, but did also experience an issue after approx. three weeks.

Anyone from Seeed around? (@Bruce.Qin?).

#7

Hi there,
Yea So click on his user profile and send him a message , same with Seeed_Serphina.
they will get you straight. PITA for sure ,It is a nice board and seems like a good idea the carrier board route, but I don’t feel they got it nailed down. Even on the others the updates from jetpack to JP6.0 were fraught with issues.
nah’ it’s NOT there Yet IMO.
HTH
GL :slight_smile: PJ :v:

#8

Hey there! I’m on my way. Could you let me know which peripherals you connected and what kind of power they need? You could also try USB with just a flash drive to see if that helps. I think the problem might be that the peripheral is asking for more current than the maximum protection current set by the board. Waitting your voice. :smiley:

#9

Hi Bruce,

I’ve tested this with an IcyBox RAID device (see original post), which was working fine on another system over USB3. However, the problem is unfortunately also present when using mice or keyboards on the USB3 port (intermittent signal, mouse appears dead, etc.) Temporarily fixed by moving to another port/re-plugging or using USB2.

The external RAID device uses its own power (seems I forgot to mention this excplicitly, but most/all RAID boxes do), which is why I thought this to be a USB3 chipset issue or a BIOS issue (however, I haven’t found any mention of this elsewhere)

So the port resets are seen with other devices as well, not just the storage device. They just happen less frequently.

#10

And yeah @PJ_Glasso – definitely a great board (too bad it encountered this issue). I have two Odysseys–the first J4105 and this one. The former, with the exception of a fan issue which was replaced, has been working well.

It would be good if it had USB3 ports on the back though, but hopefully that will make it to the third version. (I’d ask for e-SATA as well, but that is probably more unlikely)

1 Like
#11

Ok, our support team will contact you later.

2 Likes
#12

Sorry I’m late, we’ll be communicating follow-up repairs within [email protected]

1 Like
#13

Ok, thanks. So just to be clear–I should be expecting an email from [email protected] (or should I send another request there?)

#14

As it is not clear from my side whether your email address on the forum is the one you are currently using. So I suggest you can send an email to [email protected] and I will match you on it.
(include a link to the forum in your message)

#15

My email address on the forum is still current (not sure why it would be expected not to be?). In any case, I’ve also forwarded the details of this to the techsupport address.

#16

Hi everyone,
Did you solve your problem? I’m currently experiencing the same problem. I first changed my usb card but still the same disconnection. Great to have your back.

#17

Hi,

Sorry for the inconvenience, please send an email to [email protected], describe the occurrence in detail and attach this forum link. My colleague will continue to help you with this issue.