Thread dongle RESET_UNKNOWN errors

I have a Sonoff ZBDongle-E with the RCP firmware flashed for my Thread network. It has been working OK (other than the general difficulty of setting up Thread/Matter), but recently I have been having issues where all of my Thread devices (Thread over Matter and ESPHome Thread) are not reachable. At the same time, the below logs just keep repeating in the OTBR addin logs every few seconds.
I am not finding much info on these codes. ChatGPT says it may be related to power issues with the USB or a hardware fault with the dongle.
The only thing that temporarily fixes the issue is unplugging and re-plugging the dongle. However, today, even that does not fix it.

So far I have tried:

  • Using with and without USB extension cord
  • Updating firmware

The dongle is plugged directly into my PC either way so I don’t think this would be a USB hub related power issue. It does seem likely to be a hardware fault to me since power-cycling the dongle seemed to fix it temporarily. Does anyone have any more info or ideas on this?

02:01:36.471 [C] P-RadioSpinel-: Unexpected RCP reset: RESET_UNKNOWN
02:01:36.471 [W] P-RadioSpinel-: RCP failure detected
02:01:36.471 [W] P-RadioSpinel-: Trying to recover (1/2)
02:01:36.509 [N] P-RadioSpinel-: RCP recovery is done
02:01:36.511 [C] P-RadioSpinel-: RCP => [C] Platform------: Reset info: 0xb (BRO)
02:01:36.513 [C] P-RadioSpinel-: RCP => [C] Platform------: Extended Reset info: 0xB06 (DP )
02:01:36.835 [N] Mle-----------: Different partition (peer:597339313, local:1735904559)
02:01:37.221 [C] P-RadioSpinel-: Unexpected RCP reset: RESET_UNKNOWN
02:01:37.221 [W] P-RadioSpinel-: RCP failure detected
02:01:37.221 [W] P-RadioSpinel-: Trying to recover (1/2)
02:01:37.264 [N] P-RadioSpinel-: RCP recovery is done
02:01:37.266 [C] P-RadioSpinel-: RCP => [C] Platform------: Reset info: 0xb (BRO)
02:01:37.268 [C] P-RadioSpinel-: RCP => [C] Platform------: Extended Reset info: 0xB06 (DP )
02:01:37.971 [C] P-RadioSpinel-: Unexpected RCP reset: RESET_UNKNOWN
02:01:37.971 [W] P-RadioSpinel-: RCP failure detected
02:01:37.971 [W] P-RadioSpinel-: Trying to recover (1/2)
02:01:38.011 [N] P-RadioSpinel-: RCP recovery is done
02:01:38.013 [C] P-RadioSpinel-: RCP => [C] Platform------: Reset info: 0xb (BRO)
02:01:38.015 [C] P-RadioSpinel-: RCP => [C] Platform------: Extended Reset info: 0xB06 (DP )
02:01:39.221 [C] P-RadioSpinel-: Unexpected RCP reset: RESET_UNKNOWN
02:01:39.221 [W] P-RadioSpinel-: RCP failure detected
02:01:39.221 [W] P-RadioSpinel-: Trying to recover (1/2)
02:01:39.262 [N] P-RadioSpinel-: RCP recovery is done
02:01:39.264 [C] P-RadioSpinel-: RCP => [C] Platform------: Reset info: 0xb (BRO)
02:01:39.266 [C] P-RadioSpinel-: RCP => [C] Platform------: Extended Reset info: 0xB06 (DP )
02:01:40.471 [C] P-RadioSpinel-: Unexpected RCP reset: RESET_UNKNOWN
02:01:40.471 [W] P-RadioSpinel-: RCP failure detected
02:01:40.471 [W] P-RadioSpinel-: Trying to recover (1/2)
02:01:40.512 [N] P-RadioSpinel-: RCP recovery is done

Hello, please provide the following information to help identify the issue:

  1. RCP firmware and version.
  2. OTBR Addon version.
  3. Hardware platform information (e.g., Raspberry Pi version, virtual machine, or others?).
  4. Production batch number of ZBDongle-E (can be found on the back of the product).

Hi! Here is the requested info:

RCP Firmware and version: At first I was using the firmware from the Sonoff flasher website. I believe it was version 2.4.4. After it started having this issue, I switched to this (ot-rcp-v2.4.5.0-zbdonglee-460800) but still have the same issue.

OTBR Addon Version: 2.13.0

Hardware platform: HAOS on a VM with 2 cores and 5GB RAM.

Production batch number ( I think this is the right thing): 4482103DE9

I ended up ordering the official Home Assistant Zigbee Dongle and using it for Thread. So far so good, it’s been working again with that setup.
I assume the Sonoff dongle was defective.