Mark II won't boot

So I may have killed my Mark 2 and I am not sure why or how to fix it. After finally getting documentation on Dinkum, I decided to try using a different image as described in the documents. I flashed a new thumb drive and shut down the Mark 2. However, after shutting down the Mark 2, the fan kept running for 15 minutes even after the screen was blank. I had accidentally unplugged the Mark 2 before while it was on and it had booted back up, so I didn’t think anything off cutting the power at this point.
Then I tried removing the old flash drive with Dinkum and putting in a new one before reconnecting the power. The fans came on full blast again, but the screen remained black. Deciding I must not have done something right after waiting 30 minutes, I unplugged it again since I didn’t have any other way to shut it off before putting the Dinkum flash drive back in. Then I switch back to the Dinkum flash drive, but it’s doing the same thing where the fan is on at full blast for 30 minutes and nothing is happening on screen.
Is there an SD card that might be used in the booting process that possibly got corrupted? Is it possible the flash drive itself got corrupted?

I’m at a total loss. I’d like to repair it but don’t know where to start.

1 Like

I am in the same boat. I ran into an issue where my Mark II wouldn’t show the SSH option on the Devices page. So after rebooting a couple times, I tried to re-flash the Dinkum image on the USB that came with my Mark II. Aaaaaand, I had the same result as you with the fan on and blank screen.

Has anyone tried and successfully re-image their Mark II?

Is there a link somewhere to the Dinkum image? I did make a backup of my USB before starting to tinker, but it would be nice to know there’s a clean/factory image of what shipped with the Mark II that I could restore in a pinch.

@Rudism Here’s the link: https://mycroft.ai/to/mark-ii_stable

2 Likes

Hey there, can you please confirm the filename and SHA256 hash of the file you used to flash the USB?

$ sha256sum Downloads/myc200-pv-stable-2022-10-18.img.xz 
6c0e54bc63bfcffedc3a507657f1b49c5768148ad9c8ad1bcb75c4e837f0f428  Downloads/myc200-pv-stable-2022-10-18.img.xz

I just downloaded a fresh copy to verify that we didn’t accidentally push a bad image to the CDN, and that is booting fine.

It’s possible that your display cable has come loose. We discovered this on a small number of early units and have made changes to our assembly process to mitigate this. If you’re comfortable doing so, you can open the front of the Mark II with a philips head screw driver and check the display DSI ribbon cable by giving it a gently tug. Some more detailed instructions for that here:

Hey @gez-mycroft,

The SHA256 hash of the file is
6C0E54BC63BFCFFEDC3A507657F1B49C5768148AD9C8AD1BCB75C4E837F0F428

I opened up my Mark II and the display ribbon cable was not loose, even after giving it a little tug. I even unplugged the cable and re-seated it.

However, the behavior of a blank screen and the fan on high blast continues.

Don’t suppose you have a media usb in as well ? Mine won’t boot with both in. Just a thought

Nah, just one usb in at a time when I tested it.

Hi all, if anyone runs into this same issue - I’m very keen to know the filenames of all images you’ve booted on the device. Particularly if you tried using the old Dev Kit Stable image which would be named:
MarkII_arm-rpi64_Prod_013-stable_2021-12-13.img.gz

NOTE: Do not try to use this image in a production Mark II. It may be the cause of this issue, and even if it’s not - it won’t work on the newer hardware. All the new software however works on all previous iterations of the hardware.

For at least one device in this state it is the Raspberry Pi itself that is borked. Eg that Pi will no longer boot anything at all, even in isolation of the rest of the Mark II.

It might be just a bad Pi but given there are a couple of people reporting similar behaviour and they all seem to have booted fine the first time, and then failed after reflashing attempts, it seems far more likely that something is causing the failures.