Report of red flashing lights on bridges 🚨 about sirius

[notjosh]

Okay, took a quick look, and think I figured it out.

claiming.process_claim_code returns a base64'd string, and encoders.encode_bridge_command was incorrectly updated to wrap that value in base64. Yo dawg.

I'll add a test, fix it, and push a release tomorrow, then hopefully we should be seeing printers back online.

[notjosh]

That's...actually a really good question. I don't think it's trivially accessible from your end. Maybe if we start with one of:

• printer ID (from the URL, like https://littleprinter.nordprojects.co/printer/:id)
• your claim URL (like https://littleprinter.nordprojects.co/:id/:username/claim)
• the claim code you used fro the printer

Any one of those should be enough to help track it down.

You're welcome to email it over, if you're not stoked on having those details floating around on GitHub too! (email in profile).

[jeremyfranklin]

Sent it your way.

[notjosh]

Hiya @jeremyfranklin, I've merged/deployed a potential fix. Can you let me know if it gets your printer back in action?

[dickverbunt]

Mine is back alive! Awesome work @notjosh.

[jeremyfranklin]

@notjosh It works! Amazing work and thank you.

[jeremyfranklin]

My Little Printer says it is online, but I have the red light. Where do I find the ID?

[notjosh]

Welp, frustratingly, I added some tooling that seemed to work just fine!

I manually dumped the event payload, looks something like:

{
  "type": "DeviceCommand",
  "bridge_address": "some-bridge-address",
  "command_id": 1,
  "timestamp": "0",
  "binary_payload": "[base64 blob]",
  "device_address": "[redacted]"
}

And it decoded in my sandbox as:

(images get sent upside down for printing, so this is correct).

SO, that means potentially one of a couple of things:

1. the failed encoding is in one of the values I stubbed out here, which seems unlikely?
2. I'm looking at the wrong thing. which seems more likely :)

[notjosh]

OKAY, yes, duh, I'm looking at the wrong thing. The error mentions executeBridgeCommand, and "unpack requires a string argument of length 16", so I'm guessing that's pointing us at this line, which makes way more sense.

I'll take another look tomorrow when I get a chance!

[jeremyfranklin]

@notjosh I can let you know tonight.