Electron LED - solid cyan?

Ok here’s a new one:

My Electron enters this state where the LED is solid cyan, and the Electron is non-responsive. This only happens after I leave it my device on overnight - it’s working fine (breathing cyan) when I go to bed, and when I wake up it’s zombified (solid cyan). I can push the reset button, and everything goes back to normal, but I had to at least ask the question:

What the heck does solid cyan mean? Can’t find it anywhere in the docs or community.

Cheers,
L3

Solid cyan is not an “official” state–it’s usually that the device locked up or stopped running firmware after some period of time. What firmware are you running on the device? Do you see any red colors from the RGB LED on the Electron?

If you’d be willing to post the code that’s running on the device, that’d be most helpful.

@will,
I really appreciate the offer - but it may be more than you bargained for:

https://github.com/leo3linbeck/particle/tree/master/BrevitestV10

I’d understand if you decided that wading through this code is not worth it. But if you want more color on what we’re trying to do, feel free to shoot me an email: leo3 at linbeck dot com.

Cheers,
L3

@will,
And in answer to your questions:

Firmware: Electron - not sure how to find the version but the cli compile command generates this message:

downloading binary from: /v1/binaries/5701eee66cb8481512d5c7f5

LED: no change to red that I ever see (but then I'm asleep when this is happening...).

Cheers,
L3

UPDATE: woke up this morning with the LED flashing blue. Go figure. Cheers, L3

This happens if you use SYSTEM_THREAD (ENABLED);
It is still in beta and does not work well with the electron.
Disable that option and will be fine :wink:

Hmmm, good to know.

Actually, when I don’t use SYSTEM_THREAD(ENABLED), I have a different problem: I keep losing connection to the cloud and it stops execution while reconnecting. There are a number of situation where I have long-running tasks, and stopping execution to reconnect to the cloud is not a good thing. This problem went away when I enabled the system thread. I may keep the system thread for now - it runs just fine during the day, and I can simply restart each morning.

Cheers,
L3

This info might interest you in that connection

1 Like

@ScruffR,

Ah, thanks so much for the link. Didn’t know that about the Electron and multithreading. The stack size issue now makes perfect sense - it’s the kind of problem you’d expect in this set of circumstances:

  1. Problem emerges after running for a long time.
  2. Big, complex user code.
  3. Second-rate hacker (i.e. me).

I look forward to the 0.5.0 release with baited breath. All the best,

L3

1 Like

@leo3linbeck as a heads up, v050 is currently in prerelease! You should be able to experiment with it here before it’s officially released: