Jason Axelson

Oooh, that would be nice! Another option would be for `mix firmware.burn` to not allow overwriting an existing nerves image and instead require `mix firmware.burn --overwrite` to actually overwrite the...

Are there any breaking changes in 4.0?

I have confirmed that this regression was caused by: https://github.com/nerves-project/nerves/commit/40a77d8a9511fa9f14b0c5204ebda3c3ce0d3c76

A weird thing is that if I use nerves as a `:path` dependency ec5ae3f2639ededd229ee159d5ec9aa8e23b5470 then the project still always gets recompiled, but without the make output. MIX_DEBUG=true mix compile --verbose...

So it looks like the main issue is that `elixir_make` always returns `{:ok, []}` ([source link](https://github.com/elixir-lang/elixir_make/blob/2016764ff1ce9cb2fe9fa88404eaac9612234690/lib/mix/tasks/compile.make.ex#L153)), which is indicating that `nerves` was "recompiled", which means `mix will_recompile` is run on...

Okay, the above info is correct, but I was thrown off for a bit because `:path` dependencies are treated rather differently when it comes to detecting if recompilation is needed....

In addition to @epinault's fork there is this hex package: https://hex.pm/packages/wobserverfork which I believe originates from https://github.com/alfonsoperez/wobserver

+1 to this! I also wanted to note that the "The Road to 2 Million Websocket Connections in Phoenix" article looked at mailbox length as a major indicator in one...

I think that a workflow-based section of the README would be helpful, although I don't think it would supplant the reference portion (although perhaps the reference portion could shrink a...

Does anyone have a workaround for this? I've been bitten by this multiple times now.