start script isn't working

[lightx]

The start.sh script isn't working but the ./bin/lifx-http is.

It gives me this error when I try starting the start.sh script

/usr/local/rvm/gems/ruby-2.1.1@global/gems/bundler-1.5.3/lib/bundler/rubygems_integration.rb:240:in   `block in replace_gem': thin is not part of the bundle. Add it to Gemfile. (Gem::LoadError)
        from /usr/local/rvm/gems/ruby-2.1.1/bin/thin:22:in `<main>'
        from /usr/local/rvm/gems/ruby-2.1.1/bin/ruby_executable_hooks:15:in `eval'
        from /usr/local/rvm/gems/ruby-2.1.1/bin/ruby_executable_hooks:15:in `<main>'

Im trying to demonize the process on my Pi. Im sure I'm just missing something easy. Help.

[chendo]

Whoops, forgot to take out start.sh when I converted it into a gem. Right now, there's no built-in support to daemonize. Thin was a bit too heavy, and I didn't have enough time to try to shoehorn in the MDNS stuff nicely.

[chendo]

I'm running mine in a screen session at this point until I can write get some clean daemonizing thing in.

[lightx]

Makes sense..

PS: I got the lifx-http to communicate with https://github.com/Neal/PebblFX (its pretty cool) that's why I'm trying to get it to run in the background.

[chendo]

I'm using PebblFX myself :) It's just running in a screen session at this point. Daemonizing would require background logging etc, and I wanted to ship the MDNS stuff before the end of the weekend heh. If you can figure out a clean way, I'd be happy to pull it in

[justinatomatic]

Ive got lifx-http running in the background on a raspberry pi using supervisor and RVM wrapper scripts.

http://omaticlabs.com/homepage/24/lifx-http-supervisor

[ristomatti]

@justinatomatic, I have it running (in Raspberry) basically with just executing it in the background with "bin/lifx-http &". Have you had any issues of it freezing over time like every 0.5-3 days? I had the exact same issue with a virtual machine on an Intel platform. I've now "solved" the issue by creating a shell script that kills and restarts the process every 4 hours and now it's almost always responding...

[justinatomatic]

I've been running it in a detached screen session up until now and I've never had to restart it. So months of uptime.

Supervisor version up two days now, see how we go.

[ristomatti]

@justinatomatic, good luck then(?) :) I've been having this issue (and commented on it also) https://github.com/chendo/lifx-http/issues/16

As a side note- I've also went through a similar "forced" crash course on Ruby/RVM as you describe on your page.

[chendo]

We have an official HTTP API coming if you fellas want to wait for that. Similar API (much improved actually)

[ristomatti]

@chendo, that's great news! Will it be Ruby based also? I'm kind of hoping for some Node.js or native option since Ruby takes a lot of extra resources on memory constrained devices like Raspberry and I'm not using it for anything else. But in any case if it's more robust it's very welcome.

[chendo]

It's actually going to be hosted, so you'll hit an endpoint with a token. No need to host anything yourself!

[chendo]

As for a self-hosted version, I'll have to work on that in my own time, which is already split across so many personal projects heh

[justinatomatic]

This is rapidly going to get off topic, but in the absence of a decent LIFX forum :)

Local non cloud communication is what's important to me. I'm working on a LIFX-MQTT bridge in Ruby for bidirectional communication with my code/home.

Is the upcoming clouded based API related to the move to AllJoyn?

[chendo]

Ah cool, I'd like to see that if you plan to release it.

Nope, not related to AllJoyn

[justinatomatic]

Going to to be open source but i'm using your private API's as i need event callbacks so i don't have to poll for state changes. Be good to chat I'll get in touch during work hours as it's work related.