Field Guide


A while ago, I started building this:

The idea is that it’s collaborative documentation about how Something New works and how to do things here. It’s an idea shamelessly borrowed from work (hence it being on WordPress). I’ve not touched it for months and I’m conscious it requires folks to be given a user account, so it’s not ideal.

So tonight I decided to do some dog-fooding and set up based on the exact same system used for the OpenPolitics Manifesto.

The Votebot did work at first but now I’m getting an “Application Error” so we can’t actually edit anything. Once I get it working, we can start using this for things like the excellent guide @andrewdwilliams wrote about being a council candidate, or a how-to on recording expenses for a campaign following @Floppy’s fine example from the 2015 GE.

@Floppy I’ll probably need your help with this Heroku app :slight_smile: Your instructions were great, and I got it set up, and it appeared to work and then suddenly broke. There are no errors in the build and deployment is successful but I’m getting that Application Error and I can’t see anything that tells me what’s wrong. Any ideas?


We should also use the address for the new thing.


We should also use the address for the new thing.

I’ve just done this, so don’t be surprised if you don’t see an incomplete WordPress site when you go to now :slight_smile:


Correction: it didn’t work! Maybe I’ll just go to bed :laughing:


heroku logs -t will show you the app log, so you can see the errors. If you share access to [email protected] I’ll take a look at it myself, too.


Fantastic idea to dogfood the reusable votebot, too, nice work.


Should this include things like our candidate selection process, etc, as mentioned in the constitution? I think it probably should…


Well, I was going to try and fix this up now but Amazon killed all the internets :frowning:


I just ran it again and this is what I see in the log:

2017-03-01T10:16:32.404229+00:00 heroku[router]: at=error code=H14 desc=“No web processes running” method=GET path="/" request_id=50e981c1-30d9-47e3-bac2-e50c48dbe339 fwd=“” dyno= connect= service= status=503 bytes=
2017-03-01T10:16:33.649439+00:00 heroku[router]: at=error code=H14 desc=“No web processes running” method=GET path="/favicon.ico" request_id=b3426f8a-4fc9-4132-9ee3-b5e281181484 fwd=“” dyno= connect= service= status=503 bytes=

It’s scheduled to deploy overnight, and I also see this in the logs from last night:

2017-03-01T03:30:20.714369+00:00 heroku[scheduler.3254]: Starting process with command bundle exec rake update
2017-03-01T03:30:21.298057+00:00 heroku[scheduler.3254]: State changed from starting to up
2017-03-01T03:30:23.373140+00:00 heroku[scheduler.3254]: State changed from up to complete
2017-03-01T03:30:23.349169+00:00 heroku[scheduler.3254]: Process exited with status 1
2017-03-01T03:30:23.256182+00:00 app[scheduler.3254]: rake aborted!
2017-03-01T03:30:23.256211+00:00 app[scheduler.3254]: No Rakefile found (looking for: rakefile, Rakefile, rakefile.rb, Rakefile.rb)
2017-03-01T03:30:23.256421+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/rake-12.0.0/exe/rake:27:in <top (required)>' 2017-03-01T03:30:23.256423+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli/exec.rb:74:inload’
2017-03-01T03:30:23.256424+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli/exec.rb:74:in kernel_load' 2017-03-01T03:30:23.256425+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli/exec.rb:27:inrun’
2017-03-01T03:30:23.256425+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli.rb:332:in exec' 2017-03-01T03:30:23.256427+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/vendor/thor/lib/thor/command.rb:27:inrun’
2017-03-01T03:30:23.256428+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/vendor/thor/lib/thor/invocation.rb:126:in invoke_command' 2017-03-01T03:30:23.256429+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/vendor/thor/lib/thor.rb:359:indispatch’
2017-03-01T03:30:23.256430+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli.rb:20:in dispatch' 2017-03-01T03:30:23.256431+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/vendor/thor/lib/thor/base.rb:440:instart’
2017-03-01T03:30:23.256431+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/cli.rb:11:in start' 2017-03-01T03:30:23.256432+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/exe/bundle:34:inblock in <top (required)>'
2017-03-01T03:30:23.256433+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/lib/bundler/friendly_errors.rb:100:in with_friendly_errors' 2017-03-01T03:30:23.256433+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/gems/bundler-1.13.7/exe/bundle:26:in<top (required)>'
2017-03-01T03:30:23.256434+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/bin/bundle:22:in load' 2017-03-01T03:30:23.256438+00:00 app[scheduler.3254]: (See full trace by running task with --trace) 2017-03-01T03:30:23.256434+00:00 app[scheduler.3254]: /app/vendor/bundle/ruby/2.2.0/bin/bundle:22:in

Curious how it worked before I started making changes to the repo. I notice that the Votebot repo has a Rakefile but not the template - is that likely to be the cause?

Added you as a collaborator anyway so you can take a look if you have a moment.


Weird, it looks like the deploy hasn’t happened right for some reason. I’ll take a closer look.


This site could take some of the deeper content off the main site, I think - stuff like the candidate guides, and so on. That stuff isn’t really accessible from the main navigation anyway, it’s a bit buried.


Ah, I’ve had a look @philipjohn and the heroku app is deploying the jekyll code from, not the votebot code. That’s probably something to do with it :slight_smile:


Hahaha did I just paste the wrong repo URL in?


I think so, yes :slight_smile:

Was that when setting up auto-deployment, not on the initial deploy with the heroku button, presumably?


On the domain, I can set up if you like, but at the moment it’s coming up at by default, thanks to GitHub’s default URL mapping. What would you prefer?


I’m easy, either works. The sub domain might help separate out a bit.

I deleted the app this morning to start again but didn’t get around to it so I’ll try first thing tomorrow!


Double-checking I’m doing this right, @Floppy

the heroku app is deploying the jekyll code from, not the votebot code.

Here’s what I’m entering into the Heroku app settings:

…because the instructions say:

Enter the repository path of your cloned template site - it should be something like your-username/repository-name.

So, the Heroku app will end up deploying the SomethingNewUK/field-guide repo based on that setting, right?


Ah, no, that’s all correct. It was this bit I got wrong last time, right?:


Seems to be okay!


Looks good. Can you see where I should make the instructions better?