“Command failed with status ()” when precompiling

Question

I am having similiar problems like a lot of users when compiling assets on their productive box. The only difference is that I can't get any hint out of the trace to solve the problem.

rake assets:precompile RAILS_ENV=production --trace
** Invoke assets:precompile (first_time)
** Execute assets:precompile
/usr/local/rbenv/versions/1.9.3-p362/bin/ruby /usr/local/rbenv/versions/1.9.3-p362/bin/rake assets:precompile:all RAILS_ENV=production RAILS_GROUPS=assets --trace
** Invoke assets:precompile:all (first_time)
** Execute assets:precompile:all
** Invoke assets:precompile:primary (first_time)
** Invoke assets:environment (first_time)
** Execute assets:environment
** Invoke environment (first_time)
** Execute environment
** Invoke tmp:cache:clear (first_time)
** Execute tmp:cache:clear
** Execute assets:precompile:primary
rake aborted!
Command failed with status (): [/usr/local/rbenv/versions/1.9.3-p362/bin/r...]
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils.rb:53:in `block in create_shell_runner'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils.rb:45:in `call'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils.rb:45:in `sh'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils_ext.rb:39:in `sh'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils.rb:80:in `ruby'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/file_utils_ext.rb:39:in `ruby'
/home/app/application/ruby/1.9.1/gems/actionpack-3.2.12/lib/sprockets/assets.rake:12:in `ruby_rake_task'
/home/app/application/ruby/1.9.1/gems/actionpack-3.2.12/lib/sprockets/assets.rake:21:in `invoke_or_reboot_rake_task'
/home/app/application/ruby/1.9.1/gems/actionpack-3.2.12/lib/sprockets/assets.rake:29:in `block (2 levels) in <top (required)>'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:205:in `call'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:205:in `block in execute'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:200:in `each'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:200:in `execute'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:158:in `block in invoke_with_call_chain'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/monitor.rb:211:in `mon_synchronize'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:151:in `invoke_with_call_chain'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/task.rb:144:in `invoke'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:116:in `invoke_task'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:94:in `block (2 levels) in top_level'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:94:in `each'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:94:in `block in top_level'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:133:in `standard_exception_handling'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:88:in `top_level'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:66:in `block in run'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:133:in `standard_exception_handling'
/usr/local/rbenv/versions/1.9.3-p362/lib/ruby/1.9.1/rake/application.rb:63:in `run'
/usr/local/rbenv/versions/1.9.3-p362/bin/rake:32:in `<main>'
Tasks: TOP => assets:precompile

There is effectively no status code, just the fail. It also does not make any difference if I call rake directly or via bundle exec.

About the environment debian squeeze box with a global rbenv installation (/usr/local/rbenv like you can see from the trace). Ruby 1.9.3 2012-12-25 patchlevel 362.

Any hints / ideas on this?

Answer 1

I also had same issue. I fixed it by adding swap(in my case 1gb for 512RAM available on my server)

Answer 2

Try running htop while your assets are compiling to see if you run out of RAM. Adding swap fixed it for me.

See this guide for adding swap: https://www.digitalocean.com/community/articles/how-to-add-swap-on-ubuntu-12-04

Answer 3

I am going to answer this myself because I could solve it more or less. If someone has additions, don't hesitate to write an own answer or comment this answer/question.

What I've found out so far:

If you play around with assets:precompile by e.g just compiling the primary assets (assets:precompile:primary) or explicitly calling assets:precompile:all you might end up with a hint about the source of your problem. In my case, I ran into Errno::EACCES on both public/ and tmp/. Somehow that was not displayed, so make sure your user has full rights to create/delete files and folders there.

In my case, it worked sometimes because I've shut down the rails app and precompiled while it was down. Since precompiling allocates a lot of memory, I did some trial and error and eventually got the known Killed message when rake tries to execute asstes:precompile:primary. The task simply was killed because of using too much memory.

Another problem was that sprockets could not find bootstrap to put it into the assets pipeline on precompilation. Moving gem 'bootstrap' outside of group :assets solved this. This was also something I got hinted to when I played around with the commands.

The best way to solve – or better: work around – this problem is to simply compile your assets locally. Just throw a rake assets:precompile RAILS_ENV=development into your terminal and then deploy public/assets. Remember to delete that folder in your development environment after deploying it or you will end up debugging on why your changes on app/assets/* are not taking effect in development. At least that works for me(tm).

Alternatively, growing your swap partition may also work. However, compilation on a VPS where you need to use swap may take its while, so I'd stick to the local way.