resque-web crashes with segfault. Ruby 2.0.0-p0, Resque 1.24.1, Redis 2.6.12 #966

Closed
vaharoni opened this Issue Apr 22, 2013 · 14 comments

Comments

Projects
None yet
5 participants

See the repository for more details of the issue and how to reproduce it:
https://github.com/vaharoni/resque_test

Contributor

tarcieri commented Apr 22, 2013

Looks like a crash in erubis... not technically resque-web's fault but I'm not sure where you should report it

Thanks tarcieri. I just confirmed that this is also reproducible on a fresh precise32 Vagrant box (Ubuntu 12.04.2).
Does anyone have an idea where this should be reported?

Contributor

tarcieri commented Apr 22, 2013

Barring anything else, it could probably be reported as a Rails bug.

@steveklabnik any ideas?

Oh wait, just noticed this was the Sinatra version, never mind :|

Actually, mounting it in routes.rb is unnecessary. It produces the same stack trace when running:
$ resque-web -F

Update

I just reproduced it in a clean folder, without a Rails application. To reproduce (assuming you have a redis-server running):
$ rvm use 2.0@clean
$ gem install resque
$ gem install redis
$ resque-web -F

Click on 'live-poll' and wait 2-5 minutes to watch it fail.

  • This only happens with Ruby 2.0. When doing rvm use 1.9.3, everything works.
  • resque-web -s thin -F also fails with the same issue.
Owner

steveklabnik commented Apr 22, 2013

Hmm. We don't rely on any cexts, iirc, so it shouldn't' be our failure...

Contributor

tarcieri commented Apr 22, 2013

It looks like what's happening here is @vaharoni is trying to run the Sinatra version of Resque from within the Rails resque-web, and it's crashing in erubis? I'm confused

Thank you for your responses. @tarcieri I updated the issue description above with shorter steps to reproduce it. There is no need to use Rails to reproduce this issue. It seems that simply running resque-web -F after installing resque and redis gems with a minimal Ruby 2.0 environment fails. I tried this on two machines and reproduced it. Is this not happening in your environment? Perhaps it's a problematic gem version? Here's my gem list:
bigdecimal (1.2.0)
bundler (1.3.5)
daemons (1.1.9)
eventmachine (1.0.3)
io-console (0.4.2)
json (1.7.7)
minitest (4.3.2)
mono_logger (1.0.1)
multi_json (1.7.2)
psych (2.0.0)
rack (1.5.2)
rack-protection (1.5.0)
rake (10.0.4, 0.9.6)
rdoc (4.0.0)
redis (3.0.3)
redis-namespace (1.2.1)
resque (1.24.1)
rubygems-bundler (1.1.1)
rvm (1.11.3.7)
sinatra (1.4.2)
test-unit (2.0.0.0)
thin (1.5.1)
tilt (1.3.7)
vegas (0.1.11)

+1 on this issue with ruby 2.0.0-p0. I just ran "resque-web -F" and waited for 4 minutes and I got a segmentation fault as well. I am developing with OS X 10.8.2. Happy to share any environment info you feel could be relevant here.

*** LOCAL GEMS ***
bigdecimal (1.2.0)
bundler (1.3.5)
io-console (0.4.2)
json (1.7.7)
minitest (4.3.2)
mono_logger (1.0.1)
multi_json (1.7.2)
psych (2.0.0)
rack (1.5.2)
rack-protection (1.5.0)
rake (10.0.4, 0.9.6)
rdoc (4.0.0)
redis (3.0.4)
redis-namespace (1.2.1)
resque (1.24.1)
rubygems-bundler (1.1.1)
rvm (1.11.3.7)
sinatra (1.4.2)
test-unit (2.0.0.0)
tilt (1.3.7)
vegas (0.1.11)

moser commented May 21, 2013

I also experience this.
Did any of you find a workaround? Are there news from the maintainers side? How can we help fixing the problem?

I just tried it with Ruby 2.0.0-p195 and it seems to be working. @moser and @nicolasmeunier, can you please confirm it works for you too?

moser commented May 28, 2013

Will try tomorrow.

moser commented Jun 6, 2013

Seems to be working on 2.0.0-p195.

vaharoni commented Jun 7, 2013

Thanks! Closing the issue.

@vaharoni vaharoni closed this Jun 7, 2013

Owner

steveklabnik commented Jun 7, 2013

Yay!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment