Closed lucascs closed 14 years ago
This sort of thing is highly specific to your action, or individual pieces of your action, and shouldn't be handled at the broad level of the job.
If you'd like to time out an action, just wrap the relevant code in a timeout block.
http://ruby-doc.org/stdlib/libdoc/timeout/rdoc/index.html
The Timeout::Error
will propagate back up through CloudCrowd, and will cause the work unit and the job to be marked as "failed", and then call your application at the callback_url
to tell it. Your application can take it from there.
Sound alright?
Sounds good to me =)
Thanks
Hi,
It should be possible to specify a timeout for jobs, when adding it:
so if an input takes more than 60 seconds to process, it is canceled...
and on merge, this input should be either ignored or have an error/default value.
Thanks