Dead-simple task manager for "fire and forget" jobs. Has two interchangeable pools - processes and threads. The API of those two is the same, so you can play at will and figure out which one works better.
The only thing Chantier checks for is that the spun off tasks have completed. It also limits the number of tasks active at the same time. Your code will block until a slot becomes available for a task.
manager = Chantier::ProcessPool.new(slots = 4) # You can also use ThreadPool
jobs_hose.each_job do | job |
manager.fork_task do # this call will block until a slot becomes available
Churner.new(job).churn # this block runs in a subprocess
end
manager.still_running? # => most likely "true"
end
manager.block_until_complete! #=> Will block until all the subprocesses have terminated
If you have a finite Enumerable at hand you can also launch it into the
ProcessPool/ThreadPool, like so:
manager = Chantier::ThreadPool.new(slots = 4)
manager.map_fork(job_tickets) do | job_ticket | # job_tickets has to be an Enumerable
# this block will run in a thread
Churner.new(job_ticket).churn
...
end
Chantier does not provide any built-in IPC or inter-thread communication features - this should stimulate you to write your tasks without them having to do IPC in the first place.
Chantier implements what it calls FailurePolicies. A Policy is an object that works
like a counter for failed and successfully completed jobs. After each job, the policy
object will be asked whether limit_reached? is now true. If it is, calls to fork_task()
on the Pool using the failure policy will fail with an exception. There is a number of
standard FailurePolcies which can be applied to each specific Pool, by supplying it in
the failure_policy keyword argument.
For example, to stop the Pool from accepting jobs if more than half of the jobs fail
(either by raising an exception within their threads or by exiting the forked process with
a non-0 exit code):
fp = Chantier::FailurePolicies::Percentage.new(50)
pool = Chantier::ThreadPool.new(num_threads = 5, failure_policy: fp)
4.times { pool.fork { puts "All is well"} }
6.times { pool.fork { raise "Drat!"} } # Will only run 4 times and fail after
To allow only a specific number of failures within a time period wrap the policy in
a WithInterval object:
# Only allow 5 failures per second
counter = Chantier::FailurePolicies::Count.new(5)
fp = Chantier::FailurePolicies::WithinInterval.new(counter, within_seconds=3)
You can use those to set fine-grained failure conditions based on the runtime behavior of
the Pool you are using and job duration/failure rate. Chantier pools are made to run in
very long loops, sometimes indefinitely - so a FailurePolicy can be your best friend. You
can also bundle those policies together.
- Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
- Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
- Fork the project.
- Start a feature/bugfix branch.
- Commit and push until you are happy with your contribution.
- Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
- Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.
Copyright (c) 2014 Julik Tarkhanov. See LICENSE.txt for further details.