[0.9.0] Pool size metrics

What's Changed

Pool size metrics (using telemetry) were added. How to use them is described here: Working with metrics guide.

To enable pool size metrics, you need to set the pool_size_metrics parameter to true on the pool initialization:

children = [
  {Poolex, 
    pool_id: :worker_pool,
    worker_module: SomeWorker,
    workers_count: 5,
    pool_size_metrics: true}
]

[0.8.0] Merge `run` interfaces

What's Changed

Breaking changes

• Option :timeout renamed to :checkout_timeout.

  • Reason: This option configures only the waiting time for worker from the pool, not the task's work time. This naming should be more understandable on the call site.

    # Before
    Poolex.run(:my_awesome_pool, fn worker -> some_work(worker) end, timeout: 10_000)
    
    # After
    Poolex.run(:my_awesome_pool, fn worker -> some_work(worker) end, checkout_timeout: 10_000)

• Poolex.run/3 returns tuple {:error, :checkout_timeout} instead of :all_workers_are_busy.

  • Reason: It is easier to understand the uniform format of the response from the function: {:ok, result} or {:error, reason}.

• Poolex.caller() type replaced with struct defined in Poolex.Caller.t().

  • Reason: We need to save unique caller references.

• Poolex.run!/3 was removed in favor of Poolex.run/3. The new unified function returns {:ok, result} or {:error, :checkout_timeout} and not handles runtime errors anymore.

  • Reason: We should not catch errors in the caller process. The caller process itself must choose how to handle exceptions and exit signals.

Fixed

• Fixed a bug when workers get stuck in busy status after checkout timeout.

[0.7.6] Handling caller's death

What's Changed

Fixed

• Fixed a bug with workers stuck in busy status. Added caller monitoring. #56

[0.7.5] Fixed breaking bug

What's Changed

Fixed

• Fixed breaking bug with releasing workers in #55. Version 0.7.4 retired.

[0.7.4] Fixed processing of restarting a fallen worker

What's Changed

Fixed

• Fixed a bug where a restarted worker was not automatically dispatched to pending callers #54.

Changed

• Upgraded ex_doc from 0.29.4 to 0.30.3

[0.7.3] Fixed bugs

What's Changed

Fixed

• Fixed a bug with an incorrect number of running workers when specifying a zero or negative number in the workers_count parameter by @alexcastano in #49

[0.7.2] Improve the shutdown process

What's Changed

Fixed

• Trap exit to shutdown correctly by @alexcastano in #46

Changed

• Implementation settings are stored in the pool process state instead of the ETS table. This makes the testing process easier and removes unnecessary entities.

New Contributors

• @alexcastano made their first contribution in #46

[0.7.1] Improve the shutdown process

Fixed

• Fix the shutdown process: stop workers before the pool. Issue: #44.

[0.7.0] `FIFO` workers implementation

Added

• Added FIFO worker's implementation. About implementations. May be needed if you used the :fifo strategy in poolboy.

[0.6.1] Documentation improvements

Documentation updates

• Refactored some custom types and added some typedocs.
• Added diff syntax highlighting support.
• Updated guides:
  • The pages were merged into a group called Guides.
  • Tried to refresh some pages with cheatmd.