If the caller dies, the worker is returned to the pool, even if it's still busy

[rlipscombe]

I might be reading the source code wrongly, but it appears to me that if I do something like this:

x() ->
  W = poolboy:checkout(P, true),
  gen_server:call(W, Req),
  poolboy:checkin(P, W).

If x dies while inside gen_server:call (can it actually do that?), then W is returned to the pool, but it's still busy. That means that when someone checks it out later, their gen_server:call will be queued behind this one.

Is there any way to get around this?

[devinus]

Do you want the worker to continue it's work even when the caller has died?

[rlipscombe]

Well, that's a good question. Sometimes yes, sometimes no. In our particular case, usually no. Either the write is not that important, or it's a read, in which case it's wasted.

[rlipscombe]

I had two thoughts:

1. Get the worker to check itself back in. Requires adding some stuff to handle_call, which is awkward.
2. link the two processes. If the caller dies, the worker dies. Poolboy will create a replacement worker, no?

[devinus]

@rlipscombe You have many options, I don't find no. 1 awkward at all.

You can kill the worker as soon as it captures the caller EXITs. Many options are available to you, they all depend on your use case.