[🐛 BUG]: RR doesn't respect http.pool.allocate_timeout

[Warxcell]

No duplicates 🥲.

• I have searched for a similar issue in our bug tracker and didn't find any solutions.

What happened?

I have

http:
    address: ${HOST}:${PORT}
    pool:
        debug: ${RR_DEBUG}
        num_workers: ${RR_WORKERS}
        max_jobs: 0
        allocate_timeout: 120s

but it tells

php 2022-07-25T13:53:05.610Z    ERROR    http            execute    {"start": "2022-07-25T13:52:05.602Z", "elapsed": "1m0.007540715s", "error": "failed to spawn a worker, possible reasons: https://roadrunner.dev/docs/known-issues-allocate-timeout/2.x/en"}

elapsed is 1m, so it doesn't respect 120s (2m)

and there is a typo in the docs: https://roadrunner.dev/docs/http-available-middleware/2.x/en

it should be http.pool.allocate_timeout not pool.allocate_timeout.

Version (rr --version)

2.11.0-beta.1

Relevant log output

php 2022-07-25T13:53:05.610Z    ERROR    http            execute    {"start": "2022-07-25T13:52:05.602Z", "elapsed": "1m0.007540715s", "error": "failed to spawn a worker, possible reasons: https://roadrunner.dev/docs/known-issues-allocate-timeout/2.x/en"}

[rustatian]

Hey @Warxcell 👋🏻
The workers' pool is not an http plugin property.
But allocate_timeout is the property of the pool. It can be jobs.pool.allocate_timeout, http.pool..., grpc.pool..., etc. Thus in the docs, we used a more general term, pool.allocate_timeout.

[rustatian]

Regarding the first part of the issue, about allocate_timeout not taking into account, I'll have a look; Thanks.

[rustatian]

@Warxcell Is this all configuration? Might be you have a supervisor section?

[Warxcell]

This is whole conf:

version: "2.7"

server:
    command: "php /app/server.php"
    relay: "unix:///tmp/rr.sock"

http:
    address: ${HOST}:${PORT}
    pool:
        debug: ${RR_DEBUG}
        num_workers: ${RR_WORKERS}
        max_jobs: 0
        allocate_timeout: 120s
    middleware: [ "http_metrics", "static", "sendfile", "gzip" ]
    # Settings for "static" middleware (docs: https://roadrunner.dev/docs/http-static).
    static:
        dir: "/app/public/"
        forbid: [ ".php" ]
        calculate_etag: false
        weak: false

rpc:
    listen: tcp://127.0.0.1:6001

logs:
    mode: ${RR_LOGS}
    output: stderr

metrics:
    address: 0.0.0.0:8080

[rustatian]

Could you please try to use relay: pipes?

[Warxcell]

Could you please try to use relay: pipes?

Yeah, with pipes it seems to work.

[rustatian]

Nice. I fixed the issue for the sockets/tcp. It will be later today in the beta.2.