What's the right way to reload files/server in hypervel?

[foremtehan]

Ive put hypervel into supervisor like so:

[program:hypervel]
command=php artisan serve
stdout_logfile=/dev/stdout
stdout_logfile_maxbytes=0
stderr_logfile=/dev/stderr
stderr_logfile_maxbytes=0
autorestart=true
autostart=true
startretries=3
startsecs=5
stopwaitsecs=10

I do restart the whole process via: exec('supervisorctl restart hypervel'); In order to apply code changes in production

But it cause 4second down time, is there better approach ? i tried many ways to do it like using kill command but not working like supervisorctl command

It would be great if we had something like in octane: php artisan octane:reload

[albertcht]

Hi @foremtehan , there's an equivalent command in Hypervel: server:reload

[foremtehan]

It doesn't doing anything in my side:

this the server.php config:

<?php

declare(strict_types=1);

use App\Http\Kernel as HttpKernel;
use Hyperf\Server\Event;
use Hyperf\Server\Server;
use Swoole\Constant;

return [
    'mode' => SWOOLE_PROCESS,
    'servers' => [
        [
            'name' => 'http',
            'type' => Server::SERVER_HTTP,
            'host' => env('HTTP_SERVER_HOST', '0.0.0.0'),
            'port' => (int) env('HTTP_SERVER_PORT', 9501),
            'sock_type' => SWOOLE_SOCK_TCP,
            'callbacks' => [
                Event::ON_REQUEST => [HttpKernel::class, 'onRequest'],
            ],
        ],
    ],
    'kernels' => [
        'http' => HttpKernel::class,
    ],
    'settings' => [
        'document_root' => base_path('public'),
        'enable_static_handler' => true,
        Constant::OPTION_ENABLE_COROUTINE => true,
        Constant::OPTION_WORKER_NUM => env('SERVER_WORKERS_NUMBER', swoole_cpu_num()),
        Constant::OPTION_PID_FILE => base_path('runtime/hypervel.pid'),
        Constant::OPTION_OPEN_TCP_NODELAY => true,
        Constant::OPTION_MAX_COROUTINE => 100000,
        Constant::OPTION_OPEN_HTTP2_PROTOCOL => true,
        Constant::OPTION_MAX_REQUEST => 100000,
        Constant::OPTION_SOCKET_BUFFER_SIZE => 2 * 1024 * 1024,
        Constant::OPTION_BUFFER_OUTPUT_SIZE => 2 * 1024 * 1024,
    ],
    'callbacks' => [
        Event::ON_WORKER_START => [Hyperf\Framework\Bootstrap\WorkerStartCallback::class, 'onWorkerStart'],
        Event::ON_PIPE_MESSAGE => [Hyperf\Framework\Bootstrap\PipeMessageCallback::class, 'onPipeMessage'],
        Event::ON_WORKER_EXIT => [Hyperf\Framework\Bootstrap\WorkerExitCallback::class, 'onWorkerExit'],
    ],
];

Also i manually kill the master pid but it doesnt actually kill the process:

the posix function return true but don't know why it didn't do anything about it

this is the docker image from hyperf/hyperf:8.4-alpine-vedge-swoole-v6.0.2

[albertcht]

Unable to reproduce it

[foremtehan]

If the server:reload command is supposed to terminate the Master process (which had PID 57008, as seen in your image), then after you run it, why does the Master process still have the same PID? So it didn't kill it correctly, or am I wrong? Did you run an HTTP server and change the code to see if the changes were applied?

[foremtehan]

nevermind it's working now