Queue worker dies after a while.

[jericopulvera]

I have deployed a queue worker via AWS Copilot Backend Service. It does work but it consistently dies after a while.

[27-Dec-2022 05:54:37] NOTICE: Terminating ...
--
[27-Dec-2022 05:54:37] NOTICE: exiting, bye-bye!

Dockerfile

# syntax = docker/dockerfile:experimental

# Default to PHP 8.1, but we attempt to match
# Valid version values are PHP 7.4+
ARG PHP_VERSION=8.0
FROM serversideup/php:${PHP_VERSION}-fpm-nginx-v2.0.2:latest as base

# PHP_VERSION needs to be repeated here
# See https://docs.docker.com/engine/reference/builder/#understand-how-arg-and-from-interact
ARG PHP_VERSION

# See https://docs.docker.com/engine/reference/builder/#understand-how-arg-and-from-interact
RUN apt-get update && apt-get install -y \
    git curl zip unzip rsync ca-certificates vim htop cron \
    php${PHP_VERSION}-mysql php${PHP_VERSION}-bcmath \
    php${PHP_VERSION}-swoole php${PHP_VERSION}-xml php${PHP_VERSION}-mbstring \
    php${PHP_VERSION}-intl php${PHP_VERSION}-memcache php${PHP_VERSION}-memcached \
    jpegoptim optipng pngquant gifsicle webp \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /var/www/html
COPY . /var/www/html

RUN composer install --optimize-autoloader --no-dev \
    && mkdir -p storage/logs \
    && php artisan optimize:clear \
    && chown -R webuser:webgroup /var/www/html \
    && sed -i 's/protected \$proxies/protected \$proxies = "*"/g' app/Http/Middleware/TrustProxies.php \
    && echo "MAILTO=\"\"\n* * * * * webuser /usr/bin/php /var/www/html/artisan schedule:run" > /etc/cron.d/laravel \
    && rm -rf /etc/cont-init.d/* 

ENV SSL_MODE=off

CMD ["su", "webuser", "-c", "php artisan queue:work --tries=3"]

What outputs this Notice?

[jaydrogers]

This will happen if exit 1 is returned. I wonder if this is an application specific error in your RecalclateOrdersJob?

Unfortunately I won't be able to debug unless if I have repeatable steps to reproduce and diagnose the issue.

[jericopulvera]

I see then it must be on the container health check.

https://aws.github.io/copilot-cli/docs/manifest/backend-service/

image:
  # Docker build arguments. For additional overrides: https://aws.github.io/copilot-cli/docs/manifest/backend-service/#image-build
  build: Dockerfile.worker
  port: 80
  healthcheck:
    command: ["CMD-SHELL", "curl -f http://localhost:80/ping || exit 1"]
    retries: 10
    interval: 21s
    timeout: 20s
    start_period: 30s

That must mean that /ping do not respond when it's processing a job. Does moving to -fpm only without nginx be better?

[jaydrogers]

Are you referencing my container healthcheck? If so, my healthcheck is required for even the container to start: https://github.com/serversideup/docker-php/blob/main/src/fpm-nginx/etc/s6-overlay/s6-rc.d/nginx/data/check

So it would be weird for the S6 Overlay healthcheck to pass, then fail later on.

Are you sure something in your Laravel job isn't causing an exit code of 1?

[jericopulvera]

it turns out the issue was with using Planetscale database as the queue driver

[jaydrogers]

Thanks for sharing your solution!