2

I'm currently trying to link the host's database with php-fpm inside the docker container.

However, I have to ask you a question because the area is not working properly.

I didn't use php offical base image. so, i created a docker file as shown below and built an image.

RUN apt-get update --fix-missing && \ 
    apt-get -qq -y install aptitude software-properties-common 
RUN aptitude purge `dpkg -l | grep php| awk '{print $3}' |tr "\n" " "` 
RUN add-apt-repository ppa:ondrej/php
RUN apt-get update

# install php7 fpm
RUN apt-get -qq -y install build-essential wget && \
    apt-get -qq -y install php-pear \
        php7.0-fpm php7.0-cli php7.0-gd php7.0-curl php7.0-mbstring php7.0-xml php7.0-simplexml \ 
        php7.0-iconv php7.0-json php7.0-mcrypt libapache2-mod-php7.0 php7.0-mysql 

# clean up
RUN apt-get -qq -y --purge remove git subversion

# remove package
RUN apt-get -qq -y --purge remove build-essential wget && \
    apt-get -qq -y autoremove && \
    apt-get clean && \
    rm -rf /tmp/src 

# add config
ADD php.ini /etc/php/7.0/fpm/php.ini
ADD php-fpm.conf /etc/php/7.0/fpm/php-fpm.conf
ADD www.conf /etc/php/7.0/fpm/pool.d/www.conf

# log configure
RUN mkdir -p /var/log/php && \
    touch /var/log/php/fpm-php.www.log && \
    chown -R www-data:www-data /var/log/php/fpm-php.www.log
RUN ln -sf /dev/stderr /var/log/php/fpm-php.www.log 


# RUN
COPY entrypoint.sh /usr/local/bin/
RUN chmod +x /usr/local/bin/entrypoint.sh && \
    ln -s usr/local/bin/entrypoint.sh /entrypoint.sh
ENTRYPOINT ["entrypoint.sh"]
VOLUME /var/run/php

CMD ["/usr/sbin/php-fpm7.0", "-F"]

I originally built an image of the website with php5-fpm instead of php7-fpm, and there was no operational problem with php5-fpm. But, there was a problem when I uploaded it to version 7.

Thinking that this might be a problem for the database, I started checking for possible problems with the database.

  1. Verify that the bind-address of the host's database is set to 0.0.0.0.0. This part is thought to cause serious security problems, but we set it up because we're in a hurry to operate.
  2. Created 'user'@'%' remote access enabled.
  3. Give database permissions to the user.

However, there was an error 'Connection time out'. Then, I set the log level to a maximum of 9 to start checking all logs.

At that time, I saw the log below.

2019-05-07 1:10:34 57 [Warning] Aborted connection 57 to db: 'unconnected' user: 'unauthenticated' host: 'localhost' (CLOSE_CONNECTION)

I did googling and found that you were solving this problem by adding '--skip-name-reslove' from mysql, but i couldn't resolve the problem.

I uploaded my docker-compose source file just in case.

version: '3'
services:
  nginx:
    build:
      context: ./nnet-nginx/
    restart: always
    image: seansin/nnet-nginx:latest
    depends_on:
      - app
    volumes:
      - app:/var/www/html/zbxe:rw
      - php:/var/run/php
      - log:/var/log/php
    network_mode: "bridge"
    ports:
      - 80:80
  app:
    build:
      context: ./nnet-app-php7/
    restart: always
    image: seansin/nnet-app-php7:latest
    environment:
      - XE_DBCONFIG_FILE=${XE_DBCONFIG_FILE}
      - DB_USER=${DB_USER}
      - DB_PORT=${DB_PORT}
      - DB_PASS=${DB_PASS}
      - DB_NAME=${DB_NAME}
    volumes:
      - app:/var/www/html/zbxe:rw
      - php:/var/run/php
      - log:/var/log/php 
volumes:
  app:
  php: 
  log:

Also, my development environment is two things.

  1. Windows 10 / MariaDB 10.3 (host's DB) / Docker-Engine 18.09.1
  2. MacOS Mojave / MariaDB 10.3 (host's DB) / Docker-Engine 18.09.1

Inside the docker container, you are trying to connect using the address 'host.docker.internal'.

Linux uses the command '/sbin/ip route|awk '/default/ {print $3 }'

The same problem occurs in both development environments. What's the problem?

Liquid.Bear
  • 333
  • 6
  • 21

0 Answers0