sulu / sulu-docker Goto Github PK

The reason being that the target folder is not empty.

Hey,
I think it could be nice to have a cli tool in order to setup project, run our stack, manage sulu app...

For instance here is a list of command the cli tool will handle:

• sulu-cli install: execute composer create-project "sulu/sulu-minimal" app and other stuff to install a sulu new app
• sulu-cli start: Detect you OS and launch the correct stack (docker-sync + docker-compose.{os})
• sulu-cli stop: Stop the current running stack
• sulu-cli exec sulu:build dev --destroy: execute a Sulu command
• sulu-cli fix-permission: fix permissions if it's needed
• ...

I think we can develop this sulu-cli tool in Python or NodeJs or Golang.

What do you think about this?

In my opinion, the worst decision for #51 is to replace nginx+fpm with apache.
I would suggest keeping the previous configuration ideally here and create apache-based environment as a separate project / separate docker-compose file here.

Is apache still a thing in php world in 2020?

I think the correct domain should be sulu.localhost, no?

At this time the mysql image does not support ARM processors. Or in other words, it does not work on the new M1 from Apple .
I tested the set up on an M1 once with Maria DB.
With Maria DB it works and seems to run the same way as with mysql.

So what do you think about changing the setup from mysql to maria db for supporting the ARM stack with no disadvantage for other machines.

Proposed solution

db:
    image: mariadb:10.5.8
    command: --default-authentication-plugin=mysql_native_password
    ports:
      - ${PORT_MYSQL}:3306
    volumes:
      - ./data/mysql:/var/lib/mysql
    environment:
      MYSQL_ALLOW_EMPTY_PASSWORD: 'yes'
      MYSQL_DATABASE: ${MYSQL_DATABASE}
      MYSQL_USER: ${MYSQL_USER}
      MYSQL_PASSWORD: ${MYSQL_PASSWORD}

Running docker-compose up after a fresh git pull results in an elastisearch error and failure:

> [sulu-docker] ⁁(master)% docker-compose up
[+] Running 0/2
⠿ elasticsearch Error
⠴ mysql Pulling
no matching manifest for linux/arm64/v8 in the manifest list entries
> [sulu-docker] ⁁(master)%

The Login to Kibana fails because of this message:

Login is currently disabled because the license could not be determined. Please check that
Elasticsearch has the X-Pack plugin installed and is reachable, then refresh this page.

I googled a bit for it and it seems https://github.com/elastic/kibana-docker is abandoned and the docker images by Elastic Inc. should be used now.

Hi,

I tried to launch your sulu docker repo on my dev computer.

During first launch, I received this issue for elasticsearch image like explain here :
https://stackoverflow.com/questions/65668188/elastic-search-accessdeniedexception-usr-share-elasticsearch-data-nodes-0-a

Like in post, I've to change default volume filepath in docker-compose :

volumes:
  - ./data/elasticsearch:/usr/share/elasticsearch/data

with

volumes:
  - ./data/elasticsearch:/**var**/**lib**/elasticsearch/data

Works well for me after this modification.

I let you modify in your template if you think it's necessary.

Thanks for your great job with sulu 👍

Regards,
B.

Hey,

I think it could be nice to have documentation on how to run the stack as daemon:

https://github.com/EugenMayer/docker-sync/wiki/2.3-daemon-mode

START

#docker-sync-daemon start # not needed anymore
docker-sync start
docker-compose up -d

STOP

docker-compose stop
docker-sync stop

REMOVE

docker-compose down
docker-sync stop

Installation of Fish Shell failed:

RUN apt-get install software-properties-common -y
RUN apt-add-repository ppa:fish-shell/release-2
RUN apt-get update
RUN apt-get install fish

# After set the user
ENV SHELL /usr/bin/fish

Problem with the ppa occurs (can't found the fish-shell release2).

Failed to fetch http://ppa.launchpad.net/fish-shell/release-2/ubuntu/dists/jessie/main/binary-amd64/Packages  404  Not Found

And get it to work with remote enable

example content of var/logs/nginx/project_error.log file:

2018/10/02 13:32:02 [error] 9#9: *49 upstream sent too big header while reading response header from upstream, client: 192.168.0.1, server: app.lo, request: "GET / HTTP/1.1", upstream: "fastcgi://192.168.0.7:9000", host: "app.lo:10080"
2018/10/02 13:34:26 [error] 8#8: *1 upstream sent too big header while reading response header from upstream, client: 192.168.0.1, server: app.lo, request: "GET / HTTP/1.1", upstream: "fastcgi://192.168.0.4:9000", host: "app.lo:10080"

Hi guys,

container elasticsearch exist by itself every once in a while.. No log for errors, something other is problem...

Config: OSX, 16GB RAM, 2.8 GHz Quad-Core Intel Core i7
Docker desktop: 3.3.3, Engine 20.10.6, Compose 1.29.1

add the domain and the ports and add some documentation for this file

line 2-11 of the following code-snippet are ineffective right now:

/var
!/var/data/elasticsearch/.gitkeep
!/var/data/indexes/.gitkeep
!/var/data/media/var/.gitkeep
!/var/data/media/web/.gitkeep
!/var/data/mysql/.gitkeep
!/var/data/php/.gitkeep
!/var/logs/project/.gitkeep
!/var/logs/nginx/.gitkeep
!/var/tmp/bootstrap.php.cache
!/var/tmp/SymfonyRequirements.php

this is because the parent directory (or an ancestor-directory) of the files is ignored:
https://stackoverflow.com/questions/26189082/track-file-inside-ignored-directory

Hi,

if I set the PHP_USER_ID in the env file to the id of the host system user, the php container crashes with the following errors:

php_1    | AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 192.168.208.3. Set the 'ServerName' directive globally to suppress this message
php_1    | (13)Permission denied: AH00072: make_sock: could not bind to address 0.0.0.0:80
php_1    | no listening sockets available, shutting down
php_1    | AH00015: Unable to open logs

If I set PHP_USER_ID to root, the containers are created without any issue but I obviously can't change files in the project folder without sudo.

I'm on Ubuntu 20.10, VS Code with Docker version 19.03.8. Any help is welcome.

Thanks.

The Initialize Sulu part of the readme does not work with the current dev-branch (and I assume Version 2 of Sulu). Since the readme specifies to use the dev-branch during install, this should be changed.

Hey,

I would like to use this on my Linux Mint (Ubuntu based) but there is some issue with unison sync on linux. Add to that the native linux sync is already performant.

If I remove the sync_strategy in docker-sync.yml https://github.com/wachterjohannes/sulu-docker/blob/master/docker-sync.yml#L11 this will use the native linux sync and works for linux too!

Can you test if on MacOS and Windows this line is necessary? According to docker-sync documentation by default on MacOS and Windows, it will use unison.

Thanks!

Note: another solution will be to create several docker-compose.{os}.yml