Modify nginx proxy settings in Elastic Beanstalk options

nginx.config

files:
  "/tmp/proxy.conf":
    mode: "000644"
    owner: root
    group: root
    content: |

      proxy_send_timeout 600;
      proxy_read_timeout 600;
      send_timeout       600;
container_commands:
  00-add-config:
    command: cat /tmp/proxy.conf >> /etc/nginx/conf.d/00_elastic_beanstalk_proxy.conf
  01-restart-nginx:
    command: /sbin/service nginx restart

Hmm, so when I use this as is, the file is indeed modified, but I get an error on the nginx restart:

Command failed on instance. Return code: 137 Output: Stopping nginx: /sbin/service: line 66: 23734 Killed env -i PATH="$PATH" TERM="$TERM" "${SERVICEDIR}/${SERVICE}" ${OPTIONS}.

If I either change nginx restart to nginx reload, or add an "ignoreErrors" to the restart-nginx command, then the deploy passes, but the configuration file is not modified. This seems to imply that the container_commands is running too early.

Any clue?

EDIT: Ok, managed to solve it with a post deploy folder hook. So my script looks like this (put my nginx cofig file in my app root):

files:
  "/opt/elasticbeanstalk/hooks/appdeploy/post/90_copy_nginx_config.sh":
    mode: "000755"
    owner: root
    group: root
    content: |
        #!/usr/bin/env bash
        cat /var/app/current/00_elastic_beanstalk_proxy.conf > /etc/nginx/conf.d/00_elastic_beanstalk_proxy.conf
        /sbin/service nginx reload

According to the doc https://docs.aws.amazon.com/elasticbeanstalk/latest/dg/platforms-linux-extend.html
Now we should place the nginx conf files in .platform/nginx

|-- web.jar
|-- .ebextensions/
|-- .platform/
    |-- nginx/                # Proxy configuration
    |   |-- nginx.conf
    |   `-- conf.d/
    |       `-- custom.conf

This is actually a lot easier because aws will copy nginx conf files and reload nginx process automaticlly so that we don't need to write any command.

Tested at Sept 2021.