Docker compose links section can't support

[xianlubird]

About compose introduce links https://docs.docker.com/compose/compose-file/compose-file-v2/#links .
I have a compose like this:

web:
  image: registry.aliyuncs.com/acs-sample/wordpress:4.5
  ports:
    - '80'
  environment:
    WORDPRESS_AUTH_KEY: changeme
    WORDPRESS_SECURE_AUTH_KEY: changeme
    WORDPRESS_LOGGED_IN_KEY: changeme
    WORDPRESS_NONCE_KEY: changeme
    WORDPRESS_AUTH_SALT: changeme
    WORDPRESS_SECURE_AUTH_SALT: changeme
    WORDPRESS_LOGGED_IN_SALT: changeme
    WORDPRESS_NONCE_SALT: changeme
    WORDPRESS_NONCE_AA: changeme
  restart: always
  links:
    - 'db:mysql'
db:
  image: registry.aliyuncs.com/acs-sample/mysql:5.7
  environment:
    MYSQL_ROOT_PASSWORD: password
  restart: always
  labels:
    aliyun.logs: /var/log/mysql

kompose will convert db service to db deployment, but the wordpress deplotment will connect db by mysql hostname. So wordpress pods will call that and it can't connect the db database. Because the links can't convert to k8s yaml.

I have two kinds suggestions:

1. kompose convert db service to mysql service, just rename deployment name for links, and the new service name is mysql, the wordpress can connect mysql by service name corrently.

2. print warning or error message if compose has links section which alias name is different from service name.

   Can anyone give me some suggestion, and I will create a PR to fix this, Thanks.

[surajnarwade]

I think this will be bit confusing since name of deployment and service will change,

kompose convert db service to mysql service, just rename deployment name for links, and the new service name is mysql, the wordpress can connect mysql by service name corrently.

This second option, I think will be better,

print warning or error message if compose has links section which alias name is different from service name.

@cdrage thought here ?

[cdrage]

Yeah, we should add the second option since the links key isn't able to be effectively converted to Kubernetes. We should clarify more about why it doesn't work and what the alternatives are.

Thanks @xianlubird for pointing this out!

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[surajnarwade]

/remove-lifecycle stale

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale