Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Properties renaming #513

Closed
thibaultcha opened this issue Aug 26, 2015 · 9 comments
Closed

Properties renaming #513

thibaultcha opened this issue Aug 26, 2015 · 9 comments
Milestone

Comments

@thibaultcha
Copy link
Member

0.5.0 was decided to introduce many breaking changes because we want to correct a few mistakes made along the way. So far, it will introduce those breaking changes:

The database migration is taken care of by the migration script written for the occasion, but the configuration file changes will need to be done manually by users. Everything is explained in UPDATE.yml.

First, it would really help if all those (code, script and documentation) were reviewed since this will be a very important steps for users. @thefosk @Tieske @ahmadnassri


The question now is that we had decided on more things to rename, should we do it now or not?

  • The hosts property in the configuration file to contact_points, as generally used for Cassandra clusters.
  • An API's public_dns to inbound_dns.
  • An API's path to inbound_path.
  • An API's strip_path to strip_inboud_path.
  • An API's target_url property to upstream_url or outbound_url.
  • plugins_configurations (the table and name of the entity itself (in docs etc) to plugins (even those they are indeed configurations of the said plugins.

Please consider each of them individually, and wether or not it is worth renaming it. Some are easy, some will require heavy work on the migration script. The thing is now is a really good time to proceed to those changes to avoid having other painful updates for users in the future.

@thefosk @Tieske @ahmadnassri

@thibaultcha thibaultcha added this to the 0.5.0 milestone Aug 26, 2015
@sonicaghi
Copy link
Member

contact_points
inbound_dns
upstream_url
plugins

@thibaultcha
Copy link
Member Author

Also in plugins_configurations (or plugins now), renaming value to config seems like a priority. All those updates will require huge work on the documentation, yet alone the code itself and the migration script.

@sonicaghi
Copy link
Member

better sooner rather than later.

@thibaultcha
Copy link
Member Author

What about inboud_host instead of inboud_dns? Since we are talking about the Host header with this resolver?

@thibaultcha thibaultcha mentioned this issue Aug 28, 2015
19 tasks
@sonicaghi
Copy link
Member

most of the users get confused only on dns and target url. i would prioritize those.

@thibaultcha
Copy link
Member Author

Yeah hence why I suggest inboud_host instead of inboud_dns, but I'm not sure about it. Would like some other opinions.

@sonicaghi
Copy link
Member

not sure. good to have other point of view. @ahmadnassri @Tieske @thefosk

@subnetmarco
Copy link
Member

+1 for inbound_dns and upstream_url

All of this changes will delay 0.5.0 - but as @thibaultcha we have to do them sooner or later, so let's do them now.

All those updates will require huge work on the documentation

Massive changes indeed.

@thibaultcha
Copy link
Member Author

So far all those updated have been done in the code and documented in CHANGELOG.md and UPDATE.md in #518. The update instructions allow users to migrate and update to 0.5.0 without downtime. Developers (@thefosk @Tieske) will have to run a few commands as describe in the PR.

Reviews and testing is highly appreciated, especially the migration script but also the state of Kong once migrated.

A typical use case would be:

  • Kong in 0.4.2 with a some APIs and a few plugins that were renamed (eg: rate-limiting + key-auth)
  • Follow the UPDATE.yml instructions, making sure there is no downtime between steps
  • Run Kong in 0.5.0 and make sure it has the same behaviour as 0.4.2

The only work left is udpating the website now (and drop the support for the old rate-limiting schema @thefosk).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants