The default status of a service configuration |
Then, we test backup, alter, and restore Kong configurations.
To test, let's create a service definition.
$ curl -i -X POST --url http://localhost:8001/services/ --data 'name=v4' --data 'url=http://172.20.11.223:9099/services/v4'
Check the services via a browser (or a curl command, if you prefer it that way).
http://localhost:8001/services
Notice the timeouts are 60,000. This is 60,000 ms. i.e., 1 min by default. Let's increase it to 10 min, with Deck!
Steps:
1. Dump the configuration:
$ deck dump
This creates a kong.yaml in the current directory.
$ cat kong.yaml
_format_version: "1.1"
services:
- connect_timeout: 60000
host: 172.20.11.223
name: v4
path: /services/v4
port: 9099
protocol: http
read_timeout: 60000
retries: 5
write_timeout: 60000
Modify the connect_timeout, read_timeout, and write_timeout as 600000 (from their default value of 60000) via a text editor.
2. Check the changes:
$ deck diff
updating service v4 {
- "connect_timeout": 60000,
+ "connect_timeout": 600000,
- "created_at": 1.593636546e+09,
"host": "172.20.11.223",
"id": "5c25f92f-d3d0-486f-9a8e-9c406a130501",
"name": "v4",
"path": "/services/v4",
"port": 9099,
"protocol": "http",
- "read_timeout": 60000,
+ "read_timeout": 600000,
"retries": 5,
- "updated_at": 1.593636546e+09,
- "write_timeout": 60000
+ "write_timeout": 600000
}
Summary:
Created: 0
Updated: 1
Deleted: 0
3. Commit the changes:
$ deck sync
Confirm the changes are reflected, from the browser. Yep, it worked. Everything updated with zero downtime.
Timeouts are set to 10 mins (600,000) now, from 1 mins! |
No comments:
Post a Comment
You are welcome to provide your opinions in the comments. Spam comments and comments with random links will be deleted.