-
Notifications
You must be signed in to change notification settings - Fork 30
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
improve doc #64
improve doc #64
Conversation
docs/en/setup/Configuration-File.md
Outdated
@@ -53,6 +53,8 @@ The `KinD` environment follow these steps: | |||
1. Wait until all steps are finished and all services are ready with the timeout(second). | |||
1. Expose all resource ports for host access. | |||
|
|||
Notice, when you create resources such as webhooks, they need time to set up. If you get some error, you can try `sleep` to wait for the resources established. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is there any command kubectl
that can wait for webhook ready?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I haven't found it yet.
docs/en/setup/Configuration-File.md
Outdated
@@ -150,6 +152,12 @@ trigger: | |||
|
|||
The Trigger executed successfully at least once, after success, the next stage could be continued. Otherwise, there is an error and exit. | |||
|
|||
If you get the following error message when using the trigger, You should check whether the port is useful. A possible reason may be that the exporting service needs time to set up, so you can add a larger interval to avoid the error. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Did you use the wait
setting to wait for the resource ready?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, the resouce is ready before port-forward, I can use curl to get data.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Does the e2e framework support NodePort
or LoadBalancer
? They are the more canonical ways to expose services than port-forward
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@kezhenxu94 How does the swctl
connect to UI in current e2e cases?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We have settings to expose the ports in cluster
skywalking-infra-e2e/docs/en/setup/Configuration-File.md
Lines 43 to 46 in 5adf908
expose-ports: # Expose resource for host access | |
- namespace: # The resource namespace | |
resource: # The resource name, such as `pod/foo` or `service/foo` | |
port: # Want to expose port from resource |
And you can reference via ${<resourceType>_<resourceName>_host}:${<resourceType>_<resourceName>_<port>}
, such as ${service_skywalking_ui_host}:${service_skywalking_ui_80}
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @zhenxu
And FWIK, the item is not special to the e2e framework. The user should ensure a service accessible before accessing it. It's common sense more than a FAQ which the e2e user should care about.
These added words seem like a FAQ. How about adding a FAQ section rather than this kind of If there are other ways, all these should be considered as a FAQ. |
Ok, I will add them to the FAQ section. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The health checker was introduced by kubernetes-sigs/controller-runtime#1588 which is not part of swck
's dependencies though.
@dashanji Once swck
upgrades to the latest controller-runtime, we could check the webhook server's health endpoint to ensure it. I hope it will happen in the next release.
If this exists, then FAQ should not be added. Instead, this is a potential feature for SWCK. |
Yep, to ensure webhook server before applying CRs is the e2e framework users' job. In another word, it's the behavior of kubernetes. If the e2e framework declared it's built on it, the user should accept this result. Based on that, this tip should be submitted to the kubernetes communication, not here 😄 . |
Yes,the two problems are to avoid users stepping on the same pit, how about adding them to the discussion's Q&A? |
Please be clear what bug it is, and how do you fix. |
When I use
Then the error occurred like follows. This bug is the yaml cannot recognize |
Could you submit a new pull request? Which make sure the commit logs and PR content are clear. As a new bug fix should be added in the 1.1.0 in https://github.com/apache/skywalking-infra-e2e/blob/main/CHANGES.md |
OK. |
Sorry, this is not a bug, I forgot to add |
The PR is to improve the document to avoid repeated pitfalls. When using the infra e2e to test skywalking-swck, I encountered some problems and solved them.
Problem 1
When the controller's pod is ready, the controller's webhook may not be ready.
Solution
Use the command
sleep 30s
to wait for the webhook establishing.Problem 2
If the interval is small, we may get the error.
Solution
Add a larger interval such as 30.