Skip to content

KubePlus chart - 4.2.0#1441

Merged
devdattakulkarni merged 251 commits intomasterfrom
develop
Jul 15, 2025
Merged

KubePlus chart - 4.2.0#1441
devdattakulkarni merged 251 commits intomasterfrom
develop

Conversation

@devdattakulkarni
Copy link
Copy Markdown
Contributor

Removed setting of nodeName in the OpenAPISpec of the registered Kind

Also, modified provider-kubeconfig generator to include https
in the provided API server IP only if it is not already present.
In the original design, we were using the output of kubectl connections
to find out the Pods and then running metrics and logs on those Pods.
However, for workloads in which Pods get created at runtime, connections
is not able to find all the newly created workload Pods. This leads to
incomplete results for metrics and logs. A simple way to handle this is
to use all the pods in the namespace in which the application pods are running
for collecting logs and metrics. The runtime Pods are typically
created in the same namespace in which the application is running.
This will give accurate results (atleast more accurate than using the output
of kubect connections for the list of Pods).

Fixes: #1190
Added a field ('error') to the status object.
This is used by helmer to store any errors that
are encountered when performing helm upgrade.
It provides a UI for managing application instances.
But it does not run on the cluster.
The open-source equivalent is consumerUI, which is
deployed as part KubePlus deployment.
- Added link to Contributing guidelines. These guidelines
  contain pointer to setting up development environment
- Extracted Architecture in a separate section
- Updated link to CNCF Application definition section
- Added Getting Started section

Fixes: #1221
Updated component versions to use Ubuntu 22.04 as the base image
When collecting pods in a namespace for an app instance,
we want to ensure that the namespace is indeed for the
Kind whose app instance is being passed to the kubectl metrics command.
Without this Kind name check, kubeplus may gets wrong Pods if
app instances for two different kinds are given the same name.
- Removed Namespace parameter
- Updated logic of constructing the app url (Fixes: #1351)
In the kubectl metrics plugin, displaying individual Pod metrics
in the Prometheus output format
Removed adding nodeName to the spec of registered Kind
as we are no longer supporting Pod-level policies
Removed setting of nodeName in the OpenAPISpec of the registered Kind
@devdattakulkarni devdattakulkarni merged commit dffd5ab into master Jul 15, 2025
1 check passed
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

Successfully merging this pull request may close these issues.

1 participant