## What this PR does
Replaces Helm lookup functions with FluxCD valuesFrom mechanism for
passing configuration to HelmReleases. This provides cleaner config
propagation and eliminates the need for force reconcile controllers.
### Changes:
**Platform/Tenant charts:**
- Add Secret `cozystack-values` creation in platform chart (for
tenant-root and system namespaces)
- Add Secret `cozystack-values` creation in tenant chart (for child
namespaces)
**cozystack-api:**
- Add `valuesFrom` references to HelmRelease when creating applications
- Filter keys starting with `_` when returning Application specs
- Validate that user values don't contain `_` prefixed keys
**cozystack-controller:**
- Add validation that HelmRelease contains correct valuesFrom
configuration
- Remove `CozystackConfigReconciler` (no longer needed)
- Remove `TenantHelmReconciler` (no longer needed)
**Helm charts (40+ files):**
- Add helper templates in cozy-lib for `_cluster`/`_namespace` access
- Replace ConfigMap lookups with `.Values._cluster.*`
- Replace Namespace annotation lookups with `.Values._namespace.*`
### Architecture:
```
Secret cozystack-values (in each namespace)
├── _cluster: YAML with data from ConfigMaps (cozystack, cozystack-branding, cozystack-scheduling)
└── _namespace: YAML with namespace service references (etcd, host, ingress, monitoring, seaweedfs)
HelmRelease
└── spec.valuesFrom:
├── Secret/cozystack-values → _namespace → .Values._namespace
└── Secret/cozystack-values → _cluster → .Values._cluster
```
### Release note
```release-note
[platform] Replace Helm lookup functions with FluxCD valuesFrom mechanism for configuration propagation
```
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Helm releases and namespaces now source centralized cluster/namespace
configuration via a new secret (cozystack-values), and many templates
read values from chart-provided _cluster/_namespace entries.
* **Bug Fixes**
* API now rejects application specs containing reserved keys prefixed
with "_" to prevent invalid configurations.
* **Refactor**
* Two background reconciler controllers were removed from startup,
simplifying controller initialization.
<sub>✏️ Tip: You can customize this high-level summary in your review
settings.</sub>
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
This change is extracted from
- https://github.com/cozystack/cozystack/pull/1641
and reworked to work standalone
## What this PR does
This PR extracts changes from the
https://github.com/cozystack/cozystack/pull/1641. It adds application
metadata labels to HelmReleases and updates the filtering mechanism to
use labels instead of chart/sourceRef matching.
Changes:
- Add three application metadata labels
(`apps.cozystack.io/application.kind`,
`apps.cozystack.io/application.group`,
`apps.cozystack.io/application.name`) when creating/updating HelmRelease
via Cozystack-API
- Replace `shouldIncludeHelmRelease` filtering with label-based
filtering in Get, List, and Update methods
- Always add kind and group label requirements in List for precise
filtering
- Update CozystackResourceDefinitionController to watch only
HelmReleases with `cozystack.io/ui=true` label
- Update LineageControllerWebhook to extract metadata directly from
HelmRelease labels instead of using chart mapping configuration
- Add functionality to update HelmRelease chart from
CozystackResourceDefinition using label selectors
### Release note
```release-note
[registry] Add application labels and update filtering mechanism to use label-based filtering instead of chart/sourceRef matching
```
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added application metadata labels (kind, group, name) and exported
label keys for HelmRelease identification.
* New reconciler keeps HelmRelease charts in sync with application CRDs.
* **Refactor**
* Mapping, listing and selection moved to label-driven logic;
reconciliation responsibilities split so core reconciler focuses on
restart/debounce while a separate reconciler updates HelmReleases.
* **Chores**
* Migration script to backfill application labels on existing
HelmReleases.
<sub>✏️ Tip: You can customize this high-level summary in your review
settings.</sub>
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## What this PR does
Fixes the logic for generating CustomFormsOverride schema to properly
nest properties under `spec.properties` instead of directly under
`properties`.
Changes:
- Updated `buildMultilineStringSchema` to check for `spec` property in
OpenAPI schema
- Process `spec.properties` instead of root `properties`
- Create schema structure as `schema.properties.spec.properties.*`
- Updated tests to reflect the new nested structure
### Release note
```release-note
[dashboard] Fix CustomFormsOverride schema generation to nest properties under spec.properties
```
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Refactor**
* Updated custom form schema generation to organize form fields within a
nested structure for improved organization.
* **Tests**
* Expanded test coverage to validate the new schema organization and
field type handling across nested levels.
<sub>✏️ Tip: You can customize this high-level summary in your review
settings.</sub>
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## What this PR does
To use the new VPC feature, users need to pass the subnet ID to the VMs
they wish to launch in a given VPC/subnet. As the dashboard cannot
compute the subnet ID in the same manner as the Helm template, a helper
configmap is created, containing the details of the subnets attached to
a given VPC. This configmap is queried by the dashboard frontend to
render those details to the user.
### Release note
```release-note
[vpc,dashboard] Expose subnet details to end-user in the dashboard.
```
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
## What this PR does
In a previous patch (#1555) the reconciliation loop for the OpenAPI UI
resources was accidentally removed. This patch reintroduces a separate
controller, which handles updates to CozystackResourceDefinitions and
creates, updates, or deletes the dashboard's custom resources.
### Release note
```release-note
[dashboard] Reintroduce the accidentally removed reconciler that
autoconfigures custom dashboard resources for the OpenAPI UI.
```
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
This patch drops the custom caching of the Cozystack resource
definitions in favor of the informer cache and adds a flag to the
Cozystack controller to select, whether it restarts the cozystack-api
deployment or the cozystack-api daemonset.
As with the new default behavior of using a local endpoint for the k8s
API by the lineage webhook and the Cozystack API, the Cozystack
controller now also defaults to restarting a Cozystack API DaemonSet
instead of a Deployment. To revert to the old behavior, disable the
local k8s API endpoint on the webhook and cozystack API and set the
`cozystackController.cozystackAPIKind` value in the Cozystack controller
system Helm chart to "Deployment".
```release-note
[controller] Use informer cache instead of the older bespoke
implementation and add support for running the Cozystack API as a
DaemonSet.
```
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
Fix an incorrect JSON path that prevented Service LoadBalancer IPs from
rendering in the table view.
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
[cozystack-controller] Introduce new dashboard-controller
[dashboard] Introduce new dashboard based on openapi-ui
Co-authored-by: kklinch0 <kklinch0@gmail.com>
Signed-off-by: kklinch0 <kklinch0@gmail.com>
Signed-off-by: Andrei Kvapil <kvapss@gmail.com>
This patch implements external monitoring of the Kube-OVN cluster. A new
reconciler timed to run its reconcile loop at a fixed interval execs
into the ovn-central pods and collects their cluster info. If the
members' opinions about the cluster disagree, an alert is raised. Other
issues with the distributed consensus are also highlighted.
```release-note
[kubeovn,cozystack-controller] Implement the KubeOVN plunger, an
external monitoring agent for the ovn-central cluster.
```
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
Workloads tracking an object undergoing deletion can be reconciled when
the object is marked for deletion, but is not yet removed. After the
object is deleted, there is no event to trigger another reconciliation
of the workload and it might never get deleted until a global reconcile
happens or the controller is restarted. This patch ensures they are
requeued in the reconciliation loop.
<!-- Thank you for making a contribution! Here are some tips for you:
- Start the PR title with the [label] of Cozystack component:
- For system components: [platform], [system], [linstor], [cilium],
[kube-ovn], [dashboard], [cluster-api], etc.
- For managed apps: [apps], [tenant], [kubernetes], [postgres],
[virtual-machine] etc.
- For development and maintenance: [tests], [ci], [docs], [maintenance].
- If it's a work in progress, consider creating this PR as a draft.
- Don't hesistate to ask for opinion and review in the community chats,
even if it's still a draft.
- Add the label `backport` if it's a bugfix that needs to be backported
to a previous version.
-->
## What this PR does
### Release note
<!-- Write a release note:
- Explain what has changed internally and for users.
- Start with the same [label] as in the PR title
- Follow the guidelines at
https://github.com/kubernetes/community/blob/master/contributors/guide/release-notes.md.
-->
```release-note
[platform] Fix stale workloads not being deleted
```
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Improvements**
* Added a delay before reprocessing items that are being deleted,
resulting in more efficient handling of deletions.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Workloads tracking an object undergoing deletion can be reconciled when
the object is marked for deletion, but is not yet removed. After the
object is deleted, there is no event to trigger another reconciliation
of the workload and it might never get deleted until a global reconcile
happens or the controller is restarted. This patch ensures they are
requeued in the reconciliation loop.
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
When the WorkloadMonitor is reconciled and child Workload objects are
created, they will now get additional labels in the
`workloads.cozystack.io` namespace, containing metadata about the
workload. This particular commit checks if a pod targeted by a Workload
is owned by a VirtualMachineInstance (i.e. it launches a KubeVirt VMI)
and, if so, gets the VMI instance type and puts it in the
`kubevirt-vmi-instance-type` label.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Workload objects created for Pods now include additional labels
extracted from their owner references, specifically for
VirtualMachineInstance resources.
- If a VirtualMachineInstance has a relevant annotation, its instance
type is now reflected as a label on the associated Workload.
- **Chores**
- Updated and added several dependencies to improve compatibility and
maintainability.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
When the WorkloadMonitor is reconciled and child Workload objects are
created, they will now get additional labels in the
`workloads.cozystack.io` namespace, containing metadata about the
workload. This particular commit checks if a pod targeted by a Workload
is owned by a VirtualMachineInstance (i.e. it launches a KubeVirt VMI)
and, if so, gets the VMI instance type and puts it in the
`kubevirt-vmi-instance-type` label.
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Introduced a new controller to synchronize tenant HelmReleases and
propagate configuration changes.
- Added dynamic host value overrides in multiple Helm templates by
conditionally retrieving values from the "tenant-root" HelmRelease.
- Updated RBAC permissions to allow management of HelmRelease resources.
- **Improvements**
- Added support for Helm v2 API integration.
- Enhanced HelmRelease reconciliation logic and configuration
propagation for tenant environments.
- **Bug Fixes**
- Fixed periodic reconciliation for the "tenant-root" HelmRelease by
setting its interval to zero.
- **Version Updates**
- Incremented version numbers for the "info" and "ingress" packages.
- **Chores**
- Updated version mappings and commit references.
- Improved .gitignore to exclude the .vscode directory.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
* Count Workload resources for pods by requests, not limits
* Do not count init container requests
* Prefix Workloads for pods with `pod-`, just like the other types to
prevent possible name collisions (closes#787)
The previous version of the WorkloadMonitor controller incorrectly
summed resource limits on pods, rather than requests. This prevented it
from tracking the resource allocation for pods, which only had requests
specified, which is particularly the case for kubevirt's virtual machine
pods. Additionally, it counted the limits for all containers, including
init containers, which are short-lived and do not contribute much to the
total resource usage.
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
When populating the WorkloadMonitor objects, the status field is now
populated with a specially formatted string, mimicking the keys of
ResourceQuota.spec.hard, e.g.
`<storageclassname>.storageclass.storage.k8s.io/requests.storage` or
`<ipaddresspoolname>.ipaddresspool.metallb.io/requests.ipaddresses`
so the storage class or IP pool in use can be tracked. Part of #788.
Signed-off-by: Timofei Larkin <lllamnyp@gmail.com>
Workload object counts were previously getting out of control as the recreation of a related Pod would spawn a new workload, while the old one would never get deleted (except for StatefulSets, where the names of Pods are stable). Workloads without a matching object are now deleted.
