Releases: headlamp-k8s/plugins
keda v0.1.2
New Changes
- 8c8b499 keda: make the pause-scaling dialog one translatable sentence
- 01bc925 keda: fix "Defaults" rendering in stabilization window
- b2d8bd2 KEDA: implement i18n
- d5c9a71 keda: Add pause action for ScaledObject and ScaledJob
- 5e9e408 artifacthub: Move to folder based releases
- e5a12a6 keda: Fix logo with new URL
- 40c2dcd keda: Update artifacthub-pkg.yaml for version 0.1.1-beta
Dependency Updates
- ce6cb10 keda: Bump version to 0.1.2
- 3e2f65e keda: Override ws to 8.21.0
- 52084f5 keda: Bump headlamp-plugin to 0.14.0
- 8b68fae keda: Bump headlamp-plugin to 0.13.1
- 9734acc Bump lodash to 4.17.23
- c6063f1 Bump glob to 10.5.0
- 9fdd18e Bump tar to 7.5.3
- 703da85 keda: Update headlamp-plugin to 0.13.0
- aa36439 KEDA: Bump @babel/runtime to 7.27.1
- 3d12e1c KEDA: Bump @babel/core to 7.27.1
- 288d34f KEDA: Bump vite to 6.4.1
- 118e84c KEDA: Bump vite-plugin-static-copy to 2.3.2
- f67bfa8 KEDA: Bump tmp to 0.2.4
- 9893e46 KEDA: Bump form-data to 4.0.4
- 54da007 KEDA: Bump sha.js to 2.4.12
Tarball checksum (sha256): 979996aa0d0efdbc852baf6977e178c955264e346fea2c82df94b2d58ef57501
prometheus v0.9.0
Highlights
This release expands the Prometheus plugin's metrics coverage to several popular Kubernetes ecosystem CNCF projects
✨ New CRD metrics support
Volcano — Metrics charts for Queue resources: batch scheduling, queue utilization, and resource allocation.
Knative — Metrics charts for Knative workloads, with improved rendering and translation support.
Cluster API (CAPI) — Metrics support for CAPI resources, with new chart configs and queries for cluster lifecycle and infrastructure health.
Strimzi (Kafka) — Metrics charts for Strimzi-managed Kafka clusters, covering broker and cluster-level health.
New Changes
- 1b85ea3 prometheus: Add Volcano Queue metrics charts
- 309bf6c prometheus: knative: Fix metrics render and add translation
- 51eaf0e Prometheus: Cluster-Api: Add metrics support for cluster-api
- f0054d8 Prometheus:Cluster-Api: Enhance CAPI metrics support by adding new chart configurations and queries
- 31b1a30 prometheus: charts: Add Knative charts
- d7c91b1 prometheus: Add Kafka cluster metrics charts
Fixes:
- 2886bee prometheus: gate Job metrics/actions on batch/v1
- 247b8a8 fix(settings): handle plain text response in Prometheus connection test
- 44e21eb prometheus: Chart: Guard merge against out-of-bounds series access
Dependency Updates
- 21717b1 prometheus: Bump version to 0.9.0
- 7d892cc prometheus: Override ws to 8.21.0
- 5608e23 prometheus: Bump @kinvolk/headlamp-plugin to ^0.14.0
- e8f9c09 prometheus: Bump headlamp-plugin to 0.14.0
Chore:
plugin-catalog v0.4.4
Dependency Updates
- 2c84f45 plugin-catalog: Bump version to 0.4.4
- fcf5f14 plugin-catalog: Fix svg import
- a637b6a plugin-catalog: Override ws to 8.21.0
- 0e4bc01 plugin-catalog: Allow version selection in Detail view
- d68c09a plugin-catalog: Bump headlamp-plugin to 0.14.0
- d2e90c7 plugin-catalog: Bump headlamp-plugin to 0.13.1
- 9734acc Bump lodash to 4.17.23
- c6063f1 Bump glob to 10.5.0
- 9fdd18e Bump tar to 7.5.3
Tarball checksum (sha256): 03bf713e4b4acb790fc2f10dcb985c9a66ce8c0828c69ffddcfb476aa28470fb
backstage v0.1.0-beta-3
New Changes
- b9f70c8 backstage: Bump version to 0.1.0-beta-3
- d802ff6 backstage: Add info to plugin settings and fix lint
- 1941de2 backstage: Override ws to 8.21.0
- 44967ab backstage: Bump tar to ^7.5.15
- d1f9590 backstage: Bump @kinvolk/headlamp-plugin to ^0.14.0
- 3c214c9 backstage: Bump headlamp-plugin to 0.13.1
- 9734acc Bump lodash to 4.17.23
- c6063f1 Bump glob to 10.5.0
- 9fdd18e Bump tar to 7.5.3
- aad268d backstage: Update headlamp-plugin to 0.13.0
- 27adccb backstage: Bump tar to 7.5.2
- f8478ff backstage: Bump headlamp-plugin 0.13.0-alpha.11
Tarball checksum (sha256): 62544c020176e2f75fce420c70a038d8f9c2ea787a2fe0abfabed2d823bc0b25
app-catalog v0.9.0
v0.9.0
This release focuses on making the Installed releases list view faster to work with: you can now find releases quickly, act on them without leaving the list, and clean up several releases at once.
✨ Highlights
Bulk delete in the installed list view
Select multiple releases with the row checkboxes and uninstall them in a single action. A bulk-actions toolbar appears with the selected count and a delete button, and a confirmation dialog lists exactly what will be removed before anything is uninstalled.
Quick actions (Upgrade, Rollback, Delete) in the list view
Each release row now has a three-dot menu giving direct access to Upgrade, Rollback, and Delete — no need to open the detail page first. The menu is context-aware: Upgrade is enabled only when a newer chart version is available, and Rollback is disabled for releases that are still on their first revision. Rollback opens a dialog to pick the exact revision to revert to.
Filter by name and namespace in the installed list view
A filter bar in the list header lets you search releases by name and narrow the list to a specific namespace. The namespace dropdown is populated from the namespaces of the currently loaded releases, and the name search is debounced for smooth typing.
🛠 Other improvements & fixes
- Fixed the catalog list view.
- Allow sorting by
updatedAtin the installed list view. - Fixed the date label in the detail and list views.
- Fixed a delete race condition and clarified the history entry type.
New Changes
- eb0c020 app-catalog: Fix the catalog list view
- 9fc8815 app-catalog: allow sort by updatedAt in Install list view
- 16cc0fe app-catalog: fix date label in detail/list view
- 3c45046 app-catalog: fix the installed list action menu
- 8bc2e6c app-catalog: fix delete race condition and clarify history entry type
- a1029b0 app-catalog: add JSDoc to new interfaces and functions
- abe5c5e app-catalog: fix rollback to use numeric version and poll status
- 1465f5f app-catalog: add filtering, bulk delete and quick-actions menu
- b9251e7 app-catalog: Format files with prettier
- 1e8eb72 app-catalog: Fix rotted Storybooks after Vite migration
- 29ee785 app-catalog: Add i18n support
Dependency Updates
- 8cf2c42 app-catalog: Bump version to 0.9.0
- 33fa0b7 app-catalog: Override ws to 8.21.0
- d8deb5e app-catalog: Bump headlamp-plugin to 0.14.0
- 179cf99 app-catalog: Bump headlamp-plugin to 0.13.1
- 9734acc Bump lodash to 4.17.23
- 9fdd18e Bump tar to 7.5.3
Tarball checksum (sha256): 69ba1940e9d2c6e9e540f4c5f709fa2729b6b6ca1afcdad2554387b1f23e8613
Cluster API Plugin for Headlamp — 0.2.0-beta
Second release. This is a beta release. APIs and interfaces may change before v1.0. Feedback and bug reports are welcome.
About
Cluster API is a Kubernetes sub-project that brings declarative, Kubernetes-style APIs to cluster lifecycle management. The Cluster API Plugin for Headlamp (@headlamp-k8s/cluster-api) adds dedicated UI views for managing CAPI resources directly inside Headlamp.
This release builds substantially on v0.1.0. The focus is on three areas: a centralized health dashboard, expanded operational actions across all resource detail pages, and real-time Prometheus metrics embedded directly into resource views.
Demo
Cluster-Api-v0.2.0.mp4
Contributors
What's New
Centralized Cluster API Dashboard
A new top-level dashboard gives platform teams a single place to assess the health of their entire CAPI environment without navigating through individual resource pages.
The dashboard shows:
- Overall resource counts — clusters, machines ready, provisioning, and error states — in a summary bar at the top
- A per-cluster breakdown listing every cluster with its current status, the number of machines and control plane replicas it owns, and any active error conditions surfaced inline
- Common infrastructure error messages (such as
InfrastructureReady: FalseorWaitingForControlPlane) shown directly in the table row so you can see what is wrong at a glance - Quick navigation from any row into the full cluster detail view
This replaces the need to scan individual Cluster, MachineDeployment, and Machine list views separately when something is broken.
Map View Improvements & Glance Tooltips
The cluster topology map introduced in v0.1.0 has been significantly improved. CAPI resources now appear as proper nodes in Headlamp's map view with accurate status representation and richer visual context.
Glance tooltips — hovering over any CAPI node in the map now shows a compact status summary without leaving the map. Each resource kind has its own glance component registered via registerKubeObjectGlance, surfacing the information most relevant to that kind:
- Cluster — phase, infrastructure ready status, control plane ready status
- Machine — phase, node reference, provider ID
- MachineDeployment / MachineSet — replica counts (ready / desired / available)
- KubeadmControlPlane — replica counts and initialized status
Status derivation — node status in the map is now derived from the actual CAPI conditions and phase fields rather than generic Kubernetes object status. This means a Cluster node shows red whenInfrastructureReadyis false, and a Machine node reflects its provisioning phase correctly, making it easy to spot unhealthy resources in the topology view at a glance.
Map filtering — CAPI resource types are now properly registered as map sources, making them available in the map's filter/layer controls so users can toggle CAPI resources on and off independently.
Expanded Actions on Detail Pages
Every CAPI resource detail page now has a unified action bar with resource-specific operations. Previously only the kubeconfig download was available on the Cluster detail page; now each resource type exposes its own set of relevant actions.
Pause and resume reconciliation is now available on all of the following:
- Clusters
- MachineDeployments
- MachineSets
- MachinePools
- KubeadmControlPlanes
- MachineHealthChecks
Pausing reconciliation is useful when performing manual interventions or debugging state drift without the controller overwriting your changes. The pause action patches thespec.pausedfield directly from the UI. Resuming clears it.
Cluster-level scaling — the Cluster detail page now exposes a scale dialog that lets you adjust the desired replica count at the cluster topology level. This is the correct place to scale topology-managed clusters, where modifying MachineDeployment replicas directly would be overridden. The dialog validates the input and patches the resource without requiring a kubectl patch command.
Machine-specific actions — the Machine detail page now has a dedicated Actions.tsx component with operations that are meaningful at the individual machine level:
- View node — navigates directly to the Kubernetes Node object that this machine is bound to, using the node reference stored in the machine status
- Replace machine — triggers a rolling replacement of the machine by annotating it for deletion and allowing the owning MachineSet or MachineDeployment to provision a replacement
- Pause/Resume reconciliation — same as above, but scoped to the individual machine
- Open provider console — opens the cloud provider console URL for the underlying infrastructure instance, when a provider ID is available in the machine spec
All actions are now managed through per-resource action factory functions (getClusterActions, getMachineDeploymentActions, getKubeadmControlPlaneActions, etc.) that return the correct set of buttons for each resource type. This replaces the previous approach of hardcoding individual action components directly in each detail view.
Prometheus Metrics in Resource Detail Views
When the Prometheus plugin is installed and a scrape target is configured for the CAPI controller-manager, metrics charts appear automatically in the detail views of Cluster, MachineDeployment, MachineSet, Machine, MachinePool, and KubeadmControlPlane.
Seven chart types are available, each showing a different signal:
- Reconcile — success and error rates per second
- Reconcile duration — p50 and p99 latency
- Workers — active workers vs. configured maximum
- Workqueue — queue depth, add rate, and retry rate
- Webhooks — admission request rates and latency
- Cluster cache — connection status and healthcheck rates (Cluster only)
- SSA cache — Server-Side Apply cache hit and miss rates
Each resource type gets the relevant subset automatically. Queries usecapi_*metrics with fallback tocontroller_runtime_*for older CAPI versions. A reference scrape configuration (capi-prometheus-values.yaml) is included in the repository.
Localization
40+ new translation keys have been added across all 19 supported language files to cover the new metrics chart labels, workqueue terminology, webhook metrics, and cluster cache metrics.
Supported languages: English, German, Czech, Spanish, French, Hungarian, Indonesian, Italian, Japanese, Korean, Dutch, Polish, Portuguese (BR and PT), Russian, Swedish, Turkish, Chinese Simplified, Chinese Traditional.
Changelog
New features
-
Added centralized Cluster API dashboard with per-cluster health summary, resource counts, and inline error conditions
-
Added map view glance tooltips for all major CAPI resource types (Cluster, Machine, MachineDeployment, MachineSet, KubeadmControlPlane)
-
Added accurate status derivation for CAPI nodes in the map view based on CAPI conditions and phase fields
-
Added CAPI resource type registration as map sources with filter/layer support
-
Added pause and resume reconciliation action to Cluster, MachineDeployment, MachineSet, MachinePool, KubeadmControlPlane, and MachineHealthCheck detail pages
-
Added cluster-level scale dialog on the Cluster detail page for topology-managed replica management
-
Added machine-specific actions: view node, replace machine, pause/resume reconciliation, open provider console
-
Added Prometheus metrics integration for CAPI resource detail views (7 chart types: reconcile, duration, workers, workqueue, webhooks, cluster cache, SSA cache)
-
Added
capi-prometheus-values.yamlreference scrape configuration -
Added 40+ localization keys across 19 language files for all new metrics terminology
Improvements -
Refactored all detail page actions into per-resource factory functions (
getClusterActions,getMachineDeploymentActions,getKubeadmControlPlaneActions, etc.) -
Updated API version validation to use regex patterns, supporting both
v1beta1andv1beta2resources for Prometheus metric detection -
Improved kubeconfig retrieval with proper loading states and error handling
Compatibility
| Component | Version |
|---|---|
| Headlamp | >= 0.13.1 (requires @kinvolk/headlamp-plugin ^0.13.1) |
| Cluster API | v1beta1, v1beta2 |
| ... |
Contributors
Assets 3
Knative Plugin for Headlamp (v0.3.0-beta)
About
Knative brings serverless workloads to Kubernetes- handling traffic routing, autoscaling, and revision management so teams can deploy and iterate without fighting infrastructure. The Knative plugin for headlamp provides an intuitive visual interface for managing Knative resources like Kservice, Revisions and Networking.
This release builds on top of the v0.2.0-alpha release, focusing on DomainMapping, ClusterDomainClaim, Map interface for Knative CRDs, metrics and testing.
Map View
Visualizing your serverless architecture is now easier than ever. The Headlamp Map view now natively supports Knative serving CRDs, providing a graphical representation of your infrastructure. You can now clearly see the topological relationships and traffic flow linked between your DomainMapping, Service, and Revision resources.
DomainMapping
Managing custom domains is now more streamlined. The new DomainMapping list view provides a clear, sortable overview of your mappings, displaying vital information such as the associated KService URL, namespace, and overall ready status.
ClusterDomainClaim
We have added a dedicated list view for ClusterDomainClaims. You can now easily track the namespace, readiness status, and age of all your cluster-level domain claims at a glance.
Metrics and Testing
Support for Prometheus metrics has also been added via the Prometheus plugin for Headlamp. You can now view graphical metrics for both KServices and Revisions, tracking:
- Request Rate
- Request Rate per Revision (for KService)
- Latency
- Resource usage
For developers, we've included new YAML test files and shell scripts to cover a variety of edge cases. These tools make it much easier to spin up demo clusters, simulating intentional errors and generating demo traffic for Prometheus to scrape. You can find the specific use cases of each file and script in the plugin's README.
Contributors
Thanks to all the contributors who made this release possible!
@mudit06mah @skoeva @Joshna907 @kahirokunn
| Commit | Author | Message |
|---|---|---|
| 624af02 | @mudit06mah | knative: Kservice: Shift helpers to util |
| 956755d | @mudit06mah | knative: domainmappings: Add list view |
| a7a340e | @mudit06mah | knative: clusterdomainclaims: Add list view |
| 91d09dc | @skoeva | knative: Bump headlamp-plugin to 0.13.1 |
| 4f425cc | @mudit06mah | knative: revisions: Add undefined check for parentService |
| 74608d2 | @mudit06mah | knative: Improve map for knative crds |
| 153132e | @mudit06mah | knative: map: Fix glance and mapView for errors |
| 9c946e6 | @mudit06mah | knative: Add yaml test-files and guide to README |
| a1d47ee | @skoeva | knative: Bump headlamp-plugin to 0.14.0 |
| 369babd | @Joshna907 | knative: implement dynamic sidebar filtering based on CRD discovery |
| d693dde | @mudit06mah | knative: Add traffic test-files for metrics |
| 934e61c | @kahirokunn | knative: reuse installed check for sidebar |
| b5ab395 | @mudit06mah | knative: kserivce: Fix validation for revision name |
Contributors
Assets 3
cert-manager v0.1.1
New Changes
- 4e97c04 cert-manager: link SecretKeySelector names to Secret detail pages
- c5de017 fix: Update cert-manager/src/components/common/CommonComponents.tsx
- 9f281df fix: Improve cloudflareGetter to handle undefined values and clean up code
- 8689bf5 feat: Enhance ACMEChallengeSolverComponent with Cloudflare support. Thanks @YotamKorah
- 5e9e408 artifacthub: Move to folder based releases
- 7ad4a7e cert-manager: storybook: Add comprehensive stories for UI testing
- 65c7003 cert-manager: certificateRequests: Fix undefined error with CertifcateRequest.status.conditions being optional
- e90bb72 cert-manager: Display Expires In field is Cert list view
- 6ca99c2 cert-manager: Format files with prettier
- 6c37e0b cert-manager: Add artifacthub package
Dependency update
- e9ffea8 cert-manager: Bump version to 0.1.1
- 0a1b3a4 cert-manager: Override ws to 8.21.0
- ba9577b cert-manager: Bump @kinvolk/headlamp-plugin to ^0.14.0
- 94d744d chore: cert-manager: update headlamp-plugin to version 0.13.1
- 9734acc Bump lodash to 4.17.23
- c6063f1 Bump glob to 10.5.0
- 9fdd18e Bump tar to 7.5.3
- 9903f15 cert-manager: Update headlamp-plugin to 0.13.0
- 9cf9992 cert-manager: Bump vite to 6.4.1
- 354c924 cert-manager: Bump vite-plugin-static-copy to 2.3.2
- 3220bcc cert-manager: Bump tmp to 0.2.4
- 24e31b4 cert-manager: Bump form-data to 4.0.4
- a03f8df cert-manager: Bump sha.js to 2.4.12
- c3709ec cert-manager: Bump pbkdf2 to 3.1.3
- 0df027a cert-manager: Update headlamp-plugin to 0.12.0
- c4f5a54 cert-manager: Bump vite to 6.3.5
- 696afaf cert-manager: Upgrade to headlamp-plugin 0.11.0-alpha.3
- 6c12893 cert-manager: Update headlamp-plugin
Tarball checksum (sha256): 2b06caae0a207e2c30ce0d21bc4d672ae0a71a2cffc23a4585199110533032ab
Contributors
Assets 3
kubeflow 0.2.0-alpha
Kubeflow
First public release of the Pipelines component. This is an alpha release. APIs and interfaces may change before v1.0. Feedback and bug reports are welcome.
About
Kubeflow Pipelines (KFP) is a platform for building and deploying portable, scalable machine learning (ML) workflows based on Docker containers.
Managing complex DAGs, tracking experiments, and troubleshooting failed runs often requires switching between multiple dashboards. This plugin brings SRE-level observability for Kubeflow Pipelines directly into Headlamp, allowing operators to audit infrastructure health and verify RBAC without leaving their Kubernetes console.
release.mp4
Bringing Deep Pipeline Observability and Infrastructure Auditing to Headlamp
This release focuses on empowering platform operators with unified visibility into both Kubeflow Pipelines (KFP) and Katib workloads directly inside Headlamp:
- Control Plane & API Health: Instantly verify if the
ml-pipelineAPI is reachable. The dashboard auto-discovers KFP deployments and services, showing "Native API Mode" status and endpoint resolution across namespaces. - Version Evolution & Diffing: Track how pipelines change over time. The plugin provides a built-in Version Comparison tool to diff the YAML/JSON specs of the latest vs. previous versions directly in the UI.
- Cross-Namespace Resource Auditing: Identify expensive or failing experiments across the entire cluster using the 'Failure Tracking' table, which consolidates errors from Pipelines, Versions, and Runs into a single prioritized view.
- Storage & Artifact Roots: Visibility into the storage backends (S3/MinIO/GCS) being used by different pipeline runs, helping operators validate artifact propagation and storage configuration.
- Experiment Convergence & Health Triage: Instantly identify which experiments are active, converged, or failing. The dashboard provides a dedicated "Experiment Health" triage view to isolate failing tuning loops.
- Trial Progression & Budget Tracking: Monitor how far each experiment has progressed against its trial budgets. Track trial counts (current vs. max), parallel execution limits, and early stopping status in real-time.
- Metric-Driven Optimization Success: Track the "Best Trial So Far" for any experiment. The plugin surfaces optimal metrics and parameter assignments directly in the UI, eliminating the need to parse raw JSON status fields.
Features in v0.2.0-alpha
Pipelines Dashboard Overview
A high-level command center for KFP. Features include resource aggregation cards (Runs, Experiments, Recurring Runs), a control plane health monitor, and an RBAC audit table to verify access across all KFP custom resources.
Pipeline Version Comparison
Audit changes between workflow iterations. When multiple versions of a pipeline exist, use the "Compare Latest vs Previous Version" action to launch a side-by-side diff viewer, identifying exactly what changed in the pipeline spec.
Pipeline Definition & Spec Summary
Deep inspection of pipeline templates. Shows the underlying SDK Version, Package URLs, and a summary of Tasks and Executors defined within the workflow without needing to parse the raw spec manually.
Deep Run Observability
Rich detail views for Pipeline Runs. Features include automated duration tracking, runtime configuration inspection (viewing parameters passed to the run), and a conditions table for real-time status updates.
Experiment and Version Management
Deep-linking between Pipelines and their Versions. Track which Experiments are active and view detailed metadata for every versioned workflow template.
Recurring Run Schedules
Visibility into scheduled workflows. Monitor cron-based triggers and recurring pipeline instances to ensure your automated retraining or batch jobs are running as expected.
Katib Overview Dashboard
A command center for AutoML. Features include summary cards for total Experiments and Trials, an Experiment Health triage section, and a "Service Accounts & Worker Types" table to expose underlying infrastructure dependencies.
Experiment Management & Deep Inspection
Comprehensive visibility into tuning loops. List views show algorithm types, objectives, and trial counts at a glance. Detail views provide deep inspection of the search space, trial templates, and the current optimal parameters.
Deep Trial Observability
Rich detail views for individual Katib Trials. Features include automated timing tracking, metric result reporting, and condition tables. Operators can launch worker logs directly from the Trial list or detail views for rapid troubleshooting.
Suggestion Lifecycle Tracking
Visibility into the Suggestion engine. Monitor the status of suggestion algorithms and track requested vs. assigned suggestions to ensure the Katib controller is successfully generating new trial parameters.
Compatibility
| Component | Version |
|---|---|
| Headlamp | >= 0.13.1 |
| Kubeflow Pipelines API | pipelines.kubeflow.org/v2beta1 |
| Kubeflow Notebooks API | kubeflow.org/v1 |
| Katib API | kubeflow.org/v1beta1 |
| Training API | trainer.kubeflow.org/v1alpha1 |
Known Limitations (Alpha)
- Experiment creation, trial submission, and hyperparameter search visualization (e.g., Parallel Coordinates) are not yet supported.
- Log viewing requires the trial worker pods to be currently present in the cluster.
- Pipeline creation, run submission, and scheduling are not yet supported via the UI (Read-only observability).
- DAG visualization (Graph view) for pipeline steps is not supported in this release.
Installation & Quick Start
For detailed setup instructions and local developer testing, please refer to the official README.
Quick start, try it out:
git clone https://github.com/headlamp-k8s/plugins
cd plugins/kubeflow
npm install
npm run buildContributors
This plugin was developed by the Headlamp community.
@alokdangre
Feedback
This is an alpha release and community feedback shapes what comes next.
- Bug reports: Open an issue
- Feature requests: Start a discussion
- Contributing: CONTRIBUTING.md
Volcano 0.2.0 Beta Release
Volcano Headlamp Plugin 0.2.0-beta
Volcano is a cloud-native batch scheduler for Kubernetes, built for high-performance computing and AI/ML workloads.
This beta release of the Volcano Headlamp plugin improves the core Volcano experience in Headlamp with richer Job, Queue, and PodGroup views, lifecycle actions, log access, and better troubleshooting support.
Thanks @mahmoudmagdy1-1
Included in this release:
- Improved Job details with lifecycle actions
- Job logs viewer for pods created by a Volcano Job
- Pod Issues section in Job details
- Complete Queue details with queue actions and richer resource visibility
- PodGroup detail improvements including
minTaskMember - Better alignment of Job and Queue list views with Volcano CLI
- Added install guidance when Volcano is not detected
- Official Volcano logo in the Headlamp sidebar
Jobs (batch.volcano.sh/v1alpha1)
The Jobs view is now much more useful for day-to-day operations and troubleshooting.
- List view: better alignment with Volcano CLI
- Detail view: richer task rendering, plugins, policies, conditions, lifecycle actions
- Troubleshooting: Pod Issues section for pod related failures
- Logs: view logs for pods created by a Volcano Job directly from Headlamp
The new Job logs viewer supports:- All Pods and single-pod log viewing
- Container selection
- standard log controls:
- lines
- show previous
- timestamps
- follow
Queues (scheduling.volcano.sh/v1beta1)
The Queues view now provides much more complete scheduling and allocation visibility.
- List view: better alignment with Volcano CLI
- Detail view: capacity and allocation summaries, deserved and guaranteed resources, reservation details
- Actions: queue actions directly from the UI
- Navigation: child queue visibility and related queue structure
PodGroups (scheduling.volcano.sh/v1beta1)
The PodGroups view now includes additional gang-scheduling details.
- List view: unchanged core scheduling summary
- Detail view: improved progress and condition visibility
- Additional detail:
minTaskMembersupport
This makes PodGroup state easier to interpret when debugging gang-scheduled workloads.
Navigation and UX
The main focus of this release is improving the usability of the existing core CRD views rather than adding new CRDs.
Additional UX improvements in this release include:
- official Volcano sidebar icon
- install guidance when Volcano is not detected
- more consistent list/detail experience across the core resources
- direct Job log access from the Job detail page
Feedback and Community
If you try the Volcano plugin, we’d love to hear how it works in your cluster!
- Report bugs or request features: https://github.com/headlamp-k8s/plugins/issues/new
- Join Kubernetes Slack: https://slack.k8s.io
- Chat with us in
#headlampon Kubernetes Slack
