Helm Chart Templates For Katib #2553

kunal-511 · 2025-06-17T13:15:18Z

What this PR does / why we need it:
Helm Templates For Katib

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #

Checklist:

Docs included if any changes are user facing

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

google-oss-prow · 2025-06-17T13:15:36Z

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign johnugeorge for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

kunal-511 · 2025-06-17T13:15:59Z

cc: @juliusvonkohout

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

google-oss-prow · 2025-06-22T12:53:37Z

@kunal-511: Cannot trigger testing until a trusted user reviews the PR and leaves an /ok-to-test message.

In response to this:

/retest

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

juliusvonkohout · 2025-06-23T14:39:56Z

/ok-to-test

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

kunal-511 · 2025-06-24T04:51:11Z

/retest

varodrig · 2025-07-01T01:11:06Z

@juliusvonkohout should this PR be linked to kubeflow/community#832 ?

cc @kunal-511

varodrig · 2025-07-01T01:12:20Z

@kunal-511 let us know once this PR is ready for review. It seems there's a hold + two checks failing.

Additionally, can you add some description and comments on the PR, we have 58 files to review and it'll be great to have more comments on it to help us with the approval.

Another thing I think it will be great if we can involve the Katib working group on this review.
thoughts @juliusvonkohout
cc @andreyvelich

varodrig · 2025-07-01T01:16:26Z

charts/katib/templates/database/postgres-deployment.yaml

+            {{- end }}
+          ports:
+            - name: postgres
+              containerPort: 5432


my opinion is that ports should be configurable unless it's hard-coded somewhere in katib

varodrig · 2025-07-01T01:17:21Z

charts/katib/templates/database/mysql-deployment.yaml

+                - "/bin/bash"
+                - "-c"
+                - "mysqladmin ping -u root -p${MYSQL_ROOT_PASSWORD}"
+            periodSeconds: 15


this configurations should be customizable it could depend on the cluster that it's running and resources available.

initialDelaySeconds - periodSeconds - failureThreshold

@Electronic-Waste to add thoughts

varodrig · 2025-07-01T01:18:12Z

charts/katib/templates/controller/deployment.yaml

+          args:
+            - --katib-config=/katib-config.yaml
+          ports:
+            - containerPort: 8443


in my opinion ports should be added in the values files unless it's hardcoded somewhere in katib.

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

juliusvonkohout · 2025-07-01T12:43:42Z

/ok-to-test

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

kunal-511 · 2025-07-07T12:45:57Z

The tree structure of the Kubeflow Katib Helm chart directory with explanations for each file and folder.

katib(helm charts)/
├── Chart.yaml                                 # Helm chart metadata and version information
├── README.md                                  # Chart documentation and usage instructions
├── values.yaml                                # Default configuration values for the chart
├── ci/                                        # CI/CD configuration files for testing
│   ├── values-cert-manager.yaml               # Test values for cert-manager integration
│   ├── values-enterprise.yaml                 # Test values for enterprise deployment
│   ├── values-external-db.yaml                # Test values for external database configuration
│   ├── values-kubeflow.yaml                   # Test values for Kubeflow integration
│   ├── values-leader-election.yaml            # Test values for leader election setup
│   ├── values-openshift.yaml                  # Test values for OpenShift platform
│   ├── values-postgres.yaml                   # Test values for PostgreSQL database
│   ├── values-production.yaml                 # Test values for production deployment
│   └── values-standalone.yaml                 # Test values for standalone deployment
├── crds/                                      # Custom Resource Definitions for Katib
│   ├── experiment.yaml                        # CRD for hyperparameter tuning experiments
│   ├── suggestion.yaml                        # CRD for algorithm suggestions
│   └── trial.yaml                             # CRD for individual experiment trials
├── templates/                                 # Helm template files for Kubernetes resources
│   ├── _helpers.tpl                           # Template helper functions and common definitions
│   ├── autoscaling/                           # Horizontal Pod Autoscaler configurations
│   │   └── hpa.yaml                           # HPA template for scaling components
│   ├── config/                                # Configuration files and ConfigMaps
│   │   └── configmap.yaml                     # Main configuration settings for Katib
│   ├── controller/                            # Katib controller deployment resources
│   │   ├── deployment.yaml                    # Controller deployment specification
│   │   ├── leader-election-rbac.yaml          # RBAC for leader election functionality
│   │   ├── rbac.yaml                          # Role-based access control for controller
│   │   ├── service.yaml                       # Service for controller communication
│   │   ├── serviceaccount.yaml                # ServiceAccount for controller pod
│   │   └── trial-templates-configmap.yaml     # Trial template configurations
│   ├── database/                              # Database deployment resources
│   │   ├── mysql-deployment.yaml              # MySQL database deployment
│   │   ├── mysql-pvc.yaml                     # MySQL persistent volume claim
│   │   ├── mysql-service.yaml                 # MySQL service for database access
│   │   ├── postgres-deployment.yaml           # PostgreSQL database deployment
│   │   ├── postgres-pvc.yaml                  # PostgreSQL persistent volume claim
│   │   ├── postgres-service.yaml              # PostgreSQL service for database access
│   │   └── secret.yaml                        # Database credentials and secrets
│   ├── db-manager/                            # Database manager component
│   │   ├── deployment.yaml                    # DB manager deployment specification
│   │   └── service.yaml                       # Service for DB manager communication
│   ├── db/                                    # Additional database configurations
│   ├── istio/                                 # Istio service mesh integration
│   │   ├── authorization-policy.yaml          # Istio authorization policies
│   │   └── virtual-service.yaml               # Istio virtual service configuration
│   ├── monitoring/                            # Monitoring and observability resources
│   │   └── servicemonitor.yaml                # Prometheus ServiceMonitor for metrics
│   ├── namespace/                             # Namespace creation resources
│   │   └── namespace.yaml                     # Namespace definition for Katib
│   ├── openshift/                             # OpenShift-specific configurations
│   ├── rbac/                                  # Additional RBAC configurations
│   │   └── kubeflow-roles.yaml                # Kubeflow-specific role definitions
│   ├── security/                              # Security-related configurations
│   │   ├── network-policy.yaml                # Network policies for pod communication
│   │   └── pod-disruption-budget.yaml         # Pod disruption budget for availability
│   ├── ui/                                    # Katib UI component resources
│   │   ├── deployment.yaml                    # UI deployment specification
│   │   ├── rbac.yaml                          # RBAC for UI component
│   │   ├── service.yaml                       # Service for UI access
│   │   └── serviceaccount.yaml                # ServiceAccount for UI pod
│   └── webhook/                               # Admission webhook configurations
│       ├── certificate.yaml                   # TLS certificates for webhook
│       ├── mutating-webhook.yaml              # Mutating admission webhook
│       └── validating-webhook.yaml            # Validating admission webhook

@juliusvonkohout @varodrig

varodrig

@kunal-511 nice job! I've reviewed the standalone install including namespace and controller folder.
I'll continue with the rest during this week. If you can take a look at this comments it'll be great. thank you
cc @juliusvonkohout

varodrig · 2025-07-16T03:28:02Z

charts/katib/templates/namespace/namespace.yaml

+  labels:
+    {{- include "katib.labels" . | nindent 4 }}
+    katib.kubeflow.org/metrics-collector-injection: enabled
+    {{- if .Values.podSecurityStandards.enforced }}


I don't see this code, pod security Standards on katib namespace.yaml file. is this coming from the implementation from manifests?
Ensure we are validating any changes to the current katib manifests with the katib team.

This is Optional security hardening in helm charts which can be easily disable

@kunal-511 great security practice. I'd like the katib team to be aware of this updates
cc @Electronic-Waste to approve.

varodrig · 2025-07-16T03:34:40Z

charts/katib/Chart.yaml

+- https://github.com/kubeflow/katib
+
+annotations:
+  category: Machine Learning


validate this category with the katib team

varodrig · 2025-07-16T03:35:42Z

charts/katib/templates/controller/deployment.yaml

@@ -0,0 +1,123 @@
+{{- if .Values.controller.enabled }}


I'd keep the same file name as it's on the katib manifests, controller.yaml

varodrig · 2025-07-16T03:38:17Z

charts/katib/templates/controller/deployment.yaml

+        {{- end }}
+      annotations:
+        prometheus.io/scrape: "true"
+        prometheus.io/port: "{{ .Values.controller.ports.metrics }}"


I couldn't find this value on the values-standale.yaml file.

Values.yaml is the main file

varodrig · 2025-07-16T03:45:38Z

charts/katib/templates/controller/deployment.yaml

+      {{- end }}
+      containers:
+        - name: katib-controller
+          image: {{ include "katib.image" (dict "registry" .Values.global.imageRegistry "repository" .Values.controller.image.repository "tag" (.Values.controller.image.tag | default .Values.global.imageTag) "global" .Values.global) }}


can we define a default value on the values file?

Already present in the values.yaml file

varodrig · 2025-07-16T03:56:55Z

charts/katib/templates/controller/deployment.yaml

+            - containerPort: {{ .Values.controller.ports.healthz }}
+              name: {{ .Values.controller.service.ports.healthz.name }}
+              protocol: TCP
+          readinessProbe:


I'd keep the probes here. I do not know of a reason of why this would change unless the implementation changes.

@Electronic-Waste to provide thoughts

varodrig · 2025-07-16T03:59:17Z

charts/katib/templates/controller/deployment.yaml

+            {{- toYaml . | nindent 12 }}
+            {{- end }}
+          {{- with .Values.controller.securityContext }}
+          securityContext:


line duplicated and content is also in helpers. I recommend to keep security best practices on the deployment file.

varodrig · 2025-07-16T04:02:22Z

charts/katib/templates/controller/deployment.yaml

+          args:
+            {{- toYaml .Values.controller.command.args | nindent 12 }}
+          ports:
+            - containerPort: {{ .Values.controller.ports.webhook }}


I couldn't find the ports on the values standalone file.

Values.yaml is the main file

sorry I couldn't find it before. I see now. thanks.

kunal-511 · 2025-07-16T14:46:43Z

charts/katib/templates/controller/service.yaml

+spec:
+  {{- include "katib.serviceType" (dict "type" .Values.controller.service.type "Values" .Values) | nindent 2 }}
+  ports:
+    - port: {{ .Values.controller.service.ports.webhook.port }}


It is in the values.yaml

varodrig · 2025-07-16T04:10:35Z

charts/katib/templates/controller/leader-election-rbac.yaml

@@ -0,0 +1,37 @@
+{{- if .Values.controller.leaderElection.enabled }}


it seems this is a new file. can you validate with the katib team? or let me know if I'm missing something.

In Helm chart integrated this as a configurable option rather than requiring a separate installation profile. In the original manifests, you'd need to choose between:
katib-standalone (no leader election)
katib-leader-election (with leader election)
The Helm chart makes this toggleable through controller.leaderElection.enabled, providing more flexibility in a single chart.

thanks for the clarification. I'd refer to the katib team on this.
@Electronic-Waste

kunal-511 · 2025-07-16T14:46:43Z

charts/katib/templates/controller/service.yaml

+spec:
+  {{- include "katib.serviceType" (dict "type" .Values.controller.service.type "Values" .Values) | nindent 2 }}
+  ports:
+    - port: {{ .Values.controller.service.ports.webhook.port }}


It is in the values.yaml

juliusvonkohout · 2025-07-18T11:27:12Z

@anencore94

@Electronic-Waste

can you help here ?

juliusvonkohout · 2025-07-18T11:27:23Z

/ok-to-test

varodrig · 2025-07-19T04:03:58Z

charts/katib/templates/autoscaling/hpa.yaml

looks like a new implementation. thanks!
@Electronic-Waste to review.

varodrig · 2025-07-20T15:42:26Z

charts/katib/templates/database/postgres-pvc.yaml

+    {{- end }}
+  resources:
+    requests:
+      storage: {{ .Values.database.postgres.persistence.size | quote }}


values file shows 10Gi, original file on manifests is 3Gi

the ci values override all the main file values. So
In charts/katib/ci/values-postgres.yaml (lines 35-37):

persistence: enabled: true size: 3Gi # ← This matches the manifests!

varodrig · 2025-07-20T20:10:24Z

/ok-to-test

varodrig · 2025-07-20T20:12:48Z

great job @kunal-511 thank so much for working on this. There's a lot of work and thought process on understanding the current katib kustomize files, installation and helm .

I appreciate the amazing work you have done.

We need @Electronic-Waste to do a final review. @Electronic-Waste I left a few comments for you, please review ci in detail and there are good practices implemented from security constraints, PDB, network policies and service monitoring ensure you are ok with this implementation.

Helm Chart Templates For Katib

8d00d07

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

kunal-511 marked this pull request as draft June 17, 2025 13:15

google-oss-prow bot added do-not-merge/work-in-progress size/XXL labels Jun 17, 2025

google-oss-prow bot requested review from anencore94 and Electronic-Waste June 17, 2025 13:15

Modified helm templated to match with the kuztomize ouput

18fd722

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

kunal-511 marked this pull request as ready for review June 19, 2025 15:46

google-oss-prow bot removed the do-not-merge/work-in-progress label Jun 19, 2025

google-oss-prow bot added the ok-to-test label Jun 23, 2025

Updated path

c2f8e0f

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

varodrig reviewed Jul 1, 2025

View reviewed changes

kunal-511 added 2 commits July 1, 2025 18:02

Updated hard coded values

2905923

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

fix lint error

7d95642

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

kunal-511 added 2 commits July 6, 2025 10:06

lint errors updated

3c158dc

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

Updated comparision

9fdd8f3

Signed-off-by: kunal-511 <yoyokvunal@gmail.com>

varodrig reviewed Jul 16, 2025

View reviewed changes

kunal-511 commented Jul 16, 2025

View reviewed changes

varodrig reviewed Jul 19, 2025

View reviewed changes

charts/katib/templates/autoscaling/hpa.yaml

Copy link

varodrig Jul 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks like a new implementation. thanks!
@Electronic-Waste to review.

varodrig reviewed Jul 20, 2025

View reviewed changes

		@@ -0,0 +1,37 @@
		{{- if .Values.controller.leaderElection.enabled }}

Helm Chart Templates For Katib #2553

Are you sure you want to change the base?

Helm Chart Templates For Katib #2553

Uh oh!

Conversation

kunal-511 commented Jun 17, 2025

Uh oh!

google-oss-prow bot commented Jun 17, 2025

Uh oh!

kunal-511 commented Jun 17, 2025

Uh oh!

google-oss-prow bot commented Jun 22, 2025

Uh oh!

juliusvonkohout commented Jun 23, 2025

Uh oh!

kunal-511 commented Jun 24, 2025

Uh oh!

varodrig commented Jul 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

varodrig commented Jul 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

juliusvonkohout commented Jul 1, 2025

Uh oh!

kunal-511 commented Jul 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

varodrig left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

varodrig Jul 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

varodrig commented Jul 1, 2025 •

edited

Loading

varodrig commented Jul 1, 2025 •

edited

Loading

kunal-511 commented Jul 7, 2025 •

edited

Loading

varodrig left a comment •

edited

Loading

varodrig Jul 19, 2025 •

edited

Loading

kunal-511 Jul 20, 2025 •

edited

Loading