Fix DPF Operator documentation errors and inconsistencies - #119760
gokhancetinkaya wants to merge 3 commits into
Conversation
This commit addresses errors, inconsistencies, and missing content based on end-to-end validation of the deployment procedures.
Major changes:
1. Added missing operator installation procedures
- Created cert-manager Operator installation module
- Created MetalLB Operator installation module
- Created GitOps Operator installation module
- Removed duplicate configuration sections
2. Fixed environment variables (section 12.4.1)
- Restructured with all necessary variables
- Updated MTU values to standard (NODES_MTU=1500, OVN_MTU=1400)
- Improved VTEP_CIDR and OVN_MTU descriptions
- Added complete export block example
3. Removed manual DPU CSR approval section (auto-approved by operator)
- Updated deployment flow to clarify x86 worker CSRs vs DPU CSRs
- Removed section 12.6.7
- Updated DPU provisioning verification text
4. Fixed broken commands
- ArgoCD verification: use grep instead of label selector
- Grafana verification: use grep instead of label selector
5. Removed duplicate DTS console dashboard content
- Removed from section 12.7.7
- Removed from section 12.7.8
- Kept only in section 12.7.6
6. Removed premature verification steps
- Removed DPUCluster verification (section 12.5.5)
- Removed DPU service verification (section 12.5.8)
7. Variable syntax improvements
- BareMetalHost YAML: use ${VAR} syntax for clarity
- Kubeconfig commands: use $(pwd) for absolute paths
8. Other fixes
- Changed "16 vCPUs" to "16 CPUs" (nodes can be physical or virtual)
- Changed "hosted control planes" to "hosted control plane" throughout
- Removed incorrectly labeled NVIDIA DPF uninstall guide link
- Removed broken multicluster engine Operator installation link
- Removed unnecessary BFB_FILENAME variable
- Removed unnecessary deploy: true from maintenance-operator values
All changes validated against a working BlueField-3 DPU deployment.
openshift-ci
Bot
added
size/XL
|
Hi @gokhancetinkaya. Thanks for your PR. I'm waiting for a openshift member to verify that this patch is reasonable to test. If it is, they should reply with Regular contributors should join the org to skip this step. Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. DetailsInstructions 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-sigs/prow repository. |
| [source,yaml] | ||
| ---- | ||
| apiVersion: operators.coreos.com/v1alpha1 | ||
| kind: Subscription |
There was a problem hiding this comment.
Why we remove this? It will be installed by assisted?
There was a problem hiding this comment.
It's not installed by the Assisted Installer. This moves it to a dedicated installation section with the complete configuration.
szigmon
left a comment
szigmon
left a comment
There was a problem hiding this comment.
LGTM, just need approval on small questions to Igal
| * You have access to the cluster as a user with the `cluster-admin` role. | ||
| * You have installed the OpenShift CLI (`oc`). | ||
| * You have installed the {cert-manager-operator}, MetalLB Operator, {gitops-title}, and NVIDIA Maintenance Operator. | ||
| * You have installed the Logical Volume Manager Storage Operator, multicluster engine operator, and the Node Feature Discovery Operator. You can install them by using the Assisted Installer during cluster creation. For manual installation, see link:https://docs.redhat.com/en/documentation/red_hat_advanced_cluster_management_for_kubernetes/2.17/html/install/index[Installing multicluster engine operator] and ensure that the hosted control planes component is enabled. |
There was a problem hiding this comment.
Don't we want this manual install method too? @tsorya
There was a problem hiding this comment.
I actually think we should , customer can miss this or can have a cluster up already and will want to move it to dpu one
|
/lgtm |
|
|
||
| [role="_abstract"] | ||
| The {cert-manager-operator} manages TLS certificates for DPF components. | ||
| You install this operator by using the OpenShift CLI. |
There was a problem hiding this comment.
🤖 [error] Vale.Terms: Use 'Operators?' instead of 'operator'.
|
|
||
| [role="_abstract"] | ||
| The {gitops-title} manages DPF service deployments and configurations using GitOps principles. | ||
| You install this operator by using the OpenShift CLI. |
There was a problem hiding this comment.
🤖 [error] Vale.Terms: Use 'Operators?' instead of 'operator'.
|
|
||
| [role="_abstract"] | ||
| The MetalLB Operator provides load balancing services for DPF components on the management cluster. | ||
| You install this operator by using the OpenShift CLI. |
There was a problem hiding this comment.
🤖 [error] Vale.Terms: Use 'Operators?' instead of 'operator'.
There was a problem hiding this comment.
🤖 [error] AsciiDocDITA.TaskContents: The '.Procedure' block title is missing.
|
@gokhancetinkaya: all tests passed! Full PR test history. Your PR dashboard. DetailsInstructions 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-sigs/prow repository. I understand the commands that are listed here. |
Add a tested CLI procedure for the MCE namespace, OperatorGroup, Subscription, and HyperShift-enabled MultiClusterEngine. Align the recovery patch with the documented mce resource and wait for dependent operator CRDs. Co-Authored-By: Claude Code <noreply@anthropic.com>
Document standalone MCE install for DPF PR openshift#119760
|
New changes are detected. LGTM label has been removed. |
openshift-ci
Bot
added
size/XXL
5 participants
This commit addresses errors, inconsistencies, and missing content based on end-to-end validation of the deployment procedures.
All changes validated against a working BlueField-3 DPU deployment.