Skip to content

CORS-4316: Update the Installer GCP UPI documentation.#10235

Open
barbacbd wants to merge 1 commit intoopenshift:mainfrom
barbacbd:remove-gcp-deployment-manager
Open

CORS-4316: Update the Installer GCP UPI documentation.#10235
barbacbd wants to merge 1 commit intoopenshift:mainfrom
barbacbd:remove-gcp-deployment-manager

Conversation

@barbacbd
Copy link
Contributor

** Change the documentation to reference pure bash. The deployment manager is EOL in March 2026.

Design Choices:

  1. Terraform: The deployment manager can be converted to terraform. The installer team already removed terraform, so there is no reason to go back to using terraform.
  2. KRM: The deployment manager can be converted to use kubectl. This did not work, because very few resources do not have an equivalent in kubectl (but we use those resources).
  3. CI uses pure bash now, so this will remain consistent.

Note: The python templates are not removed in this commit. They are still found in upi/gcp/

@openshift-ci-robot openshift-ci-robot added the jira/valid-reference Indicates that this PR references a valid Jira ticket of any type. label Jan 19, 2026
@openshift-ci-robot
Copy link
Contributor

openshift-ci-robot commented Jan 19, 2026

@barbacbd: This pull request references CORS-4316 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the story to target the "4.22.0" version, but no target version was set.

Details

In response to this:

** Change the documentation to reference pure bash. The deployment manager is EOL in March 2026.

Design Choices:

  1. Terraform: The deployment manager can be converted to terraform. The installer team already removed terraform, so there is no reason to go back to using terraform.
  2. KRM: The deployment manager can be converted to use kubectl. This did not work, because very few resources do not have an equivalent in kubectl (but we use those resources).
  3. CI uses pure bash now, so this will remain consistent.

Note: The python templates are not removed in this commit. They are still found in upi/gcp/

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 openshift-eng/jira-lifecycle-plugin repository.

@barbacbd
Copy link
Contributor Author

/cc @jianli-wei
/cc @patrickdillon
/cc @tthvo

@openshift-ci
Copy link
Contributor

openshift-ci bot commented Jan 19, 2026

[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 barbacbd for approval. For more information see the Code Review Process.

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

Details Needs approval from an approver in each of these files:

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

@barbacbd barbacbd force-pushed the remove-gcp-deployment-manager branch from 4d238d0 to 9a4ab2b Compare January 20, 2026 12:27
@barbacbd barbacbd force-pushed the remove-gcp-deployment-manager branch from 9a4ab2b to 0782f19 Compare January 21, 2026 12:00
@barbacbd
Copy link
Contributor Author

/retest-required

@tthvo
Copy link
Member

tthvo commented Jan 27, 2026

/retest

@barbacbd barbacbd force-pushed the remove-gcp-deployment-manager branch from 0782f19 to 51b2283 Compare January 27, 2026 12:22
@barbacbd barbacbd force-pushed the remove-gcp-deployment-manager branch from 51b2283 to ee30935 Compare January 30, 2026 15:15
@tthvo
Copy link
Member

tthvo commented Jan 31, 2026

/test e2e-gcp-upi-xpn

Copy link
Member

Choose a reason for hiding this comment

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

Should we define a default instance type and disk size like we did before?

machine_type: 'n1-standard-4'
root_volume_size: '128'

Or similar to https://github.com/openshift/release/blob/master/ci-operator/step-registry/upi/install/gcp/upi-install-gcp-ref.yaml?

Copy link
Member

@tthvo tthvo left a comment

Choose a reason for hiding this comment

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

It takes quite a lot of efforts but I got the install correctly following this new doc. There are a few more questions/comments above, but overall, it looks good 👍

@tthvo
Copy link
Member

tthvo commented Jan 31, 2026

@jianli-wei It's unrelated to this PR, but I ran ci/prow/e2e-gcp-upi-xpn to cross-reference the commands. It seems like there is a mismatch in region between the subnet's region and the gcloud cli --region. Is it an issue we should look into?

** Change the documentation to reference pure bash. The deployment manager is EOL in March 2026.

Design Choices:
1. Terraform: The deployment manager can be converted to terraform. The installer team already removed terraform, so there is
no reason to go back to using terraform.
2. KRM: The deployment manager can be converted to use kubectl. This did not work, because very few resources do not have an
equivalent in kubectl (but we use those resources).
3. CI uses pure bash now, so this will remain consistent.

Note: The python templates are not removed in this commit. They are still found in upi/gcp/
@barbacbd barbacbd force-pushed the remove-gcp-deployment-manager branch from ee30935 to ec42623 Compare February 2, 2026 15:12
Copy link
Member

Choose a reason for hiding this comment

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

Suggested change
gcloud compute instances create ${INFRA_ID}-${machine_role}-${index} --boot-disk-size=${WORKER_DISK_SIZE} --boot-disk-type=${WORKER_DISK_TYPE} --image=${CLUSTER_IMAGE} --metadata=^#^user-data="${WORKER_IGNITION}" --machine-type=${WORKER_MACHINE_TYPE} --zone=${zone} --no-address --service-account=${WORKER_SERVICE_ACCOUNT} --scopes=https://www.googleapis.com/auth/cloud-platform --tags=${INFRA_ID}-${machine_role} --subnet=${COMPUTE_SUBNET}

I think WORKER_DISK_SIZE already has the suffix GB, right? ${WORKER_DISK_SIZE}GB will produce 128GBGB...

@openshift-ci
Copy link
Contributor

openshift-ci bot commented Feb 2, 2026

@barbacbd: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name Commit Details Required Rerun command
ci/prow/e2e-gcp-upi-xpn ee30935 link false /test e2e-gcp-upi-xpn
ci/prow/e2e-gcp-custom-dns ec42623 link false /test e2e-gcp-custom-dns

Full PR test history. Your PR dashboard.

Details

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-sigs/prow repository. I understand the commands that are listed here.

@patrickdillon
Copy link
Contributor

I didn't realize UPI was being converted to Bash. We can discuss with PM, but I have some concerns about this direction: using Bash seems like it might be difficult to maintain and to convert into our docs structure.

Deployment Manager is being replaced by Infrastructure Manager, which uses HCL/Terraform Templates. I thought we were aligned on using that, so let's regroup and discuss this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

jira/valid-reference Indicates that this PR references a valid Jira ticket of any type.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants