⚠️ Improve priority and validation of input variables consumed in wait_for_interface_or_ip

[mchiappero]

What this PR does / why we need it:
The current logic in wait_for_interface_or_ip deals with the input IP addresses or interfaces, according to the following order: PROVISIONING_IP, IRONIC_IP, PROVISIONING_INTERFACE. The improvements this PR introduces are:

• make the test case for IRONIC_IP explicit rather than implicit, improving code readability before the following change
• "group" PROVISIONING_IP and PROVISIONING_INTERFACE in terms of priority by prioritizing IRONIC_IP instead, resulting in the following order: IRONIC_IP, PROVISIONING_IP, PROVISIONING_INTERFACE
• include input validation for the externally provided IRONIC_IP

What the PR does not address is logging a warning in case of multiple input variables set and one taking inevitably precedence over the another(s).

NOTE: based on PR #787

Checklist:

• Documentation has been updated, if necessary.
• Integration tests have been added, if necessary.

[metal3-io-bot]

[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 rozzii 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:

• OWNERS

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

[metal3-io-bot]

Hi @mchiappero. Thanks for your PR.

I'm waiting for a github.com member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

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.

[tuminoid]

Some minor nitting, and also please squash the commits.

As for the actual logic, I need another look with fresher eyes tomorrow.
/cc @Rozzii @dtantsur
PTAL.

[tuminoid]

/ok-to-test

[tuminoid]

Also, isn't this the same as #787 ?

[mchiappero]

Also, isn't this the same as #787 ?

The PRs are stacked, as far as I can see it's not possible to set a different branch, but I should have mentioned that in the description. Let me fix it... along with the proposed changes (thanks for reviewing!).

[mchiappero]

Some minor nitting, and also please squash the commits.

As for the actual logic, I need another look with fresher eyes tomorrow. /cc @Rozzii @dtantsur PTAL.

Sure. But before squashing I'd like to make sure the change is fully understood, though. By the way, IRONIC_IP is not documented in https://github.com/metal3-io/ironic-image/blob/main/README.md, while it probably should. Also, which variables take precedence is also not documented either, nor logged by the scripts. I guess maybe something to discuss at the weekly meeting. Let me know what you think!

[elfosardo]

/hold
let's hold on to this until we deal with #787

[metal3-io-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues will close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[Rozzii]

/remove-lifecycle stale
Not stale, it is the second in a chain of PRs, it was waiting for the first PR to get merged.

[mchiappero]

/retest

[tuminoid]

/test metal3-centos-e2e-integration-test-main metal3-ubuntu-e2e-integration-test-main

[tuminoid]

[2026-02-25T06:37:47.688Z]    - Waiting for task completion (up to 2400 seconds)  - Command: 'check_container httpd-infra'

[2026-02-25T07:17:57.752Z] FAIL - Container httpd-infra running

/retest

[Rozzii]

/test metal3-centos-e2e-integration-test-main

[mchiappero]

Sorry, I didn't have much time to look into it, and I'm not sure I understand why it's failing; but it looks like it's some Ansible task:

11:09:08  + for name in ipa-downloader vbmc sushy-tools httpd-infra ipxe-builder registry
11:09:08  + sudo docker ps
11:09:08  + grep -w -q 'ipa-downloader$'
11:09:08  sudo: docker: command not found
11:09:08  + sudo docker ps --all
11:09:08  + grep -w -q 'ipa-downloader$'
11:09:08  sudo: docker: command not found
11:09:08  + for name in ipa-downloader vbmc sushy-tools httpd-infra ipxe-builder registry
11:09:08  + sudo docker ps
11:09:08  + grep -w -q 'vbmc$'
11:09:08  sudo: docker: command not found
[...]

What do you think?

[elfosardo]

/unhold

[elfosardo]

/retest

let's give it another try

[metal3-io-bot]

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

Test name	Commit	Details	Required	Rerun command
metal3-centos-e2e-integration-test-main	32b7266	link	true	/test metal3-centos-e2e-integration-test-main

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.

[metal3-io-bot]

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

Test name	Commit	Details	Required	Rerun command
metal3-centos-e2e-integration-test-main	32b7266	link	true	/test metal3-centos-e2e-integration-test-main

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.

[tuminoid]

   - Waiting for task completion (up to 2400 seconds)  - Command: 'check_container httpd-infra'

That could have direct relation to change being made.

[tuminoid]

   - Waiting for task completion (up to 2400 seconds)  - Command: 'check_container httpd-infra'

That could have direct relation to change being made.

My bot says:

  Root Cause Analysis

  The failure is DIRECTLY CAUSED by PR #788. Two changes combine to break the httpd-infra container:

   1. get_provisioning_interface(): Default changed from interface="provisioning" to interface="". When no PROVISIONING_MACS are set, the function now returns empty 
  string instead of "provisioning".
   2. wait_for_interface_or_ip(): New else error branch added: else
        echo "ERROR: cannot determine an interface or an IP for binding and creating URLs"
        return 1
    fi

  The httpd-infra podman container receives NO environment variables (IRONIC_IP, PROVISIONING_IP, PROVISIONING_INTERFACE, PROVISIONING_MACS are all empty). The
  execution path is:

   - IRONIC_IP="" → skip
   - PROVISIONING_IP="" → skip  
   - PROVISIONING_INTERFACE="" (was "provisioning" before PR) → skip
   - Hits new else → return 1 → set -e kills container

  Before the PR: PROVISIONING_INTERFACE would default to "provisioning" via get_provisioning_interface(), and wait_for_interface_or_ip() would loop waiting for that
  interface to get an IP. In CI, the provisioning bridge exists, so it would succeed.

  Kubernetes Ironic pods are unaffected because they receive PROVISIONING_IP=172.22.0.2 via pod spec environment variables.

Ie. it fails on ubuntu which has local (ie. non-k8s) Ironic, while passing on centos (k8s ironic).

[metal3-io-bot]

PR needs rebase.

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.

[tuminoid]

This was superseded by above linked PR.

/close

[metal3-io-bot]

@tuminoid: Closed this PR.

Details

In response to this:

This was superseded by above linked PR.

/close

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.

[mchiappero]

Let's reopen, PR #935 addresses a different problem.

If I understand it correctly the only problem is the final else, I wasn't aware of the possibility of running ironic without any of those variables defined. What should the logic be in such case?

[tuminoid]

Let's reopen, PR #935 addresses a different problem.

Yes true, we still have prioritization etc left here. My bad.
/reopen

If I understand it correctly the only problem is the final else, I wasn't aware of the possibility of running ironic without any of those variables defined. What should the logic be in such case?

We either or maybe even both:

• keep the "provisioning" default here (backwards compatibility)
• add PROVISIONING_INTERFACE=provisioning default in dev-env (sane testing defaults)

WDYT @elfosardo ?

[metal3-io-bot]

@tuminoid: Reopened this PR.

Details

In response to this:

Let's reopen, PR #935 addresses a different problem.

Yes true, we still have prioritization etc left here. My bad.
/reopen

If I understand it correctly the only problem is the final else, I wasn't aware of the possibility of running ironic without any of those variables defined. What should the logic be in such case?

We either or maybe even both:

• keep the "provisioning" default here (backwards compatibility)
• add PROVISIONING_INTERFACE=provisioning default in dev-env (sane testing defaults)

WDYT @elfosardo ?

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.

[elfosardo]

@tuminoid that should work

[Rozzii]

Hi @mchiappero, please rebase the PR.

[mchiappero]

Sorry for the late reply, I had to take some time off this week.

Before applying any change, I would like to double check:

1. my assumptions, as I guess I didn't have a full understanding of the logic when I wrote this. So, is the "provisioning" interface name specific to non metal3 environments? Because otherwise if none of the input variables are provided it fails in a confusing way. Is the case where no variables are provided valid?

2. If it's still a good idea to make IRONIC_IP higher priority. So whether to:

   • to apply this change and use the following evaluation order IRONIC_IP > PROVISIONING_IP > PROVISIONING_INTERFACE; but then is there a case none of them would be set?
   • simply keep it as is as there are other reasons I can't see

I think it would be good to document this logic, from a developer perspective, but also from the user point of view. Maybe I can add some edits to README.md in this PR if we have a decision on the outcome.

[mchiappero]

Is the case where no variables are provided valid?

This cannot be valid indeed, otherwise no IRONIC_IP & URLs are generated. I'm not sure I understand why it is not passing the tests.

[tuminoid]

Is the case where no variables are provided valid?

This cannot be valid indeed, otherwise no IRONIC_IP & URLs are generated. I'm not sure I understand why it is not passing the tests.

It is not valid, but it is just not defined anywhere with local Ironic case. To fix that, the suggestion made earlier stands.

We either or maybe even both:

• keep the "provisioning" default here (backwards compatibility)
• add PROVISIONING_INTERFACE=provisioning default in dev-env (sane testing defaults)

As for the question 2 about IRONIC_IP priority, I need to defer to @elfosardo or @Rozzii .