OCPEDGE-1485: Add initial version of etcd kill and recovery test

[fonta-rh]

Add etcd process crash recovery test for TNF clusters

Add test to validate Two Node with Fencing (TNF) cluster recovery after etcd process crashes.

The test kills the etcd process/container on one node and validates that Pacemaker automatically detects the failure and restarts etcd, with the cluster returning to a healthy state (both nodes as started voting members).

Test approach:

• Kills etcd using multiple methods (podman kill, pkill, systemctl stop)
• Retries validation every 2 minutes for up to 6 minutes total
• Uses GinkgoRecover() to properly handle validation panics within Eventually retry logic
• Validates both etcd member state and operator health

Related: OCPBUGS-59238 (etcd process crash recovery issue in TNF clusters)

NOTE: This test doesn't work on an un-patched, un-reconfigured TNF cluster at the moment. It's set to be skipped until that point

[openshift-ci]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[openshift-ci-robot]

@fonta-rh: This pull request references OCPEDGE-1485 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 "4.21.0" version, but no target version was set.

In response to this:

Add test for TNF etcd recovery from fatal failure.

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.

[openshift-ci-robot]

@fonta-rh: This pull request references OCPEDGE-1485 which is a valid jira issue.

In response to this:

Add test for TNF etcd recovery from fatal etcd pod failure.

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.

[openshift-trt]

Job Failure Risk Analysis for sha: 51df2c4

Job Name	Failure Risk
pull-ci-openshift-origin-main-e2e-aws-csi	IncompleteTests Tests for this run (18) are below the historical average (1830): IncompleteTests (not enough tests ran to make a reasonable risk analysis; this could be due to infra, installation, or upgrade problems)
pull-ci-openshift-origin-main-e2e-gcp-csi	IncompleteTests Tests for this run (19) are below the historical average (1831): IncompleteTests (not enough tests ran to make a reasonable risk analysis; this could be due to infra, installation, or upgrade problems)

[openshift-ci-robot]

@fonta-rh: This pull request references OCPEDGE-1485 which is a valid jira issue.

In response to this:

Add etcd process crash recovery test for TNF clusters

Add test to validate Two Node with Fencing (TNF) cluster recovery after etcd process crashes.

The test kills the etcd process/container on one node and validates that Pacemaker automatically detects the failure and restarts etcd, with the cluster returning to a healthy state (both nodes as started voting members).

Test approach:

• Kills etcd using multiple methods (podman kill, pkill, systemctl stop)
• Retries validation every 2 minutes for up to 6 minutes total
• Uses GinkgoRecover() to properly handle validation panics within Eventually retry logic
• Validates both etcd member state and operator health

Related: OCPBUGS-59238 (etcd process crash recovery issue in TNF clusters)

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.

[openshift-ci-robot]

@fonta-rh: This pull request references OCPEDGE-1485 which is a valid jira issue.

In response to this:

Add etcd process crash recovery test for TNF clusters

Add test to validate Two Node with Fencing (TNF) cluster recovery after etcd process crashes.

The test kills the etcd process/container on one node and validates that Pacemaker automatically detects the failure and restarts etcd, with the cluster returning to a healthy state (both nodes as started voting members).

Test approach:

• Kills etcd using multiple methods (podman kill, pkill, systemctl stop)
• Retries validation every 2 minutes for up to 6 minutes total
• Uses GinkgoRecover() to properly handle validation panics within Eventually retry logic
• Validates both etcd member state and operator health

Related: OCPBUGS-59238 (etcd process crash recovery issue in TNF clusters)

NOTE: This test doesn't work on an un-patched, un-reconfigured TNF cluster at the moment. It's set to be skipped until that point

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.

[clobrano]

The test is logically correct for me. I just left a couple of notes.

[jaypoulz]

Just a few notes from me. Otherwise looking good!

[clobrano]

I left a couple of comments

[clobrano]

/lgtm

[clobrano]

Giving time to @jaypoulz to complete his review

/hold

[openshift-ci]

@fonta-rh: 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
ci/prow/okd-scos-e2e-aws-ovn	51df2c4	link	false	/test okd-scos-e2e-aws-ovn

Full PR test history. Your PR dashboard.

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.

[jaypoulz]

/approve
/unhold

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: clobrano, fonta-rh, jaypoulz

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

The pull request process is described here

Needs approval from an approver in each of these files:

• test/extended/two_node/OWNERS [jaypoulz]

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

[fonta-rh]

/retest