Start building etcd v3.6.0-alpha.0 image for scalability tests

[serathius]

/kind feature

NONE

/sig scalability

[serathius]

/sig release

[serathius]

/assign @dims

[pacoxu]

During reading the changelog of etcd v3.6, there is a breaking change and has it any influences to Kubernetes upgradetion?

• Breaking Changes etcd will no longer start on data dir created by newer versions (for example etcd v3.6 will not run on v3.7+ data dir). To downgrade data dir please check out etcdutl migrate command.

[serathius]

During reading the changelog of etcd v3.6, there is a breaking change and has it any influences to Kubernetes upgradetion?

• Breaking Changes etcd will no longer start on data dir created by newer versions (for example etcd v3.6 will not run on v3.7+ data dir). To downgrade data dir please check out etcdutl migrate command.

This change is in data compatibility policy for downgrades. Etcd never really supported downgrades as it recommenced to restore backup from before upgrade, however it prevented users from doing so. Fact that data was backward compatible was never guaranteed nor validated.

With introduction of downgrade support to etcd we are trying to provide a downgrade guarantees, however this first comes with restrictions about removing unsafe operations. With v3.6 etcd will no longer allow administrators to just downgrade the binary and run on the same data directory. On the other side etcd will provide an tooling and documentation about how a proper etcd downgrade process should be executed.

As this change is in downgrade behavior to v3.6, there is no immediate impact of this PR to K8s as it only changes ability of downgrading from v3.7 to v3.6. As for future actions that K8s should do for v3.7, it will require updating K8s https://github.com/kubernetes/kubernetes/blob/master/cluster/images/etcd/migrate-if-needed.sh script to utilize official etcd tooling like etcdutl migrate.

[pacoxu]

Thanks for the clarification.

[dims]

/approve
/lgtm

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: dims, serathius

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:

• build/OWNERS [dims]
• cluster/images/etcd/OWNERS [dims,serathius]

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

[k8s-triage-robot]

The Kubernetes project has merge-blocking tests that are currently too flaky to consistently pass.

This bot retests PRs for certain kubernetes repos according to the following rules:

• The PR does have any do-not-merge/* labels
• The PR does not have the needs-ok-to-test label
• The PR is mergeable (does not have a needs-rebase label)
• The PR is approved (has cncf-cla: yes, lgtm, approved labels)
• The PR is failing tests required for merge

You can:

• Review the full test history for this PR
• Prevent this bot from retesting with /lgtm cancel or /hold
• Help make our tests less flaky by following our Flaky Tests Guide

/retest

[fedebongio]

/triage accepted