Comment 3 for bug 2030901

Reviewed: https://review.opendev.org/c/starlingx/vault-armada-app/+/923068
Committed: https://opendev.org/starlingx/vault-armada-app/commit/f994829fcbea79b5b79193ba3f49e40e6c498709
Submitter: "Zuul (22348)"
Branch: master

commit f994829fcbea79b5b79193ba3f49e40e6c498709
Author: Michel Thebeau <email address hidden>
Date: Thu Jan 4 20:04:01 2024 +0000

    override pre-existing anti-affinity

    commit f7a37e6a, "Removing default injector anti-affinity rules",
    disables anti-affinity for the injector pod. This is sufficient for
    future application updates. However, during application-update an old
    pod that still has anti-affinity will still prevent scheduling of a new
    pod. This is observed on AIO-SX when testing application-update in
    preparation for USM.

    Using injector.strategy.rollingUpdate.maxUnavailable
    (DeploymentStrategy) set to 100% the old pod will immediately terminate
    while the new pod waits for its termination.

    This is the workaround described in the original in starlingx bug:
    https://bugs.launchpad.net/starlingx/+bug/2030901.

    Test Plan:
    PASS AIO-SX vault sanity
    PASS application-update

    Partial-Bug: 2030901
    Story: 2011073
    Task: 50484

    Change-Id: I66fe336ece7f1ccd68caa665aabc693f1b9a5c18
    Signed-off-by: Michel Thebeau <email address hidden>