Comment 9 for bug 1838805

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to ansible-playbooks (r/stx.2.0)

Reviewed: https://review.opendev.org/677506
Committed: https://git.openstack.org/cgit/starlingx/ansible-playbooks/commit/?id=05f677d0993604a0b1a1e40b2d64f91c24c43c5c
Submitter: Zuul
Branch: r/stx.2.0

commit 05f677d0993604a0b1a1e40b2d64f91c24c43c5c
Author: Tee Ngo <email address hidden>
Date: Sat Aug 3 07:52:53 2019 -0400

    Prevent admin password regeneration upon bootstrap replay

    Currently admin password is regenerated in bootstrap replay if
    the initial bootstrap failed for whatever reason. This results in
    Authorization failure after controller-0 unlock as the admin
    password, which got regenerated in the replay and stored via
    python keyring, no longer matches with the admin password postgres
    was bootstrapped with.

    In this commit, admin password is not regenerated in a replay after
    the bootstrap puppet manifest has been applied regardless of the
    initial bootstrap status.

    This logic will be revised once the decision has been made to
    continue storing the admin password using python keyring or to switch
    to barbican as the required changes to support password generation/
    regeneration in the bootstrap playbook differ considerably between
    the two storage methods. In the meantime, the user can always change
    the admin password at any time after the system is up (e.g. via
    Horizon).

    Closes-Bug: 1837696
    Closes-Bug: 1838805
    Change-Id: I672a7e97a9183f33732b3c979188c5db9b560296
    Signed-off by: Tee Ngo <email address hidden>