Lower the loss of API access during OpenStack upgrade

[David Ames]

Investigate and improve API access during OpenStack upgrade

[David Ames (thedac)]

Currently testing using http checks for backend services.

https://review.openstack.org/#/c/451603/
https://code.launchpad.net/~thedac/charm-helpers/haproxy-options/+merge/321390

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/457757

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/457757
Committed: https://git.openstack.org/cgit/openstack/charm-nova-cloud-controller/commit/?id=e94f7882cb7603723f993600b044cf9dcf544760
Submitter: Jenkins
Branch: master

commit e94f7882cb7603723f993600b044cf9dcf544760
Author: David Ames <email address hidden>
Date: Tue Apr 18 10:56:09 2017 -0700

    Avoid shared-db change when using access-network

    When the percona-cluster charm sets an access-network but the default
    unit-get address is not on that network extra shared-db relations get
    executed. This is specifically a problem when running upgrades and
    trying to avoid API downtime.

    The root cause is that the access-network is not checked until the
    SharedDBContext is consulted. But then db_joined function will
    change it back to the wrong ip on subsequent runs.

    This change adds a check for access-network on the relation during the
    db_joined function and pushes IP selection off to get_relation_ip.

    Charm helpers sync to pull in changes to get_relation_ip.

    Partial-bug: #1677647
    Change-Id: Ifd4e975d9abbb9f7a8b0f12c8f0a8cf8f78595b6

[OpenStack Infra (hudson-openstack)]

Change abandoned by David Ames (<email address hidden>) on branch: master
Review: https://review.openstack.org/451603
Reason: Test only

[Alex Kavanagh (ajkavanagh)]

Closing as it has been overtaken by events.