unable to login in fuel dashboard after reboot

[Daniele Pizzolli]

1. get fuel-community-5.1-9-2014-08-26_21-42-16.iso
2. install on a virtual machine
3. create a environment
4. reboot the virtual machine
5. wait for the login panel
6. try to login
7. you will see login failed even if you are using the right username password

Inspecting the http traffic reveals that the api response is 502 bad gateway.

Working workaround suggested in #fuel-dev by Vladimir Kuklin

dockerctl restart keystone
dockerctl restart nailgun

[Mike Scherbakov (mihgen)]

Daniele, thanks for the report,
did you try to just wait? I actually saw such issue, but I waited for about 5 min after master node booted back, and then Fuel CI started to work again.

[Matthew Mosesohn (raytrac3r)]

This looks related to a keystone db_sync bug I fixed today. The following error in the logs is exactly the same as this bug: https://bugs.launchpad.net/bugs/1362139

2014-08-27 11:00:33.431 547 TRACE keystone.common.wsgi ProgrammingError: (ProgrammingError) column service_1.enabled does not exist
2014-08-27 11:00:33.431 547 TRACE keystone.common.wsgi LINE 1: ...S service_1_id, service_1.type AS service_1_type, service_1....
2014-08-27 11:00:33.431 547 TRACE keystone.common.wsgi ^
2014-08-27 11:00:33.431 547 TRACE keystone.common.wsgi 'SELECT endpoint.id AS endpoint_id, endpoint.legacy_endpoint_id AS endpoint_legacy_endpoint_id, endpoint.interface AS endpoint_interface, endpoint.region AS endpoint_region, endpoint.service_id AS endpoint_service_id, endpoint.url AS endpoint_url, endpoint.enabled AS endpoint_enabled, endpoint.extra AS endpoint_extra, service_1.id AS service_1_id, service_1.type AS service_1_type, service_1.enabled AS service_1_enabled, service_1.extra AS service_1_extra \nFROM endpoint LEFT OUTER JOIN service AS service_1 ON service_1.id = endpoint.service_id \nWHERE endpoint.enabled = %(enabled_1)s' {'enabled_1': True}

I'm marking this as a duplicate. A new community ISO should have this fix.

[Matthew Mosesohn (raytrac3r)]

Confirming this is fixed in the latest ISO. I can reboot and still log out and log back in with no issues. Keystone container works.

[Matthew Mosesohn (raytrac3r)]

The related bug fixed the problem of failing out keystone container if puppet apply fails, but supervisor config for keystone container is missing. I'm adding this in now.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/119744

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/119744
Committed: https://git.openstack.org/cgit/stackforge/fuel-library/commit/?id=32e70577990d25ba474f2bd57bb425ed3a2b1079
Submitter: Jenkins
Branch: master

commit 32e70577990d25ba474f2bd57bb425ed3a2b1079
Author: Matthew Mosesohn <email address hidden>
Date: Mon Sep 8 16:28:06 2014 +0400

    Add missing keystone supervisor config

    This file is necessary for supervisor to
    manage keystone container and restart it.

    Change-Id: I541d65e2dc8c4dee25c317c75dc332fdb1c8733f
    Closes-Bug: #1362207