Better error reporting in solar-worker when database is wrongly configured

[Jedrzej Nowak]

When worker has different solar_db than cli then after `solar o run-once last` in solar worker we can see following output:
-----
vagrant@solar-dev:/vagrant$ solar-worker
INFO (gevent_runner.py::29)::Spawning scheduler, system log and tasks workers.
INFO:solar:Spawning scheduler, system log and tasks workers.
DEBUG (__init__.py::26)::DB session for <solar.orchestration.workers.system_log.SystemLog object at 0x7f93451a1910>
DEBUG:solar:DB session for <solar.orchestration.workers.system_log.SystemLog object at 0x7f93451a1910>
DEBUG (scheduler.py::162)::Scheduler subscribes to tasks hooks
DEBUG:solar:Scheduler subscribes to tasks hooks
DEBUG (system_log.py::31)::System log subscribes to tasks hooks
DEBUG:solar:System log subscribes to tasks hooks
DEBUG (__init__.py::26)::DB session for <solar.orchestration.workers.scheduler.Scheduler object at 0x7f933d6975d0>
DEBUG:solar:DB session for <solar.orchestration.workers.scheduler.Scheduler object at 0x7f933d6975d0>
DEBUG (locking.py::217)::Create lock UID system_log:07b5db08-a4f0-4355-9192-d1e030f5514b for 140270387568464
DEBUG:solar:Create lock UID system_log:07b5db08-a4f0-4355-9192-d1e030f5514b for 140270387568464
DEBUG (locking.py::168)::Lock for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b acquired by 140270387568464
DEBUG:solar:Lock for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b acquired by 140270387568464
DEBUG (scheduler.py::52)::Received *next* event for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b
DEBUG:solar:Received *next* event for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b
DEBUG (scheduler.py::58)::Scheduled tasks []
DEBUG:solar:Scheduled tasks []
DEBUG (locking.py::180)::Release lock system_log:07b5db08-a4f0-4355-9192-d1e030f5514b with 140270387568464
DEBUG:solar:Release lock system_log:07b5db08-a4f0-4355-9192-d1e030f5514b with 140270387568464
DEBUG (locking.py::217)::Create lock UID system_log:07b5db08-a4f0-4355-9192-d1e030f5514b for 140270387568464
DEBUG:solar:Create lock UID system_log:07b5db08-a4f0-4355-9192-d1e030f5514b for 140270387568464
DEBUG (locking.py::168)::Lock for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b acquired by 140270387568464
DEBUG:solar:Lock for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b acquired by 140270387568464
DEBUG (scheduler.py::52)::Received *next* event for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b
DEBUG:solar:Received *next* event for system_log:07b5db08-a4f0-4355-9192-d1e030f5514b
DEBUG (scheduler.py::58)::Scheduled tasks []
DEBUG:solar:Scheduled tasks []
DEBUG (locking.py::180)::Release lock system_log:07b5db08-a4f0-4355-9192-d1e030f5514b with 140270387568464
DEBUG:solar:Release lock system_log:07b5db08-a4f0-4355-9192-d1e030f5514b with 140270387568464
-----

It says nothing about objects not found, or job not found. Nothing. Just does some work and then stops. We definitely need better error reporting there.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/283632

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/283632
Committed: https://git.openstack.org/cgit/openstack/solar/commit/?id=72876bf2150e9cfe62f94238416107d6617d5d6f
Submitter: Jenkins
Branch: master

commit 72876bf2150e9cfe62f94238416107d6617d5d6f
Author: Maciej Kwiek <email address hidden>
Date: Tue Feb 23 16:27:21 2016 +0100

    Raise an error for empty plan

    If the execution graph that solar-worker retrieves from db is empty,
    ValueError is raised (it's handled by zerorpc executor).

    Additionally, db string used for db connection is logged on start of
    solar-worker.

    Change-Id: I2acafec0e94c0d2792b34d0485415782eae1b9dd
    Closes-bug:1545813