catch another cinder call, when cinder unavailable

[Matthias Runge]

During review of backport of fix for bug https://code.launchpad.net/bugs/1084137 it has been discovered, that there is another call to cinderclient, which will fail, when cinder is unavailable.
volume_list from horizon/openstack_dashboard/api/cinder.py should become:
def volume_list(request, search_opts=None):
    """
    To see all volumes in the cloud as an admin you can pass in a special
    search option: {'all_tenants': 1}
    """
    c_client = cinderclient(request)
    if c_client is None:
        return []
    return c_client.volumes.list(search_opts=search_opts)

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/19378

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/19378
Committed: http://github.com/openstack/horizon/commit/d0aff3a1a79ae51b33653a944c4f0fc127dd517f
Submitter: Jenkins
Branch: master

commit d0aff3a1a79ae51b33653a944c4f0fc127dd517f
Author: Matthias Runge <email address hidden>
Date: Thu Jan 10 12:00:20 2013 +0100

    Catch another breakage, when cinder is unavailable

    During review of backporting avoid cinder calls, this
    has been discovered and should be applied to master branch.

    Fixes bug 1098146

    Change-Id: I74c7fabd31ed333fc9f44ec92b2c391dd7c9b9ef