Drop unused PostgresSQL sequences from Folsom

[Dan Prince]

The Nova Folsom database migrations left a couple unused sequences in the PostgreSQL schema. Specifically the snapshots_id_seq and volumes_id_seq are unused and when using PostgreSQL should probably be dropped from the schema.

The reason we don't need these sequences anymore is that the 'id' columns in the snapshots and volumes tables are now handled as UUID strings instead of auto increasing integers.

The fix seems to be to run the following for PostgreSQL:

DROP SEQUENCE snapshots_id_seq CASCADE;
DROP SEQUENCE volumes_id_seq CASCADE;

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/16443

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/16443
Committed: http://github.com/openstack/nova/commit/8e9e5c7965d50113dda4e69c9ec0248cd3db0037
Submitter: Jenkins
Branch: master

commit 8e9e5c7965d50113dda4e69c9ec0248cd3db0037
Author: Dan Prince <email address hidden>
Date: Mon Nov 19 11:53:23 2012 -0500

    Drop unused PostgreSQL sequences from Folsom.

    In Folsom the snapshots and volumes tables were converted to use
    UUID's. When we performed this conversion in Folsom a couple of
    unused PostgreSQL sequences were left behind.

    This migration removes the snapshots_id_seq and volumes_id_seq
    if they exist in the schema when using PostgreSQL.

    Fixes LP Bug #1080786.

    Change-Id: I075e2afebcb5236f96ab5d6ab13e249d078da86b