Description of problem:
This bug is based on BZ 1434471. If Satellite server is stopped during the cdn-sync, the sync process continues to 100% progress, and ends with flurry of cursor errors and finally a traceback. 

Version-Release number of selected component (if applicable):
spacewalk-backend-cdn-2.5.3-133

How reproducible:


Steps to Reproduce:
1. Sync some big channel from the CDN
2. Wait for the "Importing packages" phase to run
3. Stop the Satellite server

Actual results:
Importing packages goes all the way to 100% mark (takes a few minutes, not too long), and then there is following ouptut:

Importing packages:     |##################################################| 100.0%
13:48:21 Linking packages to channel.
13:48:22 ERROR: cursor already closed
13:48:22 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.9/x86_64/kickstart
13:48:25 ERROR: cursor already closed
13:48:25 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.8/x86_64/kickstart
13:48:26 ERROR: cursor already closed
13:48:26 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.6/x86_64/kickstart
13:48:27 ERROR: cursor already closed
13:48:27 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.3/x86_64/kickstart
13:48:28 ERROR: cursor already closed
13:48:28 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.1/x86_64/kickstart
13:48:29 ERROR: cursor already closed
13:48:29 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.4/x86_64/kickstart
13:48:30 ERROR: cursor already closed
13:48:30 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.7/x86_64/kickstart
13:48:31 ERROR: cursor already closed
13:48:31 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.2/x86_64/kickstart
13:48:32 ERROR: cursor already closed
13:48:32 Repo URL: https://cdn.redhat.com/content/dist/rhel/server/6/6.5/x86_64/kickstart
13:48:33 ERROR: cursor already closed
Traceback (most recent call last):
  File "/usr/bin/cdn-sync", line 219, in <module>
    sys.exc_info()[2])
  File "/usr/bin/cdn-sync", line 164, in <module>
    error_messages = cdnsync.sync(channels=args.channel)
  File "/usr/lib/python2.6/site-packages/spacewalk/cdn_tools/cdnsync.py", line 466, in sync
    cur_time, failed_packages = self._sync_channel(channel)
  File "/usr/lib/python2.6/site-packages/spacewalk/cdn_tools/cdnsync.py", line 407, in _sync_channel
    return sync.sync()
  File "/usr/lib/python2.6/site-packages/spacewalk/satellite_tools/reposync.py", line 484, in sync
    """, channel_id=int(self.channel['id'])) or []
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/__init__.py", line 291, in fetchall_dict
    h = prepare(sql)
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/__init__.py", line 270, in prepare
    return db.prepare(sql, blob_map=blob_map)
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/driver_postgresql.py", line 224, in prepare
    return Cursor(dbh=self.dbh, sql=sql, force=force, blob_map=blob_map)
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/driver_postgresql.py", line 270, in __init__
    sql_base.Cursor.__init__(self, dbh, sql, force)
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/sql_base.py", line 113, in __init__
    self._real_cursor = self._prepare(force=force)
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/sql_base.py", line 124, in _prepare
    cursor = self._prepare_sql()
  File "/usr/lib/python2.6/site-packages/spacewalk/server/rhnSQL/driver_postgresql.py", line 281, in _prepare_sql
    cursor = self.dbh.cursor()
Exception: SYNC ERROR: attempting to display as much information as possible
 connection already closed

After restarting the Satellite server, the synced channel is present, but is't empty.

Expected results:
Graceful exit from the sync process, no traceback.

Additional info: