Daemons don't restart cycles after zenhub connection failure

[cluther]

When the collector daemons timeout connecting to zenhub, then later reconnect they will not automatically start their collection cycles. This leaves them running, but doing nothing. Fortunately they aren't even sending heartbeats so the failure is known.

This is most prevalent when the daemons fail to connect when they are first started and before they start their first collection cycle.

[cluther]

I created a phony hub service called Busy that calls time.sleep(600) and used it in zenping. After the 10 minutes are expired and the daemon tries to reconnect to zenhub the following exception is thrown. After which the daemon stays running, but does nothing.

ERROR:zen.zenperfsnmp:Timeout connecting to zenhub: is it running?

WARNING:zen.zenperfsnmp:Reconnected to ZenHub
Unhandled error in Deferred:
Traceback (most recent call last):
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 317, in _runCallbacks
    self.result = callback(self.result, *args, **kw)
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 507, in _cbDeferred
    self.callback(self.resultList)
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 239, in callback
    self._startRunCallbacks(result)
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 304, in _startRunCallbacks
    self._runCallbacks()
--- <exception caught here> ---
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 317, in _runCallbacks
    self.result = callback(self.result, *args, **kw)
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 239, in callback
    self._startRunCallbacks(result)
  File "/opt/zenoss/lib/python/twisted/internet/defer.py", line 290, in _startRunCallbacks
    raise AlreadyCalledError
twisted.internet.defer.AlreadyCalledError:

[ecn]

(In [9915]) * fixes #3588: reviewed by chet

[ian]

(In [9991]) Refs #3588: Backporting r9915 to the zenoss-2.2.x branch

[cholden]

Verified on 2.2.4 (build 59) native rpm. zenping and zenmodeler reconnected after initial timeouts.