[Varnish] #977: Client and hash-directors doesn't retry on different backends

[Varnish]

#977: Client and hash-directors doesn't retry on different backends
----------------------+-----------------------------------------------------
 Reporter:  kristian  |       Owner:       
     Type:  defect    |      Status:  new  
 Priority:  normal    |   Milestone:       
Component:  varnishd  |     Version:  trunk
 Severity:  normal    |    Keywords:       
----------------------+-----------------------------------------------------
 (Ticket to get a bug number)

 Using the client- or hash-directors, the retry mechanism doesn't consider
 any other backends than the canonical backend when retrying after a failed
 backend connection. This is caused by faulty retry-logic.

 If the random-director is used, the backend to try is randomized for each
 retry, but such is not the case for client/hash.

 Test and fix coming up.

-- 
Ticket URL: <http://www.varnish-cache.org/trac/ticket/977>
Varnish <http://varnish-cache.org/>
The Varnish HTTP Accelerator