Retry failed rebind attempts

When we receive a migration, if rebinding to the new listen address and
port fails for a reason which might be fixable, rather than killing the
server we retry once a second.  Also in this patch: non-overlapping log
messages and a fix for the client going away halfway through a sendfile
loop.

tests/acceptance/flexnbd/fake_source.rb

@@ -54,12 +54,28 @@ module FlexNBD
       send_request( 2, handle )
     end
 
+    def write_read_request( from, len, handle="myhandle" )
+      send_request( 0, "myhandle", from, len )
+    end
+
 
     def write_data( data )
       @sock.write( data )
     end
 
 
+    # Handy utility
+    def read( from, len )
+      timing_out( 2, "Timed out reading" ) do
+        send_request( 0, "myhandle", from, len )
+        read_raw( len )
+      end
+    end
+
+    def read_raw( len )
+      @sock.read( len )
+    end
+
     def send_mirror
       read_hello()
       write_write_request( 0, 8 )